Linguifex - User contributions [en]

Kämpya

2017-05-24T08:35:30Z

Linguist Wannabe: /* Verb Subordination */ Direct Reported Speech

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí dò̰k=gó θú=gḭ̂p/

guardian Q.ACC=SEC dog=DAT ANTIP=give

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/ˈlḭ̀zád=gó θú=ˈbíʔ dô̰k éi=ˈhḛ̂b/

lizard=DAT ANTIP=bite dog Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/ˈlḭ̀zád=gó θú=ˈbíʔ bá=dô̰k/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù dò̰k=gó ˈbíʔ=kà bá=pʰò̤wè /

lizard=ERG dog=DAT bite=LOC which=forest

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=péiʔk tí=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=péiʔk tí=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ tí=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=ˈpéiʔkðjíʔ tí=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

====Indirect Reported Speech====

In this case, when reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾá mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC.EMP=SEC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy had departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾá mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC.EMP=SEC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy had departed (but I didn't see it, I just heard about it).

====Direct Reported Speech====

In this case, there is no superordinate verb marking. The quoted speech is places at the end of the sentence, preceded by the particle /ðéʔ/ and followed by the particle /tḛ̂/ e.g.

/ˈkʰḭ̂d mí=ˈpéiʔk ðéʔ ˈḛ̂nèm déˈpʰáʔɾ tḛ̂/

child 1PS.EXCL.ACC=speak QUOTE enemy depart QUOTE

The child told me that the enemy had departed

In the above sentence, the speaker is saying the the child had used the direct evidential when reporting the enemy's departure, meaning that the speaker is reporting that the child implied that they had witnessed it themselves. This sentence conveys no information about whether or not the speaker had also seen the departure.

Compare:

/ˈkʰḭ̂d mí=ˈpéiʔk ðéʔ ˈḛ̂nèm sái=déˈpʰáʔɾ tḛ̂/

child 1PS.EXCL.ACC=speak QUOTE enemy REP=depart QUOTE

The child told me that he had heard that the enemy had departed (and not witnessed it himself).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-05-24T07:48:16Z

Linguist Wannabe: /* Antipassive Voice */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí dò̰k=gó θú=gḭ̂p/

guardian Q.ACC=SEC dog=DAT ANTIP=give

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/ˈlḭ̀zád=gó θú=ˈbíʔ dô̰k éi=ˈhḛ̂b/

lizard=DAT ANTIP=bite dog Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/ˈlḭ̀zád=gó θú=ˈbíʔ bá=dô̰k/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù dò̰k=gó ˈbíʔ=kà bá=pʰò̤wè /

lizard=ERG dog=DAT bite=LOC which=forest

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=ˈpéiʔkðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:33:42Z

Linguist Wannabe: /* Verb Subordination */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí dò̰k=gó θú=gḭ̂p/

guardian Q.ACC=SEC dog=DAT ANTIP=give

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/ˈlḭ̀zád=gó θú=ˈbíʔ dô̰k éi=ˈhḛ̂b/

lizard=DAT ANTIP=bite dog Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/ˈlḭ̀zád=gó θú=ˈbíʔ bá=dô̰k/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù dò̰k=gó ˈbíʔ=kà bá=pʰò̤wè /

lizard=ERG dog=DAT bite=LOC which=forest

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí ˈkʰḭ̀d=gó θú=péiʔk wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá ˈkʰḭ̀d=gó θú=ˈpéiʔkðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC child=DAT ANTIP=speak 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:31:15Z

Linguist Wannabe: /* Wh-Questions */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí dò̰k=gó θú=gḭ̂p/

guardian Q.ACC=SEC dog=DAT ANTIP=give

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/ˈlḭ̀zád=gó θú=ˈbíʔ dô̰k éi=ˈhḛ̂b/

lizard=DAT ANTIP=bite dog Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/ˈlḭ̀zád=gó θú=ˈbíʔ bá=dô̰k/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù dò̰k=gó ˈbíʔ=kà bá=pʰò̤wè /

lizard=ERG dog=DAT bite=LOC which=forest

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:27:26Z

Linguist Wannabe: /* Relative Clauses */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:26:47Z

Linguist Wannabe: /* Applicative Voice */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the postpositonal object "slot" in the sentence (and marking it with the dative clitic /=gó/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=góˈbíʔ=kà/

forest dog=ERG lizard=DAT bite=LOC

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì ˈè̤ném=gó déˈpáʔɾ=hwàm/

valley enemy=DAT depart=from

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù dò̰k=góˈgḭ̂p=tá/

bone guardian=ERG dog=DAT give=SEC

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè dò̰k=gó té=ˈbíʔ=kà/

forest dog=DAT 2PS.ACC=bite=LOC

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù lḭ̀zád=gó bíʔ ˈˈjě̤tài=kà/

forest dog=ERG lizard=DAT bite yesterday.ADV=LOC

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:22:03Z

Linguist Wannabe: /* Allomorphy */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá dò̰k=gó θú=gḭ̂p/

guardian snare=SEC dog=DAT ANTIP=give

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:21:06Z

Linguist Wannabe: /* Allophones of /h/ */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:20:33Z

Linguist Wannabe: /* Consonant Neutralisations */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the alienable genitive clitic /jì/ to form /nâḭm=jì/ - "of the name", which is pronounced [nâḭmjì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:19:34Z

Linguist Wannabe: /* Consonant Neutralisations */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the genitive clitic /ì/ to form /nâḭm=ì/ - "of the name", which is pronounced [nâḭmì].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:17:51Z

Linguist Wannabe: /* Ditransitive Sentences */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:16:42Z

Linguist Wannabe: /* Ditransitive Sentences */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá dò̰k=gó θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC dog=DAT ANTIP=give

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2017-04-09T11:11:32Z

Linguist Wannabe: /* Antipassive Voice */ Changed antipassive marker to -go

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced using the dative postposition, /-gó/ (which changes the noun into Tone Class 2). These nouns are re-introduced into the same syntactic "slot" as postpositional objects e.g.

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ /

dog lizard=DAT ANTIP=bite

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k ˈlḭ̀zád=gó θú=ˈbíʔ áˈwâḭ/

dog lizard=DAT ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Hnyengu

2016-10-16T09:25:33Z

Linguist Wannabe:

Hnyengu (pronounced [ˈɲ̥əŋgu]) is another language of my far-future post-apocalypse Antarctican conword, spoken along the Ross Sea coast to the north and west of where [[Kämpya]] is spoken. Genetically, it is a Japonic language, however it has been heavily influenced by other languages. In the generations after the initial colonisation, Hnyengu speakers were usually multilingual, and the language absorbed a huge amount of loanwords, especially from English and Spanish. Over the centuries, it continued to be subject to areal influence. A strong Sprachbund has emerged on the Ross Sea coast, and there are also a number of Antarctic areal features.

There are four major dialect groups: Eastern, Central, Southwestern and Northwestern. The Central dialect is spoken in the major city in the region, Myinatu. For a long time it had been the prestige dialect, but after the city was conquered by Kämpya speakers, the inhabitants of the city switched to Kämpya, and within the city there is now a strong stigma attached to the Central dialect of Hnyengu. However, this stigma does not extend to the other dialects, and there is still a large amount of bilingualism between them and Kämpya.

== Brief Description ==

Like modern Japanese, Hnyengu is a topic comment language. However, a succession of sound changes has led to the morphology becoming more fusional.

The morphosyntactic alignment is split ergative (the ergative suffix /-ŋ/ being an erosion of the Japanese genitive /no/). The exact rules of the split vary between dialects, but depend on animacy and relative topicality of the subject.

The TAM system depends on the dialect. The dialects use roughly the same set of verb affixes, however different dialects use them to mark different things tense/aspect/evidentiality.

The syntax also depends on the dialect. Most dialects have SVO word order, except for the Eastern dialect which is syntactically ergative. In all dialects though, nouns can be brought to the front of the sentence as topics.

The grammar is fairly typical for the region, secundative ditransitive alighment, postpositions, indirect objects coming before the verb, split ergativity (the details of which depend on the dialect), pronouns marked by verbal prefixes, a reciprocal voice marked by a reduplicative prefix, a large number of grammatical moods, and no obligatory plural marking.

The phonology is also fairly typical for the region, with a three way aspirated vs. tenuis vs. voiced contrast in stops, a voicing contrast in sonorants consonants, moderately restrictive phonotactics, and a phonation contrast in vowels.

==Phonology==

===Vowels===

There are 6 monophthongs /a i u e o ə/. Schwa cannot occur in stressed syllables. There are also 4 diphthongs /ai au əu əi/. In the Central dialect /əu/ has the allophone [ou] after velar consonants, and /əi/ has the allophone [ei] after palatal consonants. The Eastern dialects have taken this sound change even further, shifting /əu/ to /ou/ and /əi/ to /ei/ in all cases. Meanwhile, the Northwestern dialects have shifted the diphthongs in a different way, with /əu/ becoming /eu/ and /əi/ becoming /oi/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Alveolo-Palatal !! Palatal !! Velar !! Glottal
|-
| Nasal || m̥ m || || n̥ n || || (ɲ̥) (ɲ) || ŋ̊ ŋ* ||
|-
| Stop || pʰ p b || || tʰ t d || || (cʰ) (c) (ɟ) || kʰ k g* || ʔ
|-
| Affricate || || || tsʰ ts dz || (tɕʰ) (tɕ) (dʑ) || || ||
|-
| Fricative || || θ ð || sʰ s z || (ɕ) (ʑ) || (ç) || || h
|-
| Lateral Fricative || || || ɬ || || || ||
|-
| Approximant || ʍ w || || || || j || ||
|-
| Lateral Approximant || || l || || || (ʎ) || ||
|-
| Flap || ⱱ || || ɾ || || || ||
|}

Central Hnyengu, as well as some nearby NW and SW dialects merge /θ/ and /ʍ/ into /f/, and /ð/ and /ⱱ/ into /v/. However, the more outlying NW and SW dialects preserve these phonemes.

The NW dialect has a series of palatalised labial consonants /pʰʲ/, /pʲ/, /bʲ/, /ⱱʲ/, that correspond to sequences of labial consonant + /j/ or /i/ + labial consonant in other dialects.

The alveolo-palatal consonants could be considered to be underlyingly clusters of alveolar/dental consonants + /j/ (in the Eastern dialect that is how they are pronounced). Likewise, in all but the NW dialect the palatal consonants could be considered to be underlyingly clusters containing /j/ i.e. [ɲ̥] is underlyingly /n̥j/, [ɲ] is underlyingly /nj/, [cʰ] [c] and [ɟ] are underylingly /kʰj/, /kj/ and /gj/ respectively, [ç] is underlyingly /hj/, and [ʎ] is /lj/ underylingly.

[ŋ] and [g] are in free variation. After vowels, [ŋ] is more common, and elsewhere, [g] is more common. So the name of the language Hnyengu could be pronounced either [ˈɲ̥əŋgu] or [ˈɲ̥əŋŋu].

The pronunciation of /tsʰ/, /ts/ and /dz/ varies a lot between dialects. There is a strong tendency for them not to be sibilants, or to be less sibilant compared to /sʰ/, /s/, /z/. In SW dialects they are retroflex /ʈʂʰ/ /ʈʂ/ /ɖʐ/, in NW dialects they are dental /tθʰ/ /tθ/ /dð/ and in the East they are lateral /tɬʰ/, /tɬ/, /dɮ/.

Especially in Eastern dialects, /z/ is often pronounced as /ɹ/.

There is often a degree of velarisation to /l/, and this tendency increases the further west you go. In NW and SW dialects, /l/ is pronounced as [ɫ].

===Phonotactics===

Syllable structures can be of the form C (C) V (C), where C is a consonant and V is a vowel (monophthong or diphthong). When two consonants form the onset of a syllable, the second must be /j/. Excluding the previously discussed cases, the only permissible clusters consist of a labial consonant + /j/.

In all dialects, /p/, /t/, /k/ and /ʔ/ can occur as codas.

Nasals can also occur as codas, but in most dialects they do not contrast phonemically with one another. Before another consonant a nasal is always homorganic with that consonant, and at the end of words the only nasal coda heard is [ɲ] after front vowels and [ŋ] elsewhere (in the SW dialect this is simply nasalisation of the preceding vowel).

In the NW dialect, /m/, /n/, /ɲ/ and /ŋ/ are phonemically distinct from one another in codas. /pʲ/, /c/, /ʎ/ and /l/ can also occur as codas in these dialects.

In the Eastern dialect, the flaps /ⱱ/ and /ɾ/ can also occur as codas (in other dialects, the corresponding words end in /ⱱə/ and /ɾə/.

====Grassman's Law====

=====Eastern Dialect=====

In the Eastern dialect, a sound change analagous to Grassman's Law in Indo-European has occured. In a single word, there can only be a single aspirated consonant. In words that contain more than one, all but the last is deleted e.g. /pʰuˈɾepʰe/ - "to prepare" is /puˈɾepʰe/ in the Eastern dialect. Voiceless nasals, /ʍ/, /ɬ/ and /h/ count as aspirated consonants, and change to voiced nasals, /w/, /l/ and /ʔ/ respectively e.g. /ˈkʰoɬi/ - "coffee" is /ˈkoɬi/ in the Eastern dialect.

=====NW Dialect=====

Grassman's Law also operates in the NW dialect, albeit in the opposite direction. All but the first aspirated consonant is deleted, so /pʰuˈɾepʰe/ - "to prepare" is /pʰuˈɾepe/. Also, /ɬ/ lenites to /ʎ/ rather than /l/ e.g. /ˈkʰoɬi/ - "coffee" becomes /ˈkʰoʎi/.

==Noun Morphology==

===Case===

Hnyengu

2016-10-16T09:19:34Z

Linguist Wannabe:

Hnyengu (pronounced [ˈɲ̥əŋgu]) is another language of my far-future post-apocalypse Antarctican conword, spoken along the Ross Sea coast to the north and west of where [[Kämpya]] is spoken. Genetically, it is a Japonic language, however it has been heavily influenced by other languages. In the generations after the initial colonisation, Hnyengu speakers were usually multilingual, and the language absorbed a huge amount of loanwords, especially from English and Spanish. Over the centuries, it continued to be subject to areal influence. A strong Sprachbund has emerged on the Ross Sea coast, and there are also a number of Antarctic areal features.

There are four major dialect groups: Eastern, Central, Southwestern and Northwestern. The Central dialect is spoken in the major city in the region, Myinatu. For a long time it had been the prestige dialect, but after the city was conquered by Kämpya speakers, the inhabitants of the city switched to Kämpya, and within the city there is now a strong stigma attached to the Central dialect of Hnyengu. However, this stigma does not extend to the other dialects, and there is still a large amount of bilingualism between them and Kämpya.

== Brief Description ==

Like modern Japanese, Hnyengu is a topic comment language. However, a succession of sound changes has led to the morphology becoming more fusional.

The morphosyntactic alignment is split ergative (the ergative suffix /-ŋ/ being an erosion of the Japanese genitive /no/). The exact rules of the split vary between dialects, but depend on animacy and relative topicality of the subject.

The TAM system depends on the dialect. The dialects use roughly the same set of verb affixes, however different dialects use them to mark different things tense/aspect/evidentiality.

The grammar is fairly typical for the region, with SVO word order, secundative ditransitive alighment, postpositions, indirect objects coming before the verb, split ergativity (the details of which depend on the dialect), pronouns marked by verbal prefixes, a reciprocal voice marked by a reduplicative prefix, a large number of grammatical moods, and no obligatory plural marking.

The phonology is also fairly typical for the region, with a three way aspirated vs. tenuis vs. voiced contrast in stops, a voicing contrast in sonorants consonants, moderately restrictive phonotactics, and a phonation contrast in vowels.

==Phonology==

===Vowels===

There are 6 monophthongs /a i u e o ə/. Schwa cannot occur in stressed syllables. There are also 4 diphthongs /ai au əu əi/. In the Central dialect /əu/ has the allophone [ou] after velar consonants, and /əi/ has the allophone [ei] after palatal consonants. The Eastern dialects have taken this sound change even further, shifting /əu/ to /ou/ and /əi/ to /ei/ in all cases. Meanwhile, the Northwestern dialects have shifted the diphthongs in a different way, with /əu/ becoming /eu/ and /əi/ becoming /oi/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Alveolo-Palatal !! Palatal !! Velar !! Glottal
|-
| Nasal || m̥ m || || n̥ n || || (ɲ̥) (ɲ) || ŋ̊ ŋ* ||
|-
| Stop || pʰ p b || || tʰ t d || || (cʰ) (c) (ɟ) || kʰ k g* || ʔ
|-
| Affricate || || || tsʰ ts dz || (tɕʰ) (tɕ) (dʑ) || || ||
|-
| Fricative || || θ ð || sʰ s z || (ɕ) (ʑ) || (ç) || || h
|-
| Lateral Fricative || || || ɬ || || || ||
|-
| Approximant || ʍ w || || || || j || ||
|-
| Lateral Approximant || || l || || || (ʎ) || ||
|-
| Flap || ⱱ || || ɾ || || || ||
|}

Central Hnyengu, as well as some nearby NW and SW dialects merge /θ/ and /ʍ/ into /f/, and /ð/ and /ⱱ/ into /v/. However, the more outlying NW and SW dialects preserve these phonemes.

The NW dialect has a series of palatalised labial consonants /pʰʲ/, /pʲ/, /bʲ/, /ⱱʲ/, that correspond to sequences of labial consonant + /j/ or /i/ + labial consonant in other dialects.

The alveolo-palatal consonants could be considered to be underlyingly clusters of alveolar/dental consonants + /j/ (in the Eastern dialect that is how they are pronounced). Likewise, in all but the NW dialect the palatal consonants could be considered to be underlyingly clusters containing /j/ i.e. [ɲ̥] is underlyingly /n̥j/, [ɲ] is underlyingly /nj/, [cʰ] [c] and [ɟ] are underylingly /kʰj/, /kj/ and /gj/ respectively, [ç] is underlyingly /hj/, and [ʎ] is /lj/ underylingly.

[ŋ] and [g] are in free variation. After vowels, [ŋ] is more common, and elsewhere, [g] is more common. So the name of the language Hnyengu could be pronounced either [ˈɲ̥əŋgu] or [ˈɲ̥əŋŋu].

The pronunciation of /tsʰ/, /ts/ and /dz/ varies a lot between dialects. There is a strong tendency for them not to be sibilants, or to be less sibilant compared to /sʰ/, /s/, /z/. In SW dialects they are retroflex /ʈʂʰ/ /ʈʂ/ /ɖʐ/, in NW dialects they are dental /tθʰ/ /tθ/ /dð/ and in the East they are lateral /tɬʰ/, /tɬ/, /dɮ/.

Especially in Eastern dialects, /z/ is often pronounced as /ɹ/.

There is often a degree of velarisation to /l/, and this tendency increases the further west you go. In NW and SW dialects, /l/ is pronounced as [ɫ].

===Phonotactics===

Syllable structures can be of the form C (C) V (C), where C is a consonant and V is a vowel (monophthong or diphthong). When two consonants form the onset of a syllable, the second must be /j/. Excluding the previously discussed cases, the only permissible clusters consist of a labial consonant + /j/.

In all dialects, /p/, /t/, /k/ and /ʔ/ can occur as codas.

Nasals can also occur as codas, but in most dialects they do not contrast phonemically with one another. Before another consonant a nasal is always homorganic with that consonant, and at the end of words the only nasal coda heard is [ɲ] after front vowels and [ŋ] elsewhere (in the SW dialect this is simply nasalisation of the preceding vowel).

In the NW dialect, /m/, /n/, /ɲ/ and /ŋ/ are phonemically distinct from one another in codas. /pʲ/, /c/, /ʎ/ and /l/ can also occur as codas in these dialects.

In the Eastern dialect, the flaps /ⱱ/ and /ɾ/ can also occur as codas (in other dialects, the corresponding words end in /ⱱə/ and /ɾə/.

====Grassman's Law====

=====Eastern Dialect=====

In the Eastern dialect, a sound change analagous to Grassman's Law in Indo-European has occured. In a single word, there can only be a single aspirated consonant. In words that contain more than one, all but the last is deleted e.g. /pʰuˈɾepʰe/ - "to prepare" is /puˈɾepʰe/ in the Eastern dialect. Voiceless nasals, /ʍ/, /ɬ/ and /h/ count as aspirated consonants, and change to voiced nasals, /w/, /l/ and /ʔ/ respectively e.g. /ˈkʰoɬi/ - "coffee" is /ˈkoɬi/ in the Eastern dialect.

=====NW Dialect=====

Grassman's Law also operates in the NW dialect, albeit in the opposite direction. All but the first aspirated consonant is deleted, so /pʰuˈɾepʰe/ - "to prepare" is /pʰuˈɾepe/. Also, /ɬ/ lenites to /ʎ/ rather than /l/ e.g. /ˈkʰoɬi/ - "coffee" becomes /ˈkʰoʎi/.

Hnyengu

2016-10-15T09:55:57Z

Linguist Wannabe:

Hnyengu (pronounced [ˈɲ̥əŋgu]) is another language of my far-future post-apocalypse Antarctican conword, spoken along the Ross Sea coast to the north and west of where [[Kämpya]] is spoken. Genetically, it is a Japonic language, however it has been heavily influenced by other languages. In the generations after the initial colonisation, Hnyengu speakers were usually multilingual, and the language absorbed a huge amount of loanwords, especially from English and Spanish. Over the centuries, it continued to be subject to areal influence. A strong Sprachbund has emerged on the Ross Sea coast, and there are also a number of Antarctic areal features.

There are four major dialect groups: Eastern, Central, Southwestern and Northwestern. The Central dialect is spoken in the major city in the region, Myinatu. For a long time it had been the prestige dialect, but after the city was conquered by Kämpya speakers, the inhabitants of the city switched to Kämpya, and within the city there is now a strong stigma attached to the Central dialect of Hnyengu. However, this stigma does not extend to the other dialects, and there is still a large amount of bilingualism between them and Kämpya.

== Brief Description ==

Like modern Japanese, Hnyengu is a topic comment language. However, a succession of sound changes has led to the morphology becoming more fusional.

The morphosyntactic alignment is split ergative (the ergative suffix /-ŋ/ being an erosion of the Japanese genitive /no/). The exact rules of the split vary between dialects, but depend on animacy and relative topicality of the subject.

The TAM system depends on the dialect. The dialects use roughly the same set of verb affixes, however different dialects use them to mark different things tense/aspect/evidentiality.

The grammar is fairly typical for the region, with SVO word order, secundative ditransitive alighment, postpositions, indirect objects coming before the verb, split ergativity (the details of which depend on the dialect), pronouns marked by verbal prefixes, a reciprocal voice marked by a reduplicative prefix, a large number of grammatical moods, and no obligatory plural marking.

The phonology is also fairly typical for the region, with a three way aspirated vs. tenuis vs. voiced contrast in stops, a voicing contrast in sonorants consonants, moderately restrictive phonotactics, and a phonation contrast in vowels.

==Phonology==

===Vowels===

There are 6 monophthongs /a i u e o ə/. Schwa cannot occur in stressed syllables. There are also 4 diphthongs /ai au əu əi/. In the Central dialect /əu/ has the allophone [ou] after velar consonants, and /əi/ has the allophone [ei] after palatal consonants. The Eastern dialects have taken this sound change even further, shifting /əu/ to /ou/ and /əi/ to /ei/ in all cases. Meanwhile, the Northwestern dialects have shifted the diphthongs in a different way, with /əu/ becoming /eu/ and /əi/ becoming /oi/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Alveolo-Palatal !! Palatal !! Velar !! Glottal
|-
| Nasal || m̥ m || || n̥ n || || (ɲ̥) (ɲ) || ŋ̊ ŋ* ||
|-
| Stop || pʰ p b || || tʰ t d || || (cʰ) (c) (ɟ) || kʰ k g* || ʔ
|-
| Affricate || || || tsʰ ts dz || (tɕʰ) (tɕ) (dʑ) || || ||
|-
| Fricative || || θ ð || sʰ s z || (ɕ) (ʑ) || (ç) || || h
|-
| Lateral Fricative || || || ɬ || || || ||
|-
| Approximant || ʍ w || || || || j || ||
|-
| Lateral Approximant || || l || || || (ʎ) || ||
|-
| Flap || ⱱ || || ɾ || || || ||
|}

Central Hnyengu, as well as some nearby NW and SW dialects merge /θ/ and /ʍ/ into /f/, and /ð/ and /ⱱ/ into /v/. However, the more outlying NW and SW dialects preserve these phonemes.

The NW dialect has a series of palatalised labial consonants /pʰʲ/, /pʲ/, /bʲ/, /ⱱʲ/, that correspond to sequences of labial consonant + /j/ or /i/ + labial consonant in other dialects.

The alveolo-palatal consonants could be considered to be underlyingly clusters of alveolar/dental consonants + /j/ (in the Eastern dialect that is how they are pronounced). Likewise, in all but the NW dialect the palatal consonants could be considered to be underlyingly clusters containing /j/ i.e. [ɲ̥] is underlyingly /n̥j/, [ɲ] is underlyingly /nj/, [cʰ] [c] and [ɟ] are underylingly /kʰj/, /kj/ and /gj/ respectively, [ç] is underlyingly /hj/, and [ʎ] is /lj/ underylingly.

[ŋ] and [g] are in free variation. After vowels, [ŋ] is more common, and elsewhere, [g] is more common. So the name of the language Hnyengu could be pronounced either [ˈɲ̥əŋgu] or [ˈɲ̥əŋŋu].

The pronunciation of /tsʰ/, /ts/ and /dz/ varies a lot between dialects. There is a strong tendency for them not to be sibilants, or to be less sibilant compared to /sʰ/, /s/, /z/. In SW dialects they are retroflex /ʈʂʰ/ /ʈʂ/ /ɖʐ/, in NW dialects they are dental /tθʰ/ /tθ/ /dð/ and in the East they are lateral /tɬʰ/, /tɬ/, /dɮ/.

Especially in Eastern dialects, /z/ is often pronounced as /ɹ/.

There is often a degree of velarisation to /l/, and this tendency increases the further west you go. In NW and SW dialects, /l/ is pronounced as [ɫ].

===Phonotactics===

Syllable structures can be of the form (C) (C) V (C), where C is a consonant and V is a vowel (monophthong or diphthong). When two consonants form the onset of a syllable, the second must be /j/. Excluding the previously discussed cases, the only permissible clusters consist of a labial consonant + /j/.

In all dialects, /p/, /t/ and /k/ can occur as codas.

Nasals can also occur as codas, but in most dialects they do not contrast phonemically with one another. Before another consonant a nasal is always homorganic with that consonant, and at the end of words the only nasal coda heard is [ɲ] after front vowels and [ŋ] elsewhere (in the SW dialect this is simply nasalisation of the preceding vowel).

In the NW dialect, /m/, /n/, /ɲ/ and /ŋ/ are phonemically distinct from one another in codas. /pʲ/, /c/, /ʎ/ and /l/ can also occur as codas in these dialects.

In the Eastern dialect, the flaps /ⱱ/ and /ɾ/ can also occur as codas (in other dialects, the corresponding words end in /ⱱə/ and /ɾə/.

Hnyengu

2016-10-14T13:53:49Z

Linguist Wannabe: Created page with "Hnyengu (pronounced ˈɲ̥əŋgu) is another language of my far-future post-apocalypse Antarctican conword, spoken along the Ross Sea coast to the north and west of where K..."

Hnyengu (pronounced ˈɲ̥əŋgu) is another language of my far-future post-apocalypse Antarctican conword, spoken along the Ross Sea coast to the north and west of where [[Kämpya]] is spoken. Genetically, it is a Japonic language, however it has been heavily influenced by other languages. In the generations after the initial colonisation, Hnyengu speakers were usually multilingual, and the language absorbed a huge amount of loanwords, especially from English and Spanish. Over the centuries, it continued to be subject to areal influence. A strong Sprachbund has emerged on the Ross Sea coast, and there are also a number of Antarctic areal features.

There are four major dialect groups: Eastern, Central, Southwestern and Northwestern. The Central dialect is spoken in the major city in the region, Myinatu. For a long time it had been the prestige dialect, but after the city was conquered by Kämpya speakers, the inhabitants of the city switched to Kämpya, and within the city there is now a strong stigma attached to the Central dialect of Hnyengu. However, this stigma does not extend to the other dialects, and there is still a large amount of bilingualism between them and Kämpya.

== Brief Description ==

Like modern Japanese, Hnyengu is a topic comment language. However, a succession of sound changes has led to the morphology becoming more fusional.

The morphosyntactic alignment is split ergative (the ergative suffix /-ŋ/ being an erosion of the Japanese genitive /no/). The exact rules of the split vary between dialects, but depend on animacy and relative topicality of the subject.

The TAM system depends on the dialect. Different dialects

Seggeynni

2016-10-14T09:04:52Z

Linguist Wannabe: Changed 'Nwumer to Nwuemer

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [[Nwuemer]] was spoken. Kämpya speakers were more technologically advanced than Nwuemer speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the Nwuemer speaking community, and Nwuemer has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and Nwuemer speakers. In addition to Seggeynni, almost all have a passive command of both Nwuemer (still a common trade language in the region) and Kämpya (a liturgical language).

Like Nwuemer, there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoken language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to Nwuemer. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in Nwuemer, the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to Nwuemer. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with Nwuemer, Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ χɘ́ˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ χɘ́ˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from Nwuemer, which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːdʑə̀n/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

==Noun Phrases==

Most of the noun phrase grammar is very similar to Kämpya. This section will only discuss the differences.

===Articles and Demonstratives===

Unlike Kämpya, Seggeynni has no articles.

The demonstrative system is also somewhat simplified. The specialised demonstratives for something uphill or downhill from the speaker have been lost (perhaps because Seggeynni is spoken in an area that is much less hilly than Kämpya. There are now only three degrees of distance, like Japanese, or English when it still used "yonder".

{| class="wikitable"
|-
! !! This !! That !! That over there
|-
| Pronominal || dâ̰ː || fṵ̂ː || xə́ˈɫà̤ːnə̀
|-
| Adnominal (Places) || dá || *xáwə́ || *xə́ˈɫánə́
|-
| Adnominal (Other cases) || dà̰ː || fṵ̀ː || xə́ˈɫà̤ːnə̀
|}

* When demonstratives beginning with /x/ are used adnominally to describe places, the /x/ changes to /ʔ/ if there is another spirant phoneme in the word e.g. ˈmʲúːʔnànə̀/ [ˈmʲʉ́ən̰àⁿ] - "mountain" -> /xáwə́=ˈmʲúːʔnànə̀/ [χɑ́uˈmʲʉ́ən̰àⁿ] - "that mountain", but /ˈfò̤ɥə̀/ [ˈfòy̤ː] - "forest" -> /ʔáwə́=ˈfò̤ɥə̀/ [ʔáuˈfòy̤ː] - "that forest", not */xáwə́=ˈfò̤ɥə̀/.

===Possession===

Possession is marked by the enclitic /jə/, which undergoes tonal change to indicate alienable and inalienable possession just like Kämpya /ja/ e.g.

/ˈdò̰gə́=jə́ báˈwə̂nə̀/

[dò̰ʢgə́i ˈbâṵnə̀]

dog=GEN.INALIENABLE bone

"The dog's bone" (in it's body)

vs.

/ˈdô̰gə̀=jə̀ báˈwə̂nə̀/

[dô̰ʢgə̀i ˈbâṵnə̀]

dog=GEN.ALIENABLE bone

"The dog's bone" (e.g. that it's chewing on, burying etc.)

==Mood/Evidentiality==

This also works in the exact same way as Kämpya, with particles that are mostly cognate. These have reduced forms which are used as clitics to mark a sentence for mood/evidentiality, and also full forms which are responses to polar questions.

{| class="wikitable"
|-
! Mood / Evidential !! Clitic Form !! Full Form !! Etymology
|-
| Negative || ɲawə || náˈwə̰̂ || English "no"
|-
| Necessitative || gʷa || góʔ || English "gotta"
|-
| Optative || wana || ˈwóʔnà || English "wanna"
|-
| Conditional || wədə || ˈwə̰̂də̀ || English "would"
|-
| Hypothetical || fəjə || fə́ˈjə̰̂ || English "if"
|-
| Direct || Unmarked || Repeat Verb || N/A
|-
| Reportative || sfajə|| sˤáˈjə̰̂ || English "say"
|-
| Inferential || ɟə || ɟè̤ || English "guess"
|-
| Assumptive || matə || mà̤tə̀ || English "must"
|-
| Interrogative || dəwə || də́ˈwə̰̂ || English "do"
|-
| Challenging a presupposition || N/A|| páˈwə̤̀ || Thai /plàu/
|}

==Verb Subordination==

This is done in the same way as Kämpya. The particles used are:

{| class="wikitable"
|-
! Particle !! Normal !! Emphatic
|-
| Intransitive || səjə || jə̤̀tə̀
|-
| Accusative || nə || nô̰
|-
| Ergative || nəja || ɲâ̰ː
|}

==Comparative Constructions ==

This is done exactly the same as Kämpya, using the postpositional clitic /tfapə/ (often shortened to [tfaʔ]), meaning "above".

'Nwuemer

2016-10-14T08:01:13Z

Linguist Wannabe: Replaced content with "Changing the name of the conlang back to Nwuemer, because I've decided to have a phonotactic rule prohibiting word-initial glottalised sonorants (as seems common cross-lin..."

Changing the name of the conlang back to Nwuemer, because I've decided to have a phonotactic rule prohibiting word-initial glottalised sonorants (as seems common cross-linguistically).

Nwuemer

2016-10-14T07:32:50Z

Linguist Wannabe: Got rid of initial glottalised resonants

Nwuemer is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent. The initial apostrophe in the name represents glottalisation, meaning that the name of the language is pronounced [ˈn̰ʷy̏ːmə̄], with the underlying phonemic representation being /ˈʔnʷiːmə/ (with a Low word tone). Because each group of nomads speak their own dialect, Nwuemer is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that Nwuemer has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads).

Like many other Antarctican languages (but definitely not all), Nwuemer is descended from English. However, many ancestors of Nwuemer speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the word /ˈxəⁿda/ - "shepherd" is phonemically /əⁿ/, but is almost always heard as [ən].

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate.

Some dialects merge /ɫʷ/ into /ɫ/. While both of these are strictly speaking coronal laterals with secondary articulation, for the purposes of the phonology they pattern with the uvular consonants (e.g. regarding how they affect the pronunciation of surrounding vowels).

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

/x/ is pronounced [χ] word initially. A small number of speakers pronounce it as [ħ] or [h].

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low, and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /waˈjəʔma/ (H) - "llama" (genitive) is pronounced [wɞ́ˈjɪ̋m̰á], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /waˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [wāˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /waˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [wáˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules (which are identical to Hindi stress rules):

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases, vowels are always considered short.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

Voiced obstruents cannot occur word initially, nor can clusters containing /ʔ/.

===Vowel Allophony===

The vowels of Nwuemer have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any Nwuemer adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /tʷəˈtʷoʀa/ (L) - [tʷʊ̄ˈtʷòʀɑ̄] - "shamans". However, in Nwuemer, pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, Nwuemer has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

Nwuemer has three cases, Absolutive (unmarked), Genitive (marked with the prefix /wa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /waˈfaza/ (H) [wáˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /waˈdʷoʀa/ (L) [wāˈdʷòʀɑ̀] - "of the shaman" (not */waˈtʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

===Possession===

Like most Antarctican languages, Nwuemer uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ wa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ wá-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, Nwuemer makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In Nwuemer, prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ɲəʀə/ [ɲɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ɲəʀə=ˈɕəsə/ (HL) [ɲɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

Nwuemer borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa mʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ mʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈnəʔmʲəⁿ/ (L) [ˈnɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈnətsəⁿ/ (HL) [ˈnə̋tsə́ⁿ]
|-
| 2PS Singular || /jəˈqo/ (H) [jɘ́ˈqɔ̋] || /jəˈʔɥo/ (H) [jʏ́ˈɥ̰ʉ̋] || /mo/ (H) [mő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /jəˈʔɲʷo/ (H) [jʏ́ˈɲ̰ʷʉ̋] || /nəˈʔɲʷo/ (H) [nʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈɲəʔja/ (H) [ˈɲɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ˈtɕaɽaⁿ/ (HL) - [tɕɐ́ɽɑ̀ⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtɕaɽaⁿ ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ tɕɐ́ɽɑ̀ⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || mʲaɲə- || ʎə- || na- || tʷaⁿ-
|-
| Antipassive || mʲaⁿma- || ʎama- || nama- || N/A
|-
| Benefactive Applicative || mʲaɲəqa- || ʎəka- || naka- || tʷaⁿka-
|-
| Instrumental Applicative || mʲaɲəʔpaⁿ- || ʎəpaⁿ- || napaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /mʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /mʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

Nwuemer is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa mʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/mʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[mʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa mʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ mʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/mʲaɲə-ˈpfaɫə/ (HL)

[mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa mʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá mʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎæmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎæmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎæmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎæmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/jəˈqo mʲaⁿma-ˈɕuːwə/ (H) (HL)

[jɘ́ˈqɔ̋ mʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /mʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎæmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/mʲaɲə-ˈpfaɫə ˈxəⁿda/ (HL) (H)

[mʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχʌ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/mʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxəⁿda/ (HL) (L) (H)

[mʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈxə̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxəⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /mʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎɪ́káˈbɛ́dˤʌ̀ ˈχʌ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə tə=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The Nwuemer copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /mʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxəⁿda ʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxə̋ndá ʎɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxəⁿda ʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχʌ̋ndá ʎɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, Nwuemer speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʎəˈzadˤə ˈjiːtə ɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈjêːdˤʌ̀ ɲɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the Nwuemer word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa mʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa mʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ mʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/mʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[mʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/mʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ mʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, Nwuemer is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

Nwuemer's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʎəfiː/ (L) (H) [cȅⁿ ʎɪ́ˈfʲi̋ː] - "why" (literally "what reason")

Nwuemer has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /wa-/, becoming /wa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ wa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈwɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in Nwuemer, the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in Nwuemer, treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə wa-ˈdʷogə ʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈtantəʔ wɔ́ˈdʷúgə̀ ʎɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Seggeynni

2016-09-29T07:37:59Z

Linguist Wannabe: /* Alternate Second Person Pronouns */

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [['Nwuemer]] was spoken. Kämpya speakers were more technologically advanced than 'Nwuemer speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the 'Nwuemer speaking community, and 'Nwuemer has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and 'Nwuemer speakers. In addition to Seggeynni, almost all have a passive command of both 'Nwuemer (still a common trade language in the region) and Kämpya (a liturgical language).

Like 'Nwuemer, there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoken language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to 'Nwuemer. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in 'Nwuemer, the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to 'Nwuemer. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with 'Nwuemer, Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ χɘ́ˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ χɘ́ˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from 'Nwuemer, which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːdʑə̀n/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

==Noun Phrases==

Most of the noun phrase grammar is very similar to Kämpya. This section will only discuss the differences.

===Articles and Demonstratives===

Unlike Kämpya, Seggeynni has no articles.

The demonstrative system is also somewhat simplified. The specialised demonstratives for something uphill or downhill from the speaker have been lost (perhaps because Seggeynni is spoken in an area that is much less hilly than Kämpya. There are now only three degrees of distance, like Japanese, or English when it still used "yonder".

{| class="wikitable"
|-
! !! This !! That !! That over there
|-
| Pronominal || dâ̰ː || fṵ̂ː || xə́ˈɫà̤ːnə̀
|-
| Adnominal (Places) || dá || *xáwə́ || *xə́ˈɫánə́
|-
| Adnominal (Other cases) || dà̰ː || fṵ̀ː || xə́ˈɫà̤ːnə̀
|}

* When demonstratives beginning with /x/ are used adnominally to describe places, the /x/ changes to /ʔ/ if there is another spirant phoneme in the word e.g. ˈmʲúːʔnànə̀/ [ˈmʲʉ́ən̰àⁿ] - "mountain" -> /xáwə́=ˈmʲúːʔnànə̀/ [χɑ́uˈmʲʉ́ən̰àⁿ] - "that mountain", but /ˈfò̤ɥə̀/ [ˈfòy̤ː] - "forest" -> /ʔáwə́=ˈfò̤ɥə̀/ [ʔáuˈfòy̤ː] - "that forest", not */xáwə́=ˈfò̤ɥə̀/.

===Possession===

Possession is marked by the enclitic /jə/, which undergoes tonal change to indicate alienable and inalienable possession just like Kämpya /ja/ e.g.

/ˈdò̰gə́=jə́ báˈwə̂nə̀/

[dò̰ʢgə́i ˈbâṵnə̀]

dog=GEN.INALIENABLE bone

"The dog's bone" (in it's body)

vs.

/ˈdô̰gə̀=jə̀ báˈwə̂nə̀/

[dô̰ʢgə̀i ˈbâṵnə̀]

dog=GEN.ALIENABLE bone

"The dog's bone" (e.g. that it's chewing on, burying etc.)

==Mood/Evidentiality==

This also works in the exact same way as Kämpya, with particles that are mostly cognate. These have reduced forms which are used as clitics to mark a sentence for mood/evidentiality, and also full forms which are responses to polar questions.

{| class="wikitable"
|-
! Mood / Evidential !! Clitic Form !! Full Form !! Etymology
|-
| Negative || ɲawə || náˈwə̰̂ || English "no"
|-
| Necessitative || gʷa || góʔ || English "gotta"
|-
| Optative || wana || ˈwóʔnà || English "wanna"
|-
| Conditional || wədə || ˈwə̰̂də̀ || English "would"
|-
| Hypothetical || fəjə || fə́ˈjə̰̂ || English "if"
|-
| Direct || Unmarked || Repeat Verb || N/A
|-
| Reportative || sfajə|| sˤáˈjə̰̂ || English "say"
|-
| Inferential || ɟə || ɟè̤ || English "guess"
|-
| Assumptive || matə || mà̤tə̀ || English "must"
|-
| Interrogative || dəwə || də́ˈwə̰̂ || English "do"
|-
| Challenging a presupposition || N/A|| páˈwə̤̀ || Thai /plàu/
|}

==Verb Subordination==

This is done in the same way as Kämpya. The particles used are:

{| class="wikitable"
|-
! Particle !! Normal !! Emphatic
|-
| Intransitive || səjə || jə̤̀tə̀
|-
| Accusative || nə || nô̰
|-
| Ergative || nəja || ɲâ̰ː
|}

==Comparative Constructions ==

This is done exactly the same as Kämpya, using the postpositional clitic /tfapə/ (often shortened to [tfaʔ]), meaning "above".

'Nwuemer

2016-09-29T07:06:15Z

Linguist Wannabe: /* Consonants */

'Nwuemer is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent. The initial apostrophe in the name represents glottalisation, meaning that the name of the language is pronounced [ˈn̰ʷy̏ːmə̄], with the underlying phonemic representation being /ˈʔnʷiːmə/ (with a Low word tone). Because each group of nomads speak their own dialect, 'Nwuemer is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that 'Nwuemer has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads).

Like many other Antarctican languages (but definitely not all), 'Nwuemer is descended from English. However, many ancestors of 'Nwuemer speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the word /ˈxəⁿda/ - "shepherd" is phonemically /əⁿ/, but is almost always heard as [ən].

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate.

Some dialects merge /ɫʷ/ into /ɫ/. While both of these are strictly speaking coronal laterals with secondary articulation, for the purposes of the phonology they pattern with the uvular consonants (e.g. regarding how they affect the pronunciation of surrounding vowels).

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

/x/ is pronounced [χ] word initially. A small number of speakers pronounce it as [ħ] or [h].

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋m̰á], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [w̰áˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules (which are identical to Hindi stress rules):

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases, vowels are always considered short.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of 'Nwuemer have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any 'Nwuemer adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in 'Nwuemer, pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, 'Nwuemer has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

'Nwuemer has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /ʔwaˈfaza/ (H) [w̰áˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, 'Nwuemer uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, 'Nwuemer makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In 'Nwuemer, prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

'Nwuemer borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

'Nwuemer is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxəⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχʌ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxəⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈxə̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxəⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχʌ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The 'Nwuemer copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxə̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχʌ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, 'Nwuemer speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the 'Nwuemer word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, 'Nwuemer is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

'Nwuemer's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

'Nwuemer has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in 'Nwuemer, the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in 'Nwuemer, treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə ʔwa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈtantəʔ w̰ɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Kämpya

2016-09-29T06:57:49Z

Linguist Wannabe: /* Vowels */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /sʰâ̰n/ - "son". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile".
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced at the end of the sentence using the dative clitic /-àuŋ/ (note that this does not change the tone class of the noun) e.g.

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/

dog ANTIP=bite lizard=DAT

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ áˈwâḭ/

dog ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

Note that the dative clitic has the allomorph /jàuŋ/ after a vowel e.g.

/ˈlḭ̂zàd θú=ˈbíʔ kʰjâṵ=jàuŋ/

lizard ANTIP=bite cow=DAT

The lizard bit the cow.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

'Nwuemer

2016-09-29T06:53:08Z

Linguist Wannabe: /* Consonants */ Allophones of /x/

'Nwuemer is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent. The initial apostrophe in the name represents glottalisation, meaning that the name of the language is pronounced [ˈn̰ʷy̏ːmə̄], with the underlying phonemic representation being /ˈʔnʷiːmə/ (with a Low word tone). Because each group of nomads speak their own dialect, 'Nwuemer is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that 'Nwuemer has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads).

Like many other Antarctican languages (but definitely not all), 'Nwuemer is descended from English. However, many ancestors of 'Nwuemer speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the word /ˈxəⁿda/ - "shepherd" is phonemically /əⁿ/, but is almost always heard as [ən].

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate.

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

/x/ is pronounced [χ] word initially. A small number of speakers pronounce it as [ħ] or [h].

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋m̰á], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [w̰áˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules (which are identical to Hindi stress rules):

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases, vowels are always considered short.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of 'Nwuemer have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any 'Nwuemer adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in 'Nwuemer, pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, 'Nwuemer has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

'Nwuemer has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /ʔwaˈfaza/ (H) [w̰áˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, 'Nwuemer uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, 'Nwuemer makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In 'Nwuemer, prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

'Nwuemer borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

'Nwuemer is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxəⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχʌ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxəⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈxə̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxəⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχʌ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The 'Nwuemer copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxə̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχʌ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, 'Nwuemer speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the 'Nwuemer word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, 'Nwuemer is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

'Nwuemer's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

'Nwuemer has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in 'Nwuemer, the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in 'Nwuemer, treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə ʔwa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈtantəʔ w̰ɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Seggeynni

2016-09-29T06:48:44Z

Linguist Wannabe: Changed Thangha' to 'Nwuemer

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [['Nwuemer]] was spoken. Kämpya speakers were more technologically advanced than 'Nwuemer speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the 'Nwuemer speaking community, and 'Nwuemer has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and 'Nwuemer speakers. In addition to Seggeynni, almost all have a passive command of both 'Nwuemer (still a common trade language in the region) and Kämpya (a liturgical language).

Like 'Nwuemer, there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoken language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to 'Nwuemer. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in 'Nwuemer, the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to 'Nwuemer. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with 'Nwuemer, Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ χɘ́ˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ χɘ́ˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from 'Nwuemer, which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːtèn/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

==Noun Phrases==

Most of the noun phrase grammar is very similar to Kämpya. This section will only discuss the differences.

===Articles and Demonstratives===

Unlike Kämpya, Seggeynni has no articles.

The demonstrative system is also somewhat simplified. The specialised demonstratives for something uphill or downhill from the speaker have been lost (perhaps because Seggeynni is spoken in an area that is much less hilly than Kämpya. There are now only three degrees of distance, like Japanese, or English when it still used "yonder".

{| class="wikitable"
|-
! !! This !! That !! That over there
|-
| Pronominal || dâ̰ː || fṵ̂ː || xə́ˈɫà̤ːnə̀
|-
| Adnominal (Places) || dá || *xáwə́ || *xə́ˈɫánə́
|-
| Adnominal (Other cases) || dà̰ː || fṵ̀ː || xə́ˈɫà̤ːnə̀
|}

* When demonstratives beginning with /x/ are used adnominally to describe places, the /x/ changes to /ʔ/ if there is another spirant phoneme in the word e.g. ˈmʲúːʔnànə̀/ [ˈmʲʉ́ən̰àⁿ] - "mountain" -> /xáwə́=ˈmʲúːʔnànə̀/ [χɑ́uˈmʲʉ́ən̰àⁿ] - "that mountain", but /ˈfò̤ɥə̀/ [ˈfòy̤ː] - "forest" -> /ʔáwə́=ˈfò̤ɥə̀/ [ʔáuˈfòy̤ː] - "that forest", not */xáwə́=ˈfò̤ɥə̀/.

===Possession===

Possession is marked by the enclitic /jə/, which undergoes tonal change to indicate alienable and inalienable possession just like Kämpya /ja/ e.g.

/ˈdò̰gə́=jə́ báˈwə̂nə̀/

[dò̰ʢgə́i ˈbâṵnə̀]

dog=GEN.INALIENABLE bone

"The dog's bone" (in it's body)

vs.

/ˈdô̰gə̀=jə̀ báˈwə̂nə̀/

[dô̰ʢgə̀i ˈbâṵnə̀]

dog=GEN.ALIENABLE bone

"The dog's bone" (e.g. that it's chewing on, burying etc.)

==Mood/Evidentiality==

This also works in the exact same way as Kämpya, with particles that are mostly cognate. These have reduced forms which are used as clitics to mark a sentence for mood/evidentiality, and also full forms which are responses to polar questions.

{| class="wikitable"
|-
! Mood / Evidential !! Clitic Form !! Full Form !! Etymology
|-
| Negative || ɲawə || náˈwə̰̂ || English "no"
|-
| Necessitative || gʷa || góʔ || English "gotta"
|-
| Optative || wana || ˈwóʔnà || English "wanna"
|-
| Conditional || wədə || ˈwə̰̂də̀ || English "would"
|-
| Hypothetical || fəjə || fə́ˈjə̰̂ || English "if"
|-
| Direct || Unmarked || Repeat Verb || N/A
|-
| Reportative || sfajə|| sˤáˈjə̰̂ || English "say"
|-
| Inferential || ɟə || ɟè̤ || English "guess"
|-
| Assumptive || matə || mà̤tə̀ || English "must"
|-
| Interrogative || dəwə || də́ˈwə̰̂ || English "do"
|-
| Challenging a presupposition || N/A|| páˈwə̤̀ || Thai /plàu/
|}

==Verb Subordination==

This is done in the same way as Kämpya. The particles used are:

{| class="wikitable"
|-
! Particle !! Normal !! Emphatic
|-
| Intransitive || səjə || jə̤̀tə̀
|-
| Accusative || nə || nô̰
|-
| Ergative || nəja || ɲâ̰ː
|}

==Comparative Constructions ==

This is done exactly the same as Kämpya, using the postpositional clitic /tfapə/ (often shortened to [tfaʔ]), meaning "above".

'Nwuemer

2016-09-29T06:45:44Z

Linguist Wannabe: Copied from Thangha'

'Nwuemer is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent. The initial apostrophe in the name represents glottalisation, meaning that the name of the language is pronounced [ˈn̰ʷy̏ːmə̄], with the underlying phonemic representation being /ˈʔnʷiːmə/ (with a Low word tone). Because each group of nomads speak their own dialect, 'Nwuemer is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that 'Nwuemer has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads).

Like many other Antarctican languages (but definitely not all), 'Nwuemer is descended from English. However, many ancestors of 'Nwuemer speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the word /ˈxəⁿda/ - "shepherd" is phonemically /əⁿ/, but is almost always heard as [ən].

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ χ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate.

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋m̰á], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [w̰áˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules (which are identical to Hindi stress rules):

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases, vowels are always considered short.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of 'Nwuemer have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any 'Nwuemer adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in 'Nwuemer, pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, 'Nwuemer has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

'Nwuemer has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /ʔwaˈfaza/ (H) [w̰áˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, 'Nwuemer uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, 'Nwuemer makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In 'Nwuemer, prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

'Nwuemer borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

'Nwuemer is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxəⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχʌ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxəⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈxə̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxəⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχʌ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The 'Nwuemer copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxə̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxəⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχʌ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, 'Nwuemer speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the 'Nwuemer word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, 'Nwuemer is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

'Nwuemer's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

'Nwuemer has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in 'Nwuemer, the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in 'Nwuemer, treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə ʔwa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈtantəʔ w̰ɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Thangha'

2016-09-29T06:45:30Z

Linguist Wannabe: Changed name to 'Nwuemer

Thangha'

2016-09-28T09:55:19Z

Linguist Wannabe: /* Comparative Constructions */

Thangha' is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent.. Because each group of nomads speak their own dialect, Thangha' is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that Thangha's has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads). To give an example of the diversity, the initial consonant in the name of the language can be pronounced as [tˤ], [θˤ], [θ], [tθˤ], or [tʰ]. The consonant that begins the second syllable is usually prononuced [x] or [χ], but a few dialects pronounce it as as [ħ] or [h].

Like many other Antarctican languages (but definitely not all), Thangha' is descended from English. However, many ancestors of Thangha' speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the language name Thangha' is phonemically /aⁿ/, but is usually heard as [aŋ] (or [aɴ] for speakers who use a uvular fricative instead of a velar fricative).

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ χ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate. In these dialects, Thangha' would be pronounced [ˈtsaŋxaʔ].

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋ʔmá], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [w̰áˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules:

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases only short vowels occur.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of Thangha' have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any Thangha' adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in Thangha', pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, Thangha' has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

Thangha' has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /ʔwaˈfaza/ (H) [w̰áˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, Thangha' uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, Thangha' makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In Thangha', prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

Thangha' borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

Thangha' is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxeⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχɛ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxeⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈχɛ̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxeⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχɛ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The Thangha' copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxe̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχɛ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, Thangha' speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the Thangha' word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, Thangha' is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

Thangha's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

Thangha' has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in Thangha', the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in Thangha', treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə ʔwa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈtantəʔ w̰ɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Thangha'

2016-09-28T09:52:35Z

Linguist Wannabe: Clarified rules about preservation of coda r

Thangha' is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent.. Because each group of nomads speak their own dialect, Thangha' is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that Thangha's has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads). To give an example of the diversity, the initial consonant in the name of the language can be pronounced as [tˤ], [θˤ], [θ], [tθˤ], or [tʰ]. The consonant that begins the second syllable is usually prononuced [x] or [χ], but a few dialects pronounce it as as [ħ] or [h].

Like many other Antarctican languages (but definitely not all), Thangha' is descended from English. However, many ancestors of Thangha' speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the language name Thangha' is phonemically /aⁿ/, but is usually heard as [aŋ] (or [aɴ] for speakers who use a uvular fricative instead of a velar fricative).

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ χ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate. In these dialects, Thangha' would be pronounced [ˈtsaŋxaʔ].

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋ʔmá], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [w̰áˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules:

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases only short vowels occur.

* Diphthongs, coda consonants (including coda /r/, which has survived in stresssed syllables) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of Thangha' have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any Thangha' adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in Thangha', pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, Thangha' has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

Thangha' has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐa/ (H) [ˈfɑ̋ɖʐɑ́] - father -> /ʔwaˈfaza/ (H) [w̰áˈfɑ̋ɖʐɑ́] - "of the father" -> /pəˈfaza/ (H) [pə̄ˈfɑ̋ɖʐɑ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, Thangha' uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, Thangha' makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In Thangha', prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

Thangha' borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

Thangha' is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxeⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχɛ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxeⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈχɛ̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxeⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχɛ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The Thangha' copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxe̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχɛ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, Thangha' speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the Thangha' word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, Thangha' is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

Thangha's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

Thangha' has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in Thangha', the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in Thangha', treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈdantəʔ wɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Thangha'

2016-09-28T09:30:24Z

Linguist Wannabe: /* Phonotactics */

Thangha' is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent.. Because each group of nomads speak their own dialect, Thangha' is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that Thangha's has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads). To give an example of the diversity, the initial consonant in the name of the language can be pronounced as [tˤ], [θˤ], [θ], [tθˤ], or [tʰ]. The consonant that begins the second syllable is usually prononuced [x] or [χ], but a few dialects pronounce it as as [ħ] or [h].

Like many other Antarctican languages (but definitely not all), Thangha' is descended from English. However, many ancestors of Thangha' speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the language name Thangha' is phonemically /aⁿ/, but is usually heard as [aŋ] (or [aɴ] for speakers who use a uvular fricative instead of a velar fricative).

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ χ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate. In these dialects, Thangha' would be pronounced [ˈtsaŋxaʔ].

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋ʔmá], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [wáˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules:

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases only short vowels occur.

* Diphthongs, coda consonants (including coda /r/, which has survived in some words) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Voiced obstruents cannot occur word initially.

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of Thangha' have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any Thangha' adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in Thangha', pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, Thangha' has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

Thangha' has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐə/ (H) [ˈfɑ̋ɖʐʌ́] - father -> /ʔwaˈfazə/ (H) [w̰áˈfɑ̋ɖʐʌ́] - "of the father" -> /pəˈfazə/ (H) [pə̄ˈfɑ̋ɖʐʌ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, Thangha' uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, Thangha' makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In Thangha', prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

Thangha' borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

Thangha' is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxeⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχɛ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxeⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈχɛ̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxeⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχɛ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The Thangha' copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxe̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχɛ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, Thangha' speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the Thangha' word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, Thangha' is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

Thangha's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

Thangha' has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in Thangha', the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in Thangha', treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈdantəʔ wɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.

Seggeynni

2016-09-09T14:05:53Z

Linguist Wannabe:

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [[Thangha']] was spoken. Kämpya speakers were more technologically advanced than Thangha' speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the Thangha' speaking community, and Thangha' has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and Thangha' speakers. In addition to Seggeynni, almost all have a passive command of both Thangha' (still a common trade language in the region) and Kämpya (a liturgical language).

Like Thangha', there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoekn language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to Thangha'. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in Thangha', the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to Thangha'. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with Thangha', Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ χɘ́ˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ χɘ́ˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from Thangha', which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːtèn/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

==Noun Phrases==

Most of the noun phrase grammar is very similar to Kämpya. This section will only discuss the differences.

===Articles and Demonstratives===

Unlike Kämpya, Seggeynni has no articles.

The demonstrative system is also somewhat simplified. The specialised demonstratives for something uphill or downhill from the speaker have been lost (perhaps because Seggeynni is spoken in an area that is much less hilly than Kämpya. There are now only three degrees of distance, like Japanese, or English when it still used "yonder".

{| class="wikitable"
|-
! !! This !! That !! That over there
|-
| Pronominal || dâ̰ː || fṵ̂ː || xə́ˈɫà̤ːnə̀
|-
| Adnominal (Places) || dá || *xáwə́ || *xə́ˈɫánə́
|-
| Adnominal (Other cases) || dà̰ː || fṵ̀ː || xə́ˈɫà̤ːnə̀
|}

* When demonstratives beginning with /x/ are used adnominally to describe places, the /x/ changes to /ʔ/ if there is another spirant phoneme in the word e.g. ˈmʲúːʔnànə̀/ [ˈmʲʉ́ən̰àⁿ] - "mountain" -> /xáwə́=ˈmʲúːʔnànə̀/ [χɑ́uˈmʲʉ́ən̰àⁿ] - "that mountain", but /ˈfò̤ɥə̀/ [ˈfòy̤ː] - "forest" -> /ʔáwə́=ˈfò̤ɥə̀/ [ʔáuˈfòy̤ː] - "that forest", not */xáwə́=ˈfò̤ɥə̀/.

===Possession===

Possession is marked by the enclitic /jə/, which undergoes tonal change to indicate alienable and inalienable possession just like Kämpya /ja/ e.g.

/ˈdò̰gə́=jə́ báˈwə̂nə̀/

[dò̰ʢgə́i ˈbâṵnə̀]

dog=GEN.INALIENABLE bone

"The dog's bone" (in it's body)

vs.

/ˈdô̰gə̀=jə̀ báˈwə̂nə̀/

[dô̰ʢgə̀i ˈbâṵnə̀]

dog=GEN.ALIENABLE bone

"The dog's bone" (e.g. that it's chewing on, burying etc.)

==Mood/Evidentiality==

This also works in the exact same way as Kämpya, with particles that are mostly cognate. These have reduced forms which are used as clitics to mark a sentence for mood/evidentiality, and also full forms which are responses to polar questions.

{| class="wikitable"
|-
! Mood / Evidential !! Clitic Form !! Full Form !! Etymology
|-
| Negative || ɲawə || náˈwə̰̂ || English "no"
|-
| Necessitative || gʷa || góʔ || English "gotta"
|-
| Optative || wana || ˈwóʔnà || English "wanna"
|-
| Conditional || wədə || ˈwə̰̂də̀ || English "would"
|-
| Hypothetical || fəjə || fə́ˈjə̰̂ || English "if"
|-
| Direct || Unmarked || Repeat Verb || N/A
|-
| Reportative || sfajə|| sˤáˈjə̰̂ || English "say"
|-
| Inferential || ɟə || ɟè̤ || English "guess"
|-
| Assumptive || matə || mà̤tə̀ || English "must"
|-
| Interrogative || dəwə || də́ˈwə̰̂ || English "do"
|-
| Challenging a presupposition || N/A|| páˈwə̤̀ || Thai /plàu/
|}

==Verb Subordination==

This is done in the same way as Kämpya. The particles used are:

{| class="wikitable"
|-
! Particle !! Normal !! Emphatic
|-
| Intransitive || səjə || jə̤̀tə̀
|-
| Accusative || nə || nô̰
|-
| Ergative || nəja || ɲâ̰ː
|}

==Comparative Constructions ==

This is done exactly the same as Kämpya, using the postpositional clitic /tfapə/ (often shortened to [tfaʔ]), meaning "above".

Kämpya

2016-09-09T14:04:30Z

Linguist Wannabe: /* Comparative Constructions */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /wà̰n/ - "one". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose", /tʰêḭ/ - "two". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile"
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced at the end of the sentence using the dative clitic /-àuŋ/ (note that this does not change the tone class of the noun) e.g.

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/

dog ANTIP=bite lizard=DAT

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ áˈwâḭ/

dog ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

Note that the dative clitic has the allomorph /jàuŋ/ after a vowel e.g.

/ˈlḭ̂zàd θú=ˈbíʔ kʰjâṵ=jàuŋ/

lizard ANTIP=bite cow=DAT

The lizard bit the cow.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /tʰáp/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Seggeynni

2016-09-09T13:58:39Z

Linguist Wannabe:

Kämpya

2016-09-09T13:56:30Z

Linguist Wannabe: /* Responses to Polar Questions */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /wà̰n/ - "one". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose", /tʰêḭ/ - "two". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile"
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced at the end of the sentence using the dative clitic /-àuŋ/ (note that this does not change the tone class of the noun) e.g.

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/

dog ANTIP=bite lizard=DAT

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ áˈwâḭ/

dog ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

Note that the dative clitic has the allomorph /jàuŋ/ after a vowel e.g.

/ˈlḭ̂zàd θú=ˈbíʔ kʰjâṵ=jàuŋ/

lizard ANTIP=bite cow=DAT

The lizard bit the cow.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /mà̤t/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /mà̤t/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /téŋ/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Kämpya

2016-09-09T13:55:45Z

Linguist Wannabe: /* Assumptive Mood */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /wà̰n/ - "one". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose", /tʰêḭ/ - "two". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile"
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced at the end of the sentence using the dative clitic /-àuŋ/ (note that this does not change the tone class of the noun) e.g.

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/

dog ANTIP=bite lizard=DAT

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ áˈwâḭ/

dog ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

Note that the dative clitic has the allomorph /jàuŋ/ after a vowel e.g.

/ˈlḭ̂zàd θú=ˈbíʔ kʰjâṵ=jàuŋ/

lizard ANTIP=bite cow=DAT

The lizard bit the cow.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /mat/ (/mats/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà máts=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /dâ̰/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dâ̰/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /téŋ/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Seggeynni

2016-09-09T13:33:17Z

Linguist Wannabe: Possession

Seggeynni

2016-09-09T13:22:00Z

Linguist Wannabe:

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [[Thangha']] was spoken. Kämpya speakers were more technologically advanced than Thangha' speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the Thangha' speaking community, and Thangha' has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and Thangha' speakers. In addition to Seggeynni, almost all have a passive command of both Thangha' (still a common trade language in the region) and Kämpya (a liturgical language).

Like Thangha', there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoekn language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to Thangha'. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in Thangha', the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to Thangha'. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with Thangha', Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ χɘ́ˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ χɘ́ˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from Thangha', which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːtèn/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

==Articles and Demonstratives==

Unlike Kämpya, Seggeynni has no articles.

The demonstrative system is also somewhat simplified. The specialised demonstratives for something uphill or downhill from the speaker have been lost (perhaps because Seggeynni is spoken in an area that is much less hilly than Kämpya. There are now only three degrees of distance, like Japanese, or English when it still used "yonder".

{| class="wikitable"
|-
! !! This !! That !! That over there
|-
| Pronominal || dâ̰ː || fṵ̂ː || xə́ˈɫà̤ːnə̀
|-
| Adnominal (Places) || dá || *xáwə́ || *xə́ˈɫánə́
|-
| Adnominal (Other cases) || dà̰ː || fṵ̀ː || xə́ˈɫà̤ːnə̀
|}

* When demonstratives beginning with /x/ are used adnominally to describe places, the /x/ changes to /ʔ/ if there is another spirant phoneme in the word e.g. ˈmʲúːʔnànə̀/ [ˈmʲʉ́ən̰àⁿ] - "mountain" -> /xáwə́=ˈmʲúːʔnànə̀/ [χɑ́uˈmʲʉ́ən̰àⁿ] - "that mountain", but /ˈfò̤ɥə̀/ [ˈfòy̤ː] - "forest" -> /ʔáwə́=ˈfò̤ɥə̀/ [ʔáuˈfòy̤ː] - "that forest", not */xáwə́=ˈfò̤ɥə̀/.

Kämpya

2016-09-09T13:03:12Z

Linguist Wannabe: /* Demonstratives */

Kämpya is spoken is my far-future Antarctican conworld, where runaway global warming has melted the icecaps and made the continent habitable (while rendering most of the rest of the world uninhabitable).

It originated from the area around Ross Island, one of the first large-scale colonies on Antarctica. Many of the founders were from a number of East Asian countries, and used English as a common language while bringing their own languages with them. However the community was isolated, and within a few generations their English had developed into a creole, influenced by a number of languages, especially Burmese, Minnan Chinese, and Japanese (spoken in countries where many of them fled from). Most notably, the language became postpositional (English prepositions were replaced with Burmese postpositions). A wave of Spanish speaking migrants also heavily affected the language.

It then diverged into a number of widely divergent dialects (technology regressed, and groups of speakers were cut off from one another), followed by another period of koineisation when enough technology was re-invented for speakers of different dialects to be able to contact one other again. This koine was called Kämpya, and spread to many other parts of the continent, aided by the growth of a religion called Laikyâr (although not all Kämpya speakers follow the Laikyâr religion, and not all Laikyâr believers speak Kämpya). Indeed there is a special register of the language that is only used in religious contexts. There are also a number of regional dialects. However, this page describes what could be termed the "standard" dialect.

==Brief Description==

Kämpya has topic comment syntax with isolating morphology. The morphosyntactic alignment is basically ergative, except that there is a tripartite system on pronouns. However, genitive (alienable) and ergative pronouns are identical. Possessors are marked for alienability [http://en.wikipedia.org/wiki/Alienable_possession] using tone, and come before the nouns they modify. Kämpya (at least in the standard dialect) uses postpositions rather than prepositions, and adjectives can come either before or after the nouns they modify if they are restrictive or non-restrictive respectively [http://en.wikipedia.org/wiki/Restrictiveness].

Verbs do not inflect for agreement, tense or aspect, however there is an elaborate system of particles that indicate mood / evidentiality.

In terms of phonology, the most notable thing is a 3-way vowel phonation contrast on stressed syllables (which is not present on unstressed syllables). Kämpya distinguishes words with harsh voice (marked with a tilde e.g. /a̰/), from breathy voice (marked with a pair of dots either above or below the vowel e.g. /a̤/ or /ä/), from glottalisation (marked with a glottal stop after the vowel e.g. /aʔ/.

There are many minimal pairs of words that only contrast phonation e.g. /kʰà̤ɾ/ - "plaster cast", /kʰâ̰ɾ/ - "card", /kʰáʔɾ/ - "a cart" (the differences in tone can be predicted from the phonation).

In addition to this, there is also a tone contrast, but this is only used for grammatical purposes (e.g. to change between different parts of speech, or to mark alienable / inalienable possession), never for lexical purposes. For example, from the noun /áˈlâṵn/ - "that which is alone", which has High Tone on the first syllable and Falling Tone on the second (with harsh voice), we can derive the non-restrictive adjective /àˈláṵn/ - "by itself / solitary", which has Low Tone on the first syllable and High Tone on the second (with harsh voice).

==Typological Influences==

The contrast between alienable and inalienable possession was not present in any of the major languages of the initial settlers. However it has developed into an areal feature and is now found in most of the languages on the continent.

A similar situation happened with evidentiality, although its spread is much more restricted, only being found in the languages spoken along the Ross Sea coast. In this area, high mountains and long fjords meant that people lived in small, isolated communities. In such communities, where everyone knew one another, gossip could be highly destructive. This context favoured the development of evidentiality.

The lack of marking for number and tense / aspect is common in Antarctican languages from areas where a large portion of the founding population spoke Mainland East / SE Asian languages (which lack these features).

Topic-comment and wh-in-situ word order came into Kämpya through similar means (it is very common in East and SE Asia).

Ergativity came into Antarctica from speakers of Austronesian languages from Eastern Indonesia, the Pacific and the Philippines, three areas which were among the first to be severely affected by climate change, and therefore among the first migrants to the south. some degree of ergativity is a continent-wide areal feature. In the case of Kämpya, the English possessive 's began to be used to mark ergative subjects. By analogy, possessive pronouns then began to be used as ergative pronouns e.g. /jó/ - 2PS.ERG from English "your". Later, a different suffix /ji/ began to be used to mark possession.

Clusivity also spread throughout Antarctica by similar means, although many languages (including Kämpya) ended up conflating the first person singular and the first person exclusive plural.

Applicative constructions also entered Kämpya through influence from Austronesian languages such as Tagalog. However they did not spread as widely throughout the continent.

Marking pronouns with pre-verbal clitics originated in Spanish (indeed Kämpya's accusative pronouns are borrowed from Spanish). This has spread across Antarctica as an areal feature (except that in some languages, they are prefixes rather than clitics).

The vast majority of the founders spoke pro-drop languages (Burmese, Spanish, Chinese, Japanese etc.) so it is perhaps no surprise that Kämpya has become pro-drop (apart from the pre-verbal clitics).

Kämpya borrowed a number of its postpositions from Burmese e.g. locative /ka/. Other postpositions were originally English prepositions, but switched to be postpositions under the influence of languages such as Burmese and Japanese.

Placing postpositional phrases before the verb was borrowed from Chinese (perhaps also helped by the fact that many other founders spoke verb-final languages such as Burmese and Japanese).

Anti-logophoricity is an areal feature that developed along the Ross Sea coast after settlement.

Marking restrictiveness on adjectives by word order originated in Spanish, but became more extensive after settlement. Marking restrictiveness on adjectives (in some way, not necessarily by word order) is an Antarctic areal feature.

The complicated demonstrative system (depending on whether the object is uphill or downhill from the speaker) is an areal feature of the Ross Sea coast. Like most of the areas where such a system has developed, this area is very mountainous.

Using a separate verb for locational predication (in the case of Kämpya /éʔ/, which is cognate to English "at") is an areal feature of East and SE Asia.

Using a postposition to form a comparative construction is another areal feature of Asia which has found its way into Kämpya.

==Phonology==

===Vowels===

The vowel system is quite simple, with 5 monophthongs and 4 diphthongs.

====Monophthongs====
{| class="wikitable"
|-
! !! Front !! Central !! Back
|-
| High || /i/ || || /u/
|-
| Mid || /e/ || || /o/
|-
| Low || || /a/ ||
|}

====Diphthongs====

The following diphthongs are found /ai/, /au/, /ei/ and /ou/.

===Consonants===

{| class="wikitable"
|-
! !! Labial !! Dental !! Alveolar !! Lateral !! Palatal !! Velar !! Glottal
|-
| Stop || pʰ p b || || tʰ t d || || || kʰ k g ||
|-
| Nasal || m̥ m || || n̥ n || || || ŋ ||
|-
| Fricative || || θ ð || sʰ s z || (ɬ) || (ç) || || h
|-
| Semivowel || (ʍ) w || || || || j || ||
|-
| Other Sonorants || ⱱ~ʋ || || ɾ || l || || ||
|}

# /b/, /d/ and /g/ are often pronounced as slightly implosive.
# [ɬ] is underlyingly /hl/.
# [ç] is underlyingly /hj/.
# [ʍ] is underlyingly /hw/.
# [ⱱ] and [ʋ] are in free variation.

===Phonotactics===

In the standard dialect, only syllables of the form (C) (C) V (C) are permitted.

Initial consonant clusters can consist of any non-semivowel consonant followed by /w/ or /j/, except that /ŋj/ syllable onsets are forbidden.

In writing and formal speech, syllables can also begin with /pʰl/, /pl/, /bl/, /kʰl/, /kl/ and /gl/ (i.e. any non-alveolar stop + /l/). But outside formal speech, the /l/ is usually replaced with /j/.

Clusters of /hl/, /hj/ and /hw/ are permitted, but these are pronounced [ɬ], [ç] and [ʍ] respectively.

In unstressed syllables, the coda consonant can only be a stop or a nasal.

In stressed syllables, codas can also be /ⱱ/, /ɾ/ or /h/, as long as the vowel is a monophthong and not a diphthong.

/ŋ/ cannot occur word-initially.

===Consonant Neutralisations===

In colloquial speech, stops in the codas of unstressed syllables are debuccalised to glottal stops e.g. the city named /ˈkʰóʔnàt/ is often pronounced [ˈkʰóʔnàʔ], /ˈpʰḛ̂jòd/ - "fjord" is pronounced [ˈpʰḛ̂ːjòʔ].

If a nasal coda occurs before /h/, an approximant, or at the end of a word, it is often pronounced as nasalisation of the preceding vowel e.g. /nâḭm/ - "name" is pronounced [nâḭⁿ]. But likewise it is still present underlyingly, as can be seen when combined with the dative clitic /àuŋ/ to form /nâḭm=àuŋ/ - "to the name", which is pronounced [nâḭmàuⁿ].

===Stress and Phonation===

One syllable in each word bears stress. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are harsh /a̰/ [http://en.wikipedia.org/wiki/Harsh_voice], breathy /a̤/ or /ä/ [http://en.wikipedia.org/wiki/Breathy_voice], and glottalised /aʔ/. Harsh and breathy vowels are both pronounced long, while glottalised vowels are pronounced short, and with glottalisation of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables).

===Allophones of /h/===

Between two vowels, /h/ is voiced to [ɦ] e.g. /máháˈkáʔp/ - "eternity" is pronounced /máɦáˈkáʔp/

/h/ can occur in the coda of stressed syllables. It's realisation depends on the phonation of the previous vowel.

If the stressed vowel has breathy phonation, the /h/ manifests itself as a breathy voiced fricative [ɦ] e.g. /à̤h/ - "question" is pronounced [à̤ɦ]. Compare /gwà̤/ - "grass" which is pronounced [gwà̤].

If the stressed vowel has glottalised phonation, the /h/ manifests itself as an epiglottal stop [ʡ] e.g. /dáʔh/ - "darkness" is pronounced [dáʡ].

If /h/ occurs after vowels with harsh phonation, it is pronounced as a voiced epiglottal fricative [ʢ] e.g. /zwéˈmâ̰h/ - "to report", is pronounced [zwéˈmâ̰ʢ].

Of course, if a process such a cliticisation (e.g. with the genitive clitic /ji/ or the dative clitic /auŋ/), means that the /h/ is no longer in coda position, then these processes do not occur e.g.

/à̤h=jì/ - "of the question (alienable)" is pronounced [ˈà̤çì] (/hj/ assimilates to [ç]).

/dáʔh=jì/ - "of the darkness (alienable)" is pronounced [ˈdáʔçì].

/à̤h=àuŋ/ - "to the question" is pronounced [ˈà̤ɦàuⁿ].

/dáʔh=àuŋ/ - "to the darkness" is pronounced [ˈdáʔhàuⁿ].

Also, if a coda [ʡ] (i.e. /h/ after a glottalised stressed vowel) is followed by a consonant that can be aspirated (i.e. /p/, /t/, /k/, or /s/), there is a tendency to manifest this as aspiration on the consonant instead e.g.

/ˈbóʔhsòut/ - "bauxite", can be pronounced either [ˈbóʡsòut] or [ˈbósʰòut].

===Tone Patterns===

There are 4 phonemic tones: High á, Falling â, Low à and Rising ǎ. While it may seem that the pitch of each syllable is random, actually it is not. It changes regularly depending on how the word is being used in a sentence (e.g. restrictive vs. non-restrictive, alienable vs. inalienable possessor etc.), in conjuction with the phonation on the syllable (for stressed syllables) or the location of the syllable relative to the stressed syllable (for unstressed syllables).

In terms of assigning tone, there are three parts of speech:

'''Class 1'''
This is the default, catch-all class. It is used for most nouns and verbs.

'''Class 2'''
This is used for nouns in postpositional phrases, inalienable possessors, nouns used attributively, restrictive adjectives (modifying a noun), and infinitive verbs.

'''Class 3'''
This is used for adverbs and non-restrictive adjectives (modifying a noun).

====Stressed Syllables====

This table shows how the tone of a stressed syllable (for each of the three phonations) depends on the tone class.

{| class="wikitable"
|-
! !! Harsh Phonation a̰ !! Breathy Phonation a̤ !! Glottalised Phonation aʔ
|-
| Class 1 || (High) Falling Pitch + Harsh Voice || Low Pitch + Breathy Voice || High Pitch + Glottal Stop
|-
| Class 2 || Low Pitch + Harsh Voice || Low Pitch + Breathy Voice || Low Pitch + Glottal Stop
|-
| Class 3 || High Pitch + Tense Voice || Rising Pitch + Faucalised Voice || High Pitch + Glottal Stop
|}

====Unstressed Syllables====

Here the situation is simpler. For unstressed syllables, the tone can only be high or low, and does not depend on the phonation of the stressed syllable.

For Class 1 words, the tone is high before the stressed syllable, and low after it.

For Class 2 words, unstressed syllables always have high tone.

For Class 3 words, unstressed syllables always have low tone.

====Examples====

Here is a table with examples of how roots change classes depending on what part of speech they belong to:

{| class="wikitable"
|-
! English Translation !! Class 1 !! Class 2 !! Class 3
|-
| Kämpya || ˈkà̤mpjà || ˈkà̤mpjá || ˈkǎ̤mpjà
|-
| guardian || ˈsʰíʔtà || ˈsʰìʔtá || ˈsʰíʔtà
|-
| sister || ˈsʰì̤tà || ˈsʰì̤tá || ˈsʰǐ̤tà
|-
| alone || áˈlâṵn || áˈlàṵn || àˈláṵn
|-
| alone + Genitive clitic /ji/ || áˈlâṵn=jì || áˈlàṵn=jí || àˈláṵn=jì
|-
| question || à̤h || à̤h || ǎ̤h
|-
| question + Genitive clitic /ja/ || à̤h=jì || à̤h=jí || ǎ̤h=jì
|}

===Sound Correspondences with English===

====Phonation====

Contrastive phonation developed on stressed vowels depending on the following consonants. Breathy phonation emerged via a sound change where voiceless fricatives were lost after a stressed vowel e.g. /ˈsʰì̤tà/ - "sister". Glottalised phonation developed whenever there was a voiceless stop after a stressed vowel e.g. /áʔp/ - "rise" (from English "up". In other cases (e.g. when there was a voiced stop after a stressed vowel) the syllable received harsh phonation.

====Consonants====

Aspiration on stops became phonemic. Stops were generally unaspirated, except word initial stops and stops beginning a stressed syllable. However, if the stop was preceded by an obstruent (usually /s/), it became unaspirated and the /s/ was deleted e.g. /tóʔp/ - "stop" vs. /tʰáp/ - "above" (from English "top")

/s/ acquired phonemic aspiration in similar situations to the above e.g. /sʰḛ̂n/ - "heaven" (from English "ascend") vs. /èˈsóʔt/ - "exotic.DESC" (the /k/ that used to precede the /s/ was lost, but not before blocking that aspiration that would otherwise have occurred on an /s/ in a stressed syllable).

Clusters of /s/ + Nasal became voiceless nasals e.g. /n̥âṵ/ - "snow", /m̥òṵ/ - "small.RESTR".

/sl/ clusters usually became /hl/, which is pronounced [ɬ] e.g. /hléʔp/ - "slap".

/ʍ/ re-emerged in English via spelling pronunciations, becoming Kämpya /hw/ (pronounced as [ʍ]) e.g. /ˈhwâ̰jù/ - "whale".

English /f/ became /pʰ/ e.g. /ˈpʰḛ̂jòd/ - "fjord". The exception to this was when it was clustered with /r/, in which case it became /hw/ e.g. /hwám/ - "from".

English /v/ became /bw/ before a vowel e.g. /ˈbwḛ̂lì/ - "valley"

English /r/ became either /w/ or /zw/ e.g. /ˈzwéʔp/ - quickly (from English "rapid"), /ˈpʰò̤wè/ - "forest".

English post-alveolar consonants became sequences of alveolar consonants + /j/ e.g. /báˈsjàʔp/- "beat" (from English "bash up")

Lenition occurred of obstruents after historically long vowels (Kämpya lost its phonemic vowel length contrast). Labial obstruents lenited to /ⱱ/ e.g. /pʰ/, /p/, /b/, /f/, /v/ became /ⱱ/ e.g. /là̤ⱱ/ - "laugh". Coronal consonants lenited to /ɾ/ e.g. /déˈpʰâ̰ɾ/ - "depart". Dorsal consonants lenited to /h/ e.g. /dáʔh/ - "darkness".

Clusters of Stop + /t/ were simplified to /t/ e.g. /ˈdóʔtà/ - "doctor".

Except in the above cases, /t/ was lost word finally e.g. /béiʔ/ - "belt".

Word final alveolar affricates became stops e.g. /ˈbwḭ̂d/ - "bridge".

/l/ was lost when not before a vowel e.g. /tʰṵ̂/ - "tool".

A version of [https://en.wikipedia.org/wiki/Grassmann%27s_law Grassman's Law] occured, deaspirating all but the last occurence of an aspirated consonant in a phonological word e.g. /káukákʰôṵlà/ - "Coca-Cola", not */kʰáukákʰôṵlà/. Voiceless sonorants and /h/ count as aspirated consonants.

====Vowels====

The vowel system derives from Australian English, although in many cases spelling pronunciations are used rather than the actual Australian English pronunciation. Since colonisation of Antarctica, it has undergone a number of sound changes:

# The vowel in TRAP and MARRY underwent a three-way split. In untressed syllables, it became /a/. In stressed syllables, it underwent the bad-lad split, with the long version becoming /ai/ and causing lenition of the following consonant e.g. /bàḭɾ/ - "bad.REST", while the short version became /e/ e.g. /éʔ/ - "to be at".
# The vowel in BATH, PALM and START (Australian English is non-rhotic) became /a/ in unstressed syllables, and /aː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /a/ (Kämpya lost phonemic vowel length) e.g. /déˈpʰâ̰ɾ/ - "depart".
# The vowel in NURSE merged into the above vowel e.g. /nà̤ɾ/ - "nurse".
# The vowel in LOT, CLOTH and HOT became /a/ in unstressed syllables, and /o/ in stressed syllables e.g. /ˈóʔpà/ - "opposite". However, before /l/ it became /ou/ e.g. /bôṵ/ - "bowl".
# The vowel in THOUGHT and NORTH became /o/ in unstressed syllables, and /oː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /o/ e.g. /pʰò̰/ - "four". However, before /u/ it became /ou/ e.g. /m̥óṵ/ - "small.DESC".
# The vowel in KIT became /i/ e.g. /ˈbíʔ/ - "bite" (from English "bit"). However, before /l/ it became /ju/ e.g. /hjṵ̂/ - "hill".
# The vowel in HEAT usually became /ei/ e.g. /déiʔp/ - "deeply". However, before /j/ it became /e/ e.g. /ˈpʰḛ̂jòd/ - "fjord". And before /l/, it became /i/ e.g. /pʰḭ̂lìŋ/ - "display of emotions" (from English "feeling").
# The vowel in DRESS and MERRY normally became /e/ e.g. /ˈḛ̂nèm/ - "enemy". However, before /l/ it became /ei/ e.g. /béiʔ/ - "belt".
# The vowel in SQUARE and MARY became /e/ in unstressed syllables, and /eː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /e/ e.g. /kʰḛ̂/ - "hospital patient" (from English "care")
# The vowel in STRUT normally became /a/ e.g. /wà̰n/ - "one". However, before /l/, it became /au/ e.g. /kâṵ/ - "skull".
# The vowel is FOOT became /u/ e.g. /pʰúʔ/ - "foot".
# The vowel in GOOSE became /u/ in unstressed syllables, and usually became /ei/ in stressed syllables e.g. /gèi̤/ - "goose", /tʰêḭ/ - "two". However, before another vowel, it became /e/ e.g. /sʰḛ̂wà/ - "sewer". In stressed syllables before /l/, it became /u/ e.g. /tʰṵ̂/ - "tool"
# The vowel in FACE normally became /ai/ e.g. /nâḭm/ - "name". However, before /l/ it became /aju/ e.g. /ˈhwâ̰jù/ - "whale".
# The vowel in PRICE became /ou/ in stressed syllables, and /ai/ in unstressed syllables e.g. /pʰwòṳ/ - "price", /ˈḭ̂mwài/ - "invitation" (from an initial stressed derived form of "invite"). However, before /l/ it became /oju/ in stressed syllables, and /aju/ elsewhere e.g. /tʰô̰jù/ - "tile"
# The vowel in CHOICE became /ou/ e.g. /tʰjòi̤/ - "choice", except before /l/ when it became /oju/ e.g. /ˈbô̰jù/ - "hot spring" (from English "boil").
# The vowel in GOAT usually became /au/ e.g. /n̥âṵ/ - "snow". However, in a stressed syllable before /l/ it became /ou/ e.g. /kʰôṵlà/ - "(Coca-)Cola".
# The vowel in MOUTH became /ei/ in a stressed syllable, and /au/ in an unstressed syllable e.g. /mèi̤/ - "mouth".
# The vowel in NEAR became /i/ in unstressed syllables, and /iː/ in stressed syllables (causing lenition of the following consonant). Then these both merged into /i/ e.g. /pʰì̤ɾ/ - "fierce".
# As a very general rule, schwa became /a/, however in a lot of cases it developed into another vowel, being influenced by spelling e.g. /éˈléʔt/ - "to vote" (from English "elect").

==Pronouns==

Pronouns are not marked for singular or plural, but there are two words for "we" depending on whether the listener is included [http://en.wikipedia.org/wiki/Clusivity]. The pronouns that do not include the listener are also used to only talk about the speaker. Another way of looking at this is that Kämpya makes no distinction between "us not including you" and "me", but uses different forms for "us including you".

There is also a set of interrogative pronouns that can mean "who" or "what" depending on the context (Kämpya conflates the two). By adding postpostions, words meaning "how" and "why" can be formed.

Kämpya has a tripartite case marking system on pronouns [http://en.wikipedia.org/wiki/Tripartite_language]. However, the ergative pronouns are identical to the (alienable) genitive pronouns.

In each case, there is a different form of the word depending on whether they are immediately followed by a word beginning with a vowel. There is also an "emphatic" form, which is used to emphasise the subject. The emphatic form does not depend on whether the following word starts with a vowel or not. Each form is listed below, with the normal form listed first, the form before a vowel second, and the emphatic form last:

===Intransitive Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! Interrogative
|-
| Before a consonant || ái || wéi || jéi || bá
|-
| Before a vowel || áj- || wéj- || jéw- || báj-
|-
| Emphatic || òṵ || wèḭ || jèḭ || bà̰
|}

As you can see above, the emphatic forms all have Low Tone with Harsh Phonation (i.e. belong to Tone Class 2), while all the other forms have High Tone with no phonation (i.e. modal phonation).

===Accusative Pronouns===

{| class="wikitable"
|-
! !! 1st Person Exclusive !! 1st Person Inclusive !! 2nd Person !! 3PS !! Reflexive !! Interrogative
|-
| Before a consonant || mí || lé || té || swí || sé || ké
|-
| Before a vowel || mj- || lés- || ét- || síw- || és- || ék-
|-
| Emphatic || mḭ̀ || lè̤ || tḛ̀ || swḭ̀ || sḛ̀ || kḛ̀
|}

The emphatic forms are also used with postpositions e.g. /mḭ̀=hwám/ - "from me".

===Genitive / Ergative Pronouns===

Kämpya's genitive pronouns inflect for alienable vs. inalienable possession, however, in the emphatic form, these are not distinguished. The alienable (and emphatic) forms of the genitive pronouns are also used as ergative pronouns (i.e. the subjects of transitive verbs).

There is also a proximate vs. obviative distinction on 3rd person genitive pronouns. The proximate pronouns are used when the possessor was the last noun that was in the absolutive case (i.e. unmarked). The obviative pronouns are used in other cases. To mark a 3rd person subject of a transitive verb, only the obviative form is used, never the proximate.

{| class="wikitable"
|-
! !! 1PS Excl. !! 1PS Incl. !! 2PS !! 3PS Prox. !! 3PS Obv. !! Interrogative
|-
| Before a consonant (Inalienable) || mái || áu || jó || ðé || ðí || héi
|-
| Before a vowel (Inalienable) || máj- || áw- || józw- || ðézw- || ðj- || héw-
|-
| Alienable / Ergative || môṵ || ˈḛ̂wà || jô̰ || ðḛ̂- || ðjíʔ || hèi̤
|-
| Emphatic || mòṵ || ˈḛ̀wá || bwò̤ || swḭ̀ || ðjìʔ || hèi̤
|}

==Basic Syntax==

Kämpya has topic comment syntax [http://en.wikipedia.org/wiki/Topic-prominent_language], and is also syntactically ergative [http://en.wikipedia.org/wiki/Syntactic_ergativity#Syntactic_ergativity] and secundative [http://en.wikipedia.org/wiki/Secundative_language].

===Copula===

Kämpya has no verb meaning "to be" (something) [http://en.wikipedia.org/wiki/Zero_copula]. Instead, the two words are simply placed side by side in the sentence. e.g.

/ˈsʰíʔtà ˈḛ̂nèm/

guardian enemy

The guardian is / was / will be the enemy.

/jéi ˈsʰíʔtà/ phonetically [jéiˈsʰíʔtà]

2PS.INTR guardian

You are / were / will be the guardian(s).

Pronouns take different forms if they are followed by a vowel e.g.

/jéw ˈḛ̂nèm/ phonetically [jéˈwḛ̂ːnèⁿ]

2PS.INTR enemy

You are / were / will be the enemy.

They also take special emphatic forms e.g.

/jèḭ ˈḛ̂nèm/ phonetically [jèḭ ˈḛ̂ːnèⁿ]

2PS.INT.EMP enemy

'''You''' are the enemy.

===Intransitive Sentences===

Intransitive sentences generally have free word order. If the verb is in focus [http://en.wikipedia.org/wiki/Focus_(linguistics)], then it will come after the subject. If the subject is in focus, then it will come after the verb. So both of the following are grammatical:

/dô̰k áˈlâṵn/

dog alone

The dog is / was / will be '''alone'''.

as well as

/áˈlâṵn dô̰k/

alone dog

The '''dog''' is / was / will be alone.

====Constituent Deletion====

However, the arguments of intransitive sentences can be freely deleted, so it is perfectly possible to have one word sentences e.g.

/áˈlâṵn/

alone

He / She / It / They is / are / was / were / will be alone.

====With Postpositional Phrases====

If we include a postpositional phrase in the sentence, it must come immediately before the verb (except for any pronouns) e.g. with the postpositional phrase /ˈbwḛ̀lí=hwám/ - "from the valley", the verb /déˈpʰâ̰ɾ/ - "to set off" and the noun /ˈḛ̂nèm/ - "enemy", we can say:

/ˈbwḛ̀lí=hwám déˈpʰáʔɾ ˈḛ̂nèm/

valley=from depart enemy

The enemy departs from the valley.

or

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departs from the valley.

To say "A is located at / in / on etc. B", Kämpya uses the verb /éʔ/ and the locative clitic /ká/ e.g.

/ˈḛ̂nèm ˈbwḛ̀lí=ká éʔ/

enemy valley=LOC be.at

The enemy is at the valley.

Note that /éʔ/ cannot be used as a copula (to say something is something else). As discussed previously, in such a case, no verb is used at all.

====With Adverbs====

Unlike adverbs of place and other postpositional phrases which precede the verb, adverbs of manner and time follow it (and are placed in Tone Class 3) e.g.

/déˈpʰáʔɾ zwéʔp ˈjě̤tài ˈḛ̂nèm/

depart fast yesterday enemy

The enemy departed quickly yesterday.

====With Pronouns====

The same subject pronouns as before are also used for intransitive sentences e.g.

/jéi déˈpʰáʔɾ/

2PS depart

You are setting off.

However, the pronoun must immediately precede the verb so */déˈpʰâ̰ɾ jéi/ is ungrammatical. And postpositional phrases cannot intervene between the pronoun and the verb, so */jéi ˈkʰóʔnàt=hwám déˈpʰâ̰ɾ/ is ungrammatical. They must come before the pronoun, so it is only grammatical to say:

/ˈbwḛ̀lí=hwám jéi déˈpʰáʔɾ/

valley=from 2PS depart

You are setting off from the valley.

===Transitive Sentences===

In transitive sentences without pronouns, the basic word order is SVO, with the subject marked with the ergative clitic /-zu/ e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, immediately after a monophthong with breathy or harsh voice (and thus necessarily a stressed vowel), the ergative clitic has the allomorph /-ɾu/ e.g.

/kʰjâṵ=ɾù ˈbíʔ ˈlḭ̂zàd/

cow=ERG bite lizard

It was the cow that bit the lizard.

====With Postpositional Phrases====

The come after the ergative noun, but before the verb (or any pronouns) e.g.

/dô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

====Topicalisation====

However, Kämpya speakers very often topicalise either the subject or the object. The object is topicalised by moving it in front of the subject (i.e. making the sentence OSV) e.g.

/ˈlḭ̂zàd dô̰k=zù ˈbíʔ/

lizard dog=ERG bite

The lizard was bitten by the dog.

The subject of a transitive sentence is topicalised by deleting the ergative marker e.g.

/dô̰k ˈbíʔ ˈlḭ̂zàd/

dog bite lizard

The dog bit the lizard.

The difference between this and /dô̰k=zù ˈbíʔ lḭ̂zàd/ (i.e. with the case marker), is that, in the sentence with the case marker, the "new information" being presented to the listener is that it was the dog that did the biting. Without the case marker, it is a sentence describing the dog, and the new information is that it bit the lizard. This is analagous to the difference between "ga" and "wa" in Japanese.

Also note that topicalising both the subject and object is ungrammatical i.e. we cannot say */lḭ̂zàd dô̰k ˈbíʔ/ or anything like that.

====With Pronouns====

The (alienable) genitive pronouns discussed earlier can also be the subject of transitive verbs e.g.

/jô̰ ˈbíʔ ˈlḭ̂zàd/

2PS.ERG bite lizard

You are biting the lizard.

As before, the object can be fronted as a topic e.g.

/ˈlḭ̂zàd jô̰ ˈbíʔ/

lizard 2PS.ERG bite

You are biting the lizard.

However, the subject pronoun still come immediately before the verb, so sentences like */jô̰ gáʔɾ káiʔk/ are ungrammatical.

Prepositional phrases must come before the subject pronoun e.g.

/pʰò̤wé=ká jô̰ ˈbíʔ ˈlḭ̂zàd/

forest=LOC 2PS.ERG bite lizard

You bit the lizard in the forest.

When a verb takes an accusative pronoun as its object, it becomes intransitive i.e. the subject will be unmarked (in the absolutive case) and can come either before or after the verb e.g.

/dô̰k té=ˈbíʔ/

dog 2PS.ACC=bite

The dog will bite you.

or

/té=ˈbíʔ dô̰k/

2PS.ACC=bite dog

You will be bitten by the dog.

Using the ergative form /dô̰k=zù/ is ungrammatical here.

When both the subject and the object of a verb are pronouns, the subject comes first and is in the intransitive case e.g.

/jéi=swí=ˈbíʔ/

2PS.INTR=3PS.ACC=bite

You are biting him.

It would be ungrammatical to say */jô̰ swí=bíʔ/.

====Reciprocal Voice====

This takes a transitive verb and turns it into an intransitive verb meaning "do ... to each other / one another". It is formed by reduplicating the first syllable of the verb, and putting it as a particle in the object pronoun "slot" e.g. /ˈbíʔ/ - "to bite" -> /bíˈbíʔ/ - "to bite each other".
However coda consonants and the second elements of diphthongs are deleted e.g. /péiʔk/ - "to speak" -> /péiˈpéiʔk/ - "to speak to each other".

Also if the onset of the first syllable contains an aspirated consonant, the aspiration is lost in the reduplication e.g. /pʰṵ̂/ - "to pull" -> /púˈpʰṵ̂/ - "to pull each other".

If the onset of the first syllable contains a voiceless nasal, the it becomes voiced in the reduplication e.g. - /ˈm̥ô̰jù/ "to entertain" -> /móˈm̥ô̰jù/ - "to entertain each other". Likewise, if the onset of the first syllable contains /h/, it is lost in the reduplication e.g. /hléʔp/ - "to slap" -> /léˈhléʔp/ - "to slap each other".

====Subject Deletion====

It is perfectly acceptable to delete the subject of a transitive sentence e.g.

/ˈbíʔ ˈlḭ̂zàd/

bite lizard

He / She / It / They bit the lizard.

/ˈlḭ̂zàd ˈbíʔ/

lizard bite

The lizard was bitten.

This carries no ambiguity, because the word for lizard is always in the absolutive case.

====Antipassive Voice====

To delete the object of a transitive sentence, the antipassive voice is used [http://en.wikipedia.org/wiki/Antipassive_voice]. This is done with the clitic /θu-/. It goes in the same syntactic "slot" as an object pronoun would e.g.

/dô̰k θú=ˈbíʔ/

dog ANTIP=bite

The dog bit (someone / something).

However, if followed by a vowel, the form of the clitic is /θw-/ e.g.

/dô̰k θw=álâṵn/

dog ANTIP=abandon

The dog abandoned (someone / something).

The deleted object can be re-introduced at the end of the sentence using the dative clitic /-àuŋ/ (note that this does not change the tone class of the noun) e.g.

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/

dog ANTIP=bite lizard=DAT

The dog bit the lizard.

At first glance, this may seem pointless, as we could have quite easily have said:

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd/

dog=ERG bite lizard

It was the dog that bit the lizard.

However, the difference is that Kämpya has ergative syntax. Whatever argument of the verb is in the absolutive case is the syntactic pivot [http://en.wikipedia.org/wiki/Syntactic_pivot]. In a normal transitive sentence, this is the object of the verb. But, by using the antipassive voice, the subject of the transitive verb becomes the syntactic pivot. If we combine the two sentences above with the verb /áˈwâḭ/ - "to flee", the meaning becomes very different:

/dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ áˈwâḭ/

dog ANTIP=bite lizard=DAT flee

The dog bit the lizard and (the dog) ran away.

vs.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd áˈwâḭ/

dog=ERG bite lizard flee

The dog bit the lizard and the lizard ran away.

Note that the dative clitic has the allomorph /jàuŋ/ after a vowel e.g.

/ˈlḭ̂zàd θú=ˈbíʔ kʰjâṵ=jàuŋ/

lizard ANTIP=bite cow=DAT

The lizard bit the cow.

===Ditransitive Sentences===

For verbs such as "give", "sell", "send" etc. [http://en.wikipedia.org/wiki/Ditransitive], the normal situation is to have the donor marked in the ergative case in the usual position (i.e before the verb and any postpositional phrases), the theme (whatever is being given / sold etc. to someone) directly after it, and marked with the secundative postposition /-ta/, and the recipient in the absolutive case either after the verb or topicalised at the beginning of the sentence e.g.

/ˈsʰíʔtà=zù bàṵn=tá gḭ̂p dô̰k/

guardian=ERG bone=SEC give dog

It was the guardian that gave the bone to the dog.

or

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

The ergative argument can be topicalised in the same way e.g.

/ˈsʰíʔtà tí=bàṵn=tá gḭ̂p dô̰k/

guardian INDEF=bone=SEC give dog

The guardian gave the bone to the dog.

It is also perfectly possible to put a ditransitive sentence in the antipassive voice e.g.

/ˈsʰíʔtà bàṵn=tá θú=gḭ̂p dô̰k=àuŋ/

guardian bone=SEC ANTIP=give dog=DAT

The guardian gave the bone to the dog.

===With Postpositional Phrases===

Postpositional phrases usually come after the theme (i.e. whatever takes the secundative case) e.g.

/ˈsʰíʔtà=zù bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

guardian=ERG bone=SEC forest=LOC give dog

It was the guardian that gave the bone to the dog in the forest.

====Allomorphy====

=====Clitics beginning with /t/ and /p/=====

Immediately after a monophthong with breathy voice (and thus necessarily a vowel in an open syllable that has stress), an inital /t/ in clitics lenites to /ɾ/. For example, the secundative clitic /ta/ becomes /ɾa/ e.g.

/ˈsʰíʔtà gwà̤=ɾá gḭ̂p dô̰k/

guardian grass=SEC give dog

The guardian gave the grass to the dog.

Immediately after a monophthong with harsh voice, /t/ also lenites to /ɾ/. However, it also triggers a phonation shift on the vowel from harsh to glottalised e.g. the noun meaning "snare" is /n̥ḛ̀/ in Tone Class 2. It normally has harsh voice, but it combines with the secundative clitic to form /n̥èʔ=ɾá/, as in:

/ˈsʰíʔtà n̥ḛ̀=ɾá θú=gḭ̂p dô̰k=àuŋ/

guardian snare=SEC ANTIP=give dog=DAT

Literally "the guardian gave the snare to the dog", but "give a snare" could also be translated as "use a snare to catch".

In similar situations, /p/ lenites to /ⱱ/. For example, the instrumental postposition /piŋ/ likewise becomes /ⱱiŋ/ e.g. /pèiʔk/ - "words" becomes /ˈpèiʔk=píŋ/ - "using words", but /gwà̤/ - "grass" becomes /gwà̤=ⱱíŋ/ - "using grass", and words with harsh voice on the final vowel such as /n̥ḛ̀/ - "snare", become /n̥èʔ=ⱱíŋ/ - "using a snare".

=====Clitics beginning with /d/, /z/ and /b/=====

Immediately after a monophthong with harsh or breathy voice, /d/, /z/ also lenite to /ɾ/ (as we have seen with the ergative clitic /-zu/). /b/ also lenites to /ⱱ/. But they do not trigger any phonation changes on the vowel.

=====Clitics beginning with /g/=====

Immediately after a monophthong with harsh or breathy voice, /g/ is lenited to /h/. However this triggers deaspiration in the word it attaches to (in a similar fashion to Grassman's law in Indo-European languages). Aspirated consonants lose their aspiration, /h/ is deleted, and voiceless nasals become voiced e.g. with the postposition /-gei/ - "since the time of"

{| class="wikitable"
|-
! Gloss !! Base Form !! Form with /-gei/
|-
| class || kʰlà̤ || klà̤=héi
|-
| snare || n̥ḛ̀ || nḛ̀=héi
|-
| mother || múˈhḛ̀ || múˈḛ̀=héi (more commonly /ˈmwḛ̀=héi/)
|}

=====Clitics beginning with /k/=====

The /k/ here lenits to /h/ and triggers deaspiration in exactly the same way as /g/. However, if the monophthong had harsh voice, it changes to be glottalised e.g. from /n̥ḛ̀/ - "snare", if we add the locative clitic /-ka/, the result is /nèʔ=há/ - "at the snare".

====With Pronouns====

=====Ergative Pronouns=====

These can come in their usual position immediately before the verb e.g.

/bàṵn=tá pʰò̤wé=ká jô̰ gḭ̂p dô̰k/

bone=SEC forest=LOC 2PS.ERG give dog

You gave the bone to the dog in the forest.

Or they can come before the noun in the secundative case e.g.

/jô̰ bàṵn=tá pʰò̤wé=ká gḭ̂p dô̰k/

2PS.ERG bone=SEC forest=LOC give dog

You gave the bone to the dog in the forest.

This creates a small amount of amiguity, since /jô̰ bàṵn=tá/ can also be read as "your bone (alienable)". Thus the above sentence could also be read as "your bone (alienable) was given to the dog in the forest".

=====With Postpositions=====

The emphatic forms of the accusative pronouns are used with a postposition (e.g. the secundative /tá/). However, these pronouns trigger the same sandhi rules discussed before (changing the phonation on the vowel to glottalised), and leniting the postposition to. Here is a list of the pronouns when used with the secundative postposition:

1st Person Exclusive - /mìʔ=ɾá/

1st Person Inclusive - /lè̤=ɾá/

2nd Person - /tèʔ=ɾá/

3rd Person - /swìʔ=ɾá/

Reflexive - /sèʔ=ɾá/

===Applicative Voice===

To topicalise a noun in a postpositional phrase, Kämpya uses applicative constructions [http://en.wikipedia.org/wiki/Applicative_voice]. These are formed by first shifting the noun in the absolutive case to the end of the sentence (and marking it with the dative clitic /-àuŋ/). In turn, the noun that the postposition was attached to goes into the absolutive case (and changes to Tone Class 1), and the postposition is placed after the verb as a clitic e.g. from the sentence:

/ˈdô̰k=zù pʰò̤wé=ká ˈbíʔ ˈlḭ̂zàd/

dog=ERG forest=LOC bite lizard

It was the dog that bit the lizard in the forest.

we can apply an applicative transformation to get:

/pʰò̤wè ˈdô̰k=zù ˈbíʔ=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite=LOC lizard=DAT

In the forest, the dog bit the lizard.

As another example, the sentence

/ˈḛ̂nèm ˈbwḛ̀lí=hwám déˈpʰáʔɾ/

enemy valley=from depart

The enemy departed from the valley.

becomes the following:

/ˈbwḛ̂lì déˈpáʔɾ=hwàm ˈḛ̂nèm=àuŋ/

valley depart=from enemy=DAT

From the valley, the enemy departed.

Note here the tone change from /ˈkʰòʔnát/ (in a postpositional phrase and thus Tone Class 2) to /ˈkʰóʔnàt/ (in Tone Class 1 like most nouns). The same thing happened with the word for "forest" changing from /pʰò̤wé/ to /pʰò̤wè/.

Likewise, in ditransitive sentences, the theme (i.e. whatever is given by the donor to the recipient) can also be topicalised by the same process e.g.

/dô̰k ˈsʰíʔtà=zù tí=bàṵn=tá gḭ̂p/

dog guardian=ERG INDEF=bone=SEC give

The dog was given a bone by the guardian.

becomes

/bâṵn ˈsʰíʔtà=zù ˈgḭ̂p=tá ˈdô̰k=àuŋ/

bone guardian=ERG give=SEC dog=DAT

The bone was given to the dog by the guardian.

When a pronoun is the object of the verb to be put into the applicative voice, the subject is put into the dative case instead e.g. when we apply the applicative voice to

/dô̰k pʰò̤wé=ká té=ˈbíʔ/

dog forest=LOC 2PS.ACC=bite

The dog bit you in the forest.

we get

/pʰò̤wè té=ˈbíʔ=kà dô̰k=àuŋ/

forest 2PS.ACC=bite=LOC dog=DAT

In the forest, the dog bit you.

Notice that when the postpositions /hwam/, /ka/ and /ti/ attach to a noun, they have High Tone, since postpositional phrases are in Tone Class 2. However, when they attach to a verb, they have has Low Tone, since verbs are in Tone Class 1.

This is clearly a cliticisation process, since adverbs can come before the particle but after the verb e.g.

/pʰò̤wè ˈdô̰k=zù bíʔ ˈˈjě̤tài=kà ˈlḭ̂zàd=àuŋ/

forest dog=ERG bite yesterday.ADV=LOC lizard=DAT

In the forest, the dog bit the lizard.

==Noun Phrases==

The basic order in noun phrases is: Demonstrative - Numeral - Restrictive Adjective - Noun - Non Restrictive Adjective

===Restrictiveness in Adjectives===

When an adjective modifies a noun, Kämpya makes a distinction between restrictive and non-restrictive (descriptive) adjectives [http://en.wikipedia.org/wiki/Restrictiveness]. It does this by putting restrictive adjectives in Tone Class 2, and placing them before the noun they modify e.g.

/ˈzwèʔp dô̰k/

fast.REST dog

The fast dog(s) (choosing one or more fast dogs out of a group of other dogs).

Descriptive (non-restrictive) adjectives are in Tone Class 3 and follow the noun they modify e.g.

/dô̰k ˈzwéʔp/

dog fast.DESC

The fast dog(s).

or

/sôṵlà àˈwá̰/

sun yellow.DESC

The yellow sun.

In the last case, it would never make any sense to say /áˈwà̰ sôṵlà/, unless for some reason we were disambiguating between multiple suns.

===Articles===

There are no definite articles, but there is an indefinite article clitic. It occupies the demonstrative syntactic "slot" and has the allomorphs /tí-/ before a consonant and /ít-/ before a vowel e.g.

/tí=dô̰k/

INDEF=dog

a dog

/ít=ˈḛ̂nèm/

INDEF=enemy

an enemy

Kämpya nouns are not inflected for singular or plural, but when they take the indefinite article, they are only ever singular, so the above examples could never mean "some dogs" or "some enemies".

It is important to distinguish the indefinite article from the word for one /tìʔ/. The indefinite article is a clitic, which has no stress and is phonologically part of whatever word follows it i.e. it will have plain High Pitch no matter whether it is followed by a noun or (restrictive) adjective. On the other hand, the numeral /tìʔ/ - "one" is phonologically a separate word. In terms of the tonal morphology, it (like all other numerals) belongs to category 2, and is thus pronounced with Low Pitch.

Compare:

/tí=dô̰k/

INDEF=dog

a dog

with

/tìʔ dô̰k/

one dog

one dog

The numeral for "one" (or any other numerals) cannot be used with the indefinite article, so */tí=tìʔ dô̰k/ is ungrammatical.

However, (restrictive) adjectives can intervene between the indefinite article and the noun e.g.

/ít=áˈwà̰ dô̰k/

INDEF=yellow.REST dog

a yellow dog

===Demonstratives===

Kämpya's system of demonstratives is more complex than English. Unlike English which only makes a two-way distinction between "this" and "that", Kämpya makes a five-way contrast between "this" (on the same level as the speaker), "this" (above / uphill from the speaker), "this" (below / downhill from the speaker), "that" (far from the speaker but still visible) and "that" (invisible to the speaker). The distinction between uphill and downhill objects is particularly relevant for most Kämpya speakers, since they mostly live along a mountainous coastline. Dialects spoken in flatter areas tend to simplify the system.

It also uses tone to distinguish pronominal demonstratives (e.g. in the sentence "''This'' is a cat") from adnominal demonstratives (e.g. in the sentence "''This'' cat is here"). It also uses tone to make a further distinction in adnominal demonstratives depending on whether they are describing a place (in which case they are clitics), or something else.

For example "this mountain" is /dá=ˈméiʔnàn/, since a mountain is a place. But "this dog" is /dà̰ dô̰k/, since a dog is not a place.

Here is a table of the demonstratives:

{| class="wikitable"
|-
! !! This (same level) !! This (higher / uphill) !! This (below / downhill) !! That (visible) !! That (invisible)
|-
| Pronominal || dâ̰ || tʰéʔk || áuʔk || hôṵ || hlà̤n
|-
| Adnominal (Places) || dá || *t(ʰ)ék || áuk || *(h)óu || *hlàn
|-
| Adnominal (Other cases) || dà̰ || tʰèʔk || àuʔk || hòṵ || hlà̤n
|}

*/hóu-/ has the allomorph /*hów-/ before a vowel. Before a word containing an aspirated consonant, a voiceless nasal or another /h/, the initial /h/ is dropped e.g. /óu=pʰò̤wè/ - "that forest", not */hóu=pʰò̤wè/. Under similar conditions, the initial /h/ in /hlàn/ is lost, as well as the aspiration in /t(ʰ)ék/.

===Possession===

Kämpya distinguishes alienable and inalienable possession. In both cases, possessors are marked with the cliticised case marker /ja/, and come before the noun they possess. But inalienable possessors are in Tone Class 2 e.g.

/dò̰k=jí bâṵn/

dog=GEN.INALIENABLE bone

The dog's bone (i.e. in it's body).

While alienable possessors are in Tone Class 1 e.g.

/dô̰k=jì bâṵn/

dog=GEN.ALIENABLE bone

The dog's bone (i.e. that it is chewing on / has buried etc.)

Unlike in English, there are no restrictions on possessing a noun modified by a demonstrative or an article. So it is perfectly grammatical to say the following:

/dô̰k=jì dà̰ bâṵn/

dog=GEN.ALIENABLE this bone

This bone of the dog's (literally "the dog's this bone").

===Relative Clauses===

In the same way as with adjectives, Kämpya places relative clauses before the head noun if they are restrictive, and after the noun if they are non-restrictive e.g.

/ˈlḭ̂zàd=zù bíʔ dô̰k áˈwâḭ/

lizard=ERG bite dog flee

The dog that the lizard had bitten fled.

Or, using a non-restrictive relative clause:

/dô̰k ˈlḭ̂zàd=zù bíʔ áˈwâḭ/

dog lizard=ERG bite flee

The dog, which the lizard had bitten, fled.

However, in Kämpya and unlike in English, the accessibility hierarchy is very important [http://en.wikipedia.org/wiki/Relative_clause#Accessibility_hierarchy]. In every case, the head noun must be the absolutive argument of the verb in the relative clause. If it would not normally be so, then it must be put into the absolutive case by strategies such as antipassivisation on the verb e.g.

/θú=bíʔ ˈlḭ̂zàd áˈwâḭ/

ANTIP=bite lizard flee

The lizard that had bitten it fled.

Or in a non-restrictive relative clause:

/ˈlḭ̂zàd θú=bíʔ áˈwâḭ/

lizard ANTIP=bite flee

The lizard, which had bitten it, fled.

Or an applicative construction can be used e.g.

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè m̥ôṵ/

lizard=ERG bite=LOC forest small

The forest that the lizard bit it in is small.

And in a non-restrictive relative clause:

/pʰò̤wè ˈlḭ̂zàd=zù bíʔ=kà m̥ôṵ/

forest lizard=ERG bite=LOC small

The forest, which the lizard bit it in, is small.

====Internal Heads====

It must be noted that Kämpya restrictive relative clauses are strictly speaking internally headed [http://wals.info/chapter/90]. We can only see this in antipassives and applicatives where the noun (that used to be in the absolutive case) is re-introduced as a dative argument, it follows the head noun e.g.

/θú=bíʔ ˈlḭ̂zàd dô̰k=àuŋ áˈwâḭ/

ANTIP=bite lizard dog=DAT flee

The lizard that had bitten the dog fled. (not */θú=bíʔ dô̰g=àuŋ ˈlḭ̂zàd áˈwâḭ/)

Or in an applicative construction:

/ˈlḭ̂zàd=zù bíʔ=kà pʰò̤wè dô̰k=àuŋ m̥ôṵ/

lizard=ERG bite=LOC forest dog=DAT small

The forest that the lizard bit the dog in is small. (not */ˈlḭ̂zàd=zù bíʔ=kà dô̰g=àuŋ pʰò̤wè m̥ôṵ/)

==Mood/Evidentiality==

While Kämpya does not mark tense or aspect, mood / evidentiality is very important (the grammar conflates the two). They are marked with a proclitic, that usually comes immediately before the verb, but can occur in many other positions.

===Negative Mood===

The basic way to negate something is to place the clitic /na/ immediately before it e.g. from the sentence

/té=bíʔ dô̰k/

2PS.ACC=bite dog.

You were bitten by the dog.

We can say

/té=ná=bíʔ dô̰k/

2PS.ACC=NEG=bite dog

You weren't bitten by the dog.

as well as

/ná=té=bíʔ dô̰k/

NEG=2PS.ACC=bite dog

'''You''' weren't bitten by the dog. / It's not you that was bitten by the dog.

and

/té=bíʔ ná=dô̰k/

2PS.ACC=bite NEG=dog

You weren't bitten by '''the dog'''. / It wasn't the dog that bit you.

However, when a noun is topicalised, it cannot be attached to /na/. So */ná́=dô̰g té=bíʔ/ is ungrammatical.

As we can see, /na/ can attach to either nouns or verbs. It can also attach to adjectives e.g.

/té=bíʔ ná=m̥ôṵ dô̰k/

2PS.ACC=bite NEG=small.REST dog

You weren't bitten by the '''small''' dog (i.e. you were bitten by the big one)

And adverbs e.g.

/té=bíʔ nà=déiʔp dô̰k/

2PS.ACC=bite NEG=deep.ADV dog

You weren't bitten '''deeply''' by the dog.

However, if the word after /na/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=náw=áˈlâṵn dô̰k/

2PS.ACC=NEG=abandon dog

You weren't abandoned by the dog.

===Necessitative Mood===

This is used to indicate that something ought to / is required to happen. It is indicated using the proclitic /ga/, or /gat/ before a vowel (from English "gotta") e.g.

/ái=gá=déˈpʰáʔɾ/

1PS.EXCL=NEC=depart

I / we (not including you) need to depart.

When attached to a 1st person inclusive pronoun, it often functions similarly to English "let's" e.g.

/wéi=gá=déˈpʰáʔɾ/

1PS.INCL=NEC=depart

Let's depart / We (including you) need to depart

When attached to a 2nd person pronoun, it often functions like an imperative e.g.

/jéi=gá=déˈpʰáʔɾ/

2PS=NEC=depart

Depart / You need to depart

Like other mood particles, /ga/ does not need to attach to a verb. It can attach to other parts of speech depending on the focus of the speaker e.g.

/jéi=déˈpʰáʔɾ gà=ˈzwéʔp/

2PS=depart NEC=fast

You need to depart quickly (in this case, it is already assumed that the listener is departing, and the speaker wishes to emphasise that it should happen quickly).

====Prohibitive Forms====

Kämpya has a special pattern to say that something is forbidden. This is to use the necessitative mood, and also change the verb from Tone Class 1 (the normal class for verbs) to Tone Class 2 (for infinitives). It is then followed by the verb /nâ̰/ e.g.

/jéi=gát=áˈlàṵn nâ̰/

2PS.INTR=NEC=alone.INF PROH

Don't be alone.

This pattern can also be used for transitive verbs

/jô̰ gá=bìʔ nâ̰ dô̰k/

2PS.ERG NEC=bite.INF PROH dog

Don't you bite the dog.

It is also perfectly possible to front the argument of /nâ̰/ e.g.

/dô̰k jô̰ gá=bìʔ nâ̰/

dog 2PS.ERG NEC=bite.INF PROH

The dog must not be bitten (by you).

Other nouns can be used apart from the 2nd person pronouns e.g.

/ˈsʰíʔtà=zù gá=bìʔ nâ̰ dô̰k/

guardian=ERG NEC=bite.INF PROH dog

The guardian must not bite the dog.

===Optative Mood===

This is used for things that the speaker hopes will happen / have happened. It is formed with the proclitic /wana/ (or /wan/ before a vowel) e.g.

/ái=wáná=déˈpʰáʔɾ/

1PS.EXCL.INTR=OPT=depart

I / we (not including you) want to depart.

/té=wáná=bíʔ dô̰k/

2PS.ACC=OPT=bite dog

I want you to be bitten by the dog.

/té=bíʔ wáná=dô̰k/

2PS.ACC bite OPT=dog

I want you to be bitten by the '''dog''' (and not another animal).

/té=bíʔ wáná=m̥ôṵ dô̰k/

2PS.ACC bite OPT=small.REST dog

I want you to be bitten by the '''small''' dog (and not a big one).

===Conditional Mood===

This is used for situations which may not necessarily come true / have true, but are dependent on something else. It is marked with the proclitic /kau/ (or /kaw- before a vowel). The origin of this proclitic is the Thai particle /kɔ̂ː/ e.g.

/ái=káu=déˈpʰáʔɾ/

1PS.EXCL.INTR=COND=depart

I / we (not including you) would depart (if something else happens).

/ái=déˈpʰáʔɾ kò=ˈzwéʔpìd/

1PS.EXCL.INTR depart COND=fast

I/ we (not including you) would depart quickly (but unless some other event happens, it will be slow)

Sentences in the conditional mood can occur after sentences with another mood. In this case, the event described in the conditional mood only happens if the preceding sentence comes true e.g.

/jéi=wáná=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=OPT=depart COND=1PS.EXCL=alone

I want you to leave so I / we (not including you) can be alone.

===Hypothetical Mood===

The proclitic /pʰí/ (/ípʰ/ before a vowel) is used for hypothetical and counterfactual situations. It often corresponds to cases where English would use "if" (which it is indeed cognate to) e.g.

/jéi=pí=déˈpʰáʔɾ/

2PS.INTR=HYP=depart

If you depart ...

A clause in the hypothetical mood is very often followed up with a clause in the conditional mood e.g.

/jéi=pí=déˈpʰáʔɾ ái=káw=áˈlâṵn/

2PS.INTR=HYP=depart COND=1PS.EXCL=alone

If you leave, I / we (not including you) will be alone.

Notice in both of the above sentences, Grassman's law has resulted in a loss of aspiration on the prefix, so /pʰí/ becomes /pí/

===Direct Evidential===

If a speaker is reporting something that they have experienced, then no clitic is used e.g.

/dô̰k áˈlâṵn/

dog alone

The dog is alone (maybe the speaker can see it)

===Reportative Evidential===

If the speaker is reporting information that someone else told them, the proclitic /sʰái/ (/sʰáj/ before a vowel) is used. This is etymologically related to English "say" e.g.

/dô̰k sʰáj=áˈlâṵn/

dog REP=alone

I've been told that the dog is alone.

/áˈlâṵn sʰái=dô̰k/

alone REP=dog

I've been told that it is the dog that is alone.

===Inferential Evidential===

If the speaker is arriving at a judgment based on some kind of direct physical evidence, then the proclitic /ge/ is used (/ges/ before a vowel). This is etymologically related to English "guess" e.g.

/dô̰k gés=áˈlâṵn/

dog INFR=alone

The dog must be alone (maybe the speaker sees only a single set of dog footprints)

===Assumptive Mood===

This is used when the speaker is making an assertion based on their experience with similar situations, or when (at least in their judgement), the situation is general knowledge. It uses the proclitic /da/ (/daz/ before a vowel). This is etymologically related to English "does" e.g.

/sôṵlà dáz=áʔp/

sun ASS=rise

The sun rises (the speaker is referring to a well known fact).

Compare the above sentence to one without an evidential particle e.g.

/sôṵlà áʔp/

sun rise

The sun has risen (the speaker has seen it).

===Interrogative Mood===

This is used for polar questions (those expecting a yes/no answer). It is formed with the clitic /dú/ (etymologically related to English "do") e.g.

/té=dú=bíʔ dô̰k/

2PS.ACC=POLQ=bite dog

Were you bitten by the dog?

as well as

/dú=té=bíʔ dô̰k/

POLQ=2PS.ACC=bite dog

Were '''you''' weren't bitten by the dog? / Was it you that was bitten by the dog?

and

/té=bíʔ dú=dô̰k/

2PS.ACC=bite POLQ=dog

Were you bitten by '''the dog'''? Was it the dog that bit you?

Like with /na/, /du/ can also attach to adjectives e.g.

/té=bíʔ dú=m̥òṵ dô̰k/

2PS.ACC=bite POLQ=small.REST dog

Were you bitten by the '''small''' dog? / Was it the '''small''' dog that bit you?? (or were you bitten by the big dog?)

And adverbs e.g.

/té=bíʔ dù=déiʔp dô̰k/

2PS.ACC=bite POLQ=deep.ADV dog

Were you bitten '''deeply''' by the dog? (or was it a shallow bite?)

However, if the word after /du/ begins with a vowel, an epenthetic /w/ is inserted e.g.

/té=dúw=áˈlâṵn dô̰k/

2PS.ACC=POLQ=abandon dog

You weren't abandoned by the dog.

==Responses to Polar Questions==

To answer a polar question in the negative, the particle /nâ̰/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /nâ̰/

no

No

Answering a polar question in the affirmative is more complicated. There is no single word corresponding to English "yes". The answer depends on why the speaker believes the answer to be true.

If the speaker has seen or participated in the action, then they simply repeat whatever came after /du/ in the question e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /bíʔ/

bite

Yes (I saw it)

But in other cases, there are special particles (related to the evidential/mood clitics). If the speaker believes the answer to be yes because someone told them so, then the particle /hḭ̂/ is used e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /hḭ̂/

REP

Yes / I've heard so.

If the speaker believes the answer to be yes because there is direct physical evidence, they use the particle /gè̤/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /gè̤/

INFR

Yes / It looks like it (maybe there is a bite mark on the lizard)

If the speaker believes the answer to be yes because of their experience with similar situations, or because it is general knowledge, they use the particle /dâ̰/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dâ̰/

ASS

Yes / I believe so (maybe because the dog regularly bites the lizard)

If the question entails an assumption that the speaker wishes to deny the truth of, then a special particle /pàṳ/ is used e.g.

A) /jó=wòṳ jô̰ báˈsjàʔp dú=tóʔp/

2PS.GEN=wife 2PS.ERG beat.INF POLQ=stop

Have you stopped beating your wife?

B) /pàṳ/

CHALLENGE.PRESUPPOSITION

I wasn't beating my wife.

If the speaker doesn't know the answer to a question, they can answer with the particle /dṵ̂/ e.g.

A) /dô̰k dú=bíʔ ˈlḭ̂zàd/

dog POLQ=bite lizard

Did the dog bite the lizard?

B) /dṵ̂/

POLQ

I don't know either.

==Wh-Questions==

These are formed by using interrogative pronouns. There is no distinction between "who" and "what" e.g.

/báj=ˈḛ̂nèm/

Q.INTR=enemy

Who / What is the enemy?

or

/ˈbwḛ̀lí=hwám bá=ˈdéˈpʰáʔɾ/

Hkonat=from Q.INTR=depart

Who / What departed from the valley?

/hwéi=ˈbíʔ ˈlḭ̂zàd/

Q.ERG=bite lizard

Who / What bit the lizard?

/ké=ˈbíʔ ˈlḭ̂zàd/

Q.ACC=bite lizard

Who / What did the lizard bite? (because the verb takes a pronoun, the word for lizard does not take the ergative case)

/ˈsʰíʔtà ˈkèʔ=ɾí θú=gḭ̂p dô̰k=àuŋ/

guardian Q.ACC=SEC ANTIP=give dog=DAT

Who / What did the guardian give to the dog?

When English would use a word such as "where" or "when", Kämpya instead uses /kḛ̀/ (the accusative emphatic interrogative pronoun) with a postposition e.g.

/ˈḛ̂nèm ˈkḛ̀=hwám déˈpʰáʔɾ/

enemy Q.ACC=from depart

Where did the enemy depart from?

/dô̰k=zù ˈkḛ̀=wé ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=TEM bite lizard

When did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=ⱱíŋ ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=INS bite lizard

How did the dog bite the lizard?

/dô̰k=zù ˈkèʔ=há ˈbíʔ ˈlḭ̂zàd/

dog=ERG Q.ACC=LOC bite lizard

Where did the dog bite the lizard?

===Asking about Possessors===

Kämpya has no word meaning "whose". Instead it is necessary to ask "Who has ...?", combined with a relative clause e.g.

/dô̰k=zù ˈbíʔ ˈlḭ̂zàd éi=ˈhḛ̂b/

dog=ERG bite lizard Q.ERG=have

Whose lizard did the dog bite? (literally "Who has the lizard that the dog bit?")

or

/θú=ˈbíʔ dô̰k ˈlḭ̂zàd=àuŋ éi=ˈhḛ̂b/

ANTIP=bite dog lizard=DAT Q.ERG=have

Whose dog bit the lizard? (literally "Who has the dog that bit the lizard?")

===Other Interrogatives that modify noun phrases===

To ask questions like "Which lizard did the dog bite?", attach the clitic bá- (or báj- before a vowel) to the noun being asked about e.g.

/dô̰k=zù ˈbíʔ bá=ˈlḭ̂zàd/

dog=ERG bite Q.INTR=lizard

Which lizard did the dog bite?

However, bá- can only attach to a noun in the absolutive case. To ask about the subject of a transitive verb, it is necessary to antipassivise the verb. e.g.

/θú=ˈbíʔ bá=dô̰k ˈlḭ̂zàd=àuŋ/

ANTIP=bite Q.INTR=dog lizard=DAT

Which dog bit the lizard?

Attaching the clitic to a noun in the ergative case is ungrammatical e.g. */bá=dô̰k=zù ˈbíʔ lḭ̂zàd/. Also the clitic cannot attach to a topicalised noun e.g. */bá=dô̰k θú=ˈbíʔ ˈlḭ̂zàd=àuŋ/.

To ask about the object of a postposition, an applicative construction is needed e.g.

/ˈlḭ̂zàd=zù ˈbíʔ=kà bá=pʰò̤wè ˈdô̰k=àuŋ/

lizard=ERG bite=LOC forest dog=DAT which

Which forest did the lizard bite the dog in?

==Verb Subordination==

The subordinate clause is always placed at the end of the sentence. The superordinate verb takes a pronominal prefix to mark whether the subordinate verb is the accusative, ergative or intransitive argument of the verb:

{| class="wikitable"
|-
! !! Accusative !! Ergative !! Intransitive
|-
| Before a consonant || nó || njá || sí
|-
| Before a vowel || ón- || áŋ- || íts-
|-
| Emphatic || nò̰ || njà̰ || ì̤t
|}

For example, from the sentence:

/múˈhḛ̂=ɾù sʰái=ˈsâ̰pè míʔnà/

mother=ERG REP=know everyone

I've heard the mother knows everyone.

We can replace the object with a pronoun to get

/múˈhḛ̂ lé=sʰái=ˈsâ̰pè/

mother 1PS.INCL.ACC=REP=know

I've heard the mother knows us (including you).

If there is a subordinate clause, then the accusative pronoun is replaced with /nó/ e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother SBRD.ACC=REP=know dog=ERG bite lizard

I've heard the mother knows the dog bit the lizard.

If the superordinate clause's absolutive argument is also the topic of the subordinate clause, then it does not need to be mentioned twice e.g.

/múˈhḛ̂ nó=sʰái=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ/

mother SBRD.ACC=REP=know dog=ERG bite

I've heard the mother knows the dog bit her.

It would be redundant to say something like ?/múˈhḛ̂ nó=hí=ˈsâ̰pè ˈdô̰k=zù ˈbíʔ mùˈhḛ̂/.

And another example using a postposition:

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d ˈdô̰k=zù ˈbíʔ ˈlḭ̂zàd/

mother=ERG SBRD.ACC.EMP=SEC speak child dog=ERG bite lizard

The mother told the child that the dog bit the lizard.

And as before, the absolutive argument of the superordinate clause can be the topic of the subordinate clause e.g.

/múˈhḛ̂=ɾù ˈnòʔ=ɾí péiʔk kʰḭ̂d wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child INDEF=bone=SEC NEC=give dog

The mother told the child that he / she (the child) must give the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾí θú=ˈdè̤sì ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gá=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC NEC=give dog

The mother told the child that she (the mother) must give the dog a bone.

Note that the subordinate clause must be marked for evidentiality (always from the speaker's point of view). Compare the above sentence with:

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC give dog

The mother told the child that she (the mother) gave the dog a bone (the speaker saw the act of giving).

/múˈhḛ̂ ˈnòʔ=ɾí θú=péiʔk ˈkʰḭ̂d=àuŋ wá=bàṵn=tá sʰái=gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT INDEF=bone=SEC REP=give dog

The mother told the child that she (the mother) must give the dog a bone (the speaker didn't see the act of giving, but heard about it).

====Anti-Logophoricity====

If the subordinate clause contains a transitive verb, then the 3rd person obviative ergative pronoun ðjíʔ can be used as a kind of "anti-logophoric" pronoun [http://en.wikipedia.org/wiki/Logophoricity] to indicate that subject of the subordinate clause's transitive verb is not the absolutive argument of the main clause.

So if the above 2 sentences take ðjíʔ, then they become:

/múˈhḛ̂=ɾù ˈnòʔ=ɾá péiʔk ˈkʰḭ̂d ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother=ERG SBRD.ACC.EMP=SEC speak child 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the child, either the mother or someone else) gave the dog a bone.

/múˈhḛ̂ ˈnòʔ=ɾá θú=ˈpéiʔk ˈkʰḭ̂d=àuŋ ðjíʔ wá=bàṵn=tá gḭ̂p dô̰k/

mother SBRD.ACC.EMP=SEC ANTIP=speak child=DAT 3PS.OBV.ERG INDEF=bone=SEC give dog

The mother told the child that he / she (someone other than the mother, either the child or someone else) gave the dog a bone.

===Evidentials and Reported Speech===

When reporting what someone else has said, both clauses need to marked for evidentiality, both from the point of view of the speaker e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy depart

The child told me the enemy departed (and I saw it happen too).

If the speaker is reporting an event that someone told them about, but that they have not seen themselves, then they need to use an evidential in the subordinate clause e.g.

/ˈkʰḭ̂d ˈnòʔ=ɾí mí=ˈpéiʔk ˈḛ̂nèm sái=déˈpʰáʔɾ/

child SBRD.ACC 1PS.EXCL.ACC=speak enemy REP=depart

The child told me the enemy departed (but I didn't see it, I just heard about it).

==Comparative Constructions==

To say "more than ...", Kämpya speakers say ... /tʰáp/, where /téŋ/ is a postpositional clitic that also means "above" e.g.

/dô̰k ˈlḭ̀zád=tʰáp zwéʔp/

dog.TOP lizard=above fast

Dogs are faster than lizards (literally "Dogs are fast above lizards").

Seggeynni

2016-09-09T12:57:13Z

Linguist Wannabe: /* Pronouns */

Seggeynni is another conlang spoken in my far-future Antarctican conworld. It developed from language contact, when [[Kämpya]] speakers settled in lands where [[Thangha']] was spoken. Kämpya speakers were more technologically advanced than Thangha' speakers, and Kämpya was definitely the prestige language, forming the superstrate. Most of Seggeynni's vocabulary and grammar come from Kämpya. On the other hand, Kämpya speakers intermarried extensively with the Thangha' speaking community, and Thangha' has exhibited a very heavy substrate effect.

The language documented here is spoken approximately a century after the initial migration. The vast majority of speakers are mixed-race descendents of Kämpya and Thangha' speakers. In addition to Seggeynni, almost all have a passive command of both Thangha' (still a common trade language in the region) and Kämpya (a liturgical language).

Like Thangha', there are many consonants which contrast secondary articulations, which affect the quality of adjacent vowels. Additionally, Seggeynni is characterised by an underlying simple syllable structure at the phonemic level, however many speakers "drop" vowels, so that consonant clusters are heard in the spoekn language. For example, phonemically the name of the language is /ˈséʕkə̀jə̀nə̀jə̀/, but, except when speaking extremely carefully, most speakers pronounce it as [ˈséɠə̀iɲːì].

==Phonology==

===Vowels===

The vowel system is quite similar to Thangha'. In stressed syllables, there are 6 vowel phonemes /iː/, /uː/, /e/, /o/ and /aː/ and /ə/.

As in Thangha', the length of the vowel is more important than the quality in contrasting /iː/ and /uː/ vs. /e/ and /o/. Secondary articulation on surrounding consonants "pulls" the vowels around a fair bit.

In unstressed syllables, the language has a [https://en.wikipedia.org/wiki/Vertical_vowel_system Vertical Vowel System], only contrasting /a/ and /ə/ (however other vowels such as [i], [e], [o] and [u] are heard as allophones of /ə/, depending on the surrounding consonants).

===Consonants===

The consonant inventory is very similar to Thangha'. Kämpya's aspirated consonants became velarised in Seggeynni.

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Dental / Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Glottal
|-
| Nasal || m || mʲ || n || nʷ || || || ɲ || ŋ* || || (ɲʷ) || ŋʷ* ||
|-
| Voiced Stop || b || bʲ || d || dʷ || || || ɟ || g || || (ɟʷ) || gʷ ~ gb ||
|-
| Voiceless Stop || p || pʲ || t || tʷ || || || c || k || || (cʷ) || kʷ ~ kp || ʔ
|-
| Voiced Sibilant Affricate || || || || || dʑ || || || || || || ||
|-
| Voiceless Sibilant Affricate || || || || || tɕ || || || || || || ||
|-
| Spirant Affricate || || || tθ || || || ʈʂ || || || || || ||
|-
| Voiced Sibilant Fricative |||| || z || zʷ || ʑ || || || || || || ||
|-
| Voiceless Sibilant Fricative || || || s || sʷ || ɕ || || || || || || ||
|-
| Spirant Fricative || f || fʲ || || || || ʂ || ç ~ ɬ || || χ ~ x || || ||
|-
| Flap / Trill || ⱱ ~ ʀʷ || ⱱʲ || || || ɾʲ || ɽ || || || (ʀ) || || ||
|-
| Lateral || || || || || || || ʎ || || ɫ || (ʎʷ) || ɫʷ ||
|-
| Semivowel || || || || || || || j || || || ɥ || w ||
|}

* Velar nasals cannot occur word initially.

* Except for /ɥ/, labialised palatal consonants such as /cʷ/ are rare. They only occur in loanwords. Before front vowels, most speakers merge them with labialised velar consonants, and before back vowels, most speakers merge them with plain palatal consonants.

* Labialised velar stops /kʷ/ and /gʷ/ are in free variation with their doubly articulated counterparts.

* For the purposes of phonology, it is useful to draw a distinction between sibilant and spirant (non-sibilant) affricates and fricatives. Sibilants contrast for voicing while spirants do not.

* Between sonorants, /tθ/ is lenited to [θ].

* A sound change is currently in progress which is changing the palatal fricative /ç/ to a lateral fricative /ɬ/, especially in clusters.

* /χ/ has the allophone /x/ between sonorants.

* /ʀ/ is only found in loanwords. Many speakers merge it into /ɽ/.

* Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ occur as separate phonemes in a few dialects. But most dialects have merged these with the plain labials.

===Phonotactics===

Underylingly, the only syllables permitted are CV and CCV. The latter is rare, and only a few consonant clusters are permitted, namely /pf/, /tf/, /sf/, /kf/, /pχ/, /sχ/, /kχ/, /pç/ and /kç/.

* Speakers who are shifting /ç/ to /ɬ/ are especially prone to do so in clusters. In this case, they shift /kç/ to a lateral affricate /tɬ/.

* /kχ/ is much more commonly heard as [q].

* /kf/ is sometimes heard as [qʷ].

* There used to be a cluster */tχ/, but it has now shifted to /tθ/.

===Spirant Mutation===

Seggeynni has a rule that only one "spirant" consonant can occur in a phonological word. If a process such as cliticisation would cause a violation of this rule, then every spirant consonant except the last undergoes mutation, via the following patterns:

* /f/ -> /w/

* /fʲ/ -> /ɥ/

* /pf/ -> /p/

* Other clusters of Consonant + /f/ -> Labialised versions of the initial consonant, with the /f/ deleted e.g. /sf/ -> /sʷ/, /kf/ -> /kʷ/ etc.

* /tθ/ -> /t/

* /ʈʂ/ -> /tɕ/

* /ʂ/ -> /ɕ/

* /ç/ -> /j/

* /pç/ -> /pʲ/

* /kç/ -> /c/

* /χ/ -> /ʔ/

* Clusters of consonant + /χ/ -> Simple deletion of the /χ/ e.g. /pχ/ -> /p/, /sχ/ -> /s/

e.g. when the word /ˈkχòʔɲátə́/ - "Hkonat" (the name of a city) takes the postposition /fámə́/ - "from", the result is /ˈkòʔɲátə́=fámə́/ - "from Hkonat". Similarly when the verb /ˈʔà̤ːχə̀/ - "to ask a question" takes the reportative mood prefix /sfájə́/, the result is /sʷájə́=ˈʔà̤ːχə̀/ - "apparently he asked".

===Stress===

The situation here is identical to Kämpya. One syllable in each word bears stress, the location of which is unpredictable. Vowel phonation is phonemic on stressed syllables but not elsewhere. The phonations are tense /a̰/, breathy /a̤/ or /ä/, and glottalised /aʔ/. Glottalised vowels are pronounced with gemination (and glottalisation where possible) of the following consonant (or with a glottal stop [ʔ] in the case of word final syllables). For example, the first syllable of "Seggeynni" has stress with glottalised phonation, meaning that the subsequent /k/ is pronounced as an implosive /ɠː/.

Long vowels with glottalised phonation diphthongise into centering diphthongs. For example, /ˈdʑíːʔpə̀/ - "deep" (descriptive) is pronounced [ˈdʑíəɓə̀].

Short vowels with tense phonation are pronounced with an epiglottal approximant [ʢ] after them e.g. /ɲḛ̂/ - "snare" is pronounced [ɲḭ̂ʢ].

The pronunciation of short vowels with breathy phonation varies depending on a variety of factors. At the end of words, they are normally pronounced with a final /ɦ/, or occasionally [z] or [ɹ]. If the vowel precedes an obstruent (stop, affricate or fricative), then a voiceless fricative is inserted, that is homorganic with the obstruent e.g. /sə́ˈçə̤̀tà/ - "sister" is most often pronounced [síˈçì̤stà]. However this is in free variation with preaspiration, so the word for sister can also be pronounced as [síˈçì̤ʰtà]. If the vowel precedes a sonorant, then it is pronounced geminated and with breathy voice e.g. /dè̤nə̀/ - "to dance" is usually pronounced [dè̤n̤ːə̀].

===Tone Patterns===

These are also identical to Kämpya. A description of the tone patterns can be found [http://linguifex.com/wiki/K%C3%A4mpya#Stress_and_Phonation here].

===Assimilations / Epenthesis===

There are many phonetic processes that result in the deletion of schwa, often leading to consonant clusters. They apply to unstressed syllables, with two major exceptions. The first of these is word-initial syllables. The second of these is the syllable immediately after a stressed syllable.

====Nasal Consonants====

If a schwa occurs word finally and is preceded by a nasal, then the whole syllable is often deleted, with nasalisation of the preceding vowel e.g. /ˈjḛ̂ɲə̀mə̀/ - "enemy" is pronounced [ˈjḛ̂ʢɲì'''ⁿ'''] (here and below, the parts of the word that undergo assimilation have been marked in bold).

A schwa has a strong tendency to be lost if it occurs immediately after a nasal consonant, and immediately before an obstruent. In this case, the nasal changes it's place or articulation to be homorganic with the obstruent e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the locative enclitic /ka/ to become /ˈjḛ̀ɲə́mə́=ká/, which is pronounced [ˈjḛ̀ʢɲí'''ŋk'''á].

If a schwa occurs between two nasal consonants, it is elided, along with the preceding consonant. In its place, the following nasal is geminated e.g. /ˈjḛ̀ɲə́mə́/ - "enemy" (postpositional form) combines with the enclitic /nəja/ to become /jḛ̀ɲə́mə́=nə́já/ - "near the enemy", which is pronounced [jḛ̀ʢɲínːéjǽ].

The sequence /məjə/ is usually pronounced [mːʲi]. Unlike most other assimilation processes, this can occur in the syllable immediately following the stressed one e.g. /ʔáˈɫʷṳ̀ː'''mə̀=jə̀'''/ - "everyone's" (alienable) is usually pronounced [ʔɒ́ɫʷṳ̀ː'''mːʲì'''].

A similar process transforms the sequence /nəjə/ to [ɲːi], as in the name of the language, Seggeynni [ˈséɠːə̀i'''ɲːì'''], underlyingly /ˈséʕkə̀jə̀'''nə̀jə̀'''/.

====Debuccalisation of Stops====

If a word final schwa is preceded by a stop, then it is very common to delete the schwa and debuccalise the stop to a glottal stop e.g. /ˈkχóʔɲà'''tə̀'''/ - "Hkonat" (the name of a city), is usually pronounced as [ˈkχɔ́ɲ̰ːɛ̀'''ʔ'''].

If schwa occurs after a stop, and before an obstruent of identical voicing, then both schwa and the preceding stop are deleted, and the following consonant becomes geminated e.g. /ˈkχòʔɲá'''tə́=k'''á/ - "from Hkonat" is normally pronounced [ˈkχɔ̀ɲ̰ːɛ́'''kː'''á].

If a schwa is preceded by a stop, and followed by a sonorant, then both the schwa and the preceding stop are deleted, and the sonorant is glottalised and geminated e.g. /ˈkχóʔɲàtə̀/ - "Hkonat" combines with the alienable genitive enclitic /jə̀/ to become /ˈkχóʔɲà'''tə̀=j'''ə̀/ - "of Hkonat" (alienable), which is pronounced [ˈkχɔ́ɲ̰ːæ̀'''j̰ː'''è].

====Semivowels====

An extremely common phonetic process involves /j/, /w/ or /ɥ/ flanked on both sides by /a/ or /ə/. The second vowel is deleted, and the semivowel assimilates with the preceding vowel to form a diphthong. /ajə/ and /aja/ become [ai], /awə/ and /awa/ become [au], /aɥə/ and /aɥa/ become [aʉ]. Likewise /əjə/ and /əja/ become [əi], /əwə/ and /əwa/ become [əu], /əɥə/ and /əɥa/ become [əʉ]. An example of this is in the name of the language, /ˈséʕk'''ə̀jə̀'''nə̀jə̀/, which is pronounced as [ˈséɠː'''ə̀i'''ɲːì].

This differs from the previous processes in two ways. Firstly because it can delete not only schwa, but also /a/. Secondly because it can delete vowels in any syllable, even if it bears stress or comes after the stressed syllable. For example /ʔáɫáːˈwə̰̀nə̀/ [ʔɑ́ˈɫɑ̀ːṵnə̀] - "alone" (restrictive). If a stressed vowel is deleted in this way, the resulting diphthong bears stress (as in the previous example).

====Conflicting Processes====

If a word could be transformed in more than one way, then the assimilation rules work from left to right across the word. So /ˈséʕkə̀jə̀'''nə̀jə̀'''/ is pronounced as [ˈséɠːə̀i'''ɲːì'''] and not *[ˈséɠːə̀in'''ə̀i'''].

===Comparison with Kämpya===

At the phonological level, the two languages are completely different, however at the phonetic level, many words are similar. Below is a table comparing words in the two languages, as well as words in the older version of Kämpya that the two languages diverged from.

{| class="wikitable"
|-
! Gloss !! Seggeynni Phonemic !! Seggeynni Phonetic !! Kämpya !! Old Kämpya
|-
| Seggeynni || /ˈséʕkə̀jə̀nə̀jə̀/ || [ˈséɠːə̀iɲːì] || [ˈséʔkìnì] || */ˈsɛ́ʔkìnì/
|-
| forest || /ˈfò̤ɥə̀/ || [ˈfòy̤ː]] || [ˈpʰò̤wè] || */ˈfɔ̤̀wè/
|-
| sister || /sə́ˈçə̤̀tà/ || [síˈçì̤stà] || /ˈsʰì̤tà/ || */ˈsʰì̤tà/
|-
| snare || /ɲḛ̂/ || [ɲḭ̂ʢ] || [n̥ḛ̂ː] || */n̥ḛ̂/
|-
| forgetful || /jámə́ˈɲì̤ː/ || [jɛ́ˈɲːì̤ː] || [ámˈnèi̤] || */ɛ́mˈnèi̤/
|-
| rotten (restrictive) || /zə́ˈwòʔ/ || [zóˈwòʔ] || [ˈzwóʔ] || */ˈzwɔ́ʔ/
|-
| everyone || /ʔáˈɫʷṳ̀ːmə̀/ || [ʔɒ́ɫʷṳ̀ːm] || [áˈlòṳⁿ] || */áˈlòṳm/
|-
| Hkonat (the name of a city) || /ˈkχóʔɲàtə̀/ || [ˈqɔ́ɲ̰ːɛ̀ʔ] || [ˈkʰóʔnàt] || */ˈkʰɔ́ʔnɛ̀t/
|-
| at Hkonat || /ˈkχòʔɲátə́=ká/ || [ˈqɔ̀ɲ̰ːɛ́kːá] || [ˈkʰòʔnátká] || /ˈkʰɔ̀ʔnɛ́t=ká/
|-
| cave || /kʷə́ˈçḛ̂pà/ || [kʷʉ́ˈçḭ̂ʢpà] || [ˈkʰwḛ̂ːpà] || */ˈkʰwḛ̂pà/
|}

==Telicity==

Due to contact with Thangha', Seggeynni has developed a grammatical [https://en.wikipedia.org/wiki/Telicity telicity] contrast. Atelic verbs are unmarked, and in Tone Class 1. Telic verbs are in Tone Class 2, are usually take the suffix /ɕə́ŋə́/ (usually pronounced [ɕíⁿ]). This is a grammaticalised form of the verb /ˈɕéʔŋə̀/ - "to become".

For example:

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̂ːɽʌ̀ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.ATEL one month.ADV

The enemy departed from Hkonat for a month (and then they came back)

vs.

/ˈjḛ̂ɲə̀mə̀ ˈkχòʔɲátə́=fámə́ dʑə́ˈpˤâ̰ːɽə̀-ɕə́ŋə́ tə́ˈjə̀ʔ mǎ̤ːnə̀/

[ˈjîʢɲìⁿ ˈqɔ̀ɲ̰ːɛ́fːámə́ ˈdʑɘ́ˈpˤɑ̰̀ːɽɘ́ɕíⁿ tə̀iʔ mǎ̤ːnə̀]

enemy Hkónat=from depart.TEL one month.ADV

The enemy departed from Hkónat in a month (as in it took them that long to get ready).

===Semantic Shifts===

Many verbs have completely different meanings when their aspect changes from atelic to telic. For example:

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎíːʔpə̀/

[ˈjîʢɲìⁿ xíˈʎíəɓə̀]

enemy sleep.ATEL

The enemy is asleep.

vs.

/ˈjḛ̂ɲə̀mə̀ xə́ˈʎìːʔpə́-ɕə́ŋə́/

[ˈjêʢɲìⁿ xíˈʎìəɓíɕíⁿ]

enemy sleep-TEL

The enemy is dead.

This (and many other similar semantic shifts) are clearly attributable to influence from Thangha', which inflects the verb root /səʔˈʎiːpə/ into atelic /səʔˈʎiːpə/ to mean "sleep", and telic /mʲaɲəʔ-səʔˈʎiːpə/ to mean "die". In Kämpya (which does not mark telicity), the verb /hléiʔp/ (cognate with /xə́ˈʎíːʔpə̀/) is only used to mean "sleep". There is a separate verb root /dôṵ/ to mean "die". However in Seggeynni, the cognate to this word /dʷḭ̂ː/ means to die insane.

==Pronouns==

Like Kämpya, Seggeynni has tripartite alignment in pronouns. The emphatic pronouns and the ergative pronouns (also used for alienable possession) are independent words, but the others are proclitics. There is also a distinction between emphatic and non-emphatic forms of pronouns. Unlike Kämpya, there is no proximate / obviate distinction in 3rd person pronouns (null pronouns are used when Kämpya would use a proximate pronoun).

{| class="wikitable"
|-
! !! 1PS Exclusive !! 1PS Inclusive !! 2PS !! 3PS !! Interrogative
|-
| Intransitive || /ʔájə́/ || /wájə́/ || /jə́wə́/ || Null || /bá/
|-
| Intransitive Emphatic || /wḭ̀ː/ || /ɥḭ̀ː/ || /jṵ̀ː/* || Null || /bà̰ː/
|-
| Accusative || /mə́jə́/ || /ʎə́/ || /tɕə́/ || /sə́́wə́jə́/ || /cə́/
|-
| Accusative Emphatic || /mə́ˈjà̰/ || /ˈʎè̤sə́/ || /tɕḛ̀/ || /sə́́wə́ˈjà̰/ || /cḛ̀/
|-
| Genitive Inalienable || /májə́/ || /jáfə́/ || /ɥə́/ || /ⱱə́jə́/ || /jə́fə́/
|-
| Genitive Alienable / Ergative || /mḭ̂ː/ || /jṵ̂ː/* || /ɥô̰/ || /ⱱə́ˈjáʔ/ || /çṳ̀ː/
|-
| Genitive / Ergative Emphatic || /mḭ̀ː/ || /jṵ̀ː/ || /ɥò̰/ || /ⱱə́ˈjàʔ/ || /çṳ̀ː/
|}

Due to a sound change, the emphatic forms of the 2nd person intransitive pronoun and the 1st person inclusive genitive alienable / ergative pronoun have become homophones. However, due to social norms, the 2nd person emphatic pronouns are very rarely used at all. So whenever /jṵ̀ː/ is heard, it can be assumed to be the emphatic form of the 1st person genitive alienable / ergative pronoun.

===Alternate Second Person Pronouns===

In certain situations, alternate forms of the second person pronoun are used. There are a variety of these, depending on the situation:

* /ˈɥô̰mə̀/ - used by a cleric to a layperson
* /ˈmʲṵ̂ːŋə̀/ - used by a senior cleric to refer to a junior cleric
* /tɕə́ˈjṵ̂ːtèn/ - literally "student", used by a teacher to a student
* /kə́ˈçə̰̂də̀/ - literally "child", used by an adult to a child
* /wṳ̀ː/ - literally "wife", used by a husband to a wife
* /ˈtfóʔɽə̀/ - used by men to slightly younger men (within a generation) that they know well
* /ˈmóʔɽə̀/ - used by women to slightly younger women (within a generation) that they know well
* /sˤâ̰ːnə̀/ - literally "son", used by an older person to much younger men
* /ˈdóʔɽà/ - literally "daughter", used to much younger women
* /wóˈmḛ̂/ - literally "slave", used when talking to slaves.

Most of these are derived from nouns, and indeed they pattern like nouns, not prononus. Notably, they take ergative-absolutive marking (with the enclitic /-zə̀wə̀/ or /-ɽə̀wə̀/ used to mark ergative case) rather than tripartite marking.

Thangha'

2016-09-09T08:08:22Z

Linguist Wannabe:

Thangha' is another language spoken in my far future Antarctic conworld (after runaway global warming has melted most of the ice but made most of the rest of the world uninhabitable). It is spoken by nomads who inhabit the centre of the continent.. Because each group of nomads speak their own dialect, Thangha' is a [https://en.wikipedia.org/wiki/Pluricentric_language Pluricentric language]. The situation is further complicated by the fact that Thangha's has become a lingua franca across much of East Antarctica (through other ethnic groups trading with the nomads). To give an example of the diversity, the initial consonant in the name of the language can be pronounced as [tˤ], [θˤ], [θ], [tθˤ], or [tʰ]. The consonant that begins the second syllable is usually prononuced [x] or [χ], but a few dialects pronounce it as as [ħ] or [h].

Like many other Antarctican languages (but definitely not all), Thangha' is descended from English. However, many ancestors of Thangha' speakers came from the Philippines, Indonesia and the Pacifc (these areas were the first to be severely affected by global warming), and the language is heavily influenced by Austronesian languages, most notably in its split ergative alignment. But it also borrowed heavily from Hindi and various Chinese dialects. This is a similar story to Kämpya, except that speakers of the two languages were largely isolated from each other for two thousand years, meaning that they are no longer mutually intelligible.

==Phonology==

===Vowels===

In stressed syllables, most dialects distinguish 5 plain oral vowels /a/, /iː/, /uː/, /e/, /o/ and two nasal vowels /aⁿ/ and /əⁿ/. The difference between /iː/ vs. /e/ is more dependent on length than height. For example, a short [i] is more likely to be heard as /e/ than /iː/, and a long /eː/ is more likely to be heard as /iː/ than /e/. The back vowels /uː/ and /o/ are similar in this regard.

In unstressed syllables, the plain oral vowels /iː/, /uː/, /e/ and /o/ (i.e. all of the plain oral vowels except /a/) merge to /ə/ (which is still distinct from /əⁿ/.

In non-word final syllables, it is very common for nasal vowels to [https://en.wikipedia.org/wiki/Unpacking unpack]into a sequence of oral vowel + nasal consonant (which is homorganic with the following consonant). For example, the vowel of the first syllable in the language name Thangha' is phonemically /aⁿ/, but is usually heard as [aŋ] (or [aɴ] for speakers who use a uvular fricative instead of a velar fricative).

===Consonants===

There is considerable variation between dialects here:

{| class="wikitable"
|-
! !! Labial !! Pal. Labial !! Pharyn. Labial !! Laminal !! Pharyn. Alveolar !! Lab. Dental !! Alveolo-Palatal !! Retroflex !! Palatal !! Velar !! Uvular !! Lab. Palatal !! Lab. Velar !! Lab. Uvular !! Glottal
|-
| Nasal || m || mʲ || mˤ || n || || nʷ || || ɳ || ɲ || ŋ || || ɲʷ || ŋʷ || ||
|-
| Voiced Stop || b || bʲ || bˤ || d || dˤ || dʷ || || || ɟ || g || || ɟʷ || gʷ ~ gw ~ gb || ||
|-
| Voiceless Stop || p || pʲ || pˤ || t || tˤ || tʷ || || || c || k || q || cʷ || kʷ ~ kw ~ kp || qʷ ~ kf ~ qp || ʔ
|-
| Voiced Affricate || || || || dz || || || dʑ || ɖʐ || || || || || || ||
|-
| Voiceless Affricate || || || || ts || || || tɕ || ʈʂ || || || || || || ||
|-
| Voiced Fricative || || || || z || zˤ || zʷ || ʑ || || || || || || || ||
|-
| Voiceless Fricative || f || fʲ || fˤ || s || sˤ || sʷ || ɕ || ʂ || ç ~ ɬ || x ~ χ ~ ħ ~ h || || || || ||
|-
| Flap / Trill || || || || || || || ɾʲ || ɽ || || || ʀ || || || ʀʷ ~ ⱱ ||
|-
| Lateral || || || || || || || || || ʎ || || ɫ || ʎʷ || || ɫʷ ||
|-
| Semivowel || || || || || || || || || j || || || ɥ || w || ||
|}

The uvular trill is most commonly pronounced as an approximant [ʁ].

A few dialects merge the aspirated dental consonant into the voiceless alveolar affricate. In these dialects, Thangha' would be pronounced [ˈtsaŋxaʔ].

Some dialects merge /ɫʷ/ into /ɫ/.

Labialised labial consonants /mʷ/, /bʷ/, /pʷ/ (and sometimes /fʷ/~/ɸ/) occur as separate phonemese in a few dialects. But most dialects have merged these with the plain labials.

Except for flaps, all sonorant consonants have plain and glottalised forms e.g. [m] vs. [m̰], [w] vs. [w̰] etc. However, these can be analysed as sequences of glottal stop + sonorant e.g. /m/ vs. /ʔm/, /w/ vs. /ʔw/ etc.

Velar nasals /ŋ/ and /ŋʷ/ cannot occur word initially.

===Stress and Tone===

There is both a stress and a tone system, which interact with each other.

Stress falls on either the last or the second last syllable of a word. The stressed syllable is pronounced louder than unstressed syllables. The pitch depends on the tone.

Tone is assigned at the word level, rather than to individual syllables. There are three phonemic tones, High, Low and Falling.

Words with High Tone have high pitch on all syllables other the stressed syllable, which is pronounced with even higher pitch e.g. /ʔwaˈjəʔma/ (H) - "llama" (genitive) is pronounced [w̰ɞ́ˈjɪ̋ʔmá], /tɕaˈbˤaⁿ/ (H) - "Japan" is pronounced [tɕɐ́ˈbˤɑ̋ⁿ].

Words with Low Tone have mid pitch on all syllables other than the stressed syllable, which is pronounced with an extra low pitch e.g. /ʔwaˈməwaⁿ/ (L) - "mountain" (genitive) is pronounced [w̰āˈmʊ̏wāⁿ].

Words with Falling Tone always have stress on the second last syllable. Their pronunciation depends on the vowel length of the stressed vowel. If the vowel is short and non-nasalised, there is high pitch on all syllables up to and including the stressed syllable, and low pitch on the final syllable e.g./tsaˈɲeɳə/ (HL) - "channel", "strait" is pronounced [tsǽˈɲéɳʌ̀]. However, if the stressed vowel is long or nasalised, then it has falling pitch e.g. /ʔwaˈbiːdˤə/ (HL) - "boat" (genitive) is pronounced [wáˈbêːdˤʌ̀].

====Evolution====

This system evolved via the following stages:

1) English stress became entirely regular, being assigned to the heaviest of the final three syllables of a word via the following rules:

* A phonemic distiction between long and short vowels emerges. English */æ/ is always lengthened to /æː/. The only places where this distiction is not phonemic are before nasal codas and at the end of words. In both of these cases only short vowels occur.

* Diphthongs, coda consonants (including coda /r/, which has survived in some words) and long vowels each contribute an additional mora to a syllable. So for example, in the word "bullet", the first syllable has one mora, and the second has two morae.

* The final mora of a word is extrametrical. For the purposes of assigning stress, it doesn't count. So "bullet" effectively has one mora on each syllable now.

* The syllable with the most morae gets assigned stress. So "bullet" is now */buˈlet/. There are some cases where the stress stays the same e.g. "doctor" is */dokta/.

2) Phonotactics become much stricter. The only syllables permitted are CV and CVN (where N is a nasal). Wherever a violation of this rule would occur. So "bullet" is now *[buˈletə], and "doctor" is now *[ˈdokəta].

3) Pitch becomes assigned to syllables via the following rules:

* Word final syllables that end in a nasal are assigned low pitch.

* Word final syllables that ended in schwa, */a/, */i/ or */u/ are assigned low pitch (these vowels all merge to schwa, though often modifying the previous consonant).

* Other word final syllables are assigned high pitch.

* Stressed syllables are always assigned high pitch when not word final.

* Unstressed syllables before the stressed syllable receive high pitch.

* Unstressed syllables after the stress syllable take the same pitch as the final syllable of a word.

So "doctor" is now */ˈdókə̀tà/ HLL, "bullet" is now /búˈlétə̀/ HHL, "llama" is */ˈjə́má/ HH (it begins with an initial /j/ because the word is borrowed from Spanish).

4) If a word has stress on any syllable before the penultimate, the word was truncated so that stress falls on the penultimate syllable. However, this truncation does not delete tone, which is mapped to syllables from right to left. So "doctor" is */ˈdòkà/ HLL, where the initial H tone is a floating tone that only surfaces when the noun takes a prefix.

5) All tones except the final two are lost. So "doctor" now just has a LL melody, without a floating tone. Likewise "bullet" now just has a HL melody, as can be seen from its final form /pəˈʎedˤə/ HL.

===Phonotactics===

Only CCV syllables are permitted. Furthermore, the range of permissible clusters is extremely restricted, being limited to /ʔ/ + Sonorant (other than flaps), /pf/, /bf/ (heard as [bv], /pɕ/, /bʑ/, /tf/, /df/ (heard as [dv]), /kf/ and /kɕ/.

===Vowel Allophony===

The vowels of Thangha' have a number of different allophones depending on what consonants they are adjacent to. These are described in the table below. Note that a "lowering" consonant is defined as a pharyngealised, retroflex or uvular consonant (labialised or non-labialised, including /ɫ/). If two sounds are listed, the first occurs in stressed syllables, and the second in unstressed syllables:

{| class="wikitable"
|-
! Phoneme !! /a/ !! /iː/ !! /uː/ !! /e/ !! /o/ !! /aⁿ/ !! /əⁿ/ !! /ə/
|-
| Default || [a] || [iː] || [uː] || [e] || [o] || [aⁿ] || [əⁿ] || [ə]
|-
| Adjacent to a Palatal or Palatalised Consonant || [æ] || [iː] || [ʉː] || [i] || [ɵ] || [ɛⁿ] || [eⁿ] || [ɪ]
|-
| Adjacent to a Lowering Consonant (Non-Labialised) || [ɑ] || [eː] || [oː] || [ɛ] || [ɔ] || [ɑⁿ] || [ʌⁿ] || [ʌ]
|-
| Adjacent to a Labialised Consonant (Non-Lowering) || [a] || [yː] || [uː] || [ø] || [u] || [aⁿ] || [oⁿ] || [ʊ]
|-
| Adjacent to a Labialised Uvular Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Lowering Consonant and a Labialised Consonant || [ɒ] || [øː] || [oː] || [œ] || [o] || [ɒⁿ] || [ɔⁿ] || [ɔ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Consonant (Non-Lowering) || [ɞ] || [yː] || [ʉː] || [y] || [ʉ] || [œⁿ] || [øⁿ] || [ʏ]
|-
| Between a Palatal / Palatalised Consonant, and a Labialised Uvular Consonant|| [ɐ] || [øː] || [oː] || [ø] || [o] || [ɐⁿ]|| [əⁿ] || [ɵ]
|-
| Between a Palatal / Palatalised Consonant, and a Lowering Consonant (Non-Labialised) || [ɐ] || [eː] || [oː] || [e] || [o] || [ɐⁿ] || [əⁿ] || [ɘ]
|}

==Nouns==

Any Thangha' adjective can be used as a noun, and vice verca, any noun can be used as an adjective (which come before the head noun when used attributively, like English). For example, /ˈsaⁿdə/ (HL) [ˈsandə] can mean either "sand" or "sandy".

===Plurals===

====Additive Plural====

This is formed by reduplicating the first syllable of the noun, except that the vowel is changed to schwa e.g. /ˈjəʔma/ (HL) [ˈjɪ̋m̰à] - "llama" -> /jə-ˈjama/ (HL) [jɪ́ˈjɪ̋m̰à] - "llamas". However, in Thangha', pluralisation is optional. So /ˈjəʔma/ (HL) could mean either "llama" or "llamas". In general, the plural is only used to emphasise that there is a large number / quantity of something. It can even be used for uncountable nouns e.g. /ˈsaⁿdə/ (HL) [ˈsândə̀] - sand (from a desert, Thangha' has a different word for beach sand) -> /sə-ˈsaⁿdə/ (HL) [sə́ˈsândə̀] - "large amount of desert sand / dune sea".

====Associative Plural====

Distinct from the additive plural, this means "X and company", "X and his/her mob" etc. (similar to Japanese -tachi). It is formed by prefixing /ɫʷatə-/ to the noun e.g.

/ˈtʷoʀa/ (L) - [ˈtʷòʀɑ̄] - "shaman" (from English "doctor") -> /ɫʷatə-ˈtʷoʀa/ (L) - [ɫʷɒ̄tʊ̄ˈtʷòʀɑ̄] - "the shaman and his lot"

Note that this is distinct from the additive plural discussed before. /tʷəˈtʷoʀa/ (L) refers to a group of shamans, while /ɫʷatə-ˈtʷoʀa/ (L) refers to a shaman with a group of other people associated with him (who by no means need to be shamans, and could well be the shaman's assistants or followers).

===Case===

Thangha' has three cases, Absolutive (unmarked), Genitive (marked with the prefix /ʔwa-/), and Ergative (marked with the prefix /pə-/).

e.g. /ˈfaɖʐə/ (H) [ˈfɑ̋ɖʐʌ́] - father -> /ʔwaˈfazə/ (H) [w̰áˈfɑ̋ɖʐʌ́] - "of the father" -> /pəˈfazə/ (H) [pə̄ˈfɑ̋ɖʐʌ́] - "father" (ergative)

However, if the noun root begins with a voiceless obstruent, it becomes voiced e.g. /ˈtʷoʀa/ (L) [ˈtʷòʀɑ̄] - "shaman" -> /ʔwaˈdʷoʀa/ (L) [w̰āˈdʷòʀɑ̀] - "of the shaman" (not */waˈyʷoʀa/ (L)) -> /pə-ˈdʷoʀa/ (L) [pə̄ˈdʷòʀɑ̄] - "shaman" (ergative).

Also, if the noun root begins with a glottalised sonorant, it loses its glottalisation when it takes a prefix e.g. /ˈʔməʂa/ (H) [ˈm̰ʌ̋ʂɑ́] - "mother" -> /ʔwa-ˈməʂa/ (H) [w̰áˈmʌ̋ʂɑ́] - "of the mother" -> /pə-ˈməʂa/ (H) [pə́ˈmʌ̋ʂɑ́] - "mother" (ergative).

===Possession===

Like most Antarctican languages, Thangha' uses different constructions for [https://en.wikipedia.org/wiki/Inalienable_possession alienable and inalienable possession].

====Inalienable Possession====

In this case, the word order is Possessed - Possessor, with the Possessor marked with the genitive case e.g.

/ˈpəʔwəⁿ ʔwa-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ w̰á-ˈdʷúgə̀]

bone GEN-dog

The dog's bone (in it's body)

====Alienable Possession====

The construction for alienable possession is the same as that for alienable possession, except that the possessor is marked with the Ergative case e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə/ (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀]

bone ERG-dog

The dog's bone (that it is chewing / burying etc.)

===Demonstratives===

Like English, Thangha' makes a two-way distance contrast in demonstratives, and there is no distinction between pronominal demonstratives and adnominal demonstratives (which precede the noun, like in English). Unlike English, there is no singular vs. plural contrast in demonstratives:

This / These: /ˈɕəsə/ (HL) [ˈɕɪ́sə̀]

That / Those: /ˈsazə/ (HL) [ˈsázə̀]

===Prepositional Clitics===

In Thangha', prepositions are clitics that attach phonologically to the following word (whether that is a noun, demonstrative, numeral or adjective). The pronunciation of the last vowel in a preposition is affected by the first consonant in the following word e.g.

/ʔɲəʀə/ [ɲ̰ɘʀʌ] - "near"

/ˈɕəsə/ (HL) [ˈɕɪ́sə̀] - "this one"

/ʔɲəʀə=ˈɕəsə/ (HL) [ɲ̰ɘ́ʀɘ́ˈɕɪ́sə̀] - "near to this one" (not *[ɲ̰ɘ́ʀʌ́ˈɕɪ́sə̀]).

==Pronouns==

Thangha' borrowed its pronouns from Austronesian languages, and makes a distinction between [https://en.wikipedia.org/wiki/Clusivity inclusive and exclusive "we"]. Unlike the rest of the language which has ergative-absolutive alignment, pronouns have [https://en.wikipedia.org/wiki/Active%E2%80%93stative_language Fluid-S Alignment]. In transitive sentences, subjects have the agentive case, while objects have the patientive case e.g.

/ˈʔəkfa ʔʎə-ˈbedˤə qaˈɲəʔja/ (H) (HL) (H)

[ʔə̋kfá ʎ̰ɪ́ˈbɛ́dˤʌ̀ qɐ́ˈɲɪ́j̰æ̀]

1PS.SG.AGT ATEL-bite 3PS.PAT

I was biting him/her/them.

However, in intransitive sentences, pronominal subjects only have the agentive case if they perform the action of their own free will e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground (deliberately, e.g. to duck a bullet)

If the action is involitional, pronominal subjects of an intransitive verb take the patientive case e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

The various pronouns are listed in the table below:

{| class="wikitable"
|-
! !! Agentive !! Patientive !! Genitive
|-
| 1PS Singular || /ˈʔəkfa/ (H)/ [ˈʔə̋kfá] (H)] || /ˈʔaɟəⁿ/ (HL) [ˈʔæ̋ɟèⁿ] || /kfo/ (H) [kfő]
|-
| 1PS Plural Exclusive || /qaˈmʲə/ (L) [qɐ̄ˈmʲɪ̏] || /ˈʔəʔmʲəⁿ/ (L) [ˈʔɪm̰ʲeⁿ] || /ˈʔnəʔmʲəⁿ/ (L) [ˈn̰ɪm̰ʲeⁿ]
|-
| 1PS Plural Inclusive || /ˈtˤəɥa/ (H) [ˈtˤø̋ɥɞ́] || /ˈʔətsəⁿ/ (HL) [ˈʔə̋tsə́ⁿ] (HL) || /ˈʔnətsəⁿ/ (HL) [ˈn̰ə̋tsə́ⁿ]
|-
| 2PS Singular || /ʔjəˈqo/ (H) [j̰ɘ́ˈqɔ̋] || /ʔjəˈʔɥo/ (H) [j̰ʏ́ˈɥ̰ʉ̋] || /ʔmo/ (H) [m̰ő]
|-
| 2PS Plural / Polite || /qəˈʔɥo/ (H) [qǿˈɥ̰ʉ̋] || /ʔjəˈʔɲʷo/ (H) [̰jʏ́ˈɲ̰ʷʉ̋] || /ʔnəˈʔɲʷo/ (H) [̰n̰ʏ́ˈɲ̰ʷʉ̋]
|-
| 3PS || /ˈɕəʔja/ (H) [ˈɕɪ́j̰æ̀] || /qaˈɲəʔja/ (H) [qɐ́ˈɲɪ́j̰æ̀] || /ˈʔɲəʔja/ (H) [ˈɲ̰ɪ́j̰æ̀]
|-
| Reflexive || N/A || /ˈɕeɳə/ (HL) [ˈɕéɳʌ̀] || /ʔwaⁿ/ (L) [wȁⁿ]
|-
| Reciprocal || N/A || /ˈɕaⁿkfə/ (HL) [ˈɕǽŋkfə̀] || /ˈtəʔʎə/ (HL) [ˈtɪ́ʎ̰ɪ̀]
|}

The genitive pronouns are used for both alienable and inalienable possession. In both cases, they precede the noun they possess e.g. /kfo ˈpəʔwəⁿ/ (H) (HL) [kfő ˈpʊ́w̰ʊ̀ⁿ] - "my bone" (either in my body, or in my possession).

The second person plural pronouns can also be used to refer to a single person, if the speaker desires to express respect to that person. This is similar to French tu vs. vous.

The reflexive patientive pronoun is used where English would say "myself", "yourself" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕeɳə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕéɳʌ̀]

ERG-dog ATEL-bite REFL.PAT

The dog was biting itself.

The reflexive genitive pronoun is used where English would say "my own", "your own" etc. For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ʔwȁⁿ ˈpəʔwəⁿ/ (HL) (HL) (L) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ʔwaⁿ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite REFL.GEN bone

The dog was biting its own bone.

The reciprocal patientive pronoun is used where English would say "each other" or "one another". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈɕaⁿkfə/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈɕǽŋkfə̀]

ERG-dog ATEL-bite RECP.PAT

The dogs were biting each other.

The reciprocal genitive pronoun is used where English would say "each other's" or "one another's". For example:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtəʔʎə ˈpəʔwəⁿ/ (HL) (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtɪ́ʎ̰ɪ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite RECP.PAT bone

The dogs were biting each other's bones.

==Verbs==

All verb conjugation is done using prefixes. Verbs do not inflect for tense or mood, but [https://en.wikipedia.org/wiki/Telicity Telicity] plays a central role in verb conjugation. Verbs can also be inflected into a number of different grammatical voices:

{| class="wikitable"
|-
! Voice !! Telic !! Atelic !! Negative !! Prohibitive
|-
| Default || ʔmʲaɲə- || ʔʎə- || ʔna- || tʷaⁿ-
|-
| Antipassive || ʔmʲaⁿma- || ʔʎama- || ʔnama- || N/A
|-
| Benefactive Applicative || ʔmʲaɲəqa- || ʔʎəka- || ʔnaka- || tʷaⁿka-
|-
| Instrumental Applicative || ʔmʲaɲəʔpaⁿ- || ʔʎəpaⁿ- || ʔnapaⁿ- || tʷaⁿpaⁿ--
|}

However, for verb roots that begin with voiceless obstruents all the prefixes except the default telic voice prefix /ʔmʲaɲə-/ cause lenition to voiced obstruents e.g.

/ˈpedˤə/ (HL) - "to bite" (root) -> /ʔmʲaɲə-ˈpedˤə/ (HL) - "to bite" (telic), but /ʔʎə-ˈbedˤə/ (HL) - "to bite" (atelic).

The prohibitive forms are used to give negative commands e.g.

/tʷaⁿ-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[tʷámˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

PROH-bite bone.ABS

Don't bite the bone.

===Default Voice===

====Transitive Sentences====

For transitive sentences, the most common word order is [Subject] Verb [Direct Object] [Prepositional Objects] e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting a / the bone.

Note that this construction is only used when a) the subject is definite (where English would use "the") and b) the subject is not topicalised (i.e. in cases where Japanese would use "ga" and not "wa"). See https://en.wikipedia.org/wiki/Topic_and_comment, https://en.wikipedia.org/wiki/Topic-prominent_language and https://en.wikipedia.org/wiki/Japanese_grammar#Topic.2C_theme.2C_and_subject:_.E3.81.AF_wa_and_.E3.81.8C_ga.

=====Subject Topicalisation=====

In transitive sentences where the subject is definite and is also topicalised (i.e. in a case where English would mark it with "the" , and Japanese would mark it with "wa"), the subject takes the absolutive case e.g.

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting a / the bone.

This sentence differs from the previous one (which has the ergative marker pə-) in that, without pə-, it simply describes the dog and what it is doing, whereas with pə-, it is more likely to be an answer to a question (overt or implied) asking something like "Who / What was biting the bone?"

=====Object Topicalisation=====

Objects of such transitive sentences can be topicalised by moving them to the front of the sentence (this precludes subject topicalisation, a sentence can only have one topic). This is usually used in situations where English would use the passive voice e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.ABS ERG-dog ATEL-bite

The bone was being bitten by the dog.

=====Subject Deletion=====

Thangha' is a [https://en.wikipedia.org/wiki/Pro-drop_language Pro-drop language]. In particular, it is common to delete the subject of transitive sentences if it is obvious from the context e.g.

/ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL)

[ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ATEL-bite bone.ABS

It (or he/she/they) was biting a / the bone.

====Intransitive Sentences====

In intransitive sentences, word order depends on whether the subject is a pronoun, or a full noun phrase

=====Full Noun Phrases=====

In this case, subjects come before the verb if they are topicalised e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what the shaman did)

If the subject is not topicalised, it comes after the verb e.g.

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground)

=====Pronouns=====

In intransitive sentences, subject pronouns always come before the verb e.g.

/ˈʔəkfa ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.AGT TEL-fall

I dropped to the ground.

This applies even if the subject pronoun is in the patientive case (to indicate that the action was involitional) e.g.

/ˈʔaɟəⁿ ʔmʲaɲə-ˈpfaɫə/ (H) (HL)

[ˈʔæ̋ɟèⁿ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

1PS.SG.PAT TEL-fall

I fell to the ground (by accident)

=====Subject Deletion=====

As with transitive sentences, subjects of intransitive sentences can be freely deleted if they are obvious from context e.g.

/ʔmʲaɲə-ˈpfaɫə/ (HL)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

TEL-fall

I/you/he/... fell to the ground.

====Ditransitive Verbs====

The subject of a ditranstive verb takes the same case as normal, and both the direct and indirect objects take absolutive case. However, the indirect object is also marked with the prepositional clitic /tə/. Unlike the case prefixes, this does not cause lenition e.g.

/ˈʔəkfa ʔmʲaɲə-ˈtena ˈpəʔwəⁿ tə=ˈtʷogə/ (H) (H) (HL) (HL)

[ˈʔə̋kfá m̰ʲǽɲɪ́ˈte̋ná ˈpʊ́w̰ʊ̀ⁿ tʊ́ˈtʷùgə̀]

1PS.SG.AGT TEL-give bone DAT=dog

I gave a bone to the dog.

===Antipassive Voice===

This is only applicable to transitive verbs. It deletes the object of the verb, and places the subject of the verb in the absolutive case. https://en.wikipedia.org/wiki/Antipassive_voice. The situation when it is used depends on whether the subject of the verb is a full noun phrase or a pronoun.

====Full Noun Phrases====

There are two specific usages of the antipassive when the subject is a full noun phrase. One is in [https://en.wikipedia.org/wiki/Gnomic_aspect gnomic utterances], and the other is to indicate that the subject of a transitive sentence is indefinite

=====Gnomic Utterances=====

To talk about general truths, the antipassive is used, with the subject occuring before the verb e.g.

/ˈtʷogə ʔʎama-ˈbedˤə/ (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀]

dog.ABS ANTIP.ATEL-bite

Dogs bite.

The object of the sentence can resurface after verb, marked with the dative prepositional clitic /tə/ e.g.

/ˈtʷogə ʔʎama-ˈbedˤə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷùgə̀ ʎ̰æmaˈbɛ́dˤʌ̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS ANTIP.ATEL-bite DAT=bone

Dogs bite bones.

=====Indefinite Subjects=====

The default voice can only be used for transitive verbs if the subject is definite. If the subject is indefinite, then the verb is antipassivised, and the subject reappears after the verb e.g.

/ʔʎama-ˈbedˤə ˈtʷogə/ (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀]

ANTIP.ATEL-bite dog.ABS

A dog was biting (me / you / him / her / it ...)

As with before, the object of the sentence can be placed at the end, marked with /tə/ e.g.

/ʔʎama-ˈbedˤə ˈtʷogə tə=ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ʎ̰æmaˈbɛ́dˤʌ̀ ˈtʷùgə̀ tə́ˈpʊ́w̰ʊ̀ⁿ]

ANTIP.ATEL-bite dog.ABS DAT=bone

A dog was biting the bone.

====Pronominal Subjects====

Transitive verbs with pronomninal subjects are antipassivised in two cases:

=====Object De-Emphasis=====

If the subject of a transitive verb is a pronoun, and it is considered to be more integral to the action being performed than the object, then the verb is antipassivised e.g.

/ʔjəˈqo ʔmʲaⁿma-ˈɕuːwə/ (H) (HL)

[j̰ɘ́ˈqɔ̋ m̰ʲǽmmǽ-ˈɕʉ̂ːwʊ̀]

2PS.SG.AGT ANTIP.TEL-rescue

You saved it / him / her (with an feeling that only the listener could have performed the rescue)

If the verb does not take the antipassive i.e. /ʔmʲaɲə-ˈɕuːwə/, there is no nuance that the rescue was only something that the listener could have performed.

=====Involitional Action=====

If a transitive verb has a pronominal subject that performs the action involitionally, then the verb is antipassivised and the subject takes the patientive case e.g.

/qaˈɲəʔja ʔʎama-ˈbedˤə/ (H) (HL)

[qɐ́ˈɲɪ́j̰æ̀ ʎ̰æmaˈbɛ́dˤʌ̀]

3PS.PAT ANTIP.ATEL-bite

He/she/they was/were biting (with a nuance that the biting was uncontrallable, maybe they had rabies or were zombified or in an uncontrollable rage etc.)

===Benefactive Applicative Voice===

Thi sused to highlight the beneficiary of an action. It can apply to intransitive verbs i.e. changing the sentence "A did something" into "A did something for B". It can also apply to transtive verbs i.e. changing the sentence "A did something to X" into "A did something to X for B". In both cases, the orginial absolutive argument of the verb is shifted to the end of the sentence, and is marked with the prepositional clitic /dəʔ/. For example, if we take the following intransitive sentence:

/ʔmʲaɲə-ˈpfaɫə ˈxeⁿda/ (HL) (H)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈχɛ̋ndá]

TEL-fall shepherd.ABS

A shepherd dropped down.

The above sentence can take the applicative voice to become.

/ʔmʲaɲəqa-ˈbvaɫə ˈtʷoʀa tə=ˈxeⁿda/ (HL) (L) (H)

[m̰ʲǽɲɘ́qɑ́-ˈbvɑ́ɫʌ̀ ˈtʷȍʀɑ̄ tʌ́ˈχɛ̋ndá]

TEL.APPL-fall shaman.ABS DAT=shepherd

A shepherd dropped down for the shaman (maybe prostrating himself).

Either /ˈtʷoʀa/ - "shaman" or /tə=ˈxeⁿda/ - "shepherd" (dative) can be topicalised (moving to before the verb). Both can be deleted if they are obvious from context. So /ʔmʲaɲəqa-ˈbvaɫə/ could stand as a sentence by itself.

Below is an example of applying the applicative voice to a transitive verb:

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

/pə-ˈdʷogə ʔʎəka-ˈbedˤə ˈxeⁿda tə=ˈtʷoʀa/

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́káˈbɛ́dˤʌ̀ ˈχɛ̋ndá tʊ̄ˈtʷȍʀɑ̄/

ERG-dog ATEL.BEN.APPL-bite shepherd.ABS DAT=shaman

A dog was biting the shaman for the shepherd (maybe the shepherd ordered it to attack the boss).

===Instrumental Applicative Voice===

This functions in a similar way to the benefactive applicative voice. The noun that would normally take the plain absolutive case (i.e. the subject of an intransitive verb, or the direct object of a transitive verb) goes to the end of the sentence and takes the prepositional clitic /tə/. Whatever the subject used to complete the action takes the absolutive case, and is not marked with a preposition e.g.

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈtʷoʀa/ (HL) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈtʷȍʀɑ̄/

ERG-dog ATEL-bite shaman.ABS

A dog was biting the shaman.

becomes

/pə-ˈdʷogə ʔʎəpaⁿ-ˈbedˤə ˈʔɲəʔja ˈɕəʀə ˈtaⁿtə təʔ=ˈtʷoʀa/ (HL) (HL) (H) (L) (HL) (L)

[pʊ́ˈdʷùgə̀ ʎ̰ɪ́pám-ˈbɛ́dˤʌ̀ ɲ̰ɪ́j̰æ̀ʔ ˈɕɘ̏ʀɘ̄ ˈtântə̀ tʊ̄ˈtʷȍʀɑ̄]

ERG-dog ATEL.INSTR.APPL-bite 3PS.GEN sharp tooth.ABS DAT=shaman

A dog was biting the shaman with its sharp teeth.

==Copula==

The Thangha' copula /pʲə/ (L) is a verb that conjugates regularly like all of the others. Note that for it to correspond to English "to be", it must take the atelic aspect, so in the default voice it is /ʔʎə-ˈbʲə/. If it takes the telic aspect (i.e. becoming /ʔmʲaɲə-ˈpʲə/ in the default voice), it means "to become".

Note that the copula counts as a transitive verb, so the subject generally takes the ergative case e.g.

/pəˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[pə́ˈxe̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd-ERG ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "Who is the shaman?")

However, just as with other transitive verbs, the subject can be topicalised, in which case it loses the ergative marker e.g.

/ˈxeⁿda ʔʎə-ˈbʲə ˈtʷoʀa/ (H) (L) (L)

[ˈχɛ̋ndá ʎ̰ɪ̄ˈbʲɪ̏ ˈtʷȍʀɑ̄]

shepherd.TOP ATEL-be shaman.ABS

The shepherd is the shaman (answering the actual or implied question of "What does the shepherd do?")

===Predicative Adjectives===

However, when an adjective is used as a predicate, the copula verb is not used. Instead, the adjective is verbalised, taking the atelic prefix ʔʎə- and becoming an intranstive verb (so the subject will not take the ergative case) e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə/ (HL) (HL) (L)

[ˈtântə̀ wá-ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄]

tooth.TOP GEN-dog ATEL-sharp

The dog's teeth are sharp.

===Locative Verb===

Unlike English, to express location, Thangha' speakers use a different verb to the copula. This is /ˈsadˤə/ (HL), which usually takes an atelic prefix to become /ʔʎəˈzadˤə/. For example:

/ʔʎəˈzadˤə ˈʔjiːtə ʔɲəʀə=ˈtʷoʀa/ (HL) (HL) (L)

[ʎɪˈzɑ́dˤʌ̀ ˈj̰êːdˤʌ̀ ɲ̰ɘ̄ʀɔ̄ˈtʷȍʀɑ̄]

ATEL-LOC food.ABS near=shaman

The food is next to the shaman.

As can be seen in the above example, in contrast to the copula verb, the locative verb is intransitive, and the subject takes the absolutive case.

==Relative Clauses==

These always follow the noun that they modify, which they are separated from by the proclitic /xaⁿ/ e.g.

/ˈtʷogə xaⁿ=ʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ χɐ́ɲʎɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog.ABS REL=ATEL-bite bone.ABS

The dog that is biting the bone.

Inside a relative clause, subjects must always be marked with the ergative case marker e.g.

/ˈpəʔwəⁿ xaⁿ=pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ χɒ́mbʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone REL=ERG-dog ATEL-bite

The bone that the dog is biting.

==Numbers==

As in English, numbers precede the noun that they modify e.g.

/tfə ˈtʷogə/ (L) (HL)

[tfə̏ ˈtʷúgə̀]

two dog

Two dogs

To make ordinal numerals, prefix /sə/ to the numeral e.g.

/sə-tfə ˈtʷogə/ (L) (HL)

[sə̀tfə̏ ˈtʷúgə̀]

ORD-two dog

The second dog

However, the Thangha' word for "first" is irregular. Whilst the numeral for "one" is /waⁿ/ (L), the word for "first" is /ˈfəⁿɽə/ (L)/.

==Questions==

===Polar Questions===

A statement is turned into a polar question using the particle /pa/ (H). It comes at the beginning of the sentence, except if a noun is topicalised, in which case it follows that noun e.g.

/ˈtʷoʀa ʔmʲaɲə-ˈpfaɫə/ (L) (HL)

[ˈtʷòʀɑ̄ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP TEL-fall

The shaman dropped to the ground (where the main point of the sentence is to say what he did)

/ˈtʷoʀa pa ʔmʲaɲə-ˈpfaɫə/ (L) (H) (HL)

[ˈtʷòʀɑ̄ pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀]

shaman.TOP Q TEL-fall

Did the shaman drop to the ground?

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (HL) (L)

[m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

TEL-fall shaman.ABS

The shaman dropped to the ground (where the main point of the sentence is to say who dropped to the ground).

/ʔmʲaɲə-ˈpfaɫə ˈtʷoʀa/ (H) (HL) (L)

[pa̋ m̰ʲǽɲɪ́ˈpfɑ́ɫʌ̀ ˈtʷòʀɑ̄]

Q TEL-fall shaman.ABS

Did the shaman drop to the ground? (or was it somebody else).

/ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to describe what the dog was doing)

/pa ˈtʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (H) (HL) (HL) (HL)

[pa̋ ˈtʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q dog ATEL-bite bone.ABS

Was the dog biting the bone?

/pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-bite bone.ABS

The dog was biting the bone (where the main emphasis of the speaker is to point out that the dog was biting the bone, and not a cat)

/pa pə-ˈdʷogə ʔʎə-ˈbedˤə ˈpəʔwəⁿ/ (HL) (HL) (HL)

[pa̋ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ ˈpʊ́w̰ʊ̀ⁿ]

Q ERG-dog ATEL-bite bone.ABS

Was the dog biting the bone? (or was it something else)

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (HL) (HL) (HL)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

The bone was being bitten by the dog.

/pa ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈbedˤə/ (H) (HL) (HL) (HL)

[pa̋ ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀]

bone.TOP ERG-dog ATEL-bite

Was the bone being bitten by the dog?

===Wh-Phrases===

Unlike English, Thangha' is a wh-in-situ language, i.e. it lacks [https://en.wikipedia.org/wiki/Wh-movement Wh-movement] e.g.

/ˈtʷogə ʔʎə-ˈbedˤə cəⁿ/ (HL) (HL) (L)

[ˈtʷùgə̀ ʎ̰ɪ́ˈbɛ́dˤʌ̀ cȅⁿ]

dog-TOP ATEL-bite what

What was the dog biting?

Thangha's equivalents of English "wh-words" are all derived from /cəⁿ/ - "what" e.g.

/cəⁿ ˈpuːmʲə/ (L) (HL) [cȅⁿ ˈpʉ̂ːmʲɪ̀] - "where" (literally "what place")

/cəⁿ ˈɕiːdʑaⁿ/ (L) (HL) [cȅⁿ ˈɕîːdʑæ̀ⁿ] - "when" (literally "what time")

/ˈcənaⁿ/ (HL) [ˈcɪ́nàⁿ] - "who" (a contraction of /ceⁿ ʔaⁿ/ - "what person")

/cəⁿ fʲe/ (L) (H) [cȅⁿ fʲi̋] - "how" (literally "what method")

/cəⁿ ʔʎəfiː/ (L) (H) [cȅⁿ ʎ̰ɪ́ˈfʲi̋ː] - "why" (literally "what reason")

Thangha' has two ways to ask questions where English would use "whose", depending on whether the possession is alienable or inalienable. In the case of inalienable possession, /cəⁿ/ takes the genitive prefix /ʔwa-/, becoming /ʔwa-ˈɟəⁿ/ (L), and coming after the noun it possesses e.g.

/ˈpəʔwəⁿ ʔwa-ˈɟəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿˈw̰ɞ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

In the case of alienable possession, /cəⁿ/ takes the ergative prefix /pə-/, becoming /pə-ˈɟaⁿ/ e.g.

/ˈpəʔwəⁿ pə-ˈcəⁿ/ (HL) (L)

[ˈpʊ́w̰ʊ̀ⁿ pɪ̄ˈɟȅⁿ]

bone.ABS GEN-what

Whose bone? (in their body)

==Predicative Possession==

===Standard Intransitive Construction===

To say "A has B" in Thangha', the standard way is for A (the possessor) to appear at the start of the sentence as a topic (in the absolutive case), followed by the verb /ˈtaʔa/ (H) - "to exist", taking the atelic prefix /ʔʎə-/, becoming /ʔʎə-ˈdaʔa/ (H). Finally, the possessed noun appears, also in the absolutive case. For example:

/ˈtʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈtʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

dog.TOP ATEL-exist bone.ABS

The dog has a bone.

===Transitive Construction===

The above is the most common way to express predicative possession in Thangha', treating /ˈtaʔa/ as an intransitive verb. However, the language is undergoing a process of "have-drift", due to contact with other languages which have a transitive verb just like English "have". Recently, many speakers have begun to use the above construction interchangeably with a second construction, where the possessor is in the ergative case:

/pə-ˈdʷogə ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

ERG-dog ATEL-have bone.ABS

The dog has a bone.

Speakers tend to use the transitive construction when they wish to put focus on the possessor i.e. in the above case, emphasise that it is the dog that has the bone, and not a cat / human etc. that has the bone. Speakers also use the transitive construction when they wish to topicalise the possessed noun e.g.

/ˈpəʔwəⁿ pə-ˈdʷogə ʔʎə-ˈdaʔa/ (HL) (HL) (H)

[ˈpʊ́w̰ʊ̀ⁿ pʊ́ˈdʷúgə̀ ʎ̰ɪ́ˈda̋ʔá]

bone.TOP ERG-dog ATEL-have

The bone belongs to the dog.

However, /ˈtaʔa/ has not yet become a fully fledged transitive verb. For example, unlike other transitive verbs, it cannot take an antipassive prefix to become */ʔʎama-ˈdaʔa/.

===With Pronouns===

When a pronoun is the possessor, it takes the agentive case if the possession is alienable, and the patientive case if the possession is inalienable e.g.

/ˈʔəkfa ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (H) (H) (HL)

[ˈʔə̋kfá ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.AGT ATEL-have bone

I have a bone (e.g. I'm holding it).

/ˈʔaɟəⁿ ʔʎə-ˈdaʔa ˈpəʔwəⁿ/ (HL) (H) (HL)

[ˈʔæ̋ɟèⁿ ʎ̰ɪ́ˈda̋ʔá ˈpʊ́w̰ʊ̀ⁿ]

1PS.SG.PAT ATEL-have bone

I have a bone (e.g. in my body).

==Comparative Constructions==

The prepositional clitic /ʔapəⁿ/ literally means "above", but is also used to mean "more than" e.g.

/ˈtaⁿtə wa-ˈdʷogə ʔʎəˈʑəʀə ʔapəⁿ=ˈʔɲəʔja ˈpəʔwəⁿ/ (HL) (HL) (L) (H) (HL)

[ˈdantəʔ wɔ́ˈdʷúgə̀ ʎ̰ɪ̄ˈʑɘ̏ʀʌ̄ ʔápɪ́ɲ=ɲ̰ɪ̋j̰æ̀ ˈpʊ́w̰ʊ̀ⁿ]

tooth.TOP GEN-dog ATEL-sharp above 3PS.GEN bone

The dog's teeth are sharper than its bones.