Random phonology/phonemic inventory thread

[Knox Adjacent]

Germ of an idea:

/m n
m̰ n̰
pʰ tʰ kʰ
p t k ʔ
b d g ʔ̰
s h
z
ɾ/

Less germane, standard 5 vowels plus centering diphthongs contrasting modal vs creaky voice.

[DV82LECM]

Germ of an idea:

/m n
m̰ n̰
pʰ tʰ kʰ
p t k ʔ
b d g ʔ̰
s h
z
ɾ/

Less germane, standard 5 vowels plus centering diphthongs contrasting modal vs creaky voice.

I don't think I've ever seen anyone use /ʔ̰/.

Also, yes, 'tis a germ, sir!

[Creyeditor]

I had a diachronic phonology idea recently. What if we start with a bisyllabic root proto-lang and then take two radically different routes on what gets altered. Imagine three proto-lang roots *kwusa, *kita, *bome

Lang 1 does intervocalic lenition first. This gives you glottal fricatives, glottal stops and nasalized glides.

*kwusa > kwuha
*kita > kiʔa
*bome > boũe

Then Lang 1 goes through a sequisyllabic stage where the initial-syllable vowel is reduced to schwa.

kwuha > kwəha
kiʔa > kəʔa
boũe > bəũe

When the reduced vowel is eventually deleted, the consonants fuse. Previously intervocalic /h/ yields aspirated stops, glottal stop yield ejectives and nasalized glides yield prenasalized stops by metathesis.

kwəha > kʰʷa
kəʔa > k'a
bəũe > mbe

Lang 2 goes a radically different route. First, glides and vowels trigger place assimilation on preceding consonants and labiovelars become labials.

*kwusa > pusa
*kita > cita
*bome > bome

Then vowels assimilate regressively in some vowel quality features.

pusa > posa
cita > ceta
bome > bøme

Finally, labial plosives become affricates before rounded vowels and final vowels are reduced.

pusa > pfosə
cita > cetə
bome > bvømə

This leads to the following beautiful cognate pairs.

pfosə - kʰʷa
cetə - k'a
mbe - bvømə

[Arayaz]

Beautiful.

[Omzinesý]

I had a diachronic phonology idea recently. What if we start with a bisyllabic root proto-lang and then take two radically different routes on what gets altered. Imagine three proto-lang roots *kwusa, *kita, *bome

Lang 1 does intervocalic lenition first. This gives you glottal fricatives, glottal stops and nasalized glides.

*kwusa > kwuha
*kita > kiʔa
*bome > boũe

Then Lang 1 goes through a sequisyllabic stage where the initial-syllable vowel is reduced to schwa.

kwuha > kwəha
kiʔa > kəʔa
boũe > bəũe

When the reduced vowel is eventually deleted, the consonants fuse. Previously intervocalic /h/ yields aspirated stops, glottal stop yield ejectives and nasalized glides yield prenasalized stops by metathesis.

kwəha > kʰʷa
kəʔa > k'a
bəũe > mbe

Lang 2 goes a radically different route. First, glides and vowels trigger place assimilation on preceding consonants and labiovelars become labials.

*kwusa > pusa
*kita > cita
*bome > bome

Then vowels assimilate regressively in some vowel quality features.

pusa > posa
cita > ceta
bome > bøme

Finally, labial plosives become affricates before rounded vowels and final vowels are reduced.

pusa > pfosə
cita > cetə
bome > bvømə

This leads to the following beautiful cognate pairs.

pfosə - kʰʷa
cetə - k'a
mbe - bvømə

Doesn't mbe derive more easily from bome than from bəũe?

[Omzinesý]

p' t̪' tˠ' k'
pʰ t̪ʰ tˠʰ kʰ
p t̪ tˠ k
m n̪ nˠ
s sˠ
l̪ lˠ
ɟ͡ʝ~j g͡ɣ~ɣ g͡w~w

[Creyeditor]

Doesn't mbe derive more easily from bome than from bəũe?

True. I was aiming for uniformity and parallelism in sound change. Also, there is if course an ANADEW, with PIE *kʷíd yields Welsh pa /paː/, Polish cz- /t͡ʃ/, Greek τι /ti/, and Italian che /ke/.

[VaptuantaDoi]

Code: Select all

    t̪     k  ʔ
 β  ð  s
 w  ð̞  ɾ

(C that isn't /ʔ/)(ɾ)V(N,ʔ,l)

/nβ nt̪ nð ns nk/ → [mb n̪t̪ n̪d̪ nt ŋk]
/nw nð̞/ → [m n̪]
/ɾβ ɾt̪ ɾð ɾs ɾk/ → [lb l̪t̪ l̪d̪ ɾt lɡ]

or something idk

[Porphyrogenitos]

Yet another phonological sketch for a monosyllabic isolating-analytic language. Initially I had wanted to do something more Arabic or generally Semitic inspired, but then it took on more of a Turkish and, in fact, Volapük inspiration.

/m n ŋ/ <m n ŋ>
/t ʔ/ <t ∅>
/b d g/ <b d g>
/dʒ/ <j>
/s ʃ χ h/ <s ş x h>
/v z/ <v z>
/l r j/ <l r y>

/i y u/ <i ü u>
/e ø o/ <e ö o>
/a/ <a>

All syllables are CVC. /ʔ/ is not notated in the orthography, but compound words are written with spaces, as in Vietnamese.

/ŋ/ only occurs word-finally.

The only notable allophonic process is that /v/ becomes [w] in the coda.

/dʒ/ originated from historic /g/; historic /k/ split into current /χ/ (a uvular) and /g/. Possibly there was also a historic /f/, maybe ultimately from an even older /p/, which split into /h/ and a phone that merged with /v/. /ŋ/ probably originated from a lexicalized assimilatory process.

There may be some small bits of sandhi here and there; but I have not decided what, if any.

A sample generated by lingweenie's lexifier:

Rud az hit yaz giz, jis gaj gij viy jüv jin? Gir göt duz lij a zat. Va
xun gun ya xa dix jit. Het hiz gu naj gid dan jov gax jer, jet. Ji doz
xeg je la nod. Jis nit gex get nöh, lel vöx vid raj dis ad rin. Tun
giv gir git naz, git xüt jot ve is öz? Göx deg o dig get jal jer. Jax
gux hav jan. Tur get jom dix jix jih, i zöŋ. Xid den rid hez, i jit
xiv jiz? Ga an jeg tot gur vil gim jud xot, xal. Vat it rag le lin gej
rit? Gaj ger, rat nut xij na vin yiv vöy yid yar. Jaz vet miz iz gur
git in. Get vöz xit, jar hö nin nat on jin za mit yir. Raz zar u i val
a zan. Hux rün ril, joj ge gi han git rer u jir nin. Vit uh get git
nax an döx. Jij jiv söl gor dir nun jax. Jeh ha nat u xex jar, ge sin.
Lo gat yan gil öj net gi, döx nal gag? Rül jir von don dar dih tel,
de. At ge ji dal giv uz. Nüz tön sej ruj ha duy zir.

Yes, it's supposed to be rather ugly.

[VaptuantaDoi]

/t d/
/β j w/

/i y u e ɔ a/

/dytje βɔ tade tɔβei tɔdwy βwadwa atata jetwe jatatji uja jɔ βju ja dwɔtɔdβɔ tβɔtjae edɔtyta twɔdjɔdydwɔβwɔ/
[nø̃c͡ɕe ɸo tale tɔβəi̯ tɔɡuy̯ mwælwæ atʰatʰa jekʰue̯i̯ ʄatʰac͡ɕi ujɑ jo mʎu ʄa nwotʰoʎβ̞ɔ cβ̞ʌc͡ɕɑe̯ elɔtʰøtʰa kwoɟ͡ʑɔlølwob͡βwo]

[Shemtov]

I posted before about a language called Fulcok Hülç. I have scrapped much of that language, as it doesn't fit with what I want to do with it, but I have created a phonology for the second stage in the language's (meaning the core idea, and assigned to the same conculture) creation, this time called Fulcok Hülçüün or Hülçüün Caıl.

/b t d t͡ʃ d͡ʒ k/ <b t d ç c k>
/m n nʲ / <m n ņ  >
/f s z θ ð ʃ ʒ x ɣ/ <f s z ţ ḑ ș j h g>
/r ɾʲ l lʲ / <r ŗ l ļ>
/β̞ j/ <v y>


/i y ɯ u/ <i ü ı u>
/i: y: ɯ: u:/ <ii üü ıı uu>
/e̞ ø̞ ʌ̝ o̞/ <e ö å o>
/e̞: ø̞: ʌ̝: o̞:/ <ee öö åå oo>
/ɐ/ <ă>
/æ ɑ / <ä œ a å>
/æ: ɑ: / <ää œœ aa åå>
/æi æy ɑu ɑɯ / <äi äü au aı >
/ɐi ɐy ɐɯ ɐu/ <ăi ăü ăı ău>
/iɐ yɐ ɯɐ uɐ/ <iă üă ıă uă>

There is front/back harmony. High vowels have rounding harmony in affixes. Mid-vowel affixes may have rounding affixes or not, depending on the affix. Some affixes are "palatal sensitive", meaning after a non-palatal(ized) consonant or a vowel , the affix begins with /j/ <y>, but after a palatal(ized) consonant, the affix begins with /ð/ <ḑ>. The Interdentals are treated like palatals, since only realatively recently has the following sound change taken place: /θ ð/> /s z/ /θʲ ðʲ/>θ ð/. In some roots final /ɣ/ <g> may change to /nʲ/<ņ> before a vowel-initial affix, given the sound change /_ŋ/>/ɣ/ elsewhere /ŋ/>/ɲ/>/nʲ/.

[VaptuantaDoi]

I am become Nortaneous, the Norter of phonologies

Code: Select all

    t  c  k
 β  ɾ     ɣ
 m  n  ɲ  ŋ

Surface vowels may well be /i u e o a/. Also /mβ nt ɲc ŋk ββ tt cc kk/ are [mb nd ɲɟ ŋg pː tː sː kː].

The main shpiel of the language is that phonological words are entirely distinct from syntactic words. Syntactic words – minimal segments that can be moved around without producing ungrammatical forms – are invariable based on templating morphology. Phonological words however only break where two heavy syllables (of the type CVQ or CVN) come together, regardless of their syntactic connection, or at a pause. They then rearrange themselves into sequences of iambs. This produces fairly opaque surface forms; in the example below, different syntactic words are marked in alternating red and black, while phonological words get spaces:

Nahpičintitintiihti nganngiwantamimpiíowahčewahtoȝihtitannga winngi ȝonnañuntaintutempitenniihteihtomintetóo čempeempeée wahtetompo.

[ˈnaʰpˑiˌcinˑdiˌtin.diˌʔiʰtˑɨ ˈŋaŋːiˌβanˑdaˌmimˑbiˌʔiːʔoˌβaʰcˑeˌβaʰtˑoˌɣiʰtˑiˌtaŋːə ˈβiŋːɨ ˈɣonːaˌɲunˑdaˌʔinˑduˌtemˑbiˌtenːiˌʔiʰtˑeˌʔiʰtˑoˌminˑdeˌtoːʔɘ ˈcemˑbeˌʔemˑbeˌʔeːʔɘ ˈβaʰtˑeˌtomˑbɘ]

In fact coincidentally none of the boundaries line up at all here; sometimes there's better correlation. This is intended for the next conlang relay.

[Arayaz]

This is evil. I love it.