Post your conlang's phonology

[Travis B.]

Alla

Consonants

Pulmonic stops: /p t k q ʔ/ ⟨p t k q ʔ⟩
Ejective stops: /tʼ kʼ qʼ/ ⟨tʼ kʼ qʼ⟩
Pulmonic affricates: /ts tʃ/ ⟨c č⟩
Ejective affricates: /tsʼ tʃʼ/ ⟨cʼ čʼ⟩
Fricatives: /s ʃ χ h/ ⟨s š x h⟩
Nasals: /m n/ ⟨m n⟩
Liquids: /r l/ ⟨r l⟩
Semivowels: /ʋ j/ ⟨v y⟩

All consonants can be both short and long.

Vowels

Short: /æ i ə ɒ u/ ⟨a i ə å u⟩
Long: /æː iː ɒː uː/ ⟨ā ī â ū⟩

Stress

The syllable of the last three syllables in a word that is of the highest weight and which is closest to the antepenultimate position is stressed.

Phonotactics

Syllables take the form CV(C), except for final syllables, which may be CVCC. Geminates can only occur intervocalically, and count as a coda C of one syllable combined with a onset C of the following syllable (even though historically geminates could also occur stem-finally, even though this does not survive into present-day forms).

Phonology

All short obstruents except /q/ and /ʔ/, in intervocalic positions, between a semivowel and a vowel, or following a nasal and preceding a vowel except at the starts of a stressed syllables are voiced.

Also, /æ æː i iː ə u uː/ are realized as [ɑ ɑː e eː ɐ o oː] adjacent to uvular consonants.

Stem Alternations

Historically stress was phonemic, and shifted depending on the morphemes attached to a stem, resulting in significant alternations in word-form when combined with vowel reduction, syncope, and elision of unstressed intervocalic historical voiced fricatives (versus their modification into other consonants), vowel nasalization, nasalization preventing reduction, subsequent coda nasal elision and loss of vowel nasalization (obscuring the behaviour of historical nasal vowels synchronically and thus making vowel reduction behavior unpredictable synchronically), and collapse of historical voiceless lateral fricatives into /s/ versus /ʃ/ depending on whether it fell in an onset or coda position. Also historically there was both i-umlaut and u-umlaut, which affected not only the form of stems internally, but also both the form of prefixes attached to stems and the forms of stems as per suffixes attached to them. Historically there was also palatalization, taking place in many of the same environments as i-umlaut. Both of these occurred prior to vowel reduction and syncope, which hides their conditioning environments in many cases, and makes palatalized consonants phonemic.

[Shemtov]

Proto-Arctic:
/p pʰ t tʰ t' k kʰ k' q qʰ ʔ/ <b p d t t' g k k' q qh '>
/m n/ <m n>
/s z h ɦ/ <s z h ḥ>
/t͡s'/ <ts'>
/ɬ ɮ/ <ł l>
/r/ <r>
/w j/ <w y>

/ i ɨ u/ <i ï u>
/ e o/ <e o>
/ə ə̃ ɵ/ <ä ą ö>
/a/ <a>

/ɛɪ ɛʊ ɔɪ aɪ aʊ/ <ei eu oi ai au>

Phonotactics:
(s)(C)V(N)
An s-cluster cannot occur with another fricative or an ejective


Kalahŕi:
/t̪ d t̪ʼ ʈ ɖ ʈʼ k͡p g͡b k͡p' q qʼ/ <t d t' ṭ ḍ ṭ' k g k' q q'>
/n ɳ ŋ͡m/ <n ṇ m>
/θ ʂ x͡ɸ/ <th s f>
/ɬ̪ ꞎ/ <l ḷ>
/tθ~ts tθ'~ts' ʈʂ ʈʂ' tɬ tɬ' ʈꞎ ʈꞎʼ/ <z z' ẓ ẓ' x x' x̣ x̣'>
/r ʀ͡ʙ/ <r ŕ>
/w/ <w>

/ɪ̆ ʏ̆ ɔ̆ ə̆/ <ih yh oh ah>
/i y o a/ <i y o a>
/i: y: o: a:/ <ii yy oo aa>

Phonotactics:
CVC

Allophony:
/tθ tθ'/ realized as /ts ts'/ word-finally. Front rounded vowels backed after uvulars.

[Travis B.]

/r ʀ͡ʙ/ <r ŕ>

/r ʀ͡ʙ/

ʀ͡ʙ

How do you expect your conpeople to be sufficiently capable of pronouncing this, such that this will actually be stable to any significant degree?

[Shemtov]

/r ʀ͡ʙ/ <r ŕ>

/r ʀ͡ʙ/

ʀ͡ʙ

How do you expect your conpeople to be sufficiently capable of pronouncing this, such that this will actually be stable to any significant degree?

Honestly, I can produce such a coarticulated trill, actually with more ease then a coarticulated stop.

[Travis B.]

Have you actually googled for ʀ͡ʙ? Three items total come up, and they are all pertain to conlanging, i.e. this is not a real speech sound.

[Shemtov]

Have you actually googled for ʀ͡ʙ? Three items total come up, and they are all pertain to conlanging, i.e. this is not a real speech sound.

1. Kalahŕi is an exolang, ie. not spoken by humans. It is spoken by a species, that, yes, has a human-like vocal tract, but are not human. The universals of human linguistics can hardly be applied to them.
2. I notice you chose to pick on /ʀ͡ʙ/ but the phonology also included /ʈꞎʼ/, which is not found in any natlang either. Heck, /ꞎ/ is found only in one natlang, I think, and it doesn't even have a dedicated IPA smbol- this is from the extensions to the IPA.

[Pole, the]

/r ʀ͡ʙ/ <r ŕ>

/r ʀ͡ʙ/

ʀ͡ʙ

How do you expect your conpeople to be sufficiently capable of pronouncing this, such that this will actually be stable to any significant degree?

I just spat all over myself trying to pronounce that.

[Travis B.]

Have you actually googled for ʀ͡ʙ? Three items total come up, and they are all pertain to conlanging, i.e. this is not a real speech sound.

1. Kalahŕi is an exolang, ie. not spoken by humans. It is spoken by a species, that, yes, has a human-like vocal tract, but are not human. The universals of human linguistics can hardly be applied to them.

Yes, that one's conlang is not spoken by humans - the deus ex machina of conlanging. That said, regardless of the morphology of your conbeings' vocal tracts, how do you reasonably expect them to be able to trill at two different POA simultaneously reliably?

2. I notice you chose to pick on /ʀ͡ʙ/ but the phonology also included /ʈꞎʼ/, which is not found in any natlang either. Heck, /ꞎ/ is found only in one natlang, I think, and it doesn't even have a dedicated IPA smbol- this is from the extensions to the IPA.

The problem is that your conphone I commented on is two different trills simultaneously, if you haven't realized that.

[Shemtov]

Proto-Atlantic:
/p pʰ b t tʰ d k kʰ kʷ kʷʰ k͡p k͡pʰ g͡b/ <p ph b t th d k kh kw kwh kp kph gb>
/f s sʰ x xʷ h/ <f s sh x xw h>
/t͡s t͡sʰ d͡z/ <c ch z>
/m n ŋ ŋʷ ŋ͡m/ <m n ng ngw nm>
/j w/ <y w>
/l/ <l>


/i y u ɛ œ ɔ a/ <i ü u e ö o a>

Its descendants:
/p t k kʷ/
/s x xʷ/
/t͡s/
/m n -ŋ/
/j w/
/l/

/i u ɔ a/

/˧ ˩/


_______

/p b t d k ʔ/
/f θ s ʃ x h/
/t͡ʃ d͡ʒ/
/m n ŋ/
/l/

/i u ɛ ɔ a/


__________

/b t d c k/
/f s z ʃ ʒ h/
/m n ɲ ŋ/
/j w/
/l/

/i u e o a/


_______

/p pˤ t tˤ k kʷ q qʷ/
/f v θ θˤ ð s sˤ ʃ ʃˤx ɣ ħ/
/t͡ʃ t͡ʃˤ d͡ʒ/
/l/
/r/
/j w/

/i u ə æ ɑ/

[Travis B.]

A descendent of Alla

Consonants

Stops: /p pʰ b t tʰ d k kʰ g q qʰ/ ⟨p ph b t th d k kh g q qh⟩
Affricates: /ts tsʰ dz tʃ tʃʰ dʒ/ ⟨c ch j č čh ǰ⟩
Fricatives: /f s z ʃ ʒ χ ʁ/ ⟨f s z š ž x ġ⟩
Nasals: /m n/ ⟨m n⟩
Liquids: /r l/ ⟨r l⟩
Semivowels: /ʋ j/ ⟨v y⟩

All obstruents, nasals, and liquids can be both short and long.

Vowels

Short: /e ɐ ɨ o/ ⟨e a ə o⟩
Long: /æː eː iː ɒː oː uː/ ⟨ā ē ī â ō ū⟩

Stress

Stress is phonemic.

[Porphyrogenitos]

Have you actually googled for ʀ͡ʙ? Three items total come up, and they are all pertain to conlanging, i.e. this is not a real speech sound.

1. Kalahŕi is an exolang, ie. not spoken by humans. It is spoken by a species, that, yes, has a human-like vocal tract, but are not human. The universals of human linguistics can hardly be applied to them.

Yes, that one's conlang is not spoken by humans - the deus ex machina of conlanging. That said, regardless of the morphology of your conbeings' vocal tracts, how do you reasonably expect them to be able to trill at two different POA simultaneously reliably?

2. I notice you chose to pick on /ʀ͡ʙ/ but the phonology also included /ʈꞎʼ/, which is not found in any natlang either. Heck, /ꞎ/ is found only in one natlang, I think, and it doesn't even have a dedicated IPA smbol- this is from the extensions to the IPA.

The problem is that your conphone I commented on is two different trills simultaneously, if you haven't realized that.

I'm perfectly capable of pronouncing [ʀ͡ʙ], albeit not longer than a portion of a second. If I try hard enough I can even achieve [ʀ͡r͡ʙ], though only for one tap of [r]. I don't see why this is a problem for a fictional race with a slightly different vocal tract than humans. Are simultaneous trills supposed to be impossible or something?

[Porphyrogenitos]

Apologies for the double post - the tentative phonology of a somewhat unnaturalistic conlang I'm working on:

Consonants
Nasals: /m/ /n/
Stops: /p/ /t/ /k/
Fricatives: /s/ /h/

Vowels
Front: /i/ /e/ /ø/
Central: /a/ [using the IPA <a> just to represent a generalized low vowel whose value tends toward the center]
Back: /u/ /o/

Phonotactics

Maximum syllable structure: CCCVCCC

Any two consonants may form a word-medial CC cluster.

Nasals have highest resonance, fricatives second-highest resonance, and stops lowest resonance.

For CCC and CC clusters at word boundaries, higher-resonance consonants may not be "stranded" on the outside of a cluster by lower-resonance consonants closer to the "interior" of a word. However, the fricatives /s/ and /h/ are exceptions to this in CC clusters.

For any CCC cluster, higher-resonance consonants may not be "trapped" within a cluster by two lower-resonance consonants.

Geminates may only appear within a CCC cluster if they are the first two members of a CCC cluster. Geminates are also forbidden at word boundaries.

The clusters /tn/ and /pm/ are forbidden.

Allophony

The stops /p/ /t/ /k/ are lenited to [v] [ð] [ʝ] intervocalically. After nasals, they are voiced.

The distinction between /m/ and /n/ is neutralized before all consonants but /h/. They assimilate to the position of the following consonant.

/s/ is voiced to /z/ intervocalically. /h/ varies freely between [h] [ɦ] [ʕ] intervocalically. /h/ becomes [x] in all clusters.

List of permissible word-initial CC clusters, as example:

mn

nm

pn pt pk ps ph

tm tp tk ts th

km kn kp kt ks kh

sp st sk sm sn sh

hm hn hp ht hk hs

[MobiusFlip]

Saɴy - This is probably my weirdest one, in terms of phonology.

Consonants
Nasals - /m n̪͆ ɴ/ <m n ɴ>
Stops/Percussives - /ʬ ʭ c' q' ʔ/ <p t c q ʔ>
Fricatives - /ɸ β h̪͆ ɦ̪͆ ç ɟ χ ʁ h ɦ/ <f v s z ç ɟ x r h ɦ>
Approximants - /j w/ <j w>

/n̪͆/ is a bidental nasal, a sound which I was best able to represent with the symbol for an interdental nasal.
(I'm not sure how to get n̪͆ and h̪͆ to render properly, but the combining marks should be directly above and below the letters.)

Vowels
/i e a ɶ ø y/ <i e a o u y>

The consonants <m n ɴ q ʔ f v s z x r h ɦ> can all be labialized.



Djavzae

Consonants
Nasals - /m n/ <m n>
Stops - /t d k g ɢ ʔ/ <t d k g ɢ ʔ>
Fricatives - /v s z ʒ ç ʝ χ ʁ h/ <v s z j ç ʝ χ ʁ h>
Affricates - /dz dʒ kx ʔh/ <dz dj kx ʔh>
Stops (prenazalized) - /mb nd ŋg ɴɢ/ <mb nd ŋg ɴɢ>
Liquids/semivowels - /w l r ͡χl ͡ʁl j ʀ̥ ʀ/ <w l r ͡χl ͡ʁl j ʀ̥ ʀ>

<k g kx> can also be realized as /c ɟ cç/

Vowels/Diphthongs
/i ɛ a o aj ej oj aɛ æʊ/ <i e a o ai ei oi ae ao>



Xthra'nul - Another weirder one, mostly because of the amount of possible syllabic consonants.

Consonants
Nasals - /m n ŋ/ <m n ng>
Stops - /b t k g ʔ/ <b t k g '>
Affricates/Fricatives - /f θ s z ʃ ʧ χ h/ <f th s z sh ch x h>
Liquids/Semivowels - /w r l ͡χl j ʀ/ <w r l hl y rr>

Possible syllabic consonants are <r l rr m n ng f s sh h>. The semivowels <y w> are also used to represent vowels with different phonemic values.

Vowels
/æ~a ɒ~o ʌ i~ɪ~ji~jɪ u~ʊ~uw~ʊw/ <a o u y w>

[Pogostick Man]

Some of the earliest-spoken languages in my conworld:

Language 1

/m n/ <m n>
/p b t d k g ʔ/ <p b t d c g ʔ>
/s x/ <s ḫ>
/w ɹ l j/ <w r l y>

/u uː o oː a aː e eː i iː/ <u ū o ō a ā e ē i ī>

Contrastive length on consonants as well
Vowels tend to lax before geminate consonants; loanwords into and from language 2 reflect this

Language 2

/m n ŋ/ <m n ĝ>
/p t k g/ <p t c g>
/ts/ <z>
/s ʃ h/ <s š h>
/w l/ <w l>

/ʊ u ɔ o ə a ɛ e ɪ i/ <u ú o ó a á e é i í>

Language 3

/n/ <n> (<m> before /b f w/)
/b t d k g q/ <b t d c g q>
/f s/ <f s>
/w ɹ j/ <w r y>

/u ũ a ã e ẽ i ĩ/ <u ų a ą e ę i į>

Language 4

/m n ŋ/ <m n ĝ>
/t tʰ d k kʰ g q qʰ ʔ/ <t th d c ch g q qh ʔ>
/ts tʃ/ <z ž>
/f s ʃ h/ <f s š h>
/w l/ <w l>

/u ṵ a a̰ ə ǝ̰ i ḭ/ <u ủ a ả ǝ ǝ̉ i ỉ>

Language 5

/n ŋ/ <n ĝ>
/t tʼ d k kʼ g q qʼ/ <t t’ d k k’ g q q’>
/ts dz tʃ dʒ/ <c j č ǰ>
/tɬ dɮ/ <ƛ λ>
/s z ʃ ʒ x h/ <s z š ž ḫ h>
/ɬ ɮ/ <ł l>
/w ɹ j/ <w r j>

/u uː o oː a aː ɨ ɨː e eː i iː/ <u ū o ō a ā y ȳ e ē i ī>

[mezziah]

Here's a very early sketch of the phonology of my Eastern Nisha language, which is, in contrast to Western Nisha, a direct descendant of PIE. It's currently in early development stage and still needs lot of work.

Any feedback would be greatly appreciated, since I'm not quite sure whether such a phonology is realistic or not. As for now, it only follows certain sound changes.

/m n/
/p p' t t' c ɟ k k' ʔ/
/t͡s t͡s' c͡ç (ɟ͡ʝ)/
/t͡ɬ t͡ɬ'/
/θ s x h/
/ʋ~w r/
/l j/

/a e i u ɛ ɘ/
/ɨ/ might function as an allophone or as an independent phoneme. Not sure yet. /y/ could be a future vowel as well, but I'm not yet sure how that would fit in.

[Travis B.]

Here's a very early sketch of the phonology of my Eastern Nisha language, which is, in contrast to Western Nisha, a direct descendant of PIE. It's currently in early development stage and still needs lot of work.

Any feedback would be greatly appreciated, since I'm not quite sure whether such a phonology is realistic or not. As for now, it only follows certain sound changes.

/m n/
/p p' t t' c ɟ k k' ʔ/
/t͡s t͡s' c͡ç (ɟ͡ʝ)/
/t͡ɬ t͡ɬ'/
/θ s x h/
/ʋ~w r/
/l j/

/a e i u ɛ ɘ/
/ɨ/ might function as an allophone or as an independent phoneme. Not sure yet. /y/ could be a future vowel as well, but I'm not yet sure how that would fit in.

I have some typological concerns:

You have exactly one voiced stop, and it is /ɟ/, when if only some stops are voiced the typically are more front ones. (I could understand if this were derived from /j/, but you still have a /j/.) Also, why at the same POA, is there not */c'/, to go along with stops at all other POAs aside from glottal?
Also, it is very rare to have a phonemic contrast between /c/ and /c͡ç/.

[vokzhen]

Any feedback would be greatly appreciated, since I'm not quite sure whether such a phonology is realistic or not. As for now, it only follows certain sound changes.

Three things stand out. Afaik, no language contrasts /cç ɟʝ/ with /c ɟ/, and /c ɟ/ are very often [cç ɟʝ]. If the affricates are assibilated [cɕ ɟʑ] and the plosives are true palatal stops (no coronal contact, no affrication), I think it's attested, but very rare. Second, having a voice-voiceless contrast in only the palatals and a voiceless-ejective elsewhere invites an explanation.

Third is that I'm only aware of a single language that has /tɬ/ or /tɬ'/ without having [ɬ], and even then one of its dialects realizes them as [ɬ tɬ'] instead (name escapes me, it's one of the Californian languages). Though it does only have to be an allophone, not a phoneme in its own right, Nahuatl being the most well-known example where /tɬ/ is phonemic and [ɬ] shows up as a coda allophone of /l/.

[Nortaneous]

IPA's palatal notation sucks. A better convention would be to use c for kʲ and ȶ for the alveolopalatal. Yanyuwa contrasts the two, and Japhug contrasts c and tɕ, but I'm not sure how affricated tɕ really is -- less so than English tʃ, most likely.

No natlang contrasts c and cç, if by c and cç you mean kʲ and kxʲ. It's obviously possible to contrast c and ȶ.

[mèþru]

Agefaqeg:
/p b t d k ɡ q/
/m n ŋ/
/r/
/f v θ̟ ð̟ s z ʃ h/
/t̪θ ts tʃ dʒ/
/ɹ l/
/i y u ɪ ʏ ʊ e ø o ə ɛ œ ɔ/
/ɔɪ̯, ai̯, aʊ̯, aʊ̱̯/
A rare sequence of aʊ̯ and ə occasionally occurs, and it can be treated as its own vowel. The syllable break between the two is ignored in stress rules.

[Ser]

A couple alphabetlangs, where the inventory consists of the 26 letters of the alphabet. No romanization problems!

/p t tʃ k ʔ/ <p t c k q>
/b d dʒ g/ <b d j g>
/m n/ <m n>
/f s ʃ h/ <f s x h>
/v z/ <v z>
/r/ <r>
/w l y/ <w l y>
/i e a o u/ <i e a o u>

/p t k q/ <p t k q>
/b d g/ <b d g>
/m n/ <m n>
/ts/ <c>
/f s χ h/ <f s x h>
/v z/ <v z>
/r/ <r>
/w l y/ <w l j>
/i y e a o u/ <i y e a o u>

/pʰ tʰ cʰ kʰ/ <p t c k>
/p t c k ʔ/ <b d j g q>
/m n/ <m n>
/f s ç x h/ <f z s x h>
/r/ <r>
/w l y/ <v l y>
/i e ɨ a o u/ <i e w a o u>

I think I'm gonna actually use the first one.

[Nortaneous]

natqgu imo

/p b t d dZ k g/ <p b t d j k g>
/f v s h/ <f v s h>
/m n l j w/ <m n l y w>
/{ A V O e 2 o i y u/ <x a z c e r o i q u>

[opipik]

/p b t d k ɡ ᵑɡ/ <p b t d k g q>
/m n/ <m n>
/ɸ s z x h/ <f s z x h>
/ɹ l/ <r l>

/i ɪ u ʊ e o ə ɔ ɑ/ <i y u w e o v c a>

[Jonlang]

A slightly revised phonology for Quebric:

/p b t d k g/ p b t d c g
/f v θ ð s h χ/ ff f th dh s h ch
/ɬ/ lh
/ɾ/ r
/w ɹ j/ w r i
/l/ l
/m n ŋ/ m n ng

rh - aspirated /ɾ/ or /ɹ/
wh aspirated /w/
qu - /kw/

/a ɒ ɑː/ a o á
/ε ʌ ɔ/ e (é long) y ó
/ɪ/ i, u (depending on position)
/i:/ i, u, y (depending on position)
/u:/ w

/aɪ/ ai or ae
/aʊ/ aw
/εɪ/ ei
/εʊ/ ew

[Nortaneous]

/pʲ pˠ tʲ tˠ tʷ c k kʷ/ <p b j t d c k q>
/fʲ fˠ sʲ sˠ ç x xʷ/ <f v s z x h w>
/m n ɲ/ <m n g>
/lʲ lˠ/ <l r>
/j w/ <y u>
/a ɜ ɘ ɨ/ <a e o i>

[----]

Marshallese isn't a conlang, Nort