Post your conlang's phonology

[Nortaneous]

¸¸¸¸¸¸¸¸¸¸¸¸¸¸¸¸¸

[Owlglass Moot]

Very new to this, so any input would be appreciated (especially in regards to naturalistic-ness)! There's a whole lot of influence from Spanish, I guess.

CONSONANTS:


/β ð ʝ ɣ/ are realized as /b d ɟ͡ʝ g/ after a nasal consonant. Haven't really put much work into other phonological changes yet.


VOWELS:
/i ɪ e a o u ʊ/


PHONOTACTICS (INCOMPLETE):

• Basic syllable structure: (C1 (C2 (C3))) V (S) (C4)
• Optional C1 can be any consonant.
• Optional C2 is only permissible if C1 is a stop, nasal or fricative. If C1 is a stop or nasal, C2 may only be a liquid /ɹ l/. If C1 is a fricative, however, C2 may be a stop, nasal, or liquid.
• Optional C3 can be a liquid /ɹ l/ and is only permissible if C2 is a stop or nasal.
• V can be any vowel, or the syllabic consonants /ɹ l m n ŋ/. If V is a syllabic consonant, it may not have a coda.
• Optional S can be a semivowel, liquid or nasal consonant.
• Optional C4 can be a stop or fricative consonant.

IMMEDIATE CONCERNS/QUESTIONS:

• The glides. I'm only planning for falling diphthongs to be allowed, so are they even necessary? I still don't fully understand the difference between something like [aj] and [ai̯].
• Even if I do keep the glides, does /j/ contrast with /ʝ/ in any natural languages?
• I feel like the /ɹ/ makes it too "English-y", but I can't make the [r] sound to save my life. Plus a trill or flap would make it even closer to Spanish than it already is. I would prefer the Mandarin /ɻ/, but having only a single retroflex would feel out of place.
• Assuming some form of vowel harmony, is it naturalistic to have a tense-lax distinction (/i u/ vs. /ɪ ʊ/) in only the close vowels? Would it be better to have the distinction persist all the way down the trapezoid? /i e a o u/ vs. /ɪ ɛ ə ɔ ʊ/?

[Chengjiang]

The glides. I'm only planning for falling diphthongs to be allowed, so are they even necessary? I still don't fully understand the difference between something like [aj] and [ai̯].

Honestly, for the most part it's up to convention or the transcriber's choice rather than anything phonetic. Depending on the language's phonology there may be reasons to phonemicize the diphthong as one or the other, but in a phonetic transcription those don't really have different meanings. Sometimes people will use [j] to represent a slightly more closed sound than [i̯], but AFAIK that's not a systematically applied distinction.

Even if I do keep the glides, does /j/ contrast with /ʝ/ in any natural languages?

Arguably Spanish contrasts them. [j] contrasts with [ʝ] in most dialects that have [ʝ], so if you don't analyze [j] as an allophone of /i/ Spanish has a /j/-/ʝ/ contrast. (Technically, Spanish /ʝ/ is more often an approximant than a fricative, like the voiced obstruents, but it's a closer approximant than its [j] and tends to become rounded in the vicinity of rounded vowels, which [j] does not. At any rate /ʝ/ typically fortites to an affricate in the same environments where the voiced obstruents are realized as stops, while [j] remains a glide.)

I feel like the /ɹ/ makes it too "English-y", but I can't make the [r] sound to save my life. Plus a trill or flap would make it even closer to Spanish than it already is. I would prefer the Mandarin /ɻ/, but having only a single retroflex would feel out of place.

/ɹ/ is uncommon but there's nothing "wrong" with including it in a language. Also, as long as you can describe them, I don't see a real problem with having sounds in your conlang that you can't pronounce yourself. Finally, some languages do have a rhotic as the only retroflex, including some dialects of English.

Assuming some form of vowel harmony, is it naturalistic to have a tense-lax distinction (/i u/ vs. /ɪ ʊ/) in only the close vowels? Would it be better to have the distinction persist all the way down the trapezoid? /i e a o u/ vs. /ɪ ɛ ə ɔ ʊ/?

Many languages of West Africa have height/tenseness vowel harmony that only applies to a subset of the vowels. For instance, in Yoruba /e o/ can't coexist with /ɛ ɔ/, while /i a u/ are all neutral. I think there might be a language with your exact proposed vowel harmony system, in fact.

That all said, what are your language's phonotactics? Also, what are its prosodic features? Is it syllable-timed, stress-timed, moraic? Is stress contrastive, and if so how does it manifest? I'm guessing it doesn't have tone since you didn't mention it, but just in case, does it have contrastive tone?

[Owlglass Moot]

Awesome, thank you!

[Chengjiang]

Awesome, thank you!

You're welcome. I edited my post apparently as you posted that, so do look back at the new paragraph at the bottom.

[Zaarin]

Very new to this, so any input would be appreciated (especially in regards to naturalistic-ness)! There's a whole lot of influence from Spanish, I guess.

CONSONANTS:


/β ð ʝ ɣ/ are realized as /b d ɟ͡ʝ g/ after a nasal consonant. Haven't really put much work into other phonological changes yet.


VOWELS:
/i ɪ e a o u ʊ/


IMMEDIATE CONCERNS/QUESTIONS:

• I feel like the /ɹ/ makes it too "English-y", but I can't make the [r] sound to save my life. Plus a trill or flap would make it even closer to Spanish than it already is. I would prefer the Mandarin /ɻ/, but having only a single retroflex would feel out of place.

I like your consonant inventory; looks Egyptian (minus uvulars, of course). /ɹ/ is a reasonably common sound; don't shy away from it just because some varieties of English happen to have it--indeed, I'd say having just /ɻ/ would be more English-y (this is your typical American R, along with the "bunched R" which is a weird pharyngealized monstrosity--and also what I have for R). Anyway, my point is that /ɹ/ is fairly widespread--many Native American languages that have rhotic consonants have /ɹ/, some dialects of Aramaic have /ɹ/, some reconstructions of Ancient Egyptian posit /ɹ/. However, I'd say don't exclude a sound just because you can't pronounce it--I can't reliably produce pharyngeals or ejectives, yet I have quite a few languages with both.

[Owlglass Moot]

Awesome, thank you!

You're welcome. I edited my post apparently as you posted that, so do look back at the new paragraph at the bottom.

I edited in some barebones phonotactics after you posted.

Not 100% decided on prosody and suprasegmentals, but as of right now:

• No tone.
• Lexical stress similar to Latin/Spanish, maybe? The penultimate syllable is stressed unless it contains a lax vowel, in which case the antepenultimate receives the stress. Orthographically indicated stress supersedes this.
• Stress-timed, I think.

[Chengjiang]

Owlglass Moot, I just realized I didn't properly answer one of your questions.

I'm only planning for falling diphthongs to be allowed, so are [phonemic glides] even necessary?

I'd say only if you plan for the diphthongs to both a) contrast with sequences of vowels in hiatus and b) not be a small, limited set. If they don't contrast with sequences of vowels you can just phonemicize them as vowel sequences, and if there aren't a huge number of diphthongs (i.e. if your total set of diphthongs is more like Hausa's /ai au/ than Spanish's everything-can-be-a-diphthong) it probably makes more sense to analyze the diphthongs as unitary phonemes than as sequences of vowel plus glide.

Since you've added phonotactics to your post, I'd just like to say that you shouldn't worry your language sounds too much like Spanish. This syllable structure sounds nothing like it. Honestly just allowing multi-consonant codas does a lot, but this sounds like it also allows a much broader range of onsets albeit within a sonority hierarchy to the point that that aspect reminds me a bit of Slavic.

The penultimate syllable is stressed unless it contains a lax consonant, in which case the antepenultimate receives the stress.

Did you mean "lax vowel"? If not, what does "lax consonant" mean?

[Owlglass Moot]

@Zarin:

Thanks! Now that you say it, I realize that my /r/ is much closer to [ɻ] than [ɹ], huh.

---

@Chengjiang:

Oops, I did mean lax vowel.

I'm still iffy about certain parts of my phonotactics. For C3, I may remove the potential for C2 to be a nasal. And I should probably also add in that C1/C2/C3 cannot be a liquid if V is /ɹ l/, and C1/C2 cannot be nasals if V is /m n ŋ/.

For my diphthongs I was thinking the general /a e o/ followed by /i u/, resulting in [ai̯], [ei̯], [oi̯], [au̯], [eu̯], [ou̯]. So I guess if they're being realized as combos rather than unitary phonemes, I should probably end up keeping the glides?

[Chengjiang]

For C3, I may remove the potential for C2 to be a nasal.

If you want. Do you not like the sound of onsets like /smɹ/ or /nl/? I find them kind of groovy myself, and they're a nice bit of differentiation from it sounding too much like Spanish or another western European language if you're concerned about that. More something you'd find in a Slavic or Indo-Aryan language.

For my diphthongs I was thinking the general /a e o/ followed by /i u/, resulting in [ai̯], [ei̯], [oi̯], [au̯], [eu̯], [ou̯]. So I guess if they're being realized as combos rather than unitary phonemes, I should probably end up keeping the glides?

If the diphthongs are a realization of one vowel followed by another vowel, it doesn't sound like you have phonemic glides. If I'm understanding correctly /ai/ = [ai̯], yes?

[Chengjiang]

Oh why the hell not. Here's Undreve:

/m n nʷ ɲ/ <m n mw nh>
/p b t d ts dz tʷ dʷ tʃ dʒ k g kʷ gʷ/ <p b t d ts dz tw dw ch j k g kw gw>
/f v s z sʷ zʷ ʃ ʒ/ <f v s z sw zw sh zh>
/r j w/ <r y w>
/l ʎ/ <l lh>

/i ʉ u/ <i ụ u>
/e ɵ o/ <e ọ o>
/a ɑ/ <ạ a>

|N| <n>, <m> before labials
Tones shown on /a/: /a aː ǎ ǎː â âː/ <ạ ạ̄ ạ́ ạ̌ ạ̀ ậ>

The "labialized coronal" series /nʷ tʷ dʷ sʷ zʷ/ are given historically-based diaphonemic labels that do not reflect their realization in any one dialect, as their actual realization varies massively between dialects. (The following names are placeholders.) The Capital dialect has trilled affricates for the nasal and oral stops while merging the fricatives with the labial fricatives: [mʙ tʙ̩ dʙ f v]. The Western dialect has Shona-style whistled sibilants and a nasal with rounded release: [mʷ ts͎ dz͎ s͎ z͎]. The Southern has labialized apical postalveolars, with affrication for some speakers: [ṉʷ t̠ʷ d̠ʷ s̠ʷ z̠ʷ] or [ṉʷ t̠s̠ʷ d̠z̠ʷ s̠ʷ z̠ʷ]. Southerners also pronounce the central vowels as front vowels [y ø æ].

The vowels follow both closeness and frontness harmony. /ʉ u ɵ o/ obey both, while /i e/ are neutral for frontness and /a ɑ/ are neutral for closeness. This yields four vowel sets:

Front-close: /i ʉ a/
Back-close: /i u ɑ/
Front-open: /e ɵ a/
Back-open: /e o ɑ/

Inflections contain three archiphonemes |I U A| that harmonize with the root.

The language allows syllables of the form (C)V(F), where F = one of |f s N l r w j| or |ː|, a vowel-lengthening chroneme. |N| is a nasal that takes the POA of the following consonant; it is realized as [ŋ] before a vowel-initial word or a pause and an epenthetic [d] is inserted before /r/. The obstruents assimilate in voicing to the following segment; they are generally voiceless before a pause. |s l| become [ʃ ʒ ʎ] as applicable before a palatoalveolar or palatal consonant; |sr| is realized as [dr]. |j w| are only found after /a ɑ/ and are typically realized as vocalic off-glides with the frontness (and possibly closeness) of the word, e.g. /awri/ = [aʉ̯ri]. Geminates are not allowed except for a handful of cases of geminates across morpheme boundaries in compounds; all historical geminates became single consonants with compensatory lengthening of the preceding vowel, which is one of the sources of long vowels in the modern language.

Undreve has a pitch accent shamelessly ripped from Shtokavian. Most words have a stressed syllable, which can have a falling or rising tone. This is realized as a short tone on simple (C)V syllables and syllables ending with a fricative, and a long tone on syllables ending with a sonorant or a long vowel.

(Also, yes, I know the language's name appears not to follow harmony. It's one of a few words like that. It's historically a compound.)

[Owlglass Moot]

For my diphthongs I was thinking the general /a e o/ followed by /i u/, resulting in [ai̯], [ei̯], [oi̯], [au̯], [eu̯], [ou̯]. So I guess if they're being realized as combos rather than unitary phonemes, I should probably end up keeping the glides?

If the diphthongs are a realization of one vowel followed by another vowel, it doesn't sound like you have phonemic glides. If I'm understanding correctly /ai/ = [ai̯], yes?

Ohh, I think I misunderstood. I got it, now, thanks!

[Chengjiang]

PHONOTACTICS (INCOMPLETE):

• Basic syllable structure: (C1 (C2 (C3))) V (S) (C4)
• Optional C1 can be any consonant.
• Optional C2 is only permissible if C1 is a stop, nasal or fricative. If C1 is a stop or nasal, C2 may only be a liquid /ɹ l/. If C1 is a fricative, however, C2 may be a stop, nasal, or liquid.
• Optional C3 can be a liquid /ɹ l/ and is only permissible if C2 is a stop or nasal.
• V can be any vowel, or the syllabic consonants /ɹ l m n ŋ/. If V is a syllabic consonant, it may not have a coda.
• Optional S can be a semivowel, liquid or nasal consonant.
• Optional C4 can be a stop or fricative consonant.

Some questions to perhaps help you iron out the phonotactics:

• Are there any further restrictions on which liquids can follow which other consonants? As an example, a fairly common hole in languages that allow initial clusters of consonant + liquid is for clusters of coronal stop + lateral to be disallowed. This is typical in Germanic and Romance languages, but outside Europe it's also found independently in other language families, e.g. Thai and its relatives. It's also pretty common for languages that have initial clusters with liquids to not allow liquids to follow palatal/postalveolar stops/affricates.
• Does the language allow clusters with differently voiced obstruents? You mention fricative-stop clusters; are clusters like /ɕð/ or /zk/ permitted? (If you're wondering, it's very rare but not unknown for languages to allow different-voicing obstruents to cluster in onsets, and still rather uncommon to allow differently-voiced medial clusters without assimilating one of the consonants to the other's voicing.)
• For phonotactic purposes, are /β ð ʝ ɣ/ treated as stops or as fricatives?
• Are all fricatives equally able to begin three-consonant onsets? In IE languages it's generally just /s/ or things close to it (leaving aside a minority with a huge range of possible onset clusters like Russian), because the other fricatives aren't derived from fricatives in PIE, but maybe yours has a different history.
• How do the glottal consonants behave phonotactically?

[Pogostick Man]

Common Caber

/m n/ <m n>
/p b t d k g/ <p b t d c g>
/ts dz tɕ dʑ/ <č ǧ ć ǵ>
/f s z ɕ x/ <f s z ś h>
/ɾ w/ <r w>

/u ɔ ɨ ə a ɛ i/ <u o ư ơ a e i>

[Tropylium]

A couple alphabetlangs

a b c d e f g h i j k l m n o p q r s t u v w x y z
/aː β t̺s̺ ð ə ɸ ɣ ʔ ɪ x k l m n w͡ɔ p q ʁ s̺ t ʊ ⱱ uː χ iː θ/

or, reordered:
/p t t̺s̺ k q ʔ/ p t c k q h
/ɸ θ s̺ x χ/ f z s j x
/β ð ɣ ʁ/ b d g r
/m n/ m n
/ⱱ l/ v l
/w͡ɔ/ o
/iː aː uː/ y a w
/ɪ ə ʊ/ i e u

[Tropylium]

This isn's so much a comment on the phonology than the presentation, but:

• Basic syllable structure: (C1 (C2 (C3))) V (S) (C4)
• Optional C1 can be any consonant.
• Optional C2 is only permissible if C1 is a stop, nasal or fricative. If C1 is a stop or nasal, C2 may only be a liquid /ɹ l/. If C1 is a fricative, however, C2 may be a stop, nasal, or liquid.
• Optional C3 can be a liquid /ɹ l/ and is only permissible if C2 is a stop or nasal.

You could restate this much easier in terms of three independent slots: (F) (P) (L), where F can only be a fricative; P can only be a stop or a nasal; L can only be a liquid.

Also re /j ʝ/ contrast: /ʑ/ would probably work better as the voiced counterpart of /ɕ/.

[Travis B.]

a b c d e f g h i j k l m n o p q r s t u v w x y z
/aː p tɕʼ t i ɸ k χ iː tɕ kʼ l m n u pʼ qʼ ɾ s tʼ uː β w ɕ j z/

/m n/ m n
/p t tɕ k/ b d j g
/pʼ tʼ tɕʼ kʼ qʼ/ p t c k q
/ɸ s ɕ χ/ f s x h
/β z/ v z
/ɾ l/ r l
/w j/ w y
/iː i aː u uː/ i e a o u
Note that there is also [ə], which is unwritten.

[Quark8]

I am quite new to conlanging, but these are my first three phonologies/phonemic inventories, and input is appreciated and criticism encouraged, especially in regard to naturalness, or too many similarities to a natlang. Not of the phonology is done yet, so the empty spots will be filled. So, here's Ithrynian, Istarian, and Rodkogy.

Ithrynian: There is quite the influence from Irish, my favorite of the Celtic languages, but a lack of mutations.

Consonants:
Plosives: /p b t̪ˠ d̪ˠ tʲ dʲ t d k g/ <p b t d ty dy t d c g>
Fricatives: /f v θ ð ʃ s z x h~ʔ/ <f v th dh sh s~z x h>
Nasals: /m n̪ˠ nʲ n ŋ/ <m n ny n ng>
Taps*: /ɾˠ ɾʲ r/ <rh rry rr>
Approximants: /l ɹ w/ <l r w>

*The usage of taps varies from speaker to speaker.

Vowels:
Monophthongs: /a ɔ ɛ e i ɪ o ʊ u (y) a: ɔ: ɛ: e: i: o: ʊ: u: (y:) a:: e:: i:: o: u: (y::)/ <a e/ae i/y o u á é í ó ú â ê î ô û>
Diphthongs: /ai ɛi ui au ɔɛ/ <ai ei ui au oe>

Notes: <i> and <y> are pronounced /j/ before vowels. /f/ becomes /v/ word-finally. Hard consonants (i.e t̪ˠ, d̪ˠ, etc.) occur when next to either /ɛ/, /e/, /i/, or /ɪ/ (ex. nenyas "world" is /n̪ˠɛnjas/, not */nɛnjas/). All consonants (except for /l/ and /w/) can be geminated. <y> is never pronounced /ɪ/.

The hard plosive series /t̪ˠ d̪ˠ/ and the hard nasal /n̪ˠ/ are pronounced that way in Standard Ithrynian, but in other dialects it varies greatly. The Ynaen dialect has flapped affricates for the plosives and merges the nasal with the labial nasal: /tⱱ dⱱ m/. Note that there is even more variance in the Ynaen dialect, with the northern sub-dialects realizing the flap as a bilabial flap, giving it a distinct sound amongst the Ithrynian languages, and they are therefore realized as /tⱱ̟ dⱱ̟ m/. The Ithilian dialect realizes them all with rounded release: /t̪ˠʷ d̪ˠʷ n̪ˠʷ/. The Anarian dialect has no hard consonants at all, and realizes them as labialized, while some speakers see them as palatalized: /t̪ʷ d̪ʷ n̪ʷ/ or /tʲ dʲ nʲ/. Anarian's also pronounce the back vowels /u ɔ/ as near-back /ʊ ɞ/.

Allophones:

/h/ becomes /ʔ/ between two consonants (ex. imhle "word" is /imʔle/), it becomes /ʃ/ after back vowels (/u/, /ɔ/) (ex. dariuha "hand" is /daɹjuʃa/), and /x/ word-initially before back vowels (ex. har "noise" is /xɔɾ/), if it is not deleted (ex. hambys "permission" is /ɔmbis/). When geminated, /h/ becomes /f/ (ex. bahhy "twist" is /bafi/).

In some dialects, /i/ and /ɪ/ are /y/ and /ʏ/ before and after nasals. (ex. ninaes is /nynes/).

(Incomplete)

Phonotactics:

to be added later


Astarian: Influenced by Hungarian, my favorite of the European non-Indo-European languages, but it also has the consonant mutations of the Celtic languages.

Consonants:
Nasals: /m n ɲ/ <m n ny/ñ>
Plosives: /p b t d k/ <p b t d c>
Affricates: /t͡s t͡ʃ d͡ʒ c͡ç ɟ͡ʝ/ <cc cs gh ty gy>
Fricatives: /f~ɸ v~β s z ʃ x h/ <f v s z sh ch h>
Trills*: /r̥ r/ <rh r>
Approximants: /l̥ l ɹ j~w/ <lh l r j/

*The usage of trills varies from speaker to speaker.

Vowels:
Monophtongs: /a(:) ɔ ɛ e: o: i(:) u(:)/ <a(á) a e é ó i(í) u(ú)>
Diphthongs: /ju ew aj aw oj uj/ <iu eu ai au oi ui>

Allophones:

/j/ is /w/ word-initially before vowels. /f/ and /v/ are /ɸ/ and /β/ word-initially before vowels. /j/ is /ʝ/ before consonants. /p t k/ are aspirated, but /b d/ are never aspirated. Voiceless stops stops are not aspirated after /s/ and /ʃ/.

Phonotactics:

Possible Initial Starting Clusters (Does not Include Mutations):

pt pr bt br ft fr vt vr dt dr mt mr sty sry zry nry ryty ryry dyt dyr shr nyt nyr ryt ryr sc zc nc rc
sp st sc sm st sr sn shm shd shry shny
tt tr cr tyl tyry tyny pl pry pny nl bry bny
mn
tyb tyc tyf tym ryb ryc ryf rym
nyd nyv nym


more to be added

Rodkogy:

Consonants:
Pulmonic Plosives: /t d k g q qʰ ʔ~ə/ <t d k g q qh '>
Ejective Plosives: /ť kʼ qʼ/ <t' k' q'>
Pulmonic Affricates: /t͡ɕ d͡ʑ/ <j jj>
Ejective Affricates: /tɕʼ/ <j'>
Pulmonic Fricatives: /s ɬ ʃ X ʁ̝ h~ə/ <s ll sh ch h>
Ejective Fricatives: /ɬ’/ <ll'>
Pulmonic Nasals: /m n ŋ/ <m n ng>
Ejective Nasals(?): /m' n'/ <m' n'>
Pulmonic Trill: /r/ <r>
Approximants: /l~o: j~i:/ <l~ō j~ī>

Vowels:
Monophthongs: /a e e: (ə) (o:) (i:)/ <a e ē>

Allophones/Phonotactics::

Short vowels become long before a glottal. The velars /k' k g ŋ/ only occur before /a/ or a trill, and the glottal stop /ʔ/ only occurs between vowels. /l j m n ŋ m' n'/ can appear as vowels as well as consonants, but only when not next to vowels. Same for the glottal stop and glottal fricatives. /i:/ and /o:/ are allophones of /j/ and /l/. Any glottal between consonants is a schwa /ə/. /e:/ is very rare. Vowels cannot come word-initially. No diphthongs.

[Dezinaa]

Otvei



Orthography:
/m n ŋ/ <m n g>
/p t k/ <p t k>
/v s z h/ <v s z h>
/l ʀ/ <l r>
/i u ɛ ɔ ɑ/ <i u e o a>

Syllable structure:
(C)V(C)
Any consonant can start or end a syllable.

Allophony:
/p t k/ → / N_, _[v, z], [v, z]_
/s z/ → [ʃ ʒ] / _i, i_$
/h l ʀ/ → [χ ɬ r̥] / V_$
/l/ → [ɾ] / _i
/i u/ → [j w] / V_, _V
∅ → ʔ / [a, e, o]_[a, e, o]

$ = syllable boundary
V = vowel
N = nasal

Stress is not phonemic, but tends to be on heavier syllables towards the right side of a word (optimally penultimate). Coda consonants and/or [j w] make a syllable heavier.

Morphophonological changes:
These may actually be more allophonic changes, but they only happen at morpheme boundaries. They are represented in the orthography.

/np/ → /mp/
/nk/ → /ŋk/ (/ŋk/ is always written <nk>, not <gk>.)
/nv/ → /mv/
/nʀ, ŋʀ/ → [ɴʀ] <nr>
/tŋ/ → /kŋ/
/sz/ → /z/
/zs/ → /s/
/zh/ → /z/

Consonant gradation: (still on the fence about this)
Consonant gradation happens when a suffix that starts with a vowel is added to a word that ends in /p, t, k/.

/p/ → /v/
/t/ → /z/
/k/ → /h/

Hangul alternate orthography:

/mɑ nɑ ŋɑ pɑ tɑ kɑ vɑ sɑ zɑ hɑ lɑ ʀɑ/
<마 나 가 파 타 카 바 사 자 하 다 라>

/ɑ ɛ i ɔ u/
<아 애 이 오 우>

[jɑ jɛ jɔ ju wɑ wɛ wi wɔ]
<야 얘 요 유 와 왜 위 워>

안 모딯 읻태툼 나잇카우바 시 앻타알 퉄.
An molih iltetum naiskauva si ehtaar tuk.
[ˌɑn mɔˈɾiχ iɬˈtɛtum nɑjʃˈkɑwvɑ ˌʃi ʔɛχˈtɑʔɑr̥ ˈtuk]

[Nortaneous]

Sestmag: [ˈsɛɕmɒ]

/p b t ts k/ <p b t ts k>
/f s ʂ ɕ/ <f s ṣ s/st/se/sly/etc.>
/w l ʑ j/ <w l ly e>
/m n ɲ ŋ/ <m n ny ng>

|θ ð ʁ| <v d g>

/a e o i u/ <a e o i u> + length

Unstressed vowels are [ɒ ə i u], and [ɛ] in syllables with secondary stress.

/j/ is in most circumstances lowered to [e̯]. In the other circumstances, it merges with /ʑ/. /e i u/ merge with /j j w/ before a vowel, unless there's an unrealized <d g> between them.
/t n/ are flapped intervocalically. Flapped /n/ nasalizes the preceding vowel.

The three archiphonemes <v d g> behave as follows:

<v> is realized as [t] prevocalically and [f] otherwise.
<d> is realized as [ɦ] word-initially before a vowel, [0] intervocalically, and [v] word-finally. After a consonant, it triggers voicing and drops. Before a voiceless consonant, it devoices to <v>; before a voiced consonant, it merges with /v/.
<g> is realized as [q] word-initially before a vowel. Before or after a consonant, it has various complicated effects. It lowers and rounds vowels: <ag eg og ig ug> [ɒ jɒ ə jə wə].

Palatalization is pervasive. /m n/ palatalize to [ɲ]. /p f w/ palatalize to [tɕ fɕ wj]. /ts/ palatalizes to [ʂ] and /t/ palatalizes to [ts]. As for the rest, who knows. Different consonants behave differently with regard to palatalization spread; the interdentals are most resistant. It's complicated. Syncope is also pervasive: ko=ʁa=ˈju‿V > kəaˈju‿V > kəju‿V > kəjw‿V > kiw‿V; however, ko=ʁa=ˈju‿C > kəaˈju‿C > kəju‿C > kəe̯u‿C > keu‿C.

Consonant clusters frequently resolve in unusual ways. For example: ðtnj > vdɲ > vɲɲ > vɲ̥ > fɲ̥, ʁnl > ŋl > nd, wnstɲ > wnɕɲ > ɴwɲɲ > ɴwɲ̥ > ɴfɲ̥. Note that geminates regularly devoice. (It's really preaspiration.)

edit: Some processes (esp. syncope) only apply in fast speech (syncope is recent but becoming grammaticalized), so:
[fsɛndəˈe̯ɒ kiw ĩˈʑdũ t̪ũfɲɒˈiʑ səˈʑafk əf ɲãɾə]
vs.
[fstsɛŋdəˈe̯ɒ kə.əˈiw iʑməˈtunə t̪uɲɕ ɲɒˈiʑ səˈʑafək əf ˈɲanət]

[Pogostick Man]

Proto-Tlusa:

/p b t d c ɟ k g/
/u o a e i/

(C)V(C), at least for now

[Chengjiang]

Proto-Qalic:

/m n ŋ/ <m n ŋ>
/ˀm ˀn ˀŋ/ <m' n' ŋ'>
/p t k q ʔ/ <p t k q ʔ>
/p' t' k' q'/ <p' t' k' q'>
/β ð ɣ ʁ/ <ƀ đ ǥ ǥ̂>
/θ s x χ h/ <þ s x x̂ h>
/θ' s'/ <þ' s'>
/w l r j/ <w l r y>
/ˀw ˀl ˀr ˀj/ <w' l' r' y'>
/pθ ps tθ ts kθ ks qθ qs/ <pþ ps tþ ts kþ ks qþ qs>

/i ə u/ <i ə u>
/e a o/ <e a o>

Values of segments are, obviously, somewhat tentative. The glottalized obstruents were most likely ejective, although implosive, creaky voiced, or even long segments are more remote possibilities; the nature of the glottalized sonorants is difficult to reconstruct and they have been given a non-comittal transcription. The value of the non-sibilant coronal fricatives is similarly speculative, since they are only attested with this value in one of the modern branches of the family, but as their realizations in other branches range between sibilants and labiodentals, this is the simplest reconstruction. With the exception of the pulmonic fricatives, the voicing status of consonants is uncertain, and it is possible that the consonants shown as voiced fricatives here had stop realizations in some contexts.

Onsets are obligatory. They may consist of any single consonant or any cluster C₁C₂ where C₁ is any consonant except glides and /ʔ/ and C₂ is one of /w j/. If C₁ is a stop or fricative, C₂ may also be one of /l r/. C₂ matches C₁ in glottalization.

Codas are optional. Distinctions of voicing and glottalization are neutralized in the coda; beyond that all single consonants are legal codas.

The "pseudo-affricates" /pθ ps tθ ts kθ ks qθ qs/ are treated as single consonants by the phonotactics. They can be followed by a glide in onsets, and unlike all other clusters they can occur as codas. /tθ ts/ were treated the same as the others; one branch of the family metathesizes all of these sequences to put the fricative first, and this is applied to /tθ ts/ as well as the others. It is possible that /θ' s'/ were affricates, although if so they were still treated as the glottalized counterparts to /θ s/ and were not affected by metathesis where applicable.

The syllable nucleus invariably consisted of a single vowel; all apparent diphthongs were sequences of a vowel followed by a glide, with the glide counting as the coda consonant. Vowels are divided into a low set /e a o/ and a high set /i ə u/. The quality of these vowels is somewhat uncertain; /ə/ in particular may have been a high vowel [ɨ], and there is some evidence that in early Proto-Qalic the distinction may have in fact been one of laxness versus tenseness, perhaps even short versus long vowels. Alternation between the two sets is common in Proto-Qalic's morphology and has left remnants in the daughter languages.

In the surface phonology sequences of /Cʔ/ do not seem to occur word-internally. Where this sequence was created it appears to have condensed into glottalization of the /C/ if it was not already glottalized, which in a few situations (e.g. the dorsal fricatives) generated glottalized consonants that do not exist outside of this context.

Proto-Qalic had contrastive, probably pitch-based stress, which was most often although not always found before the final consonant of the root. It was most likely syllable-timed.

[Frislander]

Frislandian

Consonants

Code: Select all

p  t  k (ʔ)
   s     h
m  n
 ʋ l j
   ɾ

The glottal stop is essentially a phonetic accompaniment to a word-initial vowel.

Stops and fricatives may optionally voice intervocallically and after sonorants.

Vowels

Code: Select all

i ɨ u
 e o
  a

There is also a two-way length distinction.

All non-low vowels are lax when short in closed syllables and tense otherwise. In addition, e and o are pure mid when short in open syllable, but close-mid when long.

Phonotactics
(s)(C)V(C)

where s may only appear before stops, nasals and l and any consonant bar h may appear in the coda, however word-finally only the stops, s, the nasals and l may appear.

Phonological processes

• h deletes when adjacent to another consonant
• Geminates are only present in the language intervocalically and, additionally, stops may also be geminated after sonorants. Whenever, therefore, a consonant would be geminated when it is adjacent to obstruents, or when a sonorant would be geminated when adjacent to another sonorant, or when it is word initial, it is degeminated.
• When non final vowels may be followed by a glide in the coda. However, when two vowels come together, the result is not a diphthong. Instead, the vowels either remain in hiatus with or without a change in quality or they merge to form a single vowel. If a resulting merged vowel would be overlong, it is shortened.

  Code: Select all

     a   e   i   o   u   y
  a  aa  ai  ai  au  au  aa
  e  ea  ee  ee  eu  eu  ee
  i  ia  ie  ii  yo  yy  ii
  o  oa  oi  oi  oo  oo  oo
  u  ua  ue  yy  uo  uu  uu
  y  ya  ye  yy  yo  yy  yy

  (y represents ɨ in the orthography)

There is also a gradation system for consonant-final stems with three degrees: unaugmented, softened and geminated.

The system is sort-of meant to be Finnish light, though there are obvious differences. The language itself, however, is quite unlike Finnish in its grammar.

[opipik]

/p t ʧ k ʔ/ <p t c k ʼ>
/m n ɲ ŋ/ <m n ñ ŋ>
/s h/ <s h>
/w l r j/ <w l r y>

/i (e) ɑ o u/ <i (e) a o u>

Stress is irregular. <V́, -V>

[Αυτοβοτα]

/p t k kʷ ʡ/
/ts/
/ɸ s x ʜ/
/m n/
/ɹ l w/
/dɾ dl dn/
/ɑ ɛ i o ai ia/

Includes a set of pre-stopped sonorants. Not sure if I want to include /bm/ as well.
/t ɹ dɾ/ are dental and /ts s n dn l dl/ are alveolar.
The dorsals tend toward prevelar, especially /x/.
/ai/ cannot occur in sequential syllables - the earlier one is changed to /ɛ/ or /ɑ/.
/dɾ/ is trill /r/ postvocalically.
/w/ has the allophone [ʍ].
/o/ is generally in open syllables and before /ɸ/ in a closed syllable, [o] in closed syllables and after /ʡ ʜ/.
/s ts/ [ʃ tʃ] / _{i, ia}, {i, ai}_%
/ʡ ʜ/ [qʡ χʜ] / _o
/o/ [ʉ~y] / {t, ɹ, dɾ}_%
/o/ [ɵ~ø] / {t, ɹ, dɾ}_C+%
/p t k kʷ/ [b̥ d̥ g̊ g̊ʷ] / V_V

(C)(C)V(C)(C)
Generally no more than clusters of 2 in any position. Initial and final clusters are restricted. 3+ medial clusters generally involve a sonorant or epiglottal on the periphery.

I have epiglottals without glottals which is completely against every language I know of that has epiglottals so maybe I need to add glottals.

I'm thinking strongly agglutinative or polysynthetic morphology. Speakers are called "treejumpers" and subsist largely on hunting, gathering, and trade.