From a noob: Is this a realistic phonology?

[conlangernoob]

Hi. I’m new to conlanging, and am not sure whether my new Elvish-y (for the lack of a better term) conlang has a realistic phonology. Any advice would be appreciated. Here it is:

Consonant phonemes: p, b, t, d, k, g, f, v, s, x, h, m, n, ŋ, r, l, j, w, ʍ (Note: I would format it as a table, but it would take too much time)
Vowel phonemes: Monophthongs - a e̞ i o̞ u, Diphthongs - [aɪ], [ɛɪ]
Romanization: Consonants - p, b, f, v, m, t, d, s, n, r, l, k, g, and are spelled the same. [j] is written ⟨y⟩. [x] is written ⟨kh⟩. [ŋg] is written ⟨ng⟩. [ŋk] is written ⟨nk⟩. [ʍ] is written ⟨hw⟩. Vowels - a i u are spelled the same. [e̞] is written ⟨e⟩. [o̞] is written ⟨o⟩. [aɪ] is written ⟨î⟩. [ɛɪ] is written ⟨ê⟩.
Stress: Initial root syllable.
Phonotactics:
1. [ŋ] is considered an allophone of [n] and is used only before [k] or [g]
2. Word-Final: t, d, s, n, r, l, nt, nd
3. Word-Initial: All phonemes
4. Word-Medial: All phonemes + lb + lp + lf + lv + lm + lt + ld + ls + ln + lk + lg + mb + mp + mn + nm + nk
+ nt + nd + ps + cw + rm + rn + rp + rb + rd + rt + rk + rg + rf + rv + rs + sc + st + sw + ts + tw
5. All vowels and diphthongs can be found in any position.
6. Word Structure: (CV)CVC(VC)(VC)...
V: Any vowel phoneme
C: Any consonant, or consonant cluster that is legal in that position

[chris_notts]

The inventory isn't obviously weird. Maybe having a voice distinction in the labiodental fricatives but nowhere else is a bit odd, but I wouldn't worry about it too much, since lots of languages have one or two oddities in their inventory.

If [ŋ] is an allophone of /n/ then it's not a phoneme, and you don't need to list it in your consonant phoneme list separately.

It would be helpful to know about morphophonology. For example what happens when morphological processes would create otherwise illegal medial clusters, or vowels in hiatus, or....?

[Darren]

Hi. I’m new to conlanging, and am not sure whether my new Elvish-y (for the lack of a better term) conlang has a realistic phonology. Any advice would be appreciated. Here it is:

Consonant phonemes: p, b, t, d, k, g, f, v, s, x, h, m, n, ŋ, r, l, j, w, ʍ (Note: I would format it as a table, but it would take too much time)

Here you go:

	Labial	Alveolar	Palatal	Velar	Glottal
Stop	/p b/	/t d/		/k ɡ/
Fricative	/f v/	/s/		/x/	/h/
Nasal	/m/	/n/
Approximantl		/r l/	/j/	/ʍ w/	/h/

[conlangernoob]

Thanks!

I think I might have some sort of consonant harmony system to deal with the morphology/phonology problems.

[alice]

A bit late here, but it looks entirely sensible (and oddly like some of mine ) You might have to think about justifying /v ʍ/ when you already have /f w/, but that's not hard.

[Ares Land]

If it helps, this tool here is good for formatting tables for the ZBB:

https://www.zompist.com/phono.html

[WeepingElf]

Looks fine to me.

[Emily]

yeah just chiming in to say the /v/ thing is the only thing that stuck out to me either. i would either expect the language to also have /z/ (and possibly also the voiced equivalent of /x/), or else expect /v/ to have descended from an earlier /w/ and thus the current language to lack /w/. it's not a capital m Mistake, but i would encourage you to think about what the origin of the distinction within the language's internal history might be

[chris_notts]

yeah just chiming in to say the /v/ thing is the only thing that stuck out to me either. i would either expect the language to also have /z/ (and possibly also the voiced equivalent of /x/), or else expect /v/ to have descended from an earlier /w/ and thus the current language to lack /w/. it's not a capital m Mistake, but i would encourage you to think about what the origin of the distinction within the language's internal history might be

Or maybe to lack /b/, especially if the remaining voiced stops have fricative or approximant allophones... but that would raise the question of why b lost its stop allophone but the others didn't.

[alice]

yeah just chiming in to say the /v/ thing is the only thing that stuck out to me either. i would either expect the language to also have /z/ (and possibly also the voiced equivalent of /x/), or else expect /v/ to have descended from an earlier /w/ and thus the current language to lack /w/. it's not a capital m Mistake, but i would encourage you to think about what the origin of the distinction within the language's internal history might be

Or maybe to lack /b/, especially if the remaining voiced stops have fricative or approximant allophones... but that would raise the question of why b lost its stop allophone but the others didn't.

That's something the designer of PIE also has to explain, so you'd be in good company

[Nortaneous]

/f v w/ with no **/z ɣ/ seems not too unreasonable, but /ʍ/ seems implausibly unsystematic - where are the other voiceless sonorants?

It could instead be /xʷ/, but then you'd expect /kʷ gʷ/.

Maybe instead /ʍ w/ are featurally something like /ɸʷ βʷ/ that split from /f v/ somehow? (Before rounded vowels preceding *ø *y > e i and hiatus collapse? Could also explain why there are no diphthongs with rounded onsets - from earlier *aj *ej *oj *uj, *oj *uj > *ø *y > e i, then *aj *ej > aɪ ɛɪ.) But why didn't this produce /pʷ bʷ mʷ/? (How did the ɸ/f and β/v contrasts develop in Ewe?)

I'm also not sure about the phonotactics. Only coronals are permitted word-finally (weird but attested in Finnish), and word-medial clusters all begin with a coronal consonant except mb mp mn nk ps. (What is c in cw sc?) If mb mp are /nb np/, that leaves three clusters, distributed entirely unsystematically. Is this sort of thing attested? I assume the idea is native unstressed syllabies in Germanic, which are biased toward coronals aside from one or two morphemes with velars. (English -ing, German -ung and -ig...)

[Travis B.]

I'm also not sure about the phonotactics. Only coronals are permitted word-finally (weird but attested in Finnish)

Wasn't this also true of Old Spanish?

[foxcatdog]

I see changes like *z > *r and *ɣ > *j/w in this languages prehistory. *w + *ʍ but no other voiceless sonorants isn't unusual see english but the latter derives from earlier *kʷ through voiceless sonorants did exist at some point in english prehistory. Given the circumstances i think that it is apt you didn't include and *h+C clusters.

[bradrn]

*w + *ʍ but no other voiceless sonorants isn't unusual see english

I’d say it’s very unusual, given that I know of no language other than English with that inventory — and for that matter, it seems pretty unstable given that most dialects have lost /ʍ/. (Note that English has a somewhat odd phonology, considered cross-linguistically.)

[WeepingElf]

/f v w/ with no **/z ɣ/ seems not too unreasonable, but /ʍ/ seems implausibly unsystematic - where are the other voiceless sonorants?

It could instead be /xʷ/, but then you'd expect /kʷ gʷ/.

Doesn't Avestan have /xʷ/ but not /kʷ gʷ/? At least, Wikipedia says so.

[alice]

I'm also not sure about the phonotactics. Only coronals are permitted word-finally (weird but attested in Finnish)

Wasn't this also true of Old Spanish?

Yes.

[conlangernoob]

Thanks for the suggestions guys! I revised the phonology to clear some of the inconsistencies and also made the word medial clusters a little more rule based. Here it is:

Consonants:
Labial Alveolar Palatal Velar Glottal
p b t d k g
m n
f s x h
r
l y w

Vowels: a, e, i, o, u. No diphthongs

j is written y
x is written kh

Words of two syllables are stressed on the first syllable. In words of three or more syllables, the stress is on the penultimate syllable it has a diphthong or ends in a consonant, otherwise on the antepenultimate syllable.

[ŋ] is considered an allophone of [n] and is used only before [k] or [g]
[v] is considered an allophone of [w] and is used in some dialects.

Word-Final: t, d, s, n, r, l, nt, nd
Word-Initial: All phonemes
Word-Medial: All phonemes + the possible adjacent consonant pairs using these four rules:
1 The first consonant of a pair of adjacent consonants must be “m”, “n”, “s”, “r”, or “l”
2 Only labial plosives can follow “m”.
3 Only non-labial plosives can follow “n”. Note that /n/ is realized as [ŋ] before [k] or [g].
4 Only obstruents and nasals can follow /r/ or /l/.

Word structure: (V)(CV)CVC(VC)...(V)
V: Any vowel phoneme
C: Any consonant, or consonant cluster that is legal in that position

Thanks again,

conlangernoob

[Richard W]

I'm also not sure about the phonotactics. Only coronals are permitted word-finally (weird but attested in Finnish)

Wasn't this also true of Old Spanish?

Yes.

And Ancient Greek.

[Zju]

Word structure: (V)(CV)CVC(VC)...(V)
V: Any vowel phoneme
C: Any consonant, or consonant cluster that is legal in that position

That's more simply writtten in terms of syllable structure: (C)V(C) (+ a note that hiatuses are disallowed)
A single letter usually represents a single phoneme

[bradrn]

A single letter usually represents a single phoneme

Not always!