Random phonology/phonemic inventory thread [2011–2018]

[Dezinaa]

What about not using alien orthography at all?
/pʼ tʼ kʼ qʼ/ <p t k q>
/fʼ sʼ ɬʼ ʃʼ xʼ Xʼ/ <f s l z x h>
/ʘʔ ǀʔ ǃʔ ǁʔ/ <b d c j>

/sʼkʼǀʔʃʼkʼ qʼfʼ ɬʼǃʼtʼtʼsʼ ǁʔǁʔXʼ pʼ ʃʼqʼtʼsʼʘʔǃʔxʼXʼ pʼfʼsʼqʼ/
<Skdzk qf lctts jjh p zqtsbcxh pfsq.>

Have to admit, it still looks alien

Or even better:

/pʼ tʼ kʼ qʼ/ <p t k w>
/fʼ sʼ ɬʼ ʃʼ xʼ Xʼ/ <e s l c r d>
/ʘʔ ǀʔ ǃʔ ǁʔ/ <a u i o>

/sʼkʼǀʔʃʼkʼ qʼfʼ ɬʼǃʼtʼtʼsʼ ǁʔǁʔXʼ pʼ ʃʼqʼtʼsʼʘʔǃʔxʼXʼ pʼfʼsʼqʼ/
<Skack we litts ooh p cwtsaird pesw.>

Thanks for the suggestions. Vowel characters for consonants is definately alien.

In light of recent posts in this thread, I have to say that I think Dezinaa's original orthography worked fine. Just my opinion.

/pʼ tʼ kʼ qʼ/ <p t k q>
/fʼ sʼ ɬʼ ʃʼ xʼ Xʼ/ <f s l c x h>
/ʘʔ ǀʔ ǃʔ ǁʔ/ <p! t! c! l!>

Ahh, now I'm tempted to draw up an inventory somewhat like this (probably something a little more terrestrial but still with a higher number of non-pulmonic consonants than pulmonic consonants… and maybe voiceless vowels too).

Is that a bad thing? Sometimes when I get phonology ideas I just write it down and save it for later (even though I probably won't end up using it).

---

A simple little thing I came up with. It looks familiar...
I might remove the voiced fricatives.

/m n ɲ ŋ/ <m n ny ng>
/p t t͡ʃ k ʔ/ <p t c k q>
/b d d͡ʒ g/ <b d j g>
/f s ʃ x h/ <f s sh x h>
/v z ʒ ɣ/ <v z zh gh>
/ɾ l j w/ <r l y w>
/i u e o a/ <i u e o a>

[DesEsseintes]

Possible idea for a Hííenununóóoþa relative:

/n/ n
/p t k/ p t k
/s h/ s h
/ð̞ j w ɰ/ d y w g
/a i o/ a ı o

It's basically Blackfoot minus the glottal stop and plus more glides.

I'm wondering whether to add /m z/... The latter is probably a bad idea as /ð̞ j w ɰ/ pattern nicely with /s t p k/ respectively.

Then there's whether to have /l/.

Phonotactics would be relatively free. All sounds but glides can be geminate and cluster in all kinds of ways; I envisage words such as waayıhkssoaagodaht.

[Birdlang]

Possible idea for a Hííenununóóoþa relative:

/n/ n
/p t k/ p t k
/s h/ s h
/ð̞ j w ɰ/ d y w g
/a i o/ a ı o

It's basically Blackfoot minus the glottal stop and plus more glides.

I'm wondering whether to add /m z/... The latter is probably a bad idea as /ð̞ j w ɰ/ pattern nicely with /s t p k/ respectively.

Then there's whether to have /l/.

Phonotactics would be relatively free. All sounds but glides can be geminate and cluster in all kinds of ways; I envisage words such as waayıhkssoaagodaht.

What would you name this language? I think only the lateral works, as well as m.

[Birdlang]

A good idea for a jokelang.
/q/
/G\/
/X R/
/R\/
/G\_X/
/Q/ is the vowel.

[gach]

Over the Christmas I found myself writing down a phonology somewhat influenced by Old Tibetan. It'll be for a language with the main purpose of providing loans for Kišta and it's neighbours especially for newly imported culture items. The phoneme inventory of the classical language is:

Code: Select all

Consonants
p   t       k   q
b   d   dʑ  g
f   s       x
v   r   ʑ       ʁ
m   n
w   l   j

Vowels
i     u
e     o    with an epenthetic [ə]
   a

The basic syllable structure is

C{w,j}[ə]sCC1VC2s

with
C1 = /f s x ʑ r l w j/
C2 = /p t k f s x n(=[ŋ]) r l/

However, this should be dealt with in two parts, the main syllable sCC1VC2s and a possible prefixed sesquisyllable C{w,j}ə. These prefixed minor syllables are what's left from worn down unstressed compound elements largely serving derivation. They are attached to the main syllable by a non-phonemic [ə] which for certain consonant combinations can drop out entirely. In addition to this the main syllable also allows original onset clusters. These are more restricted and despite the fact that you can give the formula sCC1 for them, only the following are actually possible:

Code: Select all

sp st sdʑ(=[stɕ]) sk sq
ps ks qs
tf kf qf
px tx
bʑ gʑ
br dr gr
bl dl gl vl źl
Cw Cj

At the end of the main syllable the following /s/ final clusters are possible: /ps ts ks qs ns(=[ŋs])/. No vocalic onsets are allowed, the minimum syllable structure is CV.

Already at the classical stage the language also allowed suffixed unstressed syllables. These must always have the rigid structure CV without even single word final consonants and likewise the main syllable must be open before them. As the language evolves the suffixed syllables erode and new compounds develop. This will give the language proper intervocalic consonant clusters and polysyllabic words with final consonants. The prefixes will also erode typically either collapsing onto the main syllable onset or deleting altogether. Sometimes though the resulting clusters are too heavy and the epenthetic [ə] clings on eventually turning back into a full vowel.

The name of this language will be Lkal sik from l-kal si-k (MOUTH/SOUND.PREFIX-DEMONYM speak-NMLZ) but it'll evolve through [lʲə'kalʲ siʔ] and [jə'kelʲ si] into /kelsi/ in the modern colloquial language. Thus when I write a longer description of the language it'll be a Kelsi grammar.

[Thrice Xandvii]

Two things:

C{w,j}[ə]sCC1VC2s

Doesn't this mean that every word (not syllable, unless somehow this really is a syllable structure, in which case I can't imagine how one can utter it properly) has: 4 mandatory consonants, two mandatory /s/s, and at least one vowel (not including the likely glide)?

Thus when I write a longer description of the language it'll be a Kelsi grammar.

Very nice, much success!

[DesEsseintes]

...phonology...

What would you name this language? I think only the lateral works, as well as m.

I might call it Olıosstoohwa or Olıommtoohwa. Since last night, the phonology has grown into this:

/m n/ m n
/p t k/ p t k
/s h/ s h
/ð̞ j w ɰ/ d* y w g*
/l/ l
/a i o/ a ı o
*still under consideration


Some Allophony:
- /i/ is normally [ɪ] but is before /j/, [ɨ] before /ð̞/, and [ɯ] before [ɰ]
- /a/ is mostly [ɑ], but see below
- /o/ is normally [ɔ], but [o] before most glides
- /oo/ is [oː~uː]
- geminate glides /ð̞ð̞ jj ww ɰɰ/ fortify to fricatives [ðː ʝː ɣː ɣʷː~βː]

Some morfofo:
g → y / ı_ı
ı o → y w / V(h)_V
a aa aı → [æ ɛː eɪ̯] e ee eı / _{d y}i
ıı aa oo → ıyı aga owo / _#, _CC

Some Phonotactics:
- all sounds but /h/ can occur geminate
- geminate /m n s l/ can be syllabic
- words can end in p t k h or any vowel
- h can only occur in coda
- glides elide word-initially and after another consonant except /h/
- short /m n/ assimilate to PoA of following stop but geminate (and syllabic) /mm nn/ do not, so sequences such as kmmt nnp are possible

Ponderables:
- I know it makes no sense but I really want to use something other than d or z for the alveolar approximant. I dislike ð too. I wonder if I should use đ, ȷ or even ẙ. Any preferences/other ideas?
It's the same dilemma I had when I still planned to have glides in Hííenununóóoþa, and part of the reason I got rid of them in that language.

[gach]

C{w,j}[ə]sCC1VC2s

Doesn't this mean that every word (not syllable, unless somehow this really is a syllable structure, in which case I can't imagine how one can utter it properly) has: 4 mandatory consonants, two mandatory /s/s, and at least one vowel (not including the likely glide)?

No, since instead of cluttering the formula with parentheses marking optionality I opted for a verbal description. The sets of parentheses already in there stand for an inline set of glides and a non-phonemic [ə]. More than a decade of reading linguistic texts has shown me that slightly bending the strict notation rules to fit your specific needs is often much clearer for the reader than blindly following them no matter what falls in front of you. Just how often have you been OK or not even noticed when people use their own tweaked phonemic notations?

A fuller notation for words with one and two full syllables is

(C({w,j})([ə]))(s)C(C1)V(C2(s))
(C({w,j})([ə]))(s)C(C1)VCV

How you want to treat the prefixed minor syllables C{w,j}[ə] is up to personal preferences. They can only appear at the start of the word and the [ə] isn't phonemic. If you want, you can then call them either syllables of their own right or phonemically mere loose parts of the following full syllable.

Thus when I write a longer description of the language it'll be a Kelsi grammar.

Very nice, much success!

I regret nothing.

[zyma]

Thus when I write a longer description of the language it'll be a Kelsi grammar.

Very nice, much success!

I regret nothing.

I audibly groaned when I first read that. It's perhaps the greatest pun I've ever come across.

---

I might call it Olıosstoohwa

Out of the two possibilities shown, I personally prefer this name.

/m n/ m n
/p t k/ p t k
/s h/ s h
/ð̞ j w ɰ/ d* y w g*
/l/ l
/a i o/ a ı o
*still under consideration

Are the phonemes themselves /ð̞ ɰ/ in doubt, or is it simply the orthography that's under consideration? Looking at the "Ponderables" section, I'd guess the latter. What are you considering as alternatives to g for /ɰ/?

- geminate glides /ð̞ð̞ jj ww ɰɰ/ fortify to fricatives [ðː ʝː ɣː ɣʷː~βː]

I assume you mean /ð̞ð̞ jj ww ɰɰ/ > [ðː ʝː ɣʷː~βː ɣː]?

Some morfofo:
g → y / ı_ı
ı o → y w / V(h)_V
a aa aı → [æ ɛː eɪ̯] e ee eı / _{d y}i
ıı aa oo → ıyı aga owo / _#, _CC

Especially to the last two rules.

Some Phonotactics:
- all sounds but /h/ can occur geminate
- geminate /m n s l/ can be syllabic
- words can end in p t k h or any vowel
- h can only occur in coda
- glides elide word-initially and after another consonant except /h/
- short /m n/ assimilate to PoA of following stop but geminate (and syllabic) /mm nn/ do not, so sequences such as kmmt nnp are possible

I like how some geminates can be syllabic!

It's unexpected and interesting that /h/ can only appear in a coda. Can geminate stops occur word-finally? Can other consonants occur in the coda position word-medially?

When glides elide, do they disappear without a trace?

So short nasals assimilate in POA to following stops, and geminate syllabic nasals do not, but what about geminate non-syllabic nasals?

Could sequences like kmmt nnp be independent words/roots on their own, or would such sequences be bound morphemes only? I think it would be cool to have words in which the only nuclei were syllabic consonants (I especially like the look of kmmt).

Ponderables:
- I know it makes no sense but I really want to use something other than d or z for the alveolar approximant. I dislike ð too. I wonder if I should use đ, ȷ or even ẙ. Any preferences/other ideas?
It's the same dilemma I had when I still planned to have glides in Hííenununóóoþa, and part of the reason I got rid of them in that language.

It would be quite a shame if you were to get rid of /ð̞/ because of an orthographic dilemma. And, by the way, I think it makes perfect sense to want a language to have a certain aesthetic.

My vote would either go to ȷ or ẙ.

[DesEsseintes]

Wow, many questions!

I assume you mean /ð̞ð̞ jj ww ɰɰ/ > [ðː ʝː ɣʷː~βː ɣː]?

Ah, yes.

Out of the two possibilities shown, I personally prefer [Olıosstoohwa].

I am now actually leaning towards Olıommstoohwa cos I like the s but also want to show off the syllabic nasal.

Are the phonemes themselves /ð̞ ɰ/ in doubt, or is it simply the orthography that's under consideration? Looking at the "Ponderables" section, I'd guess the latter. What are you considering as alternatives to g for /ɰ/?

Only the orthography is under consideration. I definitely want all the approximants.

I was thinking of /ð̞ ɰ/ ẙ ẘ but I'm not convinced. Perhaps ȷ g would be better... I don't mind g quite as much as d.

It's unexpected and interesting that /h/ can only appear in a coda. Can geminate stops occur word-finally? Can other consonants occur in the coda position word-medially?

Unfortunately, /h/ only occurring in coda was not my idea. That's what it's like in Blackfoot with /x/. This language was born from my 5th (or 6th?) reading of Blackfoot grammar.

Geminate stops cannot occur word-finally.

I haven't worked out the exact phonotactics yet, but weird clusters will be allowed.

So short nasals assimilate in POA to following stops, and geminate syllabic nasals do not, but what about geminate non-syllabic nasals?

Yes, that's right about the assimilation. Geminate nasals are always syllabic when adjacent to consonants. They are not syllabic when V_V.

Could sequences like kmmt nnp be independent words/roots on their own, or would such sequences be bound morphemes only? I think it would be cool to have words in which the only nuclei were syllabic consonants (I especially like the look of kmmt).

Probably not, although it is tempting...

[zyma]

I am now actually leaning towards Olıommstoohwa cos I like the s but also want to show off the syllabic nasal.

I can definitely understand that. I like it!

I was thinking of /ð̞ ɰ/ ẙ ẘ but I'm not convinced. Perhaps ȷ g would be better... I don't mind g quite as much as d.

Both options look good to me! Dotless <i j> and letters with rings above them are all very aesthetically pleasing symbols, in my opinion.

Unfortunately, /h/ only occurring in coda was not my idea. That's what it's like in Blackfoot. This language was born from my 5th (or 6th?) reading of Blackfoot grammar.

Eh, I still think it's cool that you're using the idea here.

And I still have to find time to read through those grammars… damn real life responsibilities!

[DesEsseintes]

And I still have to find time to read through those grammars… damn real life responsibilities!

It's very unfair that bothersome things such as education, money, significant otherses, pets, parents, etc. should get in the way of one's conlanging, don't you find?

[zyma]

And I still have to find time to read through those grammars… damn real life responsibilities!

It's very unfair that bothersome things such as education, money, significant otherses, pets, parents, etc. should get in the way of one's conlanging, don't you find?

Oh, absolutely.

[Dezinaa]

/m n ŋ/ <m n ng>
/p t t͡ʃ k q/ <p t c k q>
/b d d͡ʒ g/ <b d jj g>
/pʰ tʰ kʰ qʰ/ <ph th kh qh>
/f s ɬ ʃ x h/ <f s lh sh x h>
/v z ʒ ɣ/ <v z zh gh>
/l j w/ <l j w>
/i ɨ u e o a/ <i y u e o a>

Who knows if I'll actually use this.

[Birdlang]

Once I had a gleb language with a bilabial trill and bilabial flap, it only had voiceless stops, including the glottal stop and the voiceless uvular plosive. The vowels were Standard Average European, and I counted only twenty consonants.

[Squall]

Vowels:
a /ä ɐə/ or /ə/
e /ɛ ej/ or silent
i /ɪ i:/
o /ɔ ow/
u /ʊ u:/

Rules:
Each vowel is long or short.
The stress is in the last long vowel, or in the last syllable if no syllable is long.
The vowel is short if the syllable ends with a consonant or a silent 'h', or the next syllable starts with doubled consonant.
The vowel is long if it is the last letter of the syllable or there is a schwa or a silent 'e' after the next consonant.
'e' and 'a' have to be doubled to not be silent.

Examples:
koh /kɔ/
ko /kow/
kok /kɔk/
koke /kowk/

kosk /kɔsk/
koske /kowsk/

koski /kɔski:/
koseki /kowski:/
koki /kowki:/
kokki /kɔki:/

kokket /kɔkɛt/
koket /kowkɛt/

kokti /kɔkti:/
koketi /kowkti:/
koketti /kowkɛti:/
kokeeti /kowkejti:/
kokeetti /kowkɛti:/ - wrong orthography

koka /kɔkə/
kokea /kowkə/

kokati /kɔkəti:/
kokeati /kowkəti:/
kokatti /kowkäti:/
kokkatti /kɔkäti:/
kokkati /kɔkɐəti:/
kokaati /kowkɐəti:/
kokeatti /kowkäti:/ - wrong orthography

koi /kɔi:/
kooi /kowi:/

[zyma]

Hmm… This has a very SAE feel to it (not meant as an insult, sorry, just a statement). Interesting how <e> can also be silent word-medially.

[zyma]

I haven't posted any inventories here for a while. I'm not sure where this one came from, or what I'll do with it, but something about it charms me. It… kind of reminds me of Dutch, but I couldn't tell you why or how. It definitely feels European, at least.

/p t k q/ p t k q
/θ s x χ/ c s h r
/t͡ʃ/ j
/n/ n
/j w/ ẙ ẘ

/i ɪ e ɛ/ iy i ei e
/ɐ ä/ ea a
/u ʊ o ɔ/ uw u ou o

Phonotactics: (S)(C)(R)V(C)

S = any coronal fricative
C = any consonant
R = any approximant or uvular fricative
V = any vowel
C = any consonant

I might restrict initial clusters somewhat at a later date, but I might not.

Allophony and such:

/χ/ will likely become a voiced uvular approximant when following other consonants.

Obstruents are glottalized/unreleased syllable-finally (or maybe just word-finally?).

The vowel nucleus of a syllable may become pharyngealized if the syllable's coda is an obstruent.


Examples:

rean /χɐn/
juwt /t͡ʃut/
ceiẘ /θew/

---

I currently have an idea for a sprachbund of sorts, spoken on the largest continent in my conworld's northwestern hemisphere. It's currently made up of three unrelated protolangs. This is the inventory of the third. Each language in the sprachbund is spoken in its own village, and all three villages are located along the southern shore of a rather sizable glacial lake. There will probably also be a fairly small mountain range to the southeast. I enjoy this rather strange romanization. As far as morphosyntax goes, I might base this language on a combination of Germanic, Algic, Siouan, and Iroquoian languages.

/p t k/ p t c
/θ s ʃ h/ th s sh h
/t͡ʃ d͡ʒ/ ch jh
/m n/ m n
/β̞ j/ w j
/r/ r
/l/ l

/i ɛ a/ i e a
/o/ u

/aɪ̯/ ej

Phonotactics: (C)(N)(R)V(C)

C = any consonant
N = any nasal
R = any approximant, trill, or lateral approximant
V = any vowel or diphthong
C = any consonant

Only obstruents can appear before a nasal in onset clusters. I'm not sure if I want to put in some kind of POA agreement rule there or not, and I'm not sure whether or not I want to allow three-consonant onset clusters. But I guess those are problems for another day.

Allophony and such TBD.


Examples:

ejthur /aɪ̯θor/
lacan /lakan/
wuh /β̞oh/

[ol bofosh]

Been a while since I've done this. Just brushing the dust off the old phonology neurons.

/m n/ <m n>
/p pʰ b t tʰ d k kʰ g/ <p ph b t th d k kh g>
/tʃ tʃʰ dʒ/ <c ch j>
/f s/ <f s>
/w j/ <w y>
/l/ <l>
/r/ <r>

Phonemic gemination for nasals, plosives and affricates between vowels
Written <mm nn pp pph bb tt tth dd kk kkh gg cc cch jj>

/i i: u u:/ <i ì u ù>
/ʊ ɪ/ <û î>
/e eː o oː/ <e è o ò>
/ɛ ɔ/ <ê ô>
/a aː ɑ/ <a à â>

Dipthongs: /wV jV/

Stress: first syllable

Syllable structure: CFVN
(F = w or j, N = m n l r)

Edit: some provisional examples of one or two syllables using Awkwords (replaced <â ê î ô û> with <aa ee ii oo uu>)
oo atjó ál phwu dwuum cil baan feetté á pwel iir wanno jiiluu dapphi júl al úrceen caagó wun jeenphé noomthá phaam kel phággi jílchi wúl jaatthol ukhir pwu chó aa roo phálnúm óm tin swo jitthá wá gjópphuu chiithjé ee phjápphuun thúddi khuul thuunniil wo koo lweekkuu weelkjiir njul jú oon suu soon ge kím julpoor pérkee mi kjoddoon joochwul cjaamnjeem jéddúm woo khwí phjíbiil u uun oppaa wiippin lúl oocchól wettoo khun wajji oom phi jechí ul chaa jaa ándwam jibii chí úm pjaa rjiir chémbí jettór mwi waacchér weecchuun bwootjii jéddó thwe ajje

[zyma]

Is stress 100% regular?