Random phonology/phonemic inventory thread [2011–2018]

A forum for all topics related to constructed languages
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random phonology/phonemic inventory thread

Post by eldin raigmore »

cromulant wrote:Onsets: a i u
These are nuclei, not onsets.
Are all of the syllables onsetless? That is, vowel-initial?

cromulant wrote:Codas: p pm t tn m mp n nt
So, there are no onsets -- only nuclei and codas?

cromulant wrote:Word-finally, i and u are also allowable syllables.
So, almost all syllables mandatorily have codas, except word-final syllables can be the bare nuclei i and u?

[hr][/hr]
[hr][/hr]

The usual way of analyzing syllables is to divide them into an (often optional, often mandatory) onset and a mandatory rime; then to further divide the rime into a mandatory nucleus and a (usually optional, sometimes mandatory) coda.
Another not-all-that-infrequent way of analyzing syllables is to divide them into a mandatory body and a (usually optional, sometimes mandatory) coda; then to further divide the body into an (often optional, often mandatory) onset and mandatory nucleus.

For most syllables in most languages, the nucleus consists of the syllable's vowels; the onset consists of all the syllable's consonants before the nucleus; and the coda consists of all the syllable's consonants after the nucleus.
Many languages have some syllables that contain no vowels. Usually these are analyzed as having their most sonorantous consonant as their nucleus; and usually they contain at most one other consonant, which, if such a consonant occurs at all, is either that syllable's one-consonant onset or its one-consonant coda.
However I have seen a discussion of some of the languages of the Pacific Northwest of North America which analyzes some of their syllables as having onsets and codas but no nuclei.

A syllable without a coda is an "open" syllable. One with a coda is a "closed" syllable. One with a consonant-cluster for its coda is (at least) "doubly closed".
If a syllable-nucleus is a vowel-cluster it is called a "polyphthong". "Diphthong" if a two-vowel cluster, "triphthong" if a three-vowel cluster, "tetraphthong" if a four-vowel cluster (these seem to be rare), and there do not appear to be (information subject to change) any five-vowel-or-longer vowel-clusters that don't span a syllable-boundary. A word-internal syllable boundary between a vowel-final syllable and the following vowel-initial syllable is called a "hiatus".

Many languages don't have any closed syllables at all.
Many languages don't have any tautosyllabic consonant-clusters at all. (I.e. if a syllable has an onset it can only be a single consonant; and if a syllable has a coda it can only be a single consonant.)
Many languages have no polyphthongs.
Many languages require an onset to each syllable.

So, many languages' syllables are all CV syllables.
And many languages' syllables are all (C)V syllables; no codas, no polyphthongs, no tautosyllabic consonant-clusters, but onsets are optional instead of mandatory.

Rare languages have mandatory codas.
Rare languages have no onsets.
If a language's syllable structure were, for instance, V(V)C(C(C)) -- no onsets, one-or-two-vowel nuclei, one-to-three-consonant codas -- it would be such a language.


[hr][/hr]
[hr][/hr]

I'm guessing you meant your twenty-six syllables to be:

Code: Select all

  ap apm at atn am amp an ant
i ip ipm it itn im imp in int
u up upm ut utn um ump un unt
; is that right? This syllable structure would be V(C(C)); non-final syllables would be VC(C). There would be no onsets; only final syllables could be open; syllables could be closed or doubly-closed, but there could not be any three-or-more-consonant clusters.

Could you have meant instead:

Code: Select all

  pa pma ta tna ma mpa na nta
i pi pmi ti tni mi mpi ni nti
u pu pmu tu tnu mu mpu nu ntu
? I think you'ld've wanted i and u confined to word-initial, rather than word-final, if that were the case. Syllable structure would've been (C(C))V; non-initial syllables would've been C(C)V.

I could be wrong, but I think apm ipm upm atn itn utn are kind of exotic as syllables in natlangs.
(I think pma pmi pmu tna tni tnu mpa mpi mpu nta nti ntu are also exotic, though I know of an African language with mpa mpi mpu nta nti ntu, and Modern Greek is at least spelled with, in effect, <mp> as a syllable onset. But not only could I be wrong; I don't think you meant any of those to be among your 26 syllables.)
Last edited by eldin raigmore on 26 Jun 2017 04:34, edited 2 times in total.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

Well some Aboriginal Australian languages have analyses like this: Arrernte and a few Paman languages (like Uw Oykangand/Kunjen), which are analysed as having obligatory codas but no onsets.
cromulant
cuneiform
cuneiform
Posts: 143
Joined: 13 Jan 2012 00:04

Re: Random phonology/phonemic inventory thread

Post by cromulant »

Eldin: yes, I meant nuclei instead of onsets. There are no onsets. Syllable structure is V(C)(C).

I know the terminology, I was just careless with it.

Beyond that, the phonology is as I stated, exotic though it may be. The 26 syllables you listed the first time are correct.

And yeah, it is kind of like a dumbed-down Arrernte.
cromulant
cuneiform
cuneiform
Posts: 143
Joined: 13 Jan 2012 00:04

Re: Random phonology/phonemic inventory thread

Post by cromulant »

New version of dumbed-down Arrernte: add velars /k kN N Nk/. Only 2 vowels, /a @/. Codaless final syllables allowed with both vowels.

X sampa because on phone.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

cromulant wrote:New version of dumbed-down Arrernte: add velars /k kN N Nk/. Only 2 vowels, /a @/. Codaless final syllables allowed with both vowels.

X sampa because on phone.
Get an app for your phone for IPA: If you're on Android I can recommend IPA Keyboard. You'll have to go looking if you're on IOs but there should be one out there.
Clio
sinic
sinic
Posts: 228
Joined: 27 Dec 2012 23:45

Re: Random phonology/phonemic inventory thread

Post by Clio »

@DV82LECM and @cedh: Thank you both very much! I didn't really think at all about allophony, I must admit, but I really like the idea of the glides surfacing at times. Perhaps they could appear as [w j ɰ] in anlaut and as voiced fricatives between vowels of same frontness. Having [sɨ] would also be super creative; I had basically just been thinking that /h/ would have a lowering effect on the vowels, but getting another totally different consonant phone is nicely outside of the box.
Niûro nCora
Getic: longum Getico murmur in ore fuit
scratchpad
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random phonology/phonemic inventory thread

Post by eldin raigmore »

Frislander wrote:Well some Aboriginal Australian languages have analyses like this: Arrernte and a few Paman languages (like Uw Oykangand/Kunjen), which are analysed as having obligatory codas but no onsets.
Those are the languages I was thinking of in the three-line paragraph starting "Rare ...".

cromulant wrote:Eldin: yes, I meant nuclei instead of onsets. There are no onsets. Syllable structure is V(C)(C).
Beyond that, the phonology is as I stated, exotic though it may be. The 26 syllables you listed the first time are correct.
And yeah, it is kind of like a dumbed-down Arrernte.
Thank you!

cromulant wrote:I know the terminology, I was just careless with it.
I guessed as much.
I thought most of the people on this thread -- including you in particular -- did know the terminology.
Apparently you took no offense at me telling you what you already knew. If you took no offense -- I'm glad! If you did take offense -- I apologize; none was intended.
Having once been a newbie, I can imagine newbies reading this thread with interest combined with a bit of confusion.
Just in case that were the case, I thought it might be worthwhile to review the terminology and a few known facts.
I hope at the worst I didn't get in anyone's way, and at the best I helped someone.


I'm interested in the homorganic voiceless-stop-followed-by-voiced-nasal codas; [ pm tn kN ]. How common are they among languages-in-general? Any idea?
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Inspired by the tonogenesis thread, a contribution to my own 26-syllable challenge:

There are three vowels, which distinguish two tones: low and high.

There are four phonemic consonants. The only POA distinction is labial/lingual; the only MOA distinction is nasal/non-nasal.

After labial consonants, /u/ dissimilates to /i/. /n/ becomes non-nasal and lateral between vowels.

Syllable structure is (C)V. Hiatus is permitted; no glides are inserted between adjacent vowels.

mi mí ma má
ɲi (ʎi) ɲí (ʎí) nu (lu) nú (lú) na (la) ná (lá)
vi ví wa wá
tʃi tʃí tsu tsú ka ká
i í u ú a á

Painstakingly typed out on my phone since I'm on vacation.
jimydog000
greek
greek
Posts: 583
Joined: 19 Mar 2016 04:14
Location: Australia

Re: Random phonology/phonemic inventory thread

Post by jimydog000 »

Proto-Rretgia-Dohavenang:

Code: Select all

/m, n, ŋ/ 
/p, t, k, ʔ/
/b, d, g/	 		 
/pʼ, tʼ, kʼ/
/l/	 	 	 	 
/r, j, w, ʕ/ 	 
 
/i, ɨ, ɯ~u/
/ɛ, ɜ/
/a/

Syllable structure: (C(G))V(F)
C: Any consonant
G: /w, r, ʢ/
V: Vowel
F: Any consonant except /j/ and /ʔ/.

There are rules against the glottal stop being used with ejectives at the syllable/suffix/root level.
/j/ is never consonant-clustered inside a root or suffix.
/ɯ/ is in free variation /u/
/a/ is central

Stress is lexical, it is on the first or second syllable.
It's getting a little stale as I'm adding words, should I add length to /i, a/ and /u/? or make a palatal series?
A signature.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

/p t t͡ʃ k ʔ/
/t’ t͡ʃ’ k’/
/s ʃ x/
/w~m ɹ~n/

/i ĩ ĕ ə̆ a ã u ũ ŏ/

The reduced vowels do not occur in stressed syllables, though the full vowels may occur in unstressed syllables.

The approximants are realised as nasals when adjacent to nasal vowels.

Syllable structure is (C)CV(C), where any consonant may appear word-initially while only plain stops and fricatives may appear word finally. Clusters consist of:
  • Plain stop plus fricative or approximant.
  • Plain stop plus another stop (plain or ejective).
  • Fricative plus any non-fricative.
In certain contexts (most notably phrase-finally and in pre-tonic position) the full vowels become reduced, with further loss of nasalisation, in this pattern:

Code: Select all

i, ĩ| ĕ
a, ã| ə̆
u, ũ| ŏ
----
jimydog000 wrote:Proto-Rretgia-Dohavenang:

Code: Select all

/m, n, ŋ/ 
/p, t, k, ʔ/
/b, d, g/	 		 
/pʼ, tʼ, kʼ/
/l/	 	 	 	 
/r, j, w, ʕ/ 	 
 
/i, ɨ, ɯ~u/
/ɛ, ɜ/
/a/

Syllable structure: (C(G))V(F)
C: Any consonant
G: /w, r, ʢ/
V: Vowel
F: Any consonant except /j/ and /ʔ/.

There are rules against the glottal stop being used with ejectives at the syllable/suffix/root level.
/j/ is never consonant-clustered inside a root or suffix.
/ɯ/ is in free variation /u/
/a/ is central

Stress is lexical, it is on the first or second syllable.
It's getting a little stale as I'm adding words, should I add length to /i, a/ and /u/? or make a palatal series?
I'd say add the long vowels, but why not use a word generator? I actually find it quite amazing how easily I get stuck into repeated patterns with word creation that are hard to break out of, a word generator would be very helpful in that regard.
jimydog000
greek
greek
Posts: 583
Joined: 19 Mar 2016 04:14
Location: Australia

Re: Random phonology/phonemic inventory thread

Post by jimydog000 »

Frislander wrote: I'd say add the long vowels, but why not use a word generator? I actually find it quite amazing how easily I get stuck into repeated patterns with word creation that are hard to break out of, a word generator would be very helpful in that regard.
I think I might not add a long /a/ though, you're right about wordgens breaking patterns.

26 syllable challenge:

/p, t, k, kʷ/
/n/
/β, s/
/w, ɾ/
/a, i, u/

CV

kʷi becomes/is: kʷu
A signature.
User avatar
LinguoFranco
greek
greek
Posts: 613
Joined: 20 Jul 2016 17:49
Location: U.S.

Re: Random phonology/phonemic inventory thread

Post by LinguoFranco »

I want to use a three vowel system, either something like the typical /a i u/, /a e o/ or a vertical vowel system along the lines of something like /a ə~ɜ ɨ/. What are some good guides for realistic allophony?

I know that alot of languages with /a i u/ tend to lower the high vowels to /e o/ when they are next to an uvular consonant like /q/, but what kind of allophony might occur in a language without uvualars?
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

LinguoFranco wrote:I want to use a three vowel system, either something like the typical /a i u/, /a e o/ or a vertical vowel system along the lines of something like /a ə~ɜ ɨ/. What are some good guides for realistic allophony?

I know that alot of languages with /a i u/ tend to lower the high vowels to /e o/ when they are next to an uvular consonant like /q/, but what kind of allophony might occur in a language without uvualars?
Have you checked out Cheyenne? Cheyenne is often listed as having /a e o/ but /e o/ have raised [ɪ ʊ] and glide [j w] realisations especially when adjacent to other vowels.

Vowels could centre or lower before peripheral consonants (i.e. anything non-coronal), like in my conlang Apwapɨrɨng (shameless plug). You could check out the natlang Yimas for where I got the inspiration from.

If your syllable structure isn't strictly CV, open and closed syllables might be a good way to have allophony. Open stressed syllables might develop schwa offglides, etc. Oh, and stress obviously gives you lots of possibilities for allophones. Perhaps unstressed syllables tend to centre vowels.
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Some minimalism.

/n/
/k kʷ/
/h hʷ/
/w/

/e i ɑ u/

k h → t s / _F
kʷ hʷ w → t͡ʃ ʃ j / _F
t → t͡s / _i

If I was actually going to take this seriously, I'd work out some more allophony.
User avatar
LinguoFranco
greek
greek
Posts: 613
Joined: 20 Jul 2016 17:49
Location: U.S.

Re: Random phonology/phonemic inventory thread

Post by LinguoFranco »

DesEsseintes wrote:
LinguoFranco wrote:I want to use a three vowel system, either something like the typical /a i u/, /a e o/ or a vertical vowel system along the lines of something like /a ə~ɜ ɨ/. What are some good guides for realistic allophony?

I know that alot of languages with /a i u/ tend to lower the high vowels to /e o/ when they are next to an uvular consonant like /q/, but what kind of allophony might occur in a language without uvualars?
Have you checked out Cheyenne? Cheyenne is often listed as having /a e o/ but /e o/ have raised [ɪ ʊ] and glide [j w] realisations especially when adjacent to other vowels.

Vowels could centre or lower before peripheral consonants (i.e. anything non-coronal), like in my conlang Apwapɨrɨng (shameless plug). You could check out the natlang Yimas for where I got the inspiration from.

If your syllable structure isn't strictly CV, open and closed syllables might be a good way to have allophony. Open stressed syllables might develop schwa offglides, etc. Oh, and stress obviously gives you lots of possibilities for allophones. Perhaps unstressed syllables tend to centre vowels.
I've checked out Cheyenne, but I was not aware that its allophones were [ɪ ʊ]. I think Amuesha (I think that is its name) of New Guinea has /a e o/ too, with [ɪ ʊ] as a result of free variation.

My conlang has palatal consonants and palatalization as a sound change, and I hear that that often leads to vowel fronting. So if I have /a i u/ for example. /a u/ could be fronted to /æ ʉ/.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

My shameless Nuer/Dinka ripoff.

/p t̪ t c k/
/b d̪ d ɟ g/
/m n̪ n ɲ ŋ/
/l̪ ɹ/

Intervocalically the voiced stops are realised as [β ð ɾ j ɣ] while [c] and are in free variation. The voiceless stops may occur geminated when intervocalic.

The vowels are /i e ɛ a u o ɔ/. This inventory is supplemented by three-way length distinction between short, long and overlong; a three-way phonation distinction between modal, breathy and creaky voice; and tone, consisting of high, low and falling, where falling only occurs on long and overlong vowels.

Syllable structure is (C)V(C), where V and VC only occur word-initially and coda consonants are restricted to /t c k n ɹ/.

----

LinguoFranco wrote:I think Amuesha (I think that is its name) of New Guinea has /a e o/ too, with [ɪ ʊ] as a result of free variation.


Actually it's spoken in Peru, you were out in your geography there.

Also in South America, we have Tehuelche in Patagonia which has this system, with the addition of a length contrast.
User avatar
LinguoFranco
greek
greek
Posts: 613
Joined: 20 Jul 2016 17:49
Location: U.S.

Re: Random phonology/phonemic inventory thread

Post by LinguoFranco »

Frislander wrote:My shameless Nuer/Dinka ripoff.

/p t̪ t c k/
/b d̪ d ɟ g/
/m n̪ n ɲ ŋ/
/l̪ ɹ/

Intervocalically the voiced stops are realised as [β ð ɾ j ɣ] while [c] and are in free variation. The voiceless stops may occur geminated when intervocalic.

The vowels are /i e ɛ a u o ɔ/. This inventory is supplemented by three-way length distinction between short, long and overlong; a three-way phonation distinction between modal, breathy and creaky voice; and tone, consisting of high, low and falling, where falling only occurs on long and overlong vowels.

Syllable structure is (C)V(C), where V and VC only occur word-initially and coda consonants are restricted to /t c k n ɹ/.

----

LinguoFranco wrote:I think Amuesha (I think that is its name) of New Guinea has /a e o/ too, with [ɪ ʊ] as a result of free variation.


Actually it's spoken in Peru, you were out in your geography there.

Also in South America, we have Tehuelche in Patagonia which has this system, with the addition of a length contrast.


Ah, idk how I got the geography mixed up.
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Random phonology/phonemic inventory thread

Post by Vlürch »

/ʔ/ <q>
/b͡ʙ/ <p>
/ɓ ᶑ/ <b d>
/ɻʷ/ <r>

/ə ɞ ɵ/ <a e o>

Consonant clusters are broken up by a schwa, which also occurs after word-final consonants. So, basically, there are no word-final consonants or consonant clusters. All vowels adjacent to one another are separated by a voiced pharyngeal fricative. Word-initial vowels are preceded by a glottal stop, so basically, there are no word-initial vowels either. The only allophony is that all vowels become [ɚ] before retroflex consonants.

hello - qpopeqbpqadq [ʔəb͡ʙɵb͡ʙɞʔəɓəb͡ʙəʔɚᶑəʔə]
bye - poqqdeeaq [b͡ʙɵʔəʔɚᶑɞʕɞʕəʔə]
what - qqa [ʔəʔə]
why - ppea [b͡ʙəb͡ʙɞʕə]
when - adqoodoe [ʔɚᶑəʔɵʕɚᶑɵʕɞ]
anus - poq [b͡ʙɵʔə]
mouth - dob [ᶑɵɓə]
life - parp [b͡ʙɚɻʷəb͡ʙə]
death - porp [b͡ʙɚɻʷəb͡ʙə]
love - barb [ɓɚɻʷəɓə]
hate - borb [ɓɚɻʷəɓə]
to come - drrdrr [ᶑɚɻʷɚɻʷɚᶑɚɻʷɚɻʷə]
to go - erdd [ʔɚɻʷɚᶑɚᶑə]
human - poo [b͡ʙɵʕɵ]
yes - ar [ʔɚɻʷə]
no - er [ʔɚɻʷə]
genius - darp [ᶑɚɻʷəb͡ʙə]
mental retardation - derp [ᶑɚɻʷəb͡ʙə]

Verbs are conjugated with the following suffixes:
1st person: p
2nd person: b
3rd person: d

There is no concept of time, so, for example, "I came", "I'm coming" and "I will come" are all the same:
drrdrrp [ᶑɚɻʷɚɻʷɚᶑɚɻʷɚɻʷəb͡ʙə]

best conalng evar 10/10 no doubt!!!1!11 [:3]
DV82LECM
sinic
sinic
Posts: 280
Joined: 16 Dec 2016 03:31

Re: Random phonology/phonemic inventory thread

Post by DV82LECM »

Vlürch wrote:/ʔ/ <q>
/b͡ʙ/ <p>
/ɓ ᶑ/ <b d>
/ɻʷ/ <r>

/ə ɞ ɵ/ <a e o>

Consonant clusters are broken up by a schwa, which also occurs after word-final consonants. So, basically, there are no word-final consonants or consonant clusters. All vowels adjacent to one another are separated by a voiced pharyngeal fricative. Word-initial vowels are preceded by a glottal stop, so basically, there are no word-initial vowels either. The only allophony is that all vowels become [ɚ] before retroflex consonants.

hello - qpopeqbpqadq [ʔəb͡ʙɵb͡ʙɞʔəɓəb͡ʙəʔɚᶑəʔə]
bye - poqqdeeaq [b͡ʙɵʔəʔɚᶑɞʕɞʕəʔə]
what - qqa [ʔəʔə]
why - ppea [b͡ʙəb͡ʙɞʕə]
when - adqoodoe [ʔɚᶑəʔɵʕɚᶑɵʕɞ]
anus - poq [b͡ʙɵʔə]
mouth - dob [ᶑɵɓə]
life - parp [b͡ʙɚɻʷəb͡ʙə]
death - porp [b͡ʙɚɻʷəb͡ʙə]
love - barb [ɓɚɻʷəɓə]
hate - borb [ɓɚɻʷəɓə]
to come - drrdrr [ᶑɚɻʷɚɻʷɚᶑɚɻʷɚɻʷə]
to go - erdd [ʔɚɻʷɚᶑɚᶑə]
human - poo [b͡ʙɵʕɵ]
yes - ar [ʔɚɻʷə]
no - er [ʔɚɻʷə]
genius - darp [ᶑɚɻʷəb͡ʙə]
mental retardation - derp [ᶑɚɻʷəb͡ʙə]

Verbs are conjugated with the following suffixes:
1st person: p
2nd person: b
3rd person: d

There is no concept of time, so, for example, "I came", "I'm coming" and "I will come" are all the same:
drrdrrp [ᶑɚɻʷɚɻʷɚᶑɚɻʷɚɻʷəb͡ʙə]

best conalng evar 10/10 no doubt!!!1!11 [:3]
[O.O] Good God! QQA the fuck? PPEA do this?! PBORB...and PBARB.
𖥑𖧨𖣫𖦺𖣦𖢋𖤼𖥃𖣔𖣋𖢅𖡹𖡨𖡶𖡦𖡧𖡚𖠨
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Vlürch wrote:human - poo [b͡ʙɵʕɵ]

[xD] Is this indicative of a misanthropic outlook on life?
Locked