Random phonology/phonemic inventory thread

A forum for all topics related to constructed languages
Post Reply
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4165
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Sōkoan strikes again!

/m n ŋ/ m n ng
/pʰ tʰ t͡sʰ t͡ʂʰ t͡ɕʰ kʰ kβʰ/ p t c ċ ch k kw
/p͈ t͈ t͈s t͈ʂ d͡ʑː gː gβː/ pp tt cc ċċ jj gg ggw
/b d d͡z d͡ʐ d͡ʑ g gβ/ b d z ż j g gw
/ɸʰ sʰ ʂʰ ɕʰ xʰ~h/ f s ṡ sh h
/ɸ s ʂ ɕ/ ff ss ṡṡ ssh
/β~β̞ ɾ j ɣ~ɰ/ w r y ġ

/o ɑ a e/ o ŏ a e
/u ɯ ɨ i/ u ŭ ĭ i

Stops and affricates come in three series: aspirated (least marked), tense, and voiced. Tense stops are unvoiced and/or held longer than other stops. Fricatives are either aspirated or tense. (And I haven’t added the tense diacritic to the tense fricatives in the IPA, so just pretend it’s there.)

I’ve simplified the vowels by getting rid of /ɛ/.

Oh, and I love the romo. Length is a thing in vowels and long vowels are indicated by double letters, so this lovely monstrosity is a legit Sōkoan word:

jjiiririgo

My poor Sōkoan, love of my youth, what have I done to thee?
Edit: Edited to include the labialised velars kw ggw gw which are characterised by spread lips and no rounding. Also rearranged the stop series.

User avatar
Shemtov
runic
runic
Posts: 3001
Joined: 29 Apr 2013 04:06

Re: Random phonology/phonemic inventory thread

Post by Shemtov »

A protolang:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ʙ r ʀ/
/ɾ/
/n n̥/
/w y/

/i u e o a/
/i: u: e: o: a:/
/ĩ ũ ẽ õ̃ ã̃/

(C)V(C)
Geminates can only occur intervocalically.

Thoughts? I feel it's a bit unnaturalistic. What can I do to improve it, but leaving the same flavor?
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

DV82LECM
cuneiform
cuneiform
Posts: 135
Joined: 16 Dec 2016 03:31

Re: Random phonology/phonemic inventory thread

Post by DV82LECM »

Shemtov wrote:
23 Jun 2019 08:27
A protolang:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ʙ r ʀ/
/ɾ/
/n n̥/
/w y/

/i u e o a/
/i: u: e: o: a:/
/ĩ ũ ẽ õ̃ ã̃/

(C)V(C)
Geminates can only occur intervocalically.

Thoughts? I feel it's a bit unnaturalistic. What can I do to improve it, but leaving the same flavor?
Very interesting. I love the idea of the extended Iroquoian look. However...

1. Labio-velar and/or labio-uvular consonant(s).
2. Retroflex fricative.
3. Perhaps, fewer trills.
4. Voiceless liquids.
(5. Changing /y/ to /j/.)

User avatar
Shemtov
runic
runic
Posts: 3001
Joined: 29 Apr 2013 04:06

Re: Random phonology/phonemic inventory thread

Post by Shemtov »

DV82LECM wrote:
23 Jun 2019 17:03
Shemtov wrote:
23 Jun 2019 08:27
A protolang:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ʙ r ʀ/
/ɾ/
/n n̥/
/w y/

/i u e o a/
/i: u: e: o: a:/
/ĩ ũ ẽ õ̃ ã̃/

(C)V(C)
Geminates can only occur intervocalically.

Thoughts? I feel it's a bit unnaturalistic. What can I do to improve it, but leaving the same flavor?
Very interesting. I love the idea of the extended Iroquoian look. However...

1. Labio-velar and/or labio-uvular consonant(s).
2. Retroflex fricative.
3. Perhaps, fewer trills.
4. Voiceless liquids.
(5. Changing /y/ to /j/.)
1. I'm not sure I need it. Do all languages without a labial stop have labio-velars?
2. I'm going for a one-retroflex language. Blust's reconstruction of Proto-Austronesian has only /ɖ/, and no other retroflexes.


Maybe:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ r r̥ ʀ ʀ̥/
/ɾ/
/m m̥ n n̥/
/w j/

/i u e o a/
/i: u: e: o: a:/
(I didn't even want the nasal vowels, I just wanted a third vowel series, I was thinking overlong, but thought with the consonant inventory that would be to kitchen-sinky. Thoughts on overlongs?)
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

DV82LECM
cuneiform
cuneiform
Posts: 135
Joined: 16 Dec 2016 03:31

Re: Random phonology/phonemic inventory thread

Post by DV82LECM »

Shemtov wrote:
23 Jun 2019 21:27
DV82LECM wrote:
23 Jun 2019 17:03
Shemtov wrote:
23 Jun 2019 08:27
A protolang:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ʙ r ʀ/
/ɾ/
/n n̥/
/w y/

/i u e o a/
/i: u: e: o: a:/
/ĩ ũ ẽ õ̃ ã̃/

(C)V(C)
Geminates can only occur intervocalically.

Thoughts? I feel it's a bit unnaturalistic. What can I do to improve it, but leaving the same flavor?
Very interesting. I love the idea of the extended Iroquoian look. However...

1. Labio-velar and/or labio-uvular consonant(s).
2. Retroflex fricative.
3. Perhaps, fewer trills.
4. Voiceless liquids.
(5. Changing /y/ to /j/.)
1. I'm not sure I need it. Do all languages without a labial stop have labio-velars?
2. I'm going for a one-retroflex language. Blust's reconstruction of Proto-Austronesian has only /ɖ/, and no other retroflexes.


Maybe:
/t tʰ k kʰ q qʰ/
/t: k: q:/
/ʈ͡ʂ/
/s h/
/s:/
/ r r̥ ʀ ʀ̥/
/ɾ/
/m m̥ n n̥/
/w j/

/i u e o a/
/i: u: e: o: a:/
(I didn't even want the nasal vowels, I just wanted a third vowel series, I was thinking overlong, but thought with the consonant inventory that would be to kitchen-sinky. Thoughts on overlongs?)
Just gave what I thought. Glad to hear your own contextual input.

As for the the labio-velar or labio-uvular, I only thought of it to fill in for /p/.

Porphyrogenitos
sinic
sinic
Posts: 254
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Did I do this already? I was thinking about something like this again.

At a "deeper" level, or rather, at an earlier stage, the consonant inventory can simply be analyzed as:

/m n ŋ/
/p t k/

Syllable structure (C)V(C).

Nasals assimilate to the following consonant and voice a following stop.

Stops undergo complete assimilation to a following consonant.

Stops are lenited intervocalically.

Word-final neutralize as [ŋ] or simply as nasalization of the preceding vowel.

At a later stage, word-final stops in unstressed syllables merge as [ʔ] and drop.

This results in a word-initial inventory of:

/m n ŋ/
/p t k/

And a word-medial inventory of:

/m n ŋ/
/mː nː ŋː/
/mb nd ŋg/
/pː tː kː/
/β̞ ð̞ ɣ̞/

Sandhi effects followed by loss of word-final stops in unstressed syllables leads to a somewhat phonemicized system of mutations, where the word-medial consonants may appear word-initially.

Stress is word-initial, with every other syllable taking secondary, tertiary, etc stress.

Vowel inventory is /a e i o u/, with a length distinction in stressed syllables only.

I haven't exactly decided yet, but there is also palatalization of either /t/ or /k/, or perhaps both, before /i/ and maybe also /e/, leading to a palatal series:

/cç ɲɟ cːç j/

Perhaps /n/ and/or /ŋ/ will also be affected, resulting in palatal nasals. The extent of palatalization may vary by dialect.

Porphyrogenitos
sinic
sinic
Posts: 254
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

/tʼ ʈʼ cʼ kʼ kʼʷ qʼ/ <t' ṭ' c' k' k'w q'>
/t ʈ c k kʷ q ʔ/ <t ṭ c k kw q '>
/d ɖ ɟ g gʷ ɢ/ <d ḍ j g gw ġ>
/l ɻ j w/ <l r y w>
/lˀ ɻˀ jˀ wˀ/ <l' r' y' w'>

/i iː a aː u uː/ <i ii a aa u uu>

Prominent vowel allophones include /i u/ as [e o] adjacent to uvulars, and /a/ as [æ] adjacent to palatals.

The glottalized resonants are variously realized as [ʔR], [Rʔ], or with creaky voice. Resonants may be syllabic in unstressed syllables; syllabic /j w/ are, of course, simply [i u] and are spelled as such in the orthography.

/k/ or /kʼ/ + /w/ is simply realized as [kʷ] or [kʼʷ].

Syllable structure is (C)V(C), though a small number of word-initial clusters may exist.

An ejective becomes plain if preceded by an ejective in the previous syllable.

There are various ablaut processes tied to stress, which is phonemic.

One challenge in phonological analysis is the case of word-final /kʷ/ and /kw/:

In a root like /takʷ/ 'man', the final /kʷ/ is clearly realized as a single consonant with a rounded release and no following vowel.

The root /lakaw/, if given initial stress, is realized as /laku/. If /k/ + /w/ is realized as /kʷ/ word-medially, and /kʷ/ is a possible word-final consonant, why not here? This may be analyzed as a case of distinct phonemic representations for /kʷ/ and /kw/; one may also posit a schwa that precedes syllabic resonants so that /laku/ is actually /lakəw/. In any case, in writing, the matter is disambiguated, as the orthography sticks to surface values.

User avatar
Shemtov
runic
runic
Posts: 3001
Joined: 29 Apr 2013 04:06

Re: Random phonology/phonemic inventory thread

Post by Shemtov »

Canereu, a Creoloid of Spanish and Guanche, spoken in the People's Republic of the Canary Islands:
/m n ɲ/ <m n ñ>
/p b t d t͡ʃ d͡ʒ k g/ <p b t d ch dj c~qu g>
/ɸ β s z ʃ ʒ h/ < f v ss~s s x j h>
/r/ <r>
/l/ <l>
/ʋ/ <ü>

/i y u e ø o ɛ ɔ a/ <i iu u e eu o ai au a>
/oi/ <oi>
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

Porphyrogenitos
sinic
sinic
Posts: 254
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

A simpler version of the nasal-less, fricative-less, labial-less consonant inventory I posted:

/t' c' k'/ <t' c' k'> or <tx cx kx>
/t c k ʔ/ <t c k '> or <t c k x>
/d ɟ g/ <d j g>
/l j w/ <l y w>
/lˀ jˀ wˀ/ <l' y' w'> or <lx yx wx>

User avatar
Frislander
mayan
mayan
Posts: 2102
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

/p t s k h/
/pː tː ts kː/
/m n/
/w ɾ j/

Non-geminated obstruents are often voiced between voiced segments.

/i u/
/e o/
/a/

Syllable structure is (C)V(n), where in /n/ assimilates to the POA of a following obstruent. Vowels hiatus is permitted, but common only over affix boundaries. /wu ji/ are not found - where morphology would create them, they are simplified to /u i/ in hiatus.

User avatar
Omzinesý
runic
runic
Posts: 2797
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: Random phonology/phonemic inventory thread

Post by Omzinesý »

Phoneme inventory
p t c k
f θ ç x
m n
s ɕ
l r
j ɰ w

i ɨ u
e ə o
ä

Phonotactics
(C)V(glide[+/- nasal])
or
(C)l/r

So there aren't real coda consonants but r/l can appear as distinct syllables in the position.
Vowels can be followed by a glide that is either nasalized or not.

Stress
This is a pitch accent language. In longer words the stress (high pitch) falls on several syllables in line. The first one and the last one of them usually have the highest pitch.

Birdlang
greek
greek
Posts: 587
Joined: 25 Dec 2014 20:17
Location: Virginia

Re: Random phonology/phonemic inventory thread

Post by Birdlang »

I made an easy to type phonetic alphabet for one of the languages of my conworld’s equivalent of the South Pacific islands, the Dot Islands with diacritic equivalents. It’s Dot Islandic
/m n ɲ ŋ/ m n ñ ng/ŋ
/p b ɓ t d ɗ c ɟ ʄ k g ɠ ʔ/ p b bh/ƀ t d dh/đ ty/c dy/j dyh/ɉ k g gh/ǥ 7/ʾ
/f v θ ð s z ɕ ʑ x ɣ ħ ʕ h/ f v tt/ţ dd/ḑ s z c/š j/ž kk/x gg/γ 3/ȝ rr/ʿ h
/ʦ ʣ ʨ ʥ/ ts/ś dz/ź tc/č dj/ǰ
/r l j w/ r l y w
/ɬ/ ł

/i y ʉ ʊ e ø ə ɔ a/ i ÿ ü u e ö ė o a

Also I gave Modern Birdish a big upgrade.
/m n ɲ ŋ/ m n ñ ŋ
/p b t d k g ʔ/ p b t d k g ʼ
/ɸ β s z ʃ ʒ x ɣ/ f v s z x j ħ ɣ
/ʦ ʣ ʧ ʤ/ ç ȝ c ĵ
/r l ʎ j w/ r l l̃ y w

/i u eː oː ɛ ɔ a/ i u é ó e o a
The letter h is present but it’s silent, or used after a vowel to lengthen them. When it is pronounced, it is as a glottal stop.
There are long vowels, marked by doubling or an h after them. É and ó are always long. Long vowels are also marked with a circumflex above them. Breves are used to shorten a vowel before an h, when that happens a glottal stop is used too. Breves aren’t used on é and ó since they are always long.
The letters x, j, and c are only found after i. If the i is before a vowel it becomes silent.
The letter q is used in loanwords or brand names, usually replaced by k.
Ꭓꭓ Ʝʝ Ɬɬ Ɦɦ Ɡɡ Ɥɥ Ɫɫ Ɽɽ Ɑɑ Ɱɱ Ɐɐ Ɒɒ Ɓɓ Ɔɔ Ɖɖ Ɗɗ Əə Ɛɛ Ɠɠ Ɣɣ Ɯɯ Ɲɲ Ɵɵ Ʀʀ Ʃʃ Ʈʈ Ʊʊ Ʋʋ Ʒʒ Ꞵꞵ Ʉʉ Ʌʌ Ŋŋ Ɂɂ Ɪɪ Ææ Øø Ð𠌜 Ɜɜ Ǝɘ

User avatar
Thrice Xandvii
runic
runic
Posts: 2696
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

How weird of a consonant inventory is this? I am woefully out of practice and wanted to get something going again:

/m n ŋ~g/
/b t tɕ k ʔ/
/f ɕ h/
/l ɾ~d/

Edit: Now with added vowels:
/i ɛ u/
/aː oː/
/ə̃/
Last edited by Thrice Xandvii on 02 Aug 2019 11:29, edited 2 times in total.
Image

User avatar
Zekoslav
sinic
sinic
Posts: 340
Joined: 07 Oct 2017 16:54

Re: Random phonology/phonemic inventory thread

Post by Zekoslav »

It's, in my opinion, a wonderfully structured asymmetric inventory: all the gaps are in the right spot, typologically speaking. It's wierd but in a good way [:D]!

1 - labial, 2 - labiodental, 3 - dental, 4 - alveopalatal, 5 - velar, 6 - glottal
A - voiceless stop, B - voiced stop, C - fricative, D - nasal, E - liquid

Code: Select all

-   1   2   3   4   5   6
A           t   tɕ  k   ʔ
B   b       d
C       f       ɕ       h
D   m       n       ŋ
E           l
1. Fricatives are offset from stops by one place of articulation as is usual: labial but labiodental, dental but palatal, velar but glottal. That the offset fricative at the dental place of articulation is palatal instead of dental is a nice touch.

2. There's a nice pattern in /t tɕ ɕ/ and /k ʔ h/: /tɕ/ and /ʔ/ patterns as alveopalatal and glottal stops respectively, linking the gap between the dental and velar stops and alveopalatal and glottal fricatives respectively.

3. There's no /p/ and /g/, which is the most usual gap in the stop system.

(Of course, this only works if one analyzes ŋ~g as primarily /g/ and ɾ~d as primarily /d/)
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who occasionally frequents the CBB.

- Guide to Slavic accentuation

User avatar
Thrice Xandvii
runic
runic
Posts: 2696
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

Thanks for the prompt and useful feedback. It had seemed to me to be plausibly weird. Good to see I wasn't too crazy and still remember a thing or two.

Now, methinks the vowels are weirder... maybe too weird.
Image

User avatar
Zekoslav
sinic
sinic
Posts: 340
Joined: 07 Oct 2017 16:54

Re: Random phonology/phonemic inventory thread

Post by Zekoslav »

Yea, that looks like a language that has just undergone some major restructuring of long and short vowels and is about to loose vowel length completely.

One could still make sense of it by operating with distinctive features. For example, the vowel system of Classical Manchu , horrendously unbalanced from a phonetic point of view, can be made sense of (is this proper English?) by a stepwise division of vowels according to distinctive features, taking into account the behavior of vowels in vowel harmony. I'll give you the full paper and a summary:

[+low] vs. nonlow: /a/, /ə/, /ɔ/ vs. /i/, /u/, /ʊ/

[+coronal] vs. noncoronal: /i/ vs. /u/, /ʊ/

[+labial] vs. nonlabial: /ɔ/ vs. /a/, /ə/

[+ATR] vs. non-attracted tongue root: /ə/ vs. /a/ and /u/ vs. /ʊ/.


If one does that with your language, you can have:

[+nasal] vs. nonnasal: /ə̃/ vs. everything else

[+labial] vs. nonlabial: /u/, /oː/ vs. /i/, /ɛ/, /aː/

[+long] vs. short: /oː/ vs. /u/ and /aː/ vs. /i/, /ɛ/

[+high] vs. nonhigh or [+low] vs. nonlow: /i/ vs. /ɛ/


But do read that paper and others from the same author (especially the big work about Algonquian vowel systems), as well as those listed in the references, and make your own decisions! This phonological theory is an awesome way to invent vowel systems by a method different from trying to fill the vowel triangle (I still have to figure out how to do it in practice)!
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who occasionally frequents the CBB.

- Guide to Slavic accentuation

ɶʙ ɞʛ
sinic
sinic
Posts: 209
Joined: 02 Aug 2019 18:47

Re: Random phonology/phonemic inventory thread

Post by ɶʙ ɞʛ »

Initial:
/m n ŋ/
/p t ts k/
/b d dz g/
/f s z x h/
/w l r j/

/i e a o u/

pp tt kk > p’ t’ k’
pt pts pk > t’ ts’ k’
tp tts tk > ɓ ts’ k’ > w ts’ k’
ts > s _C
kp kt kts > ʘ ! ǀ̠ > ɓ ɗ ɗz > w l z
p’ > ɓ > w, but before consonants epenthetic ǝ is inserted.


mn mŋ > wn wŋ
nŋ nm > ɹŋ ɹm > ǝŋ ǝm
ŋm ŋn > jm wn

nC > ǝC if C is not coronal

p > pf in onset, h in coda

pf > f > h _C, also eliminating pre-existing f there

aǝ eǝ iǝ oǝ uǝ > ai e: i: o: u:

s z > ʃ ʒ next to i e

ih eh ah oh uh > ɪh ǝh ǝh ɔh ʊh

a ɔ > ɒ, ʊ > o

Short e i > ǝ ɪ, short u > ʊ > o

ɪ > ǝ, ǝ > a

Remaining f > h, but pf word initially

g > h in coda, x in onset

/m n ŋ/
/t’ ts’ tʃ’ k’/
/pf t ts tʃ k/
/b d dz dʒ/
/s z ʃ ʒ x h/
/w l r j/

/ǝ i: a e: ɒ o: o u: ai/

Tin > ti: > tʃi:
Ti > tʃǝ
p:na > p’ǝna > wǝna > wanɒ

ɶʙ ɞʛ
sinic
sinic
Posts: 209
Joined: 02 Aug 2019 18:47

Re: Random phonology/phonemic inventory thread

Post by ɶʙ ɞʛ »

Fortis stops: /p t ʈ~k/ [p'~p t'~t ʈ'~ʈ~k'~k]
Lenis stops: /b d ɖ g ɢ ʔ/ [pʰ~b tʰ~d ʈʰ~ɖ kʰ~ɣ qʰ~ʁ ʔ~0]
Fortis continuants: /f f: s s: ʂ ʂ: χ χ:/ [f~v pf~f s~z ts~s ʂ~ʐ tʂ~ʂ χ~ʁ qχ~χ]
Lenis continuants: /ɸ ɸ: θ θ: x x: h h:/ [h~w hʷ h~ɹ h~θ h~ɣ h~ʂ~x~kx h~0 h~ʔh]

/ɪ ɵ ɛ ə ɑ/
/i: ʉ: e: ɯ: a:/

User avatar
Linguifex
roman
roman
Posts: 924
Joined: 03 Aug 2012 08:07
Location: Ohio

Re: Random phonology/phonemic inventory thread

Post by Linguifex »

Something I'm working on for a triconsonantal language:

p t tʷ tʲ tʲʷ k kʷ ʔ ʔʷ
s sʷ ʃ ʃʷ
w l lʷ j ɥ

a aː ã ə əː ə̃

a ə → ɒ u / Cʷ_ (/p/ is treated as rounded for purposes of this rule)
a ə → æ i / else
S → N / _V[+ nasal]
Cʷ → C[- round] / _# (/p w/ are exempt; I'm not sure if I should have /p/ debuccalize to [h], here or elsewhere; thoughts?)

Cannot have three plosives in a root
Cannot have three rounded obstruents in a root
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.0
Conworld megathread

AVDIO · VIDEO · DISCO

User avatar
LinguoFranco
greek
greek
Posts: 478
Joined: 20 Jul 2016 17:49
Location: U.S.

Re: Random phonology/phonemic inventory thread

Post by LinguoFranco »

/m n ɲ ŋ/
/p t t͡s c k/
/ᵐb ⁿd ⁿd͡ʒ ᶮɟ ᵑɡ/
/s ʃ ɬ h/
/z ɮ/
/l ʎ j w/

/a e i o u/

I don't know how naturalistic this is, and I'm uncertain about the fricatives as I may add /f v/. The language is syllable-timed and has an unfixed stress. The maximum syllable allowed is CGVC.

Post Reply