Random phonology/phonemic inventory thread [2011–2018]

A forum for all topics related to constructed languages
Davush
greek
greek
Posts: 671
Joined: 10 Jan 2015 14:10

Re: Random phonology/phonemic inventory thread

Post by Davush »

Trying to come up with a phonology I like for a non-Qutrussan language spoke in Qutrus.

I am quite liking this so far:

/pʰ tʰ ʈʰ kʰ/ <p t ṭ k>
/p t ʈ k/ <b d ḍ g>
/m n ŋ/ <m n ṇ>
/f θ ð s z χ ʁ/ <f th ð s z x gh>
/ɬ ɮ/ <ś ź>
/ɽ j w/ <r y w>

/æ ɑ e o i u ɪ: ʊ:/

nghâtha /nʁɑ:θæ/
ḍîmś /ʈɪ:mɬ/
ŋoź /ŋoɮ/
kâṇzâx /kɑŋzɑχ/
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Frislander wrote: 23 Nov 2017 18:56 /p t t͡ʃ k/
/ʰp ʰt ʰt͡ʃ ʰk/
/s h/
/m n j/

/i u/
/e ø o/
/æ ɑ/

Syllable structure is CV, where the pre-aspirated series is restricted to intervocalic position.
Oh, I like this. It has a somewhat Uralic flavor. I'm trying to think of a source for the origin of the preaspirated series' distributional restriction - maybe they used to be geminates? Or maybe the syllable structure was once (C)V(h), final /h/ dropped, and all non-stop hC clusters were resolved in some way.
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Oh, speaking of preaspirated consonants, I just thought of a nice allophony rule I might use for them in the future:

After a back rounded non-low vowel, preaspiration is [ʍ]

After a low vowel, preaspiration is [χ]

After a front non-low vowel, preaspiration is [ç]

After a front rounded non-low vowel, preaspiration is [çʷ]
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Porphyrogenitos wrote: 24 Nov 2017 06:55 Oh, speaking of preaspirated consonants, I just thought of a nice allophony rule I might use for them in the future:

After a back rounded non-low vowel, preaspiration is [ʍ]

After a low vowel, preaspiration is [χ]

After a front non-low vowel, preaspiration is [ç]

After a front rounded non-low vowel, preaspiration is [çʷ]
Nice. Preaspiration is present in several of my Áánene languages. Allophony in Híí Proper descends from a very similar system to what you outline in Late Híí but then palatalisation and labialisation are lost. [:)]
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 »

Porphyrogenitos wrote: 23 Nov 2017 21:47Oh, I like this. It has a somewhat Uralic flavor. I'm trying to think of a source for the origin of the preaspirated series' distributional restriction - maybe they used to be geminates? Or maybe the syllable structure was once (C)V(h), final /h/ dropped, and all non-stop hC clusters were resolved in some way.
Well it comes pretty close to Fox, but that has /ʃk/ as well.
Porphyrogenitos wrote: 24 Nov 2017 06:55 Oh, speaking of preaspirated consonants, I just thought of a nice allophony rule I might use for them in the future...
You might want to take a look at Scottish Gaelic with this, there's some interesting variation of a different kind.
opipik
mayan
mayan
Posts: 1657
Joined: 12 Mar 2015 19:41

Re: Random phonology/phonemic inventory thread

Post by opipik »

A rework of my first conlang, made to be back-compatible.
/p b t d c ɟ k ɡ/ <p b t d çç dh k g>
/m n ɲ/ <m n ñ>
/f v s z ç~ʝ x~ɣ/ <f v s z ç h>
/w ɾ l j ɰ/ <w r l j ÿ>

/i y i: ɨ u: u/ <i y~ÿ î ï û u>
/e e: ei ø ə əi o o: oɲ/ <e é ë ø â ä o ó õC>
/ai au a a: aɲ ɨa/ <æ å a ã ñC ea>
User avatar
Shemtov
runic
runic
Posts: 3284
Joined: 29 Apr 2013 04:06

Re: Random phonology/phonemic inventory thread

Post by Shemtov »

(Proto-Wanian, just to remind people:
/p b t d c ɟ k g q ɢ /
/m n ɲ ŋ/\
/m̥ n̥ ɲ̥ ŋ̥/
/ɸ s ç x h/
/l ʎ ʟ/
/j w/

/i e o ə a/
/ai oi/)


Proto-Archipelagian, a branch of the Wanian family:
/p t k ʔ/
/m n ŋ/
/ɸ β ð s x ɣ h/
/ʀ/
/l ʟ/
/j w/

/i e ø o ə ɛ œ æ ɑ/
/ĩ ẽ ø̃ õ ə̃ ɛ̃ œ̃ æ̃ ɑ̃/

Unnamed Descendant:
/ t ʔ/
/m n ŋ/
/f s h/
/ʀ/
/l /
/j ʋ/

/i e ø o ə œ ɔ a /
/ ə̃ ɛ̃ œ̃ ɔ̃ /
This is one of the few Wanian languages to have vowel harmony- though it is limited to a front triad /ø œ œ̃ / versus a back triad/o ɔ ɔ̃ /. However, otherwise it is a typical Archipelagian language Morphosyntactically.
I probably won't do any work on this beyond this post until mid-December, but I have some ideas in mind.
Last edited by Shemtov on 26 Nov 2017 01:26, edited 1 time in total.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
User avatar
Shemtov
runic
runic
Posts: 3284
Joined: 29 Apr 2013 04:06

Re: Random phonology/phonemic inventory thread

Post by Shemtov »

DesEsseintes wrote: 24 Nov 2017 08:03
Porphyrogenitos wrote: 24 Nov 2017 06:55 Oh, speaking of preaspirated consonants, I just thought of a nice allophony rule I might use for them in the future:

After a back rounded non-low vowel, preaspiration is [ʍ]

After a low vowel, preaspiration is [χ]

After a front non-low vowel, preaspiration is [ç]

After a front rounded non-low vowel, preaspiration is [çʷ]
Nice. Preaspiration is present in several of my Áánene languages. Allophony in Híí Proper descends from a very similar system to what you outline in Late Híí but then palatalisation and labialisation are lost. [:)]
Pre-aspiration also figures predominantly in my Daa Sevǔihk (It's thread is currently on the first page of the Conlangs forum), though realization is not allophonic, but rather dialect-based- some dialects even realizing, say, /ʰk/ as [sk]. Though as /s/ is phonemic, this could be seen as loss of pre-aspiration, and the addition of /sC/ clusters, and some of the con-people consider Pattǔgian a separate language.
Last edited by Shemtov on 26 Nov 2017 06:09, edited 1 time in total.
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: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Oh, here's another thing I thought of related to preaspiration (or, well, gemination, which is a possible source of preaspiration) -

So if you had a voiced and an unvoiced series, say /b d g/ and /p t k/, and developed contrastive gemination, the following shifts could very easily happen:

/b d g bː dː gː/ > /β ð ɣ b d g/
/p t k pː tː kː/ > /p t k pʰ tʰ kʰ/ (or it could be preaspiration) > /p t k ɸ θ x/

And if you had a simple system of initial mutations (or even non-initial mutations, whatever) where original singletons became original geminates, then you'd have a funny pattern where voiced series exhibited the "opposite" behavior of the voiceless series:

Base / Mutated
β / b
ð / d
ɣ / g
p / ɸ
t / θ
k / x
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Porphyrogenitos wrote: 26 Nov 2017 06:22 Oh, here's another thing I thought of related to preaspiration (or, well, gemination, which is a possible source of preaspiration) -

So if you had a voiced and an unvoiced series, say /b d g/ and /p t k/, and developed contrastive gemination, the following shifts could very easily happen:

/b d g bː dː gː/ > /β ð ɣ b d g/
/p t k pː tː kː/ > /p t k pʰ tʰ kʰ/ (or it could be preaspiration) > /p t k ɸ θ x/

And if you had a simple system of initial mutations (or even non-initial mutations, whatever) where original singletons became original geminates, then you'd have a funny pattern where voiced series exhibited the "opposite" behavior of the voiceless series:

Base / Mutated
β / b
ð / d
ɣ / g
p / ɸ
t / θ
k / x
You could also do

/b d g bː dː gː/ > /β ð ɣ b d g/
/p t k pː tː kː/ > /pʰ tʰ kʰ p t k/

I feel geminates are more likely to resist aspiration but perhaps that’s just me. Of course, in Icelandic, it’s the geminates that develop preaspiration so if you go down that route your original arrangement makes more sense.
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

DesEsseintes wrote: 26 Nov 2017 07:40
You could also do

/b d g bː dː gː/ > /β ð ɣ b d g/
/p t k pː tː kː/ > /pʰ tʰ kʰ p t k/

I feel geminates are more likely to resist aspiration but perhaps that’s just me. Of course, in Icelandic, it’s the geminates that develop preaspiration so if you go down that route your original arrangement makes more sense.
I was actually inspired by a number of Oceanic languages where word-initial voiceless geminates (resulting from reduction of reduplication) have become aspirated, in some cases progressing all the way to voiceless fricatives, producing derivational patterns like initial /p t k/ > /f θ x/. But what you suggested could also happen, of course.
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 »

Back to the geminate madness again.

/t~ɾ ʔ/ <t ‘>
/pː tː t͡ʃ/ <p tt c>
/f s h/ <f s h>
/m n/ <m n>
/mː nː/ <mm nn>

/i u/ <í ú>
/ɪ ʊ/ <i u>
/e ə o/ <é a ó>
/ɛ ɔ/ <e o>
/a/ <á>

Syllable structure is CV(C), where any consonant may occur in the onset or coda, including geminates, however when two geminates come into contact the first one shortens, and in the cases of /pː t͡ʃ/ further spirantises to /f s/. Vowels are phonetically short when adjacent to a geminate and long otherwise.

Stress is dynamic and subject to multiple constraints. It is found on the first phonetically long vowel in a word it one is present, otherwise on the syllable with the first geminate onset. This can lead to alternations in stress location in inflected forms, e.g. nennetá [nɛnːɛˈɾaː] > nennetátt [nɛˈnːɛɾatː].

----

/p t ʔ/
/t͡s t͡ʃ t͡ʂ/
/s ʃ ʂ h/
/m n/
/w ɾ j/

The postalveolar sibilants are palatalised to some degree, though not fully palatoalveolar.

/i (iː uː)/
/eː ə/
/aː/

/aː/ is phonetically slightly shorter than the other long vowels, however it patterns with them with regards to processes such as stress.

The long close vowels are rare, and likely best analysed as /əj əw/ because of their morphophonemic behaviour.

Syllable structure is CV(F), where F is restricted to /ʔ h n j w/, where [j] only occurs phonetically after /eː aː/ and [w] only after /i eː aː/. /ə/ becomes /iː/ whenever it occurs before /j/ and /uː/ when before /w/, and the glides are then lost when not before a vowel.

Stress is determined by weight. There are three weights: light, heavy and super-heavy. A heavy syllable has either a long vowel or coda consonant, a super-heavy syllable both and a light syllable neither. Stress falls first on the first super-heavy syllable if present, if not then on the first heavy syllable and if not again then on the first syllable.

----
DesEsseintes wrote: 26 Nov 2017 07:40I feel geminates are more likely to resist aspiration but perhaps that’s just me. Of course, in Icelandic, it’s the geminates that develop preaspiration so if you go down that route your original arrangement makes more sense.
Well Cypriot Greek also aspirates its geminates, so I wouldn't want to say whether one or the other is more or less likely, and both are plausible to me.
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

This is once again not a full phonemic inventory, but just an idea: It would be rather fun/obnoxious to make a language with a phonemic inventory overall similar to English, and with an overall similar phonological structure, but only allow consonant clusters that are forbidden in English and sound weird or impossible to English speakers, e.g.

fm θm hm ml mw fml θml (etc) ...
tn tl ft ...
kn km fk fkm ...
hl hr ...
θs θf ʃθ ...
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Random phonology/phonemic inventory thread

Post by Creyeditor »

I though about a language where vowel features act like tones and tones act like vowel features. Some sub-ideas:

A rule of High Spread. If there is a high vowel in a word, all following vowels become high.
ex.
//i//+//kete// -> /ikiti/
PL+dog
'dogs'

//irik//+//o// -> /iriku/

Several Vocalic Downstep processes:
  • Non-Automatic: If two high vowels are adjacent, the second vowel is lowered to a near-high vowel and all following high vowels are lowered to near high vowels. This process is iterative, so if we get another pair of high vowels the second one is lowered to a mid-high vowel and so on.
  • Automatic: If a high vowel follows a low vowel the second vowel is lowered to a near-high vowel and all following high vowels are lowered to near high vowels. This process is iterative, so if we get another high vowel after a low vowel the second one is lowered to a mid-high vowel and so on.
  • Floating: If the High Spread rule overwrites a low vowel at the end of a word, the first vowel of a following word is lowered to near high if it is a high vowel. This process is iterative again.
Non-Automatic: /ikiti/-> [ikɪte]
Automatic: /ami/ -> [amɪ]
Floating: //idoga# #ini// -> /idugɨ# #ini/ -> [idʊgɘ# #e̞nɛ]
Maybe also quality-less vowels? I don't know.

Tones have a fixed inventory (maybe H(igh), M(id), L(ow)). They can also form contours (HL, LH) Processes usually only trigger categorical changes. They interact heavily with consonants.

Pre-low voicing: An obstruent becomes voiced before a low toned vowel.

//pāp+ì// -> /pābì/

Tonal Umlaut: A high tone becomes mid, if the following syllable includes a low tone or a voiced consonant.

//á+gā// -> /āgā/
//ám+ì// -> /āmì/

If two tones come together on one vowel, either due to morphology or because the tone bearing vowels become adjacent, they fuse either to a simple tone or to a contour tone.

Code: Select all

1/2 H  M  L
H   H  HL HL
M   LH M  M
L   LH L  L
//gā+ám//->/gǎm/
//á+ì//-> /âi/

So what do you think? Does it even look interesting?
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
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 t̪ tʲ ʈ k/
/ɬ̪ sʲ ʂ x/
/m n̪ ɲ ɳ ŋ/
/w l̪ ɻ j ɰ/

/i ɯ u/
/e ɤ/
/a/

There is a syllable-tone system with a two-way contrast between high and low tones.

Syllable structure is CV(C). Underlying /ji ɰɯ wu/ sequences are realised as onsetless [i ɯ u]. Coda consonants are restricted to plosives, nasals and fricatives. Furthermore a coronal coda consonant assimilates to the POA of a following coronal, and a coda nasal in general assimilates to the POA of a following consonant. Voiceless consonants are realised as voiced after a coda nasal, before voiced consonants and in some speakers between vowels as well.
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

Image

This is a possible overhaul of the Shitullian language I have back-burnered for a billion years. I had re-tooled the vowel system a while back to some weird diagonal vowel system that I now no longer understand why I liked when a vertical system would have been more logical. However, I now think that they both sucked and just want to scrap almost everything and start over. Also, I am thinking that the dorsals will only appear as coda consonants, possibly as alternations for their coronal variants? I am also thinking of doing something in which there is some consonant lengthening as an effect of syllable stress... it was something that was posted about by Dormouse some time ago and I don't recall it precisely enough to mention it at the moment (mental note: go look that up).

Anything in here that is truly outlandish or weird? Anything that doesn't make sense that needs some justification? For anyone who remembers the original language, do you miss the aspirated series?

(OH! And there will still be three "tones" to the vowels: long, creaky, and plain. These correspond to tones akin to the Mandarin third tone, low tone and high tone, respectively. Basically, the tonal system has somewhat fallen away and in it's place are some variants of phonation and length.)
Image
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Thrice Xandvii wrote: 29 Nov 2017 07:38 Shitulian
Looks nice and balanced, I’d say.

Are v ɸ labiodental as your IPA seems to indicate? Any chance of them being bilabial? bilabial bilabial bilabial!

The idea of having all the velar segment appear only in coda is interesting. Would not even /k g/ be allowed in onset?

EDIT: oh and I love the vowels. [:D]
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

DesEsseintes wrote: 29 Nov 2017 08:00 Looks nice and balanced, I’d say.
I thought so... glad to be validated on that piece.
Are v ɸ labiodental as your IPA seems to indicate? Any chance of them being bilabial? bilabial bilabial bilabial!
Honestly, bilabial would be somewhat more logical given the system, but I always find labiodental easier to pronounce. Nothing is to say they couldn't be a bilabial stop and fricative.
The idea of having all the velar segment appear only in coda is interesting. Would not even /k g/ be allowed in onset?
Yeah, I dunno. The idea just kinda struck me. I haven't the foggiest if it is attested, or even plausible. But no, /k g/ wouldn't be allowed in the onset, only coda. Again, there could be an exception for /k/ maybe. Feasibility on that idea isn't something I have a clue about. That was one of the reasons that I posted it.
EDIT: oh and I love the vowels. [:D]
Oh? Anything in particular? It's the rhotic/syllabic vowel, isn't it? (And the unrounded back vowel...)
Edit: It's official, I changed them to bilabial.
Image
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Thrice Xandvii wrote: 29 Nov 2017 08:17
“Des” wrote:Are v ɸ labiodental as your IPA seems to indicate? Any chance of them being bilabial? bilabial bilabial bilabial!
Honestly, bilabial would be somewhat more logical given the system, but I always find labiodental easier to pronounce. Nothing is to say they couldn't be a bilabial stop and fricative.
I’ll be pronouncing them bilabial then, if you don’t mind. :mrgreen:
The idea of having all the velar segment appear only in coda is interesting. Would not even /k g/ be allowed in onset?
Yeah, I dunno. The idea just kinda struck me. I haven't the foggiest if it is attested, or even plausible. But no, /k g/ wouldn't be allowed in the onset, only coda. Again, there could be an exception for /k/ maybe. Feasibility on that idea isn't something I have a clue about. That was one of the reasons that I posted it.
Having /x ŋ/ occur only in coda is well attested, cf. English and Blackfoot (/x/ only occurs in coda in the latter). Having /k g/ behave the same way would be justified by extension. Perhaps all non-coda /k g/ became palatal? It would be a cool feature, and would definitely give the language a raison d’être.
EDIT: oh and I love the vowels. [:D]
Oh? Anything in particular? It's the rhotic/syllabic vowel, isn't it? (And the unrounded back vowel...)
Yes, the unrounded* high back, and the contrast between /ɔ ə ə~/. Good stuff.
*autocorrect suggested ‘unfounded’, so obviously my phone is not as convinced of the validity of your system as I am. [xD]
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Thrice Xandvii wrote: 29 Nov 2017 08:17
Edit: It's official, I changed them to bilabial.
Awe. Some. [<3]
Locked