Random phonology/phonemic inventory thread
- Thrice Xandvii
- runic
- Posts: 2698
- Joined: 25 Nov 2012 10:13
- Location: Carnassus
Re: Random phonology/phonemic inventory thread
Yes, /g/ and no /k/ is odd, and a matter I've discussed in other threads. I wonder if maybe adding another voiced consonant further back would make this situation weirder, or reinforce it as an odd quirk that may have had a more logical source?
![Image](http://www.frathwiki.com/images/5/57/Forum_Banner.png)
-
- greek
- Posts: 583
- Joined: 19 Mar 2016 04:14
- Location: Australia
Re: Random phonology/phonemic inventory thread
I've seen /g/ and no /k/.
-
- cuneiform
- Posts: 85
- Joined: 05 Jan 2021 04:37
Re: Random phonology/phonemic inventory thread
When I make conlangs I have a system of randomly generating likely phonologies. It goes like this:
- randomly select a number between 3 and 9; this is the number of places of articulation. Then choose these to your liking, prioritizing bilabial, velar, and dental/alveolar categories
- randomly select a number between the number of places of articulation (except 3; the lowest number here is four) and 9; this is the number of degrees of closure. You should include at least nasals, stops, some kind of fricative, and some other liquid
- randomly generate a number between 0 and 2; this is the number of different distinctions like voicing, prenasalization, ejectives, etc. Check real world languages to get some ideas for combinations if realism is a goal for you
- randomly generate a number between 2 and 5; this is the number of vowel heights. Prioritize the open and close positions
- randomly generate a number between 2 and 3; this is the number of degrees of frontness (if you generated 2 heights you can only have 2 degrees of frontness). If front vowels appear then back vowels probably will too, no front-central or central-back systems
- randomly generate a number between 0 and 2; this is the number of other distinctions like length, nasalization, diphthongs, etc.
- randomly generate a number between 1 and 2. This is the number of consonants allowed in the onset. Go through the consonants you’ve chosen and decide if you want them to both be obstruents, for example, or obstruent-resonant pairs, or something else entirely
- randomly generate a number between 0 and 2; this is the number of consonants allowed in the coda. Again decide what kind of shape these clusters will take should they generate.
And there you have it! The numbers come in a sequence like 000000-00. I randomly generated (using random.org) 472330-10 and came up with the following phonology:
m n
p b t d k g
ᵐb ⁿd ᵑɡ
ts dz
ⁿdz
s z
ɸ β x ɣ
j w
r
i u
ə
a
(C)V syllable structure
Vowels in sequence are pronounced separately
Vowels of the same quality do not appear adjacently
/w/ is inserted between like vowels
Consonant clusters do not occur
Word-initial stress
I find that resulting phonologies are familiar but unique because they often have combinations that don’t exist in reality. Hope you like my technique and maybe even give it a shot yourself!
- randomly select a number between 3 and 9; this is the number of places of articulation. Then choose these to your liking, prioritizing bilabial, velar, and dental/alveolar categories
- randomly select a number between the number of places of articulation (except 3; the lowest number here is four) and 9; this is the number of degrees of closure. You should include at least nasals, stops, some kind of fricative, and some other liquid
- randomly generate a number between 0 and 2; this is the number of different distinctions like voicing, prenasalization, ejectives, etc. Check real world languages to get some ideas for combinations if realism is a goal for you
- randomly generate a number between 2 and 5; this is the number of vowel heights. Prioritize the open and close positions
- randomly generate a number between 2 and 3; this is the number of degrees of frontness (if you generated 2 heights you can only have 2 degrees of frontness). If front vowels appear then back vowels probably will too, no front-central or central-back systems
- randomly generate a number between 0 and 2; this is the number of other distinctions like length, nasalization, diphthongs, etc.
- randomly generate a number between 1 and 2. This is the number of consonants allowed in the onset. Go through the consonants you’ve chosen and decide if you want them to both be obstruents, for example, or obstruent-resonant pairs, or something else entirely
- randomly generate a number between 0 and 2; this is the number of consonants allowed in the coda. Again decide what kind of shape these clusters will take should they generate.
And there you have it! The numbers come in a sequence like 000000-00. I randomly generated (using random.org) 472330-10 and came up with the following phonology:
m n
p b t d k g
ᵐb ⁿd ᵑɡ
ts dz
ⁿdz
s z
ɸ β x ɣ
j w
r
i u
ə
a
(C)V syllable structure
Vowels in sequence are pronounced separately
Vowels of the same quality do not appear adjacently
/w/ is inserted between like vowels
Consonant clusters do not occur
Word-initial stress
I find that resulting phonologies are familiar but unique because they often have combinations that don’t exist in reality. Hope you like my technique and maybe even give it a shot yourself!
- LinguoFranco
- greek
- Posts: 613
- Joined: 20 Jul 2016 17:49
- Location: U.S.
Re: Random phonology/phonemic inventory thread
/m n ŋ/
/p t t͡s~t͡ʃ k kʷ/
/pʰ tʰ kʰ kʷʰ/
/f s ʃ h/
/l ɾ j w/
/i u/
/ɛ ə ɔ/
/a/
The voiceless, unaspirated stops as well as /t͡s~t͡ʃ/ become voiced intervocally.
This language has CVC phonotactics and a registere tone system. The tones are rising, falling and neutral. This is my first real attempt at a tonal conlang.
I might expand this inventory later by adding voiced stops and fricatives.
/p t t͡s~t͡ʃ k kʷ/
/pʰ tʰ kʰ kʷʰ/
/f s ʃ h/
/l ɾ j w/
/i u/
/ɛ ə ɔ/
/a/
The voiceless, unaspirated stops as well as /t͡s~t͡ʃ/ become voiced intervocally.
This language has CVC phonotactics and a registere tone system. The tones are rising, falling and neutral. This is my first real attempt at a tonal conlang.
I might expand this inventory later by adding voiced stops and fricatives.
Re: Random phonology/phonemic inventory thread
p t k͡p k
ɓ ɗ
ɹ w
m n
s
i ĩ ɨ ɨ̃ u ũ
ə
a ã
Phonotactics
(C)V(C)
The coda is always a voiceles stop or /s/.
Stress on the penultimate syllable if it is closed.
Stress on the ultimate syllable if the penultimate syllable is open.
ɓ ɗ
ɹ w
m n
s
i ĩ ɨ ɨ̃ u ũ
ə
a ã
Phonotactics
(C)V(C)
The coda is always a voiceles stop or /s/.
Stress on the penultimate syllable if it is closed.
Stress on the ultimate syllable if the penultimate syllable is open.
My meta-thread: viewtopic.php?f=6&t=5760
Re: Random phonology/phonemic inventory thread
A proto-lang inspired by Dravidian:
/p m ʋ/
/t̪ n̪ l̪/
/ⁿt tˡ s r/
/ʈ ɳ ɭ/
/c ɲ ʎ j/
/k ŋ/
/ʔ h/
/i e u o a/
/i i: e: u: o: a:/
Only nasals, liquids and glottals can be codas.
/p m ʋ/
/t̪ n̪ l̪/
/ⁿt tˡ s r/
/ʈ ɳ ɭ/
/c ɲ ʎ j/
/k ŋ/
/ʔ h/
/i e u o a/
/i i: e: u: o: a:/
Only nasals, liquids and glottals can be codas.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
-JRR Tolkien
Re: Random phonology/phonemic inventory thread
I've added consonants to make it less like PDravidian:
/p m ʋ/
/t̪ n̪ l̪ ɹ̪/
/t͡s ⁿt tˡ s r/
/ʈ ɳ ɭ ɽ/
/c ɲ ʎ j/
/k ŋ/
/ʔ h/
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
-JRR Tolkien
Re: Random phonology/phonemic inventory thread
A protolang.
/p t tʲ k kʲ kʷ q/
/ɸ θ θʲ s sʲ ɬ ɬʲ x xʲ xʷ X h/
/m n nʲ/
/l lʲ r rʲ j w/
/ɪ ɪː ə əː ɑ ɑː u uː/
CV[m n nʲ l lʲ r rʲ j w]
/p t tʲ k kʲ kʷ q/
/ɸ θ θʲ s sʲ ɬ ɬʲ x xʲ xʷ X h/
/m n nʲ/
/l lʲ r rʲ j w/
/ɪ ɪː ə əː ɑ ɑː u uː/
CV[m n nʲ l lʲ r rʲ j w]
-
- sinic
- Posts: 401
- Joined: 21 Jul 2012 08:01
- Location: Buffalo, NY
Re: Random phonology/phonemic inventory thread
There are some languages that lack (pure) labials. There are some languages that lack nasals. There are some languages that lack fricatives. There is even at least one language that lacks both nasals and fricatives (Rotokas). But there is no language that lacks all three of these categories. Is this just a coincidence?
I came up with a couple inventories lacking labials, nasals, and fricatives.
#1
/t̪ ʈ c k kʷ q qʷ/
/d̪ ɖ ɟ g gʷ ɢ ɢʷ/
/l̪ ɻ j ɰ w ʁ̞ ʁ̞ʷ/
#2
/t c k kʷ/
/d ɟ g gʷ/
/ɾ j ɰ w/
I came up with a couple inventories lacking labials, nasals, and fricatives.
#1
/t̪ ʈ c k kʷ q qʷ/
/d̪ ɖ ɟ g gʷ ɢ ɢʷ/
/l̪ ɻ j ɰ w ʁ̞ ʁ̞ʷ/
#2
/t c k kʷ/
/d ɟ g gʷ/
/ɾ j ɰ w/
Re: Random phonology/phonemic inventory thread
This was a phonology I came up with a long time ago. I tried using it, but then I lost interest in the language.
m mʔ
b̥ b̥ʔ d̥ d̥ʔ d̥ʒ̊ d̥ʒ̊ʔ kʃ kʃʔ
- -- - -- -- tʃ tʃʔ k kʔ ʔ
------------ʒ ʒʔ ------ h hʔ
a e i ə~ɵ
vowels can be stressed.
(a somewhat more accurate description would be m b̥ d̥ d̥ʒ̊ tʃ k ʒ h ʔ (kʃ) with a (C)(C/ʔ)V(V)(C/ʔ) structure)
m mʔ
b̥ b̥ʔ d̥ d̥ʔ d̥ʒ̊ d̥ʒ̊ʔ kʃ kʃʔ
- -- - -- -- tʃ tʃʔ k kʔ ʔ
------------ʒ ʒʔ ------ h hʔ
a e i ə~ɵ
vowels can be stressed.
(a somewhat more accurate description would be m b̥ d̥ d̥ʒ̊ tʃ k ʒ h ʔ (kʃ) with a (C)(C/ʔ)V(V)(C/ʔ) structure)
Re: Random phonology/phonemic inventory thread
Not really sure why, but I've been thinking about this kind of phonologies a lot lately, and this is what it ended up being like. You could maybe describe it as something like "what would happen if you threw Mandarin, Japanese, Nuxalk, Inuktitut and maybe also a hint of Finnish into a blender?" but that's not how it was meant to be... well, kinda is, but tbh it started from "how far can the effects of aspirated consonants on following vowels be taken without going overboard?" in the context of a generally easily pronounceable phonology, that ended up being harder to pronounce than intended.
I think the end result is actually kinda "unique" and not too kitchen sink-y, but YMMV.
/m n ŋ/ <m n ng>
/p t k (ʔ)/ <b d g Ø~'>
/pʰ tʰ kʰ/ <p t k>
/s/ <s>
/j w h/ <y w h>
/l ɻ/ <l r>
/ɑ i u/ <a i u>
/ɴ̩/ <n>
/ʔ/ only occurs word-initially in vowel-initial words, and intervocalically to break up vowel clusters, including following /ɴ̩/
/n t tʰ k kʰ s w h l/ are [ɲ t͡ɕ t͡ɕʰ k̟ k̟ʰ ɕ ɥ ç ʎ] before /i/
/i u/ are [i̥ u̥] between voiceless consonants, including across word boundaries, and also utterance-finally
After aspirated consonants, /ɑ i u/ are [χɑ̥ː çi̥ː xu̥ː] before voiceless consonants and word-finally (even if the following word begins with a voiced consonant), and [χɑː çiː xuː] before voiced consonants word-internally
/i/ is [ʐ̍ː] after /ɻ/
/k kʰ/ are [q qʰ] before /ɑ/
/h/ is [χ x] before /ɑ u/ respectively
/i/ is [ɘ] in contact with [q qʰ χ], including its variants, but only word-internally
/ŋ/ is always geminate [ŋː] intervocalically, but treated as a single consonant; the combination /ɴ̩ŋ/ is realised as [ŋ̍kː] but is still treated as a voiced consonant
/ɴ̩/ assimilates homorganically to the following consonant. Word-finally, its quality depends on the preceding vowel, after /ɑ i u/ respectively [ɴ̩ ɲ̍ ŋ̍]
/ɴ̩k/ is written <n'g>
/p pʰ t tʰ k kʰ s l/ can be geminated intervocalically; the first half is considered to belong to the preceding syllable, and with plosives it's unreleased
Gibberish example:
Utanti anga sibbi munngu urinpu kasu pikka un'udi.
/utʰɑɴ̩tʰi ɑŋɑ sippi muɴ̩ŋu uɻiɴ̩pʰu kʰɑsu pʰikʰkʰɑ uɴ̩uti/
[ʔu̥t̪ʰχɑːɲ̍t͡ɕʰçi̥ː ʔɑŋːɑ ɕi̥p̚pʲi muŋ̍kːu ʔuɻʐ̍ːm̩pʰxu̥ː qʰχɑ̥ːs̪u̥ pʰçɘ̥ːq̚qʰχɑ̥ː ʔuŋ̍ʔu̥t͡ɕi̥]
...actually, is it just me or are all of my phonologies somehow similar to this one?![🥴](//cdn.jsdelivr.net/gh/twitter/twemoji@latest/assets/svg/1f974.svg)
![rolling eyes :roll:](./images/smilies/icon_rolleyes2.png)
/m n ŋ/ <m n ng>
/p t k (ʔ)/ <b d g Ø~'>
/pʰ tʰ kʰ/ <p t k>
/s/ <s>
/j w h/ <y w h>
/l ɻ/ <l r>
/ɑ i u/ <a i u>
/ɴ̩/ <n>
/ʔ/ only occurs word-initially in vowel-initial words, and intervocalically to break up vowel clusters, including following /ɴ̩/
/n t tʰ k kʰ s w h l/ are [ɲ t͡ɕ t͡ɕʰ k̟ k̟ʰ ɕ ɥ ç ʎ] before /i/
/i u/ are [i̥ u̥] between voiceless consonants, including across word boundaries, and also utterance-finally
After aspirated consonants, /ɑ i u/ are [χɑ̥ː çi̥ː xu̥ː] before voiceless consonants and word-finally (even if the following word begins with a voiced consonant), and [χɑː çiː xuː] before voiced consonants word-internally
/i/ is [ʐ̍ː] after /ɻ/
/k kʰ/ are [q qʰ] before /ɑ/
/h/ is [χ x] before /ɑ u/ respectively
/i/ is [ɘ] in contact with [q qʰ χ], including its variants, but only word-internally
/ŋ/ is always geminate [ŋː] intervocalically, but treated as a single consonant; the combination /ɴ̩ŋ/ is realised as [ŋ̍kː] but is still treated as a voiced consonant
/ɴ̩/ assimilates homorganically to the following consonant. Word-finally, its quality depends on the preceding vowel, after /ɑ i u/ respectively [ɴ̩ ɲ̍ ŋ̍]
/ɴ̩k/ is written <n'g>
/p pʰ t tʰ k kʰ s l/ can be geminated intervocalically; the first half is considered to belong to the preceding syllable, and with plosives it's unreleased
Gibberish example:
Utanti anga sibbi munngu urinpu kasu pikka un'udi.
/utʰɑɴ̩tʰi ɑŋɑ sippi muɴ̩ŋu uɻiɴ̩pʰu kʰɑsu pʰikʰkʰɑ uɴ̩uti/
[ʔu̥t̪ʰχɑːɲ̍t͡ɕʰçi̥ː ʔɑŋːɑ ɕi̥p̚pʲi muŋ̍kːu ʔuɻʐ̍ːm̩pʰxu̥ː qʰχɑ̥ːs̪u̥ pʰçɘ̥ːq̚qʰχɑ̥ː ʔuŋ̍ʔu̥t͡ɕi̥]
...actually, is it just me or are all of my phonologies somehow similar to this one?
Re: Random phonology/phonemic inventory thread
Where does [q] derive from?
My meta-thread: viewtopic.php?f=6&t=5760
Re: Random phonology/phonemic inventory thread
/k/ before /ɑ/, two lines up from the note about /i/ in contact with it. So, [ɘ] is an allophone of /i/ before the syllables /kɑ/, /kʰɑ/ and /hɑ/, but I worded it less specifically in case I start trying to develop that into a more fleshed-out conlang and there could be more contexts for it to pop up in.
Re: Random phonology/phonemic inventory thread
Sorry for the double post, but... I have to post this monstrosity.
/m n/ <m n>
/m̥ː mː n̥ː nː/ <mh mm nh nn>
/p t k/ <b d g>
/pː tː kː/ <pp tt cc>
/pʰː tʰː kʰː/ <p t c>
/ɓ͡βː/ <bb>
/q͡χː/ <cr>
/z/ <z>
/sː ʃː ʒː/ <s jj j>
/j w x/ <y w h>
/θː/ <ss>
/ɾ/ <r>
/rː ʀː/ <rr gr>
/b͡ʙː d͡rː/ <br dr>
/a ə i u/ <a e~o i u>
/aː əː iː uː/ <á é~ó í ú>
/aːː ɛːː iːː ɔːː uːː/ <ae ea ie au ou>
Historically, there were three length distinctions on all vowels and consonants. However, all kinds of weird stuff happened with the consonants adding quality distinctions and bringing the length distinctions back to two, creating pairs that are more like fortis/lenis distinctions and others that just make no sense like either /*bː/ or /*bːː/ becoming an implosive affricate /ɓ͡βː/ for some reason. Also, /*ɛ *ɔ *ɛː *ɔː/ merged into /ə əː/ but for some reason that distinction was retained in spelling even though there was an orthographic reform.
As for phonotactics, I guess it'd phonemically have complex ass syllables like (C)(C)(C)(C)V(C)(C)(C)(C) but phonetically they'd have epenthetic vowels that have zero consistency in quality or placement even in the speech of the same individual. Same thing with inconsistent voicing/devoicing. So, you'd get shit like zhbgrajssdrmhcrncaurgrybb /zxpʀːaʒːθːd͡rːm̥ː.q͡χːnkʰːɔːːɾʀːjɓ͡βː/ being actually pronounced at first [sxɨbʀːaʒːəθːt͡r̥ːœmːɢ͡ʁːɔŋkʰːɔːːɾæʀːəjʊɓ͡βː] and then [zæxpɑʀːaʒːðːyd͡rːɑm̥ːq͡χːnækʰːɔːːɾuʀːjæɓ͡βː] or whatever within ten seconds by the same person.
I'd think if this had to be set somewhere, it'd be some tiny river island somewhere in Western Europe and the speakers would live in dugouts (kinda like hobbits but less cosy) and it'd have lots of Germanic and Romance loanwords with wacky unpredictable sound changes to reflect the fact that they entered the language at dozens of different periods as all the weird stuff was happening with its phonemic inventory.
/m n/ <m n>
/m̥ː mː n̥ː nː/ <mh mm nh nn>
/p t k/ <b d g>
/pː tː kː/ <pp tt cc>
/pʰː tʰː kʰː/ <p t c>
/ɓ͡βː/ <bb>
/q͡χː/ <cr>
/z/ <z>
/sː ʃː ʒː/ <s jj j>
/j w x/ <y w h>
/θː/ <ss>
/ɾ/ <r>
/rː ʀː/ <rr gr>
/b͡ʙː d͡rː/ <br dr>
/a ə i u/ <a e~o i u>
/aː əː iː uː/ <á é~ó í ú>
/aːː ɛːː iːː ɔːː uːː/ <ae ea ie au ou>
Historically, there were three length distinctions on all vowels and consonants. However, all kinds of weird stuff happened with the consonants adding quality distinctions and bringing the length distinctions back to two, creating pairs that are more like fortis/lenis distinctions and others that just make no sense like either /*bː/ or /*bːː/ becoming an implosive affricate /ɓ͡βː/ for some reason. Also, /*ɛ *ɔ *ɛː *ɔː/ merged into /ə əː/ but for some reason that distinction was retained in spelling even though there was an orthographic reform.
As for phonotactics, I guess it'd phonemically have complex ass syllables like (C)(C)(C)(C)V(C)(C)(C)(C) but phonetically they'd have epenthetic vowels that have zero consistency in quality or placement even in the speech of the same individual. Same thing with inconsistent voicing/devoicing. So, you'd get shit like zhbgrajssdrmhcrncaurgrybb /zxpʀːaʒːθːd͡rːm̥ː.q͡χːnkʰːɔːːɾʀːjɓ͡βː/ being actually pronounced at first [sxɨbʀːaʒːəθːt͡r̥ːœmːɢ͡ʁːɔŋkʰːɔːːɾæʀːəjʊɓ͡βː] and then [zæxpɑʀːaʒːðːyd͡rːɑm̥ːq͡χːnækʰːɔːːɾuʀːjæɓ͡βː] or whatever within ten seconds by the same person.
I'd think if this had to be set somewhere, it'd be some tiny river island somewhere in Western Europe and the speakers would live in dugouts (kinda like hobbits but less cosy) and it'd have lots of Germanic and Romance loanwords with wacky unpredictable sound changes to reflect the fact that they entered the language at dozens of different periods as all the weird stuff was happening with its phonemic inventory.
- DesEsseintes
- mongolian
- Posts: 4331
- Joined: 31 Mar 2013 13:16
Re: Random phonology/phonemic inventory thread
The possible onsets of my latest conlang Fajanu:
ñ ng sh j y represent /ɲ ŋ ʃ d͡ʒ j/ respectively. The vowels are /a e i o u ɯ/ a e i o u eu.
Stuff:
1. Voiced obstruents b j g v z and the rhotic r normally do not occur word-initially. This is because they evolved from nC clusters (i.e. *mp *nt *nk *nf *ns *nsh → b r g v z j)
2. eu does not occur after Cw Cy onsets or adjacent to an internal w y but happily occurs after an initial w y; eu is also not allowed finally
Code: Select all
m my n ñ ng ngw
p py t sh k kw
f s sh
b by r j g gw
v z j
r y g w
Stuff:
1. Voiced obstruents b j g v z and the rhotic r normally do not occur word-initially. This is because they evolved from nC clusters (i.e. *mp *nt *nk *nf *ns *nsh → b r g v z j)
2. eu does not occur after Cw Cy onsets or adjacent to an internal w y but happily occurs after an initial w y; eu is also not allowed finally
Re: Random phonology/phonemic inventory thread
Onset consonants
p' t' k' q'
p t k q
m n ŋ-ɴ
l ʟ
s
j ʁ
Coda consonants
n ŋ-ɴ
l ʟ
j ʁ
(But coda ʁ works like Turkish ğ and usually just lengthens the vowel.)
p' t' k' q'
p t k q
m n ŋ-ɴ
l ʟ
s
j ʁ
Coda consonants
n ŋ-ɴ
l ʟ
j ʁ
(But coda ʁ works like Turkish ğ and usually just lengthens the vowel.)
My meta-thread: viewtopic.php?f=6&t=5760
Re: Random phonology/phonemic inventory thread
Arabic-inspired, and a revision of something I posted. Might actually use.
/m n ŋ/ <m n ñ> (ك ن م)
/p t k/ <p t c> (ق ت ب)
/p' t' q~ʔ/ <p' t' c'> (ڨ ث پ)
/b~v d~z g~ɣ/ <b d g> (ع ط ۏ)
/bˤ dˤ ɢ~ʕ(~ʁ)/ <b' d' g'> (غ ظ ۋ)
/f s ɬ ʃ x/ <f s r x h> (ح س ر ص ف)
/fʰ sʰ ɬʰ ʃʰ h/ <f' s' r' x' h'> (خ ش ز ض ڥ)
/w l j/ <w l y> (ي ل و)
/i~e u~o ʌ~a/ <i/e u/o a> (أَ أُ إِ)
Tri-consonant root system. V, CV, VC, CVC; vowel hiatus, no phonemic diphthongs. /b d g/ lenite intervocally to /v z ɣ/. The realization of /ɢ/, intervocally, is /ʁ/; in free variation with /ʕ/; the latter acts as a nostalgic dialectical favorite, but tends to the former. /m n ŋ p t k b d g f s ʃ x l/ may be coda. If this rule is maintained, then any cluster is acceptable. However, there are limitations: nasals cannot go after consonants, and approximants /w j/ cannot go before consonants. (Where one might expect /l/ to go with the other approximants, it is, in fact, /ɬ ɬʰ/ which take its place, only coming after consonants; /l/ is the only approximant allowed at coda.) Most manners of each place of articulation do not play well together, i.e. clusters like /pbˤ dt gx/. /ʃ ʃʰ/ are neutral, with one exception, /s sʰ/; /ɬ ɬʰ/ are the same with /l/. /ɬʰ/ is rare. These forbidden clusters are broken up with an epenthetic vowel, normally /a/. In regards to the above rule, nasals + consonant and consonant + approximant are all that are allowed, i.e. /mb pw nt ld/. Uvular realizations are only syllable initial. Ejectives, pharyngeals, and aspirates lower vowels that come before and after them, unless broken by another consonant in a cluster. <c' g' h'> cannot exist in clusters of like manner, i.e. clusters: /pq bɢ fh/. Finally, lone vowels are considered as having a consonant onset; that is why there is vowel hiatus. This came from a bygone /ɰ/.
I am looking for this to be a typological stand-alone.
/m n ŋ/ <m n ñ> (ك ن م)
/p t k/ <p t c> (ق ت ب)
/p' t' q~ʔ/ <p' t' c'> (ڨ ث پ)
/b~v d~z g~ɣ/ <b d g> (ع ط ۏ)
/bˤ dˤ ɢ~ʕ(~ʁ)/ <b' d' g'> (غ ظ ۋ)
/f s ɬ ʃ x/ <f s r x h> (ح س ر ص ف)
/fʰ sʰ ɬʰ ʃʰ h/ <f' s' r' x' h'> (خ ش ز ض ڥ)
/w l j/ <w l y> (ي ل و)
/i~e u~o ʌ~a/ <i/e u/o a> (أَ أُ إِ)
Tri-consonant root system. V, CV, VC, CVC; vowel hiatus, no phonemic diphthongs. /b d g/ lenite intervocally to /v z ɣ/. The realization of /ɢ/, intervocally, is /ʁ/; in free variation with /ʕ/; the latter acts as a nostalgic dialectical favorite, but tends to the former. /m n ŋ p t k b d g f s ʃ x l/ may be coda. If this rule is maintained, then any cluster is acceptable. However, there are limitations: nasals cannot go after consonants, and approximants /w j/ cannot go before consonants. (Where one might expect /l/ to go with the other approximants, it is, in fact, /ɬ ɬʰ/ which take its place, only coming after consonants; /l/ is the only approximant allowed at coda.) Most manners of each place of articulation do not play well together, i.e. clusters like /pbˤ dt gx/. /ʃ ʃʰ/ are neutral, with one exception, /s sʰ/; /ɬ ɬʰ/ are the same with /l/. /ɬʰ/ is rare. These forbidden clusters are broken up with an epenthetic vowel, normally /a/. In regards to the above rule, nasals + consonant and consonant + approximant are all that are allowed, i.e. /mb pw nt ld/. Uvular realizations are only syllable initial. Ejectives, pharyngeals, and aspirates lower vowels that come before and after them, unless broken by another consonant in a cluster. <c' g' h'> cannot exist in clusters of like manner, i.e. clusters: /pq bɢ fh/. Finally, lone vowels are considered as having a consonant onset; that is why there is vowel hiatus. This came from a bygone /ɰ/.
I am looking for this to be a typological stand-alone.
Last edited by DV82LECM on 27 Sep 2021 14:56, edited 91 times in total.
𖥑𖧨𖣫𖦺𖣦𖢋𖤼𖥃𖣔𖣋𖢅𖡹𖡨𖡶𖡦𖡧𖡚𖠨
Re: Random phonology/phonemic inventory thread
/n ŋ/ <n ng>
/t k ʔ/ <t k '>
/ɸ β s x h/ <f v s x h>
/l j ɾ/ <l y r>
The /ɸ β ɾ/ can be also pronounced [xʷ w ɣ].
Is it similar to any natural language(s), and do you have any idea for the vowel system?
/t k ʔ/ <t k '>
/ɸ β s x h/ <f v s x h>
/l j ɾ/ <l y r>
The /ɸ β ɾ/ can be also pronounced [xʷ w ɣ].
Is it similar to any natural language(s), and do you have any idea for the vowel system?
Re: Random phonology/phonemic inventory thread
"Hey what if we put together a phonology *after* doing the phonotactics?" - My brain, 2 hours before I was awake
Syllable structure: (C)V(N)
Certain sounds are changed to others when placed in the coda. If the sound they are changed to is a vowel then it is (sort of?) separated into its own syllable.
onset ~ coda
/p/ ~ /ɯ̹/
/t/ ~ /ə/
/k/ ~ /x/
/f/ ~ /ʋ/
/ɬ/ ~ /ɫ/
V > Ṽ when after /ǀ/
/ǀ/ onset only
/p t k b d g/ <p t k b d g>
/f v ɬ/ <f v l>
/n/ <n>
/ǀ/ <c>
/ɪ iː ʊ uː e a o/ <i ii u uu e a o>
Some gibberish examples:
teltapn [teɫ.ta.ɯ̹n]
afciinlek [aʋ.ǀĩːn.ɬex]
fatdepna [fa.ə.de.ɯ̹.na]
The vague idea in my head with the coda changes was like "what kinda sound do you sorta get if you go to articulate it but stop before you get there?" and idk it made some cool sounding words so I'm happy. I imagine it would be a lot more readable with a proper writing system - perhaps an abjad with variations on some letter forms for their different onset/coda sounds.
Anyone's welcome to do whatever with this. I can't see myself making any real use of it ^^
Syllable structure: (C)V(N)
Certain sounds are changed to others when placed in the coda. If the sound they are changed to is a vowel then it is (sort of?) separated into its own syllable.
onset ~ coda
/p/ ~ /ɯ̹/
/t/ ~ /ə/
/k/ ~ /x/
/f/ ~ /ʋ/
/ɬ/ ~ /ɫ/
V > Ṽ when after /ǀ/
/ǀ/ onset only
/p t k b d g/ <p t k b d g>
/f v ɬ/ <f v l>
/n/ <n>
/ǀ/ <c>
/ɪ iː ʊ uː e a o/ <i ii u uu e a o>
Some gibberish examples:
teltapn [teɫ.ta.ɯ̹n]
afciinlek [aʋ.ǀĩːn.ɬex]
fatdepna [fa.ə.de.ɯ̹.na]
The vague idea in my head with the coda changes was like "what kinda sound do you sorta get if you go to articulate it but stop before you get there?" and idk it made some cool sounding words so I'm happy. I imagine it would be a lot more readable with a proper writing system - perhaps an abjad with variations on some letter forms for their different onset/coda sounds.
Anyone's welcome to do whatever with this. I can't see myself making any real use of it ^^
- Frislander
- mayan
- Posts: 2088
- Joined: 14 May 2016 18:47
- Location: The North
Re: Random phonology/phonemic inventory thread
/(p) t t͡ʃ k (kːʷ)/ <(p) t c k (kw)>
/ɸ s x~h/ <f s h>
/(m) n/ <(m) n>
/j w/ <w y>
/x~h/ varies between velar and glottal depending on the dialect and speaker.
/i e a o u/
Syllable structure is CV(C), where any consonant is permitted in both onset and coda. Stress is generally trochaic with primary stress on the initial.
There are a number of assimilation processes when two consonants come into contact.
/kat/ "cat" + /-ja/ "animate plural" → [ˈkat͡ʃːa] "cats"
/xorun/ "fire" + /ɸoj/ "ring" → [ˈxorumˌpoj] "hearth"
/junt͡ʃis/ "moon" + /xuja/ "people, men" → [ˈjunt͡ʃisˌkuja] "moon people"
/mot/ "sleep" + /-wan/ "progressive" → [ˈmokːʷan] "is sleeping".
/ɸ s x~h/ <f s h>
/(m) n/ <(m) n>
/j w/ <w y>
/x~h/ varies between velar and glottal depending on the dialect and speaker.
/i e a o u/
Syllable structure is CV(C), where any consonant is permitted in both onset and coda. Stress is generally trochaic with primary stress on the initial.
There are a number of assimilation processes when two consonants come into contact.
- The nasal /n/ assimilates to the POA of a following consonant. With /ɸ/ the fricative also undergoes fortition to [p], and the resulting cluster is written <mp>. This is the only occurrence of [m] in the language.
- The fricatives undergo fortition to the corresponding stops when preceded by another fricative: /ɸ s x/ go to [p t͡ʃ k] respectively.
- The glides merge with and geminate a preceding stop: with /j/ the result is [t͡ʃː] while with /w/ the result is [kːʷ] <kw>
/kat/ "cat" + /-ja/ "animate plural" → [ˈkat͡ʃːa] "cats"
/xorun/ "fire" + /ɸoj/ "ring" → [ˈxorumˌpoj] "hearth"
/junt͡ʃis/ "moon" + /xuja/ "people, men" → [ˈjunt͡ʃisˌkuja] "moon people"
/mot/ "sleep" + /-wan/ "progressive" → [ˈmokːʷan] "is sleeping".