Click wrote:I think I like this Híí better than the last version, /b t k g/ is especially nice!
DesEsseintes wrote:- /g b/ may be descended from Proto-Híí *q *qw as voicing and fronting of uvulars is well attested. This does mean I now have to find another source for glottal stops in Híí.
Would it be possible that word-initial glottal stops are epenthetic and the rest come from *d?
No, because there were no voiced stops in Proto-Híí. That's why I derived /g b/ from *q *qw. Fear not, I'll think of something, though other suggestions are of course welcome.
DesEsseintes wrote:- /g b/ may be descended from Proto-Híí *q *qw as voicing and fronting of uvulars is well attested. This does mean I now have to find another source for glottal stops in Híí.
Would it be possible that word-initial glottal stops are epenthetic and the rest come from *d?
No, because there were no voiced stops in Proto-Híí. That's why I derived /g b/ from *q *qw. Fear not, I'll think of something, though other suggestions are of course welcome.
What is the inventory for Proto-Híí out of interest? BecauseI was wondering if perhaps debuccalisation of *t with the gap being filled by another coronal obstruent or a palatal stop might be an option.
Click wrote:I think I like this Híí better than the last version, /b t k g/ is especially nice!
DesEsseintes wrote:- /g b/ may be descended from Proto-Híí *q *qw as voicing and fronting of uvulars is well attested. This does mean I now have to find another source for glottal stops in Híí.
Would it be possible that word-initial glottal stops are epenthetic and the rest come from *d?
No, because there were no voiced stops in Proto-Híí. That's why I derived /g b/ from *q *qw. Fear not, I'll think of something, though other suggestions are of course welcome.
*p > ʔ, maybe? Or from *ŋ?
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
DesEsseintes wrote:- /g b/ may be descended from Proto-Híí *q *qw as voicing and fronting of uvulars is well attested. This does mean I now have to find another source for glottal stops in Híí.
Would it be possible that word-initial glottal stops are epenthetic and the rest come from *d?
No, because there were no voiced stops in Proto-Híí. That's why I derived /g b/ from *q *qw. Fear not, I'll think of something, though other suggestions are of course welcome.
What is the inventory for Proto-Híí out of interest? BecauseI was wondering if perhaps debuccalisation of *t with the gap being filled by another coronal obstruent or a palatal stop might be an option.
My current ideas for Proto-Híí and its ancestor Proto-Plains can be found here.
Pretty much all the changes between Proto-Híí and Híí are accounted for. Originally, *p *t *k *q were meant to become k t Ø ʔ but now *q is gonna yield g.
Edit: Just saw Sangi's idea for the velar nasal becoming glottal. Hmm, that is an idea!
And I've got an idea for yours. In compounds, the vowel of a word is "reduced" to ư e ơ if the following word starts in an implosive. This applies regardless of intervening coda consonants. (And makes this rule cool!). So
tim ḅac → tưm ḅac
This might only apply if the two words are in the same NP or in a compound or sth. I don't know enough about your language to venture further.
Syllable structure is (C)V(V)(F), where VV is a combination of any two heterorganic vowels except schwa, which is deleted when it comes into contect with another vowel. /i/ is realised as a glide /j/ when adjacent to another vowel. Additionally final consonants are restricted to /n l/.
----
DesEsseintes wrote:Pretty much all the changes between Proto-Híí and Híí are accounted for. Originally, *p *t *k *q were meant to become k t Ø ʔ but now *q is gonna yield g.
Right, OK, that does complicate things. I would agree with the ŋ > ʔ idea, if nobody else has any better ideas.
Alternatively perhaps you could reduce coda stops to ʔ and the lose a following consonant and then epenthesise it word-initially? Or would that conflict with too many already established sound changes for that to work?
Nachtuil wrote:
Do you think you'll do something with this inventory and its associated rules at some point? It makes me want to try something very similar, though maybe without the /f/ phoneme and having some voiced and voiceless aproximate pairs.
Feel free to do whatever you please with it!
Frislander wrote:/p t k q/ <p t k q>
/m n/ <m n>
/l ʁ h/ <l r h>
/i e ə a o/ <i e v a o>
Syllable structure is (C)V(V)(F), where VV is a combination of any two heterorganic vowels except schwa, which is deleted when it comes into contect with another vowel. /i/ is realised as a glide /j/ when adjacent to another vowel. Additionally final consonants are restricted to /n l/.
I like this! It's rather Austronesian.
But about syllables with two adjacent vowels - since it's one syllable, one of the vowels would necessarily be non-syllabic (right?) - so wouldn't /i/ automatically be realized as a glide even without a rule specifying that? And what rules govern which of the vowels is non-syllabic when /i/ isn't involved? Is hiatus permitted?
Frislander wrote:/p t k q/ <p t k q>
/m n/ <m n>
/l ʁ h/ <l r h>
/i e ə a o/ <i e v a o>
Syllable structure is (C)V(V)(F), where VV is a combination of any two heterorganic vowels except schwa, which is deleted when it comes into contect with another vowel. /i/ is realised as a glide /j/ when adjacent to another vowel. Additionally final consonants are restricted to /n l/.
I like this! It's rather Austronesian.
But about syllables with two adjacent vowels - since it's one syllable, one of the vowels would necessarily be non-syllabic (right?) - so wouldn't /i/ automatically be realized as a glide even without a rule specifying that? And what rules govern which of the vowels is non-syllabic when /i/ isn't involved? Is hiatus permitted?
Actually with the inventory I was actually going for a reduced version of Abipón.
Yeah, that is a bit of a tricky one. Perhaps it would be better stated in terms of a a syllabicity hierarchy of something like a > e,o > i, so /e o/ also become non-syllabic when adjacent to /a/, and when /e/ and /o/ come together they form a hieght-harmonic diphthong as in Old English, or alternatively the first of the pair assimilates to the second, creating contrastive long mid vowels.
Hm, I think for the sake of simplicty not having hiatus is the best call.
So that gives the following altered phonology.
Spoiler:
/p t k q/ <p t k q>
/m n/ <m n>
/l ʁ h/ <l r h>
/i e ə a o/ <i e v a o>
/eː~o̯e oː~e̯o/ <ee~oe oo~eo>
/aɪ̯ eɪ̯ oɪ̯/ <ai ei oi>
/i̯a i̯e i̯o/ <ia ie io>
/ae̯ ao̯/ <ae ao>
/e̯a o̯a/ <ea oa>
Syllable structure is (C)V(V)(F), where VV is a diphthong or long vowel. Additionally final consonants are restricted to /n l/.
/m n ŋ/ <m n ng>
/p b t d k g q ʔ/ <p b t d k g q ‘>
/f v s z ʃ x/ <f v s z š x/h> (<š> can be <sh> in ASCII-only alt-orthography; <x> is <h> word-initially)
/j l r/ <y l r>
/a e i o u/ <a e i o u>
Allophony
/ʔ/ can be realised as either [ʔ] or [!]
/q/ can be realised as either [q] or [χ]
/r/ dialectal variations: [r], [ʁ], [ʀ]
/x/ can be realised as either [x] or [h]
/a/ can be realised as [ɑ]
Stress: initial / penultimate / final
Irregular stress marked with an acute accent (or <Vx> in ASCII-only alt-orthography)
When multiple words of irregular stress are joined together, the stress closest to the end of the word takes precedence
(e.g harú + -ím → haru’ím and not *harú’im)
Click here to join the Common Honey server. Or click here for a general glossopoeia server.
Vowels may take one of two pitchs, high or low, and may be either long or short.
Syllable structure is (C)V(C), where vowel-initial syllables only occur word-initially and only /k x ŋ/ may occur as coda consnants, with /ŋ/ assimilating to the POA of a following consonant.
Vowels may take one of two pitchs, high or low, and may be either long or short.
Syllable structure is (C)V(C), where vowel-initial syllables only occur word-initially and only /k x ŋ/ may occur as coda consnants, with /ŋ/ assimilating to the POA of a following consonant.
Ooh, I like this. The restriction on coda consonants is especially nice, and open to the analysis that coda consonants are unspecified or underspecified for PoA, which could open a whole Pandora's box of possibilities.
Vowels may take one of two pitchs, high or low, and may be either long or short.
Syllable structure is (C)V(C), where vowel-initial syllables only occur word-initially and only /k x ŋ/ may occur as coda consnants, with /ŋ/ assimilating to the POA of a following consonant.
Ooh, I like this. The restriction on coda consonants is especially nice, and open to the analysis that coda consonants are unspecified or underspecified for PoA, which could open a whole Pandora's box of possibilities.
Interesting! So in light of this, all coda consonants will assimilate to the POA of a following consonant.
The prenasalised stop is realised as a plain nasal before nasal vowels.
/i u/
/ẽ õ/
/a ɒ/
/aɪ̯ ɒʊ̯/
Vowels may occur in one of four tones: mid, high, low/breathy and creaky.
Sylable structure is CV.
Ooh! I like this, especially the stop series and tones. Can we see some random words to get a feel of it?
OK, how about [t͡ʃáɾā ʔi̤bɒ̰ʊ̯tẽ́ ⁿdānõ̰sṳjẽ́hāɪ̯ʔā]
(I just came up with that randomly because trying to get a word-generator to do it would be just too much work. I'd probably romanise it something like <ca'ra ‘ihbooute ndanõõsuhye'hai‘a> or something like that (I leave the question of how best to represent the tones to you).