Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Yemls Phonology (long)

From:Christophe Grandsire <christophe.grandsire@...>
Date:Wednesday, June 6, 2001, 12:45
En réponse à Jeff Jones <jeffsjones@...>:

[snip]

> > 1.3.1. Syllabary > > The transliteration for each (phonemic) syllable is: > > | /I/ /E/ /U/ /O/ > -----+--------------------- > null | e E o O > /b/ | p P b B > /f/ | v V f F > /m/ | w W m M > /l/ | l L r R > -----+--------------------- > /t/ | c C t T > /d/ | j J d D > /s/ | x X s S > /z/ | i I z Z > -----+--------------------- > /n/ | y Y n N > /k/ | k K q Q > /g/ | g G a A > /x/ | h H u U > > The phonemic value of each letter (the inverse of the above) is: > > A /gO/ a /gU/ N /nO/ n /nU/ > B /bO/ b /bU/ O /O/ o /U/ > C /tE/ c /tI/ P /bE/ p /bI/ > D /dO/ d /dU/ Q /kO/ q /kU/ > E /E/ e /I/ R /lO/ r /lU/ > F /fO/ f /fU/ S /sO/ s /sU/ > G /gE/ g /gI/ T /tO/ t /tU/ > H /xE/ h /xI/ U /xO/ u /xU/ > I /zE/ i /zI/ V /fE/ v /fI/ > J /dE/ j /dI/ W /mE/ w /mI/ > K /kE/ k /kI/ X /sE/ x /sI/ > L /lE/ l /lI/ Y /nE/ y /nI/ > M /mO/ m /mU/ Z /zO/ z /zU/ >
He he, there was a discussion not long ago about how to use the Latin alphabet as a syllabary. Seems that you found a way :) . [snip]
> > 1.4.2. Diphthongization > > Phonetically, the V (i.e. null consonant) syllables combine with the > vowel of the preceding syllable to produce diphthongs. In careful > speech, this is limited to syllables within the same word. The > following table gives the approximate phonetic values for these > combinations. The columns are for the V syllables and the rows give the > vowel phoneme of the preceding syllable. > > | {e} {E} {O} {o} > -----+------------------------- > /I/ | [i:] [j&:] [jA:] [ju:] > /E/ | [EI] [&:] [&O] [EU] > /O/ | [OI] [AE] [A:] [OU] > /U/ | [wi:] [w&:] [wA:] [u:] > > 1.4.3. Consonant Mapping > > The following table gives the phonetic values for the consonant > phonemes. If an entry has 2 forms, the first is the usual value, and > the second is an assimilatory value. Consonants in the third group have > variants conditioned by the following vowel, while those in the first > group don't vary significantly between front and back versions. Of > particular note are the consonants in the second group, which are > "historically" conditioned; the values are now distinct enough so that > the vowel itself need distinguish low vs. high. > > | front back > -----+--------------------- > /b/ | [b], [p] [b], [p] > /f/ | [f], [v] [f], [v] > /m/ | [m] [m] > /l/ | [l] [l] > -----+--------------------- > /t/ | [tSH] [tH] > /d/ | [dZ] [d] > /s/ | [S] [s] > /z/ | [Z] [z] > -----+--------------------- > /n/ | [N] [n] > /k/ | [cH] [kH] > /g/ | [y] [g], [q] > /x/ | [C] [K], [G] >
I like it. My conlang Tj'a-ts'a~n has a phoneme /s/ which can have up to four different pronunciations [s], [z], [S] or [Z] depending on whether the next vowel is front unrounded or back rounded, and whether the preceeding consonnant (if any) is voiceless or voiced/voiced nasalised. But I didn't dare go as far as you :) . [snip]
> > 1.4.5. Consonant Assimilation > > Elimination of the vowel may cause consonant assimilation. > > {s} [sb] ==> [sp] > {x} [Sb] ==> [Sp] > {f} [fb] ==> [fp] > [fd] ==> [vd] > [fdZ] ==> [vdZ] > [fg] ==> [vg] > [fy] ==> [vy] > {u} [Kb] ==> [Kp] > [Kd] ==> [Gd] > [KdZ] ==> [GdZ] > > 1.4.6. Default Stress > > Word-stems of 1 syllable are stressed only if suffixed or lengthened. > Word-stems of 2 syllables are stressed on the penult. Word-stems of 3 > or more syllables may be stressed on the antepenult if the vowel of the > penult can be dropped, otherwise the penult is stressed. >
So if I take all those rules, the name of the language Yemls should really be phonemically /nEImU'lIsU/, pronounced [NEImU'lIs]. Or did you capitalize the first letter only because of English rules, and it should be yemls: /nIImU'lIsU/, pronounced [Ni:mU'lIs]? Did I understand correctly? If so, I find it nice that a word people would naturally tend to pronounce /jEml=z/ has in fact such a different pronunciation :) . Christophe. http://rainbow.conlang.free.fr

Reply

Muke Tever <alrivera@...>