Re: sabyuka : consonants, orthography, and a few things more

Roger Mills


>Coming rather late to this discussion.........
You're welcome :)

Julien wrote:
 Political
>>> authority would be given to the best grammarians.
>My first thought was:  Chomsky as King of the World????
>od.:-((((
Definitively not ;) (I still have aftereffects from Generative Grammar
:( ) But Simon C. Dik would be a good candidate :)

>>> Notes :
>>>   - /ts/ is really problematic to me, because I want my system to be
>>> first aesthetic, /ts/ might be in every position, especially final.
I
>>> had thought of |ç| and |z|, but I finally decided for |tz| : please
>>> tell
>>> me if you have any better idea :)
>It struck me as very odd to introduce a letter |z| used only in this
>digraph and nowhere else. As a digraph |ts| is certainly preferable,
and >the proposals for |z| make good sense too.  (And it would eliminate
the >necessity for a probable | t's | across a syllable boundary,
assuming >that's a possibility in the language, as appears likely.)

So, |z| seems to be very popular :). I will certainly choose it.


>>> Here is the transliteration of the vowel system :
>>> /i/ --> |i|
>>> /e/ --> |e|
>>> /ei/ --> |ê|
>>> /@/ --> |à|
[etc., along with â for /a@/ and ô for /ou/, which I like]
>
>It is to me to me too, and you cannot imagine how hard it is for me to
>write it as |à| ;).
This is a minor, though counter-intuitive, problem, but as you explain--



>>Actually, I'm planning to build a website and
>>certainly a grammar book, where I would use the a with breve accent
>>("accent bref", as in latin if it is not the good word), and long
accent
>>for diphtongs.
>
>Exactly!!  The only problem is that the breve and macron are probably
>Unicode, which may not show up (in your website) on everyone's
browsers-- >though I guess a pdf would show them; they are of course
available in most >word-processing programs.

Yes, I had heard about such problems :( Are there so many browsers now
that don't like Unicode? It seems to me that Unicode intends to be a
standard to translitterate every language, isn't it?


>I find your vowel system very interesting.  (1) only the low vowels can
be >long/diphthongized. (2) it seems ve
>ry logical that each low vowel can diphthongize only with its high
partner
>(3) from an historical POV, there are interesting possibilities as to
how
>they arose:
>
>a) from original long vowels (somewhat like the Great Vowel Shift of
>Engl.?)
>b) as regularizations of sequences of low V plus _any_ high vowel (this
>could be very complicated...)
>c) ê and ô clearly could result from earlier *[ai] and *[au] by simple
>assimilation; the origin of â however would be a little murky in that
case,
>though perhaps *[a@] was also possible (the diphthong rule:   /a/ plus
any
>of the high vowels /i @ u/??)
>d) or perhaps they could result from the loss of a consonant (or
mysterious
"laryngeal") in *...VCC..., as we see in Port. _feito_, Fr. _fait_ <
*factu-

Thank you. I think each case could be productive, especially d), where
/a?/ could become /a@/

>>But to make it readable for everyone here, I thought that
>>circumflexed letters would be good, and also that |à| could be a
>>not-so-bad approximation of schwa (as |@| would be even worse to me).
>
>I agree.
So I try to make a good compromise, and I thought of :
|i e ê u o ô ä a â|
where circumflexed letters mark diphtongs and |ä| is /@/. Would it be ok
if I used that, could everyone here read it?


>Other questions:  Please post something about the syllable
>structure/phontactics.
I promise I'll do soon ;). It's not fixed enough in my head, but the
general pattern should be
          S
         / \
        /   \
       O     R
       |    / \
       |   N   C
       |  / \  |
       x x   x x
       | |   | |
       C V   V*C*

"*" means that they are mutually exclusive. I might allow some
word-initial clusters. Branching nucleus means diphtong.
S = syllable
O = onset
R = rhyme
N = nucleus
C = coda


>What triggers gemination of consonants? (I like geminates.) (These two
>questions are motivated by my desire to see if the use of the
apostrophe >can be eliminated)

What triggers gemination is  derivation and composition : for instance,
'say' is "teq-", then 'you (singular) say' is :
//teq+ki// --> teqqi
//otl+ru// --> ollu, and orru in some dialects (not mine as I cannot
he + INSTR     pronounce double 'r' :( ).

INSTR = instrumental
Actually "otl" is "o+tl", "tl" is the fourth person marker (I was wrong
saying in another post that "tz" was the fourth person marker, it's
actually the third person marker), "o" is a prosthetic vowel which is an
ancient article.

Regards,

Julien