--- Sean Whalen <
stlatos@...> wrote:
> a-tone>0/_Z(C)$[#/C/G]
This is what I shortened to "if Z will enter nucleus";
the m, for example, will not move to fill the nucleus
unless the previous vowel was in the same syllable.
So,
pa'-dam#
pa'-d_m#
pa'_dm#
but
ra`-gja-ma`
shows no change since the a is in a different
syllable, and not followed by #/C/G (the latter is
probably a direct cause of the former).
> Z>+syllabic in nucleus
>
> a-tone>0/[#/C/G]$(C)Z_
>
> Z>+syl
The first rule and movement to nucleus is directly
followed (or occurs at the same time) as it's equal
but opposite counterpart. This is the rule I referred
to that would let *kjlab- have the same outcome as
*kjalb-
> a-tone>0/V$Z_$CV
>
> nucleus>0 if unattached.
>
> Z>mora (but not nucleus)
>
> mora>0 if unattached
>
> resyllabification
If the first rule(s) don't affect a toneless *a this
set will still delete *a following a Z, but the
nucleus is deleted first, so the Z moves to take the
mora only.
So,
bha`-wa-xa
is not affected by the Z-syllabification rules, but
bha`-w_-xa
bha`-w -xa nucleus deleted
bha`-w-xa w moves to mora
since it's not in the nucleus it can't syllabify
bha`w-xa resyllabification
bha`u-xa G>V+high before syllable break or C (lasting
rule)
>
> repeat...
These rules repeat in whole only after the entire
series is gone through.
>
> C[N/L/G]X+sylC>C[Z+syl]XC (dialectal variations)?
>
> a-tone>0/#C_C if #CC allowed (s extrasyllabic)
>
> a-tone>0/VC_CV if first C less sonorant than second
> (obstruents probably count as same, some awkward CC
> probably not allowed or soon changed back)
>
I'm thinking about modifying these rules, but there
doesn't seem to be a need, yet.
__________________________________
Do you Yahoo!?
Take Yahoo! Mail with you! Get it on your mobile phone.
http://mobile.yahoo.com/maildemo