Re: Phonotactics?

BP Jonsson wrote:

> At 08:22 9.12.2003, Roger Mills wrote:
>
> >yur-: yurun + nawus 'swim' > yundrawus 'swimming pool'
>
> Is that nr > ndr?  Each time you reveal something of
> Kash phonology there is a similarity to Sohlob
> (which I really need to get outta my head and onto
> the net!)  Don't tell me you got fricativization of
> geminates too!
>
(Where do Sohlob geminates come from? see below[1])

Sometimes the grapheme "ndr" (in Kash script the "nd" character with
subscript "r" sign) represents /n+r/, sometimes simply /nd+r/.  /nd/ in
turn, in modern Kash, results from the merger of **nt and **nd, and
sometimes /d/ in loanwords. (Basically I snitched this from Fijian, where
"d" = /nd/ and "dr" = /ndr/).

No geminates in Kash, except occasionally at morpheme boundaries, when the
suffixes -sa, -to, -ka, -po, -pi, and -nu occur after final s, t, k, p, n
resp..  Supposedly too, a post-tonic vl.stop is slightly lengthened
("emphatic")-- kota ['kot(:)@] word vs. kotani [ko'tani] 'his word ~he
said', but since I often forget to do it when reading Kash, perhaps it's a
dying rule.......

If you've been paying attention :-)), you already know that the 3d
pers.poss.sfx -ni behaves oddly after a final nasal-- any nasal +ni > -ñi:

karun 'lord, duke'
karun+mi > karumbi 'my lord'
karun+ti > karundi 'your lord'
karun+ni > karuñi 'his/her/their lord'

yomom 'foundation, basis' > yomoñi 'its/the....'

Spoken colloq. Kash is developing geminates in such cases:  karum:i (but
still karundi), karun:i, yomon:i
-------------------------
[1] If I ever create a Kash relative with word-internal geminates, they will
have to originate in some strange way, since the bulk of proto-forms are
CVCV(C). ---I fell in love with geminates during my work with the South
Sulawesi languages of Indonesia where they have at least 4 sources-- 1.
automatically after *schwa, 2. nasal+vl.stop > geminate vl.stop, 3. a few
old infixed *r or *l, 4. unexplained, sometimes in "intensified" affective
vocab.