--- Piotr Gasiorowski <
gpiotr@...> wrote:
> Sean Whalen wrote:
>
> > But this is no evidence for dhzh, ghzh, etc.,
> ever
> > existing. As Piotr says in the last message in
> that
> > thread: in PIE Ch > C before voiceless obstr.
> > word-final (I agree, also syl-final) so the
> ordering
> > of the middle stages could go either way. If
> ghzh,
> > etc., had ever existed I'd expect *ghmas/jhmas as
> gen.
> > of "earth".
>
> I have little doubt that "something like" *gHz^H
> really existed. Note
> such correspondences as Skt. ks.arati 'flows' : Av.
> Gz^ar-
I know Av. has xs^ and Gz^ and that voiced and
voiceless merge in Sanskrit.
> There are
> Middle and New
> Indo-Aryan forms more conservative than the Sanskrit
> one! Cf. Prakr.
> jHara:i and Bengali jHara:, preserving both the
> voicing and the aspiration.
But s regularly causes a touching stop to aspirate
and then s drops out causing gemination. Even in
Sanskrit s sometimes causes aspiration.
Sanskrit to Pa:li
afterwards shoulder bathe lac bear
pas^c^a:t skandHa- sna:- la:ks.a- rks.a-
pas^c^a:t skandHa- sna:- la:ska- rks.a-
pas^c^a:t skandHa- sna:- la:ska- rt.s.a-
pas^c^a:t skandHa- sna:- la:ska- at.s.a-
pas^c^Ha:t skHanHa- snHa:- la:skHa- at.Hs.a-
pas^Hc^Ha:t sHkHanHa- sHnHa: la:sHkHa- at.Hs.Ha-
pac^Hc^Ha:t kHkHanHa- nHnHa: la:kHkHa- at.Hs.Ha-
pac^Hc^Ha:t kHanHa- nHa:- la:kHkHa- at.Hs.Ha-
pac^Hc^Ha:t kHanHa- nHa:- la:kHa- at.Hs.Ha-
pac^Hc^Ha:t kHanHa- nHa:- la:kHa- ac^Hc^Ha-
pac^Hc^Ha: kHandHa- nHa:- la:kha- ac^Hc^Ha-
metathesis in la:ska-
k>t. before s.
syllabic r to V
obstr. by s to aspirated
s aspirated by aspirated
s before C to C
geminates degeminate word-initially
geminates degeminate after long V
retroflex affricates to alveopalatal
The existence of retroflex affricates is preserved
in borrowings into Khowar.
If that Prakrit form comes from a dialect that
preserved gz. (like Av. but unlike Sanskrit) there's
no problem for my theory.
__________________________________
Yahoo! Mail - PC Magazine Editors' Choice 2005
http://mail.yahoo.com