Re: [tied] -i:r-/-u:r- after labiovelars in Sanskrit?

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> On 2006-05-26 14:13, Mate Kapoviæ wrote:

> > The reflection -u:r- is otherwise found after a labial (cf. ú:rn.a:,
> > pu:r-, pu:rn.ás etc.).

> > fluctuation here? I mean, are there cases like -i:r- after normal

labials?

> I'm not sure if there are any; at any rate the preference for /u(:)r/
> after /p/, /bH/ and /v/ is overwhelming.

It's likely there's an intermediate stage after rH > r: > &:r and
r-H > &r-H and that & > u after labials. That is, no direct r. > ur, etc.

> This doesn't mean, however,
> that the occurrence of /u(:)r/ after velars is due to their lost labial
> component. /u/ is not only [+round] but also [+back]. Velarity itself
> may be responsible for retracting the epenthetic vowel, though the
> assimilatory effect is weaker in this case and what we find is /i(:)/ ~
> /u(:)/ fluctuation rather than anything approaching Neogrammarian
> regularity.

I think labiovelars had the same effect as bilabials in the central
area but there was slight dialect mixing among:

Baltic: round after plain (not after +pal/+round)

Indic: round after P and +round (not after +pal/plain)

Iranian: never round