Semivowels In Sanskrit

[Vidhyanath Rao]

"David L. White" <dlwhite at texas.net> wrote:

>         What is interesting, I think, is that /vy-/ (or /wy-/) occurred
> though /yv-/ (or /yw-) did not.  This suggest a sound annoyingly
> intermediate (to our theoretical notions) between /v/ and /w/.

Although people often seem to assume that Sanskrit v was labio-dental,
the evidence for that is mixed and late. Early evidence is discussed in
M. M. Deshpande, "The phonetics of v in Panini", Annals of Bhandarkar
Oriental Research Institute, 56(1975) 45--65. Rkpra:ti"sa:khya, for
example describes the both p-series and v as o.s.tya, "labial". Nor is
it usually described in terms
that would suggest a spirant.

Modern pronunciation varies: Post-consonantly, it is usually a bilabial
approximant. But, AFAIK, never a spirant (so some people don't like
transcribing an English v by Devanagari v, but use vh). Intervocalicaly,
it varies more, but in some mouths (including mine) it comes as bilabial
there too.