From: Piotr Gasiorowski
----- Original Message -----From: João Simões Lopes FilhoSent: Wednesday, September 20, 2000 11:39 PMSubject: Re: [tied] Re: ParvatiNot impossible, since Indic *saras and its Iranian counterpart can reflect *seros or *selos equally well. But the Greek word helos means 'marshy ground, marsh meadow' or (by semantic extension) 'backwater', so Sarasvati:, if derived from the stem *seles-, would mean 'marshy' rather than 'watery' -- which could well be the case, but why, then, was Saras-vant- the ruler of aerial waters or the pesonification of the sea? And here's the second problem: What's the real evidence for IE *selos? The Indo-Iranian cognate is ambiguous (it may well belong to the *ser- family of words), and *selos is not the only possible ancestral form of the Greek word. It could be *swelos, for example (the gain would be the tantalising possibility that it's related to some IE terms for 'willow'). On the other hand, *ser- is a very secure root, abundantly attested in Indo-Iranian (and elsewhere, also in numerous river-names). On the whole the "*seres-wnt-i:" etymology looks more credible to me.Piotr
The etymology for SARASVATI I know is *saras-vati" "watery", < IE
seles-wnti: cf. Gr. helos "marsh". This names is cognate of toponym
Arachosia < Iranian *Harahvaiti.