23-12-03 07:46, Daniel Baum wrote:
> Well, it's certainly not difficult to prove that the Rigveda is not in
> the same dialect as classical Sanskrit, or indeed, the later Vedic
> texts. However, there is no variant form of the number six beginning
> with a /k/ in the Rigveda, at least as far as I am aware.
Then it's worth keeping in mind that neither Rigvedic nor later Vedic
(not to mention Classical Sanskrit) can be completely identified with
the common ancestor of the Middle and New Indic languages (a fact too
easily forgotten even by linguists). There are some archaic features
lost in Vedic and Sanskrit but maintained in the Prakrits and in the
modern languages. One well-known example is the contrast between PIIr.
*ks^ and *g(H)z^, both giving Skt. ks., but the latter occasionally
preserving the voicing in other Indic dialects: Skt. ks.arati 'flows'
vs. Bengali jhara: 'ooze out' (cf. Av. Gz^ar-, but we're dealing with a
shared archaism, so it doesn't prove that the ancestor of Bengali was
influenced by Iranian). If the PIIr. initial cluster was *ks^(w)-,
there's nothing absurd about assuming Proto-Indic *ks.- in 'six',
simplified in Sanskrit but preserved in the Prakrits.
Piotr