--- In
cybalist@yahoogroups.com, "fournet.arnaud" <fournet.arnaud@...>
wrote:
> We know from some troublesome verses in the Illiad,
> that these verses must have been composed in an archaic form of Greek
> where the problem did not exist, but was later created because Greek
> evolved.
> I remember a andraphontos which versifies properly as nr-phontos
> (nr-hantas) otherwise there's one step too many.
> If the Rg Veda does not pose that kind of problem,
> then it means Rg Veda datation and Sanscrit datation must be the same.
> I have no idea if there is any such versification problem in Rg Veda
or not.
Yes, there are. I may be wrong in details, as my knowledge is second
hand.
1. As Piotr has remarked several times, the first syllable of
_duhita:_ 'daughter' is long by position. The explanation is that in
open syllables a vocalic laryngeal had resolved itself into consonant
plus vowel, only for the consonant to subsequently vanish leaving just
/i/.
2. In verbs, an apparently short open vowel before the thematic vowel
is long in position. I am not sure about the relative frequency of
this phenomenon - always, usually or sometimes? This has been used as
evidence that the PIE subjunctive and thematic present morpheme was
*h1e rather than merely *e.
Richard.