Graha epithets (tAra and tArakA)

[Dominique.Thillaud]

At 19:37 +0100 7/12/97, Yaroslav V. Vassilkov wrote:
>        To begin with, kAmayA, acc. to MMW, is an indeclinable word which does
>not appear in any compound and, after all, as MMW remarks, it is
>"OMLY USED WITH brUhi AND pra-brUhi".

        Yes, that prooves kAmayA is an old syntagm, coming from a 'formula'
(see E. Benveniste, Problemes de linguistique generale 1, chap. 23, on the
delocutive forms). Hence, not necessarily submitted to the standard rules !

>        On the other hand, the compound meaning "the fight for the love of
>tArA" would look like *tArA-kAmayA-yuddha (not *tAra-kAmaya-yuddha). To
>make the false cut of the compound, an ancient Indian singer of tales had
>to make the double mistake, taking two long vowels for two short ones, which
>I doubt that he did. So the suggestion made by Dominique Thillaud, though
>witty and tempting, is, I am afraid, not valid.

        Don't be afraid ! You can't simply give the compound form of a name
without knowing what is his nude form in the mind of your ancient Indian
singer. Moreover, there is the possibility of using case forms in
compounds. If we believe to a delocution, tAra can be an old vocative (see
ved. amba!).
        Following blindly the rules is not the right way : If rAdhA /
rAdhikA, why tArakA and not *tArikA ? If the last A of tArA is not coming
from an H2 (A/i), the nude form tAra is not impossible.

        I'm not sure I'm right, but I'm not sure you're right ;-)
        Best regards,
Dominique

Dominique THILLAUD
Universite' de Nice Sophia-Antipolis, France