21-04-2004 21:36, Sergejus Tarasovas wrote:
> And as to *nep-t, Martin Huld writes in EIEC:
> "...correct segmentation revealed by the feminine forms is *nep-ot-
> in which -ot- is the same nominal suffix found in Germanic *me:no:þ-
> 'month' (from 'moon') or Hit <si:w-att-> 'day' (from 'daytime sky')".
Well, the feminine doesn't reveal anything about the internal
morphological divisions. The strong allomorph *népot- alternates with
weak *nept-, but *-(o)t- functions here a quasi-suffix at best, since
*nep- doesn't correspond to any known IE root. The analysis of *nepot-
as a compound (*ne-pot-, literally = '[having] no power', defended here
some time ago by Miguel and Jens if I remember aright), has its
problems, in my opinion, but cannot be ruled out completely. I prefer to
remain agnostic about the internal structure of *nepot- and its pre-PIE
etymology.
Piotr