Re: [cybalist] Hamp and his dog, an IE shepherd

Piotr:

> Here's more food for thought, Glen:

Well, I was still feelin' hungry and went for a Big Mac. :)

> *pku- as the unstressed (zero-grade) composition form of *peku- is
> >DIRECTLY attested in Iranian,

Alright... in Avestan and Sogdian. That's it?

>The pattern is the same as in *doru (gen. *drous) 'wood, tree'

I've been starting to notice this weird genitive pattern too. It would be
explainable as a variant of the Old IE genitive *-ese as *-se
(*t:e'reu/*t:ere'u-se) without the thematic vowel in this case. Why the
thematic vowel isn't there in some cases is starting to intrigue me.

> Compounds with *peku- (e.g. Sanskrit pa��u-pa: 'shepherd') seem to
> >contain an analogically restored *e. In the 'dog' word the early loss >of
>*p apparently severed the synchronic semantic connection between >*peku and
>*(p)kuo:n, splitting the etymon in two and preventing the >restoration of
>full-grade vocalism.
>
> The word-initial simplification of *pk to *k is evident in Greek
> >kteis/ktenos 'comb' (< *pkten-; cf. Latin pecto, pecten).

The problem is that there's never any strongly attested cognate series with
*pk-. As usual, we can cite various words here and there in a piecemeal case
for *pk-. Above, Latin has pec- with inserted *e. It could just as well have
been that *pek- didn't reduce to zero-grade at all in IE.

- gLeN

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com