Re: PIE h2ei- h1ai- h4ei- 'to take/to give'

I think the answer is much simpler.

Anatolian retained inherited /?/ and /H/.

Non-Anatolian conflated /?/ and /H/ to /H/.


Patrick

----- Original Message -----
From: "Miguel Carrasquer Vidal" <miguelc@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, March 30, 2008 3:45 PM
Subject: Re: [tied] PIE h2ei- h1ai- h4ei- 'to take/to give'


> On Sun, 30 Mar 2008 10:18:49 -0000, "alexandru_mg3"
> <alexandru_mg3@...> wrote:
>
> >
> >I'm confused about different reconstruction of this PIE root
> >
> >Beekes h2ei- Grk. ainumai
> >(by the way the Greek verb is with long or with short u:?)
>
> Short (it's aínumai, with recessive accent, as all Greek
> verbs: if the /u/ had been long, it would have had to be
> ainûmai [note that -mai counts as _short_, it's *-maj not
> *-mai]).
>
> >LIV reconstructs h1ai- (Grk, Hittite pa:i/pianzi etc...)
> >
> >Adams for Tocharian reconstruct h4ei-
>
> *h2 is not lost in Hittite between vowels, so *h2ei- is
> ruled out (*pe-h2oi- would have given *pa(:)hi-, not pa:i-).
>
> It's mainly a matter of ideology. If you don't accept PIE
> "unmotivated" */a/, you have to accept */h4/ (or find some
> other way around either the lack of <h> in Hittite or the
> /a/ in Greek and Tocharian). If you do accept */a/, there's
> no objection against a root *h1ai- (two of them, in fact).
>
> =======================
> Miguel Carrasquer Vidal
> miguelc@...
>
>