Re: [tied] Re: Nominative Loss. A strengthened theory?

----- Original Message -----
From: "Miguel Carrasquer" <mcv@...>
To: <cybalist@yahoogroups.com>
Sent: Wednesday, April 21, 2004 4:07 PM
Subject: Re: [tied] Re: Nominative Loss. A strengthened theory?


> On Wed, 21 Apr 2004 15:22:19 +0200, Piotr Gasiorowski
> <piotr.gasiorowski@...> wrote:
>
> >21-04-2004 12:08, Sergejus Tarasovas wrote:
> >
> >> One would expect *-stu:. Or you don't accept this *pt > *st rule in
Slavic?
> >
> >I'd be loath not to derive Slavic *pero from *pterom. As we both know,
> >the evidence for the rare development of *pt > *st has been contested,
> >and I'm willing to accept the possibility that *pt- > *p- root-initially
>
> There's *p(&2)truh2-yo- > stryj "uncle" word-initially.
> Not that I wouldn't be loath not to derive Slavic *pero from
> *pterom either... In the case of pero/stryj an obvious
> solution is (1) pt- > t-; (2) p&2t- > pt-; (3) pt > st-, but
> I don't know if that holds for other examples.

Did you mix something up? Shouldn't that be *pt- > p- and *ph2t- > st-?
We could speculate that first *pt- > p- as in *pterom > pero, while we still
have *pHtruHyo- and after *pt- > p- we get *pHt- > *pt- and this new *pt-
yields st- now.

Mate