Re: Brugmann's Law

----- Original Message -----

From: stlatos

To: cybalist@yahoogroups.com

Sent: Wednesday, January 16, 2008 12:48 AM

Subject: [Courrier indésirable] Re: [tied] Brugmann's Law

--- In cybalist@... s.com, Piotr Gasiorowski <gpiotr@...> wrote:
> Possible counterexamples are perfects of CVC roots such as 3sg. tatá:pa
> vs. 1sg. tatá:pa (< *tetópe, *tetoph2a). Admittedly, these may be
> analogical <cakára/caká:ra> etc., but the <napa:t-> case is harder to
> explain, and for me, personally, the crucial piece of evidence is the
> 'water' root *h2ap- (nom.sg. *h2o:ps, acc. *h2opm., gen. *h2ap(V)s,
> inst. *h2apeh1, nom.pl. *h2opes, loc. *h2apsu, etc.). There is no
> quantitative levelling in this word in IIr., so we can trust the
> lengthening in the nom.pl. á:pas (vs. its absence in the gen. apás,
ins.  apá:) as evidence of Brugmann's Law.
=========
This paradigm is wrong. The word had a strong stem with *xaxpY+ and
weak *xx,pY+.

The Sanskrit x, became a in the same syl. as another velar. In
Latin gen. pl. *xx,pYo:m > *x,pYo:m > *po:m > *pu:m in:

=======

From a macro-comparative point of view,

PIE *H2_p is cognate with Arabic s?ab "to flow" and Chinese shi1 "wet" < *syap  <* s?ap

Proto-eskimo is also interesting.

Your reconstruction does not sound acceptable

and your claim "this paradigm is wrong" needs much more explanation.

Arnaud

=====================