Re: [tied] Battle of the cow

On Thu, 1 Nov 2001 17:28:34 +0100, "Piotr Gasiorowski"
<gpiotr@...> wrote:

>Don't forget good ol' Gmc. *ku:-, usually explained by assuming *o:u > *u:, though nobody has ever proved that such a development is normal in Germanic. I think something like *gW(o)u-h2- (treated as a consonantal stem, and so rmaining outside the *a:-declension) may underlie it; cf. Greek dru:s 'oak < tree' (*dr(o)u-h2-). But in these cases the hypothetical *-h2 is external to the root.

Couldn't these be collective ("broken plural") formations, with or
without *h2, of the type (pardon my pre-PIE):

*wá:dan (> wódr) -> coll. *w(a)dá:n(h2) > *udo:r "water(s)"

So:

*dá:ru (> *dóru) -> coll. *d(a)rú:(h2) > *dru:- "tree(s)"
*gú:wu (> *gWówu) -> coll. *gW(u)wú:(h2) > *gWu:- "cow(s)"