[tied] Re: Indo-European /a/

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:

> On Fri, 08 Apr 2005 04:18:07 +0000, etherman23
> <etherman23@...> wrote:
>
> >--- In cybalist@yahoogroups.com, "P&G" <G&P@...> wrote:
> >> I'm not sure which theory you mean here. Have you moved form

PIE /a/

> >> to laryngeal theory? Because, if you are doubting the existence

of

> >> laryngeals, we need a conversation.
> >
> >I'm convinced H2 existed (though I question some of its alleged
> >properties). I see nothing to convince me of H1. It doesn't appear

in

> >Hittite that I can see (except oddly, mehur).
>
> mehur is *me:h2wr. *h1 is never observed in Hittite (or
> Anatolian in general).

Yet in other languages we see e: in cognates. Why wouldn't e:H2
develope into a: in Hittite?

> > Nor am I convinced of H3,
> >but I'm more likely to believe in that than H1. The theory is
> >elegant,
> >I'll give you that. But where's the beef?
>
> The best evidence for *h1 is still the Ablaut e: ~ o: < *eh1
> ~ *oh1, as first noted by de Saussure.

The theory was invented to explain that. Since H1 isn't preserved in
Anatolian I wonder where the actual evidence is for it.

> Important support also comes from the Greek three-way
> representation of vocalized laryngeals, with *h1 > *&1 >
> /e/.

Could Greek have replaced the zero-grade with the full grade?

> Greek also shows a different development of *i/u + *h1/2/3,
> having:
>
> *ih1 > i: *uh1 > u:
> *ih2 > ya: *uh2 > wa:
> *ih3 > yo: *uh3 > wo:
>
> Plenty of evidence, I'd say.

What are the Hittite cognates to support this (not that I'm saying
they don't exist, I just like to know what they are)?