Re: Missing the sky

> Date:         Fri, 8 Mar 2002 15:50:32 -0500
> From: Andreas Johansson <and_yo@...>
>
> Joe Hill wrote:
> > [Andreas too:]
> > > Why doesn't anybody seem to believe in the "schwa
> > > indogermanicus" any longer? Even tho' initial [pxt-] would
> > > probably delight a Georgian, doesn't the fact that a vowel turns
> > > up in all daughter languages sort of suggest that there were one
> > > there right from the start?
> >Becuase of Hittite.
> >Hittite has an /x/ sound where all other IE languages have a vowel.
> >As hittite is the earliest attested IE language, and Lanyngeals had
> >been put forth before, it makes sense, no?
> Well, I guess it does - otherwise the IEists would reconstruct the
> words in question like they do. So, does Hettite have something like
> /pxter/ for "father"?
It's a bit more complex than that.

From various ablaut series, it's fairly sure that PIE had _something_,
or rather three different somethings, that developed like this between
consonants:

        PIE     Sanskrit        Greek
        Xe      a               e/a/o
        Xo      a               o
        eX      a:              e:/a:/o:
        oX      a:              o:
        X       i               e/a/o

that doesn't say anything about what sort of sounds the X's were. When
de Saussure suggested them (in 1878), he just called them coefficients
sonantiques.

Some of the evidence for the X's being h-like sounds is that PIE TXe
becomes Tha in Sanskrit, TTa in Hittite (T = p/t/k), and that some
Xe-/Xwe- become ha-/hwe- in Hittite. (The latter was what finally made
the theory acceptable, once it was realized how old Hittite was.
Before that, PIE was thought to have had unvoiced aspirates that only
survived in Sanskrit).

For instance, PIE root *h2wes- "to sojourn, remain": Hittite huis-zi,
Skt v