Jens, regarding my acceptance of a "possibility":

>Oh, no? Fine, then we got that far.

You always _were_ that far. You just weren't paying attention.
I state that the possibility is so low as to be negligible, not that
it is impossible any more than it is impossible to get struck by
lightning. Evidence from other languages only confirms to a
small degree this possibility, but not the likelihood.

To prove likelihood, good arguements for **z must be made
in the context of IE itself. So far, this has not been accomplished
because you are far too obsessed with unnecessarily
demonstrating the possibility of your theory via every language
other than IE.

>Then let us forget that it used to provoke some very strong language on
>your part telling me I belonged in a
>loony-bin for even considering such a thing.

I didn't mean to say that you belonged in the loony-bin for
considering such a thing. You belong in the loony-bin for
persisting on such a thing. There's a difference :)

>That is exactly what I believe the facts force us to accept. If we
>do not accept an earlier distinction in the order of /s/ : /z/

Your evidence still doesn't necessarily show that we need **z.
At best, the thematic vowel shows that there may have been
automatic voicing of final *-s and thus an _allophonic_ phoneme
[z]. At first, it might seem that we need a "distinct" phoneme
**z given the opposition of 2ps thematic non-indicative *-e-s
and the thematic nominative in *-o-s. However, *-e-s is closely
tied with the indicative *-e-si with non-final [s]. Analogy then
cannot be ruled out.

The only thing left is the plural *-es, however this suffix doesn't
behave normally in terms of ablaut, nor is it normal when used
with thematic stems (ie *-o-es instead of **-o-s). So using the
plural as evidence of contrast between *s and **z is suspect, to
be sure.

This mirrors the conundrum of *-t versus *-d. We may compare
the ablative *-od (which you deny for some reason to the shock
of every IEist) versus 3ps thematic non-indicative *-e-t. However
this is also not a true opposition because of the potential analogy
with indicative *-e-ti with medial *t. Just like with endings in *-s
versus a supposed **-z, there is no ending in *-t that doesn't
closely alternate with a form where the same final phoneme
occurs medially.

This is an important dilemma that you must solve.

So, *-t and *-d are ultimately allophonic, just as the forms in
*-s and those you perceive to be **-z are also ultimately
allophonic. They may be united under an automatic rule of voicing
final non-laryngeal phonemes. Nothing suggests a distinct phoneme.


- gLeN

_________________________________________________________________
STOP MORE SPAM with the new MSN 8 and get 2 months FREE*
http://join.msn.com/?page=features/junkmail