Miguel:
>There is no need for *-meni in Pre-Hittite.
I think its obvious agreement with Greek would warrant it, wouldn't you? The
fact that both *m and *w forms securely exist _together_ in IE itself with
_different_ functions (dual *w versus sg & pl *m in Sanskrit paradigm;
*wei/*me) halt dead any nonsense that they were once allophonic variants of
a *mw and that they belong to trivially interchangeable morphemes. One
should suspect that *-we- and *-me- had slightly different functions at some
point in time as well (*-we- = original 1pp stative ending?).
>>But Hittite has only -mi in the 1sg.
>
>While Luwian has only -wi. "Undoubtedly" Hitt. -mi and Luw. -wi >share a
>common origin.
No, not undoubtedly. I and many others have doubt. One might see these forms
with *m and forms with *w as ultimately deriving from a conjugation with two
different sets of endings (as was already the case in the IE singular
endings anyway), all stemming from Nostratic's suppletive
absolutive/ergative pronominal system.
Clearly *wei- cannot be construed as related phonetically to *me (nor can we
do the same with other so-called *w/*m "alternations" such as -mant/-vant in
Sanskrit) by way of any realistic linguistic processes available and yet
this is the same problem as we find in the pronominal endings. We simply
cannot logically reconcile these two very different forms, claiming that the
initials derive from one same phoneme because no linguistic phenomenon known
to exist can give credence to this silly theory, not even what Miguel
proposes in ad hoc fashion. A solution like a convenient *mw phoneme using
"blender" phonetics is a farce and needs to be scrapped for something much
better.
The *w/*m alternations are only here-and-there mostly centering around
pronominals... because that's where the source of the alternations lies,
silly! The rest is analogy.
>I forgot to mention the dual ending, which is likely to be a >secondary
>grammaticalization of the *-we- allomorph. It shows that >forms with *w
>also occurred outside of Anatolian.
Of course *w-forms must have existed in IE alongside the *m-forms, however
it's another thing to assert that *m and *w are derived from *mw, the latter
still being improbable based on your examples and the phonetics you
contrive.
>Indeed. In this case I prefer the retention scenario, with Luwian
>-(a)wi one step towards non-Anatolian PIE *-o: [and Ly{c|d}ian(*) >-u].
Problem is... Luwian isn't the ancestor of PostAnatolian IE, now is it. This
would require some Anatolian *-a-wi based on Luwian alone and further
reconstruction in IE as *-o-wi, I would suppose (again, based only on one
language called Luwian), none of it warranted or even very plausible.
What other examples of this change of intervocalic *-m- to *-w- exist? None.
Further, Hittite shows that -meni is used after -u-... This sound -u- is a
vowel, dear, ergo the -m- remains unchanged intervocalically against that
false rule :)
Absolutely nothing to hold on to here, Miguel. It would be easier to accept
that *m and *w have little to do with each other in previous stages of IE
phonetically and that the pronominal endings reflect an earlier grammar
where two different sets of verb endings existed side-by-side with two
different functions.
- gLeN
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at
http://www.hotmail.com