Mate:
>I don't think we will ever agree on this but let's go on.
No, we probably won't, but it's more fun this way :) If a few of us
primates didn't think differently and eventually walk upright, we'd
never have fastfood chains. (Supposedly it's "evolution", they say.)
>My point is this - because of the "satem" lgs we have to reconstruct
>the system of *k - *kw - *k' for the last stage of *some* of the IE
>lgs. And I see it better to reconstruct the like system for all lgs than
>different ones for kentum and satem lgs.
Um, we ARE reconstructing a common system: *[k q kW]. I hope you
realize that while Satem was a localized group of dialects that shared
a common innovation, "Centum" does not exist as a single, small dialect
area. Therefore, it is valid to reconstruct a common Satem sound
system that consisted of *[k^ k kW].
Unless you see Satem as a NON-localized group and are willing to
defend that position, I don't understand where you're going with this
arguement.
>And as there *is* direct attestation of palatalisation and there
>is no direct attestation of uvularity the first is more important in my
>view.
You're evidently saying that the presence of sibilants for earlier
velar stops is "direct attestation" of earlier palatal *k^, something
that I don't even deny since I at least accept Proto-Satem *k^.
Well, then similarily, the overwhelming scarcity of *e neighbouring
*q is "direct attestation" of an even earlier uvularity in Proto-IE itself.
>By the way, according to Melchert, Luwian shows a difference of
>*k' and *k in some words (in front of *u at least). The example is
>zuwi "dog". I just heard about, so I don't know the details but I will look
>it up.
In "some" words? Quantify "some". You are aware of possible
Indo-Iranian influence, yes?
= gLeN
_________________________________________________________________
Help STOP SPAM with the new MSN 8 and get 2 months FREE*
http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarket%3den-ca