Miguel:
> I was actually referring to PIE *-kWe.
Yes, I was referring to the same thing. IE *kWe is "and" and its
Tyrrhenian equivalent is *ke. It becomes suffixed in both groups
independently because beforehand, it was a postpositional particle
in IndoTyrrhenian, a seperate word. IndoTyrrhenian was an SOV
language if this helps you understand which is why *-kWe came to
be a suffix so easily. So *kW being initial here, becomes *k in
Tyrrhenian without problems.
> You said something about a merger of *-e and *-a.
Ugh. Whoops.
Yes, but I just realized that's an old view that I had that
still lingered in my head. Must update my brain. So that's wrong.
Sorry. Please ignore. There was no merger because *-a and *-e
remain in MIE to explain 3ps *bere becoming eLIE *ber& after
Syncope while *káwana- became *kwan-. In other words *-e becomes
*-& while *-a normally becomes null during Syncope.
Rather, *kWo- is from the eLIE enclitic stem *kW&- while its
stressed variant was *kWi-, just as enclitic *t&- pairs with
stressed *ta- and so on. In MIE, it had to have been *kWi-sa 'who'
and kWi-ta 'what' with oblique *kWe. As I said, Syncope caused
the unstressed final *-e to become *& while stressed *e remained.
Hence unstressed *kW&- > *kWo-/*kWe-. Got it? Good. Again, ignore
the merger thing.
> I've never seen that. Reference?
Okay, maybe I goofed on that one. Can you find -c before that
period of syncope? In addition, is there evidence of syncope
even before this syncope?
= gLeN