Mate:
>>>**ka -> *k
>>>**ki -> *k^
>>>**ku -> *kW
>>
>>Yes.
No, it's not that simple. We've already gone over (again and again)
that the "palatal" velars _must_ be plain due to markedness. Since
the purpose of theories is to explain things as best we can, it works
against us to ignore the implausible sound system that IE is normally
given. Theories should be optimalized as best we can, not left as
"unlikely".
Markedness forces us to reevaluate *k^ as plain *k, leaving
traditional *k to be viewed as "marked". Marked with what? Uvularity
is the best candidate given the phenomena already mentioned.
Modern linguistic theory DEMANDS this reevaluation. There is little
choice in the matter. Therefore the above palatal-based journey
into pre-IE is a complete waste of time.
On the other hand, I think we all instinctively agree that IE derives
from an earlier three-vowel system (perhaps four) that gave rise
to the velar varieties, particularly the labialized velars. In this uvular
interpretation, there is no palatalization at all and therefore both
*i and *a would yield the same result, plain *k, while labialization
would be caused by earlier *u, hence *kW.
So this begs the question where that leaves *q. My point is that the
differences between *k and *q arose AFTER the development of *k
versus *kW. It was the result of a-colouring. Piotr already mentioned
that uvularity can be seen as an "a-colouring". Well, simply put, *k
neighbouring *a was coloured to *q in an old stage of IE.
However, by early Late IE, after the event of syncope, the association
between *q (up to this point an allophone of *k) and *a was
automatically obliterated. This left *q to become more than just an
allophone. It acquired its own phonemic status.
= gLeN
_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.
http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarket%3den-ca