*G^EN-

[petegray]

>> *keubH2-, 'lie down, lay down',
>> I see no reason not to derive it from Pokorny's 2. *kew-, 'bend' Isolating
>>the initial CVC-component of PIE stems allows comparison with initial
>>CVC-segments of PAA triliteral roots.

Yes, I understand why you wish to derive PIE *keubH from **kew.

It is always tedious to rehearse arguments.   My examples were of roots with
final -CH, of which there are plenty.   For theoretical reasons, you wish to
reduce all of these to CVC roots.  You may or may not be right, but the fact
that your reason for doing it is theoretical, and not based on PIE evidence,
is why I suggest (too dismissively, and I'm sorry for that), that the theory
is actually irrelevant to PIE data.  I also suspect that PIE data will not
affect your belief, so I didn't pursue the argument further.

Incidentally you have still not responded to my interpretation of PIE, where
I suggest that it is the -u- in *keubH which is the essential vowel, not
the -e-.  If you want connections with PAA, then the triliteral root *K-U-B
is surely a better guess than the biconsonantal *K-e-W.

Peter