>>Working back to Mid IE, *q always seems to be next to *a.
>
>No it doesn't.
A matter of opinion. I didn't say that this is the state
of affairs in IE itself.
>>The word *qreux- appears to be a deverbal noun, composed
>>of a stem *qreu- and *-x- (transitive).
>
>It's actually a root *qreux- and a suffix *-(V)s.
Yes it is unanalysable (afaik) in IE. However the stem
appears strongly to be a verbal derivative which is why
I analysed it that way. The ablaut that you're questioning
is for the early Late IE verb *qar-/*qor- that is the
basis of later *qreux-. No one expects a vowel between *q
and *r in *qreux- since the accent is only found on *-eu-
or following syllables. The expected ablaut is as it is.
>We see *q in a consonant cluster all the time. Just look under kl-, kr- or
>kw- in Pokorny, or in a Sanskrit
>dictionary.
Yes I know that, but these are all verbal derivatives.
Are there any indivisible roots that contain *q in an
initial consonant cluster? Are there any of these roots
that cannot be analysed as an earlier deverbal formation?
As I showed, *qreux- appears to be composed of an extended
verb *qr-eu-x- and we see the prolific use of these same
suffixes like *-x- and *-eu- in many other extended verbs.
Thus we arrive at a verb *qar-/*qor-.
In contrast, we can't divide *kwon- the same way, since
there is just nothing verbal about it.
= gLeN
_________________________________________________________________
Add photos to your e-mail with MSN 8. Get 2 months FREE*.
http://join.msn.com/?page=features/featuredemail