*kw- words (was: River names)

From: Glen Gordon
Message: 2287
Date: 2000-04-30

Sergei:
>- what about Old Indic ��v��nt?s 'flourishing'?
>- what about Old Indic ��v??��ate 'opens (itself)'< (?) *k'uenk-?
>- what about PIE *k'uen-:k'un- 'dog'?
>- is PIE *k'uen-to- rhyming with *suen-to- 'healthy' just casually?

First, how could these words be possibly related semantically?!

Second, I have objections to *kwen- "dog" being linked to *keu-. I think
it's clear that words with *kw- (as opposed to *kW-) point to an old initial
syllable that has been zeroed through accent from earlier **kVu- but this
certainly doesn't mean that every word with *kw- is derived from *keu-.

I would explain the term as derived from Old IE *kAwenV (A = [*a,*e]) to
explain the accent and palatalization. There are apparently similar forms in
Uralic according to Illich-Svitych (Lapp ga^idne, Mordvin/Udmurt kyjon,
Cheremis/Komi kejin "wolf") and some in AfroAsiatic according to Bomhard
("dog": West Chadic *kuHen > Mogogodo kwehen, Fyer kWeeng; East Chadic
*kanya > Dangla kanya, Jegu kany-; Omotic *keHen > Dime keenu; Berber *kun-
> Guanche cuna).

All of it would suggest that either the word has been borrowed in some very
remote stage of Pre-IE or it's an inherited term. Uralic would seem to point
to a form like *ku"ina". Bomhard wants to reconstruct AA *k@... but the
reconstruction is obviously biased towards IE since *kuhin is far more
appropriate to explain the laryngeals and labialization.

To relate Uralic *ku"ina" to Old IE *kAwenV, one would require a Steppe
reconstruction such as *kuini (> Old IE *ke'wene > *kewe'ne > *kewe'n- >
*kwe'n-). Thus Old IE *kAwenV should be better reconstructed as *kewene. The
intervocalic laryngeal would have been lost early on. A tentative Nostratic
reconstruction of *kuhini "canine" might be set up. Further, it may also be
a compound word *ku-hini (*ku "wild canine" > ? Sumerian ka "fox", Semitic
*kalb- "dog")... dunno. It's all a wonderful puzzle.

Piotr:
>If a case is to be made for *k'wei- and *k'wen- being related, they >must
>be analysed as *k'w-ei- and *k'w-en-respectively, i.e. as >extended *k'eu-.

I agree with this analysis.

- gLeN
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com