Hittite walwa 'lion' WAS: Bears and why they are mostly...

[Alexander S. Nikolaev]

Miguel Carrasquer Vidal wrote:

>JoatSimeon at aol.com wrote:

>>>anaikio at mail.student.oulu.fi writes:

>>>Aryan / Iranian (e.g. Avestan vr.ka- 'wolf'). Any parallels?>>

>>-- there's the fact that Anatolian (Luvian) walwa/i means "lion" rather than
>>"wolf", as in all the other IE languages.

>I don't see how that word can be related.  G&I give walwa-,
>walwi- with question mark as Hittite, besides Luwian walwa-, but
>that could be a Luwian borrowing into Hittite.  Still, I'm not
>aware of a development *kw > w in Luwian (*k^ > zero, yes).

>The Anatolian word is much more likely to be related (as wa-lwa-
>< *lwa-l(e)wa- ?) to the general Eastern Mediterranean word for
>"lion", Egyptian <rw> (< *lw), Kartvelian *lom-, Semitic *labu?at
>("lioness").  As to Greek-Latin/Germanic-Slavic *lew-, G&I argue
>with some justification against the Germanic > Slavic word being
>< Latin, although I probably wouldn't go as far as establishing a
>PIE *lew- "lion" (well maybe, if Toch. <lu> "wild animal" is
>another reflex).

	I agree; there is no reason to connect the
	hittite word to PIE *wl.kwos (regrettably, one can't appeal to
	any authorities: the hittite word discussed is missing in
	both Puhvel and Tischler dictionaries, Neu mentions it in
	his Anitta-edition, but without any IE etymology).
	But there is to my mind little which can prevent us from
	reconstructing a PIE lexeme for 'lion' -- we must reckon with the fact,
	that in prehistorical times the species Panthera Leo was
	spread vastly, the ancestors of today's lions were dwelling
	even in Europe (Darlington P. Zoogeography: the geographical
	distribution of animals. NY, 1957.)

	As to the shape of this putative PIE lexeme, hittite walwa
	posits a lot of difficulties. I know of a hypothesis,
	which i find attractive, but the text can hardly be accessible
	to any of the list members  (it appeared in the
	"Jazgulamskij sbornik", St-Petersburg, 1996 and belongs to A.
	Ryko). I am taking the liberty to outline it briefly, as it
	is of interest. A "broken reduplication", suggested by G&I,
	is an extremely rare type, if exists at all.
	That is why the author suggests to reconstruct the word as
	a o-grade nominal formation from the root *welw/wlew, which
	om its part can be an w-enlarged root *wel- 'to tear' (some
	of the other PIE roots with the same shape *wel-, such as 'to
	see', 'to deceive', 'hair'  are compelling candidacies, too. -
	- why not trace wl.-kw-os to the same root, whatever it might
	be? And what is gr. alo:pe:ks then?)
	The initial *w- in this Schwebeablauting root can be proved
	with the help of the greek material, cf. e.g.
		Tro:es de {F}leiousin eoikotes o:mophagoisi
	"de", which stands in the beginning of the 2nd foot, should form a long
	syllable, and the length is caused by the dygamma, which
	closes the syllable.

	Any comments?

			Alexander Nikolaev