Re: [cybalist] River names

From: Sergejus Tarasovas
Message: 2269
Date: 2000-04-29

 Piotr wrote:
 A good point. My suggestion was tentative and not really worth defending. The parallel formations are very convincing. Whether what they MEAN is 'famous, glorious'. Balto-Slavic 'glory' is a long-vowelled derivative of *k'leu- (*k'lo:w-a:), and the meaning is strongly suggestive of Indo-Iranian influence. The forms you quote look more directly deverbal, especially the transparently participial Словут- and Šlavant- < *k'lew-ont-. I prefer not to venture a detailed semantic interpretation without having a closer look at similar formations in other branches. 
Well, why the Slavic verb *slyti:slovON 'be famous etc.' doesn't fit?
I told you I can't assess the dialectal evidence. I accept they are most likely 'sacred'. (Of course *k'weit- without a nasal is well represented in Balto-Slavic hydronymy.) 
The situation in Lithuanian dialects is as follows:
Some dialects of Aukštaitija implement stressed standard in as yn (i.e. prolongate i), most of them also turn standard en into in, but never vise versa. So the only way to derive Šventoji from *Švint- is to allow a hypercorrection here.
Most dialects of Žemaitija implement standard ir,il as er,el (close i-flavoured e), but NEARLY never in,im as en,em. Still a number of cases (most of them derived from historical documents) are registered. But the 'great' Šventoji I mentioned flows wholly through Aukštaitija's territory (we also have 'small' Šventoji in Žemaitija).
 
  If a genetic relation between *k'went- and *k(')weit- is possible, it must be very deep, and inevitably somewhat speculative. Let me take them apart: 
Agree. I wrote 'think', (or rather 'believe'  :)  ) not 'can prove'.
As for *k(')weit- (I understand the brackets mean that you include Slavic *cvět-, which is OK as far as I'm concerned) 
Yes, that's what I meant.
In *k'went- the *n can't be an infix, since as the Iranian cognates demonstrate, the root is *k'wen- and the *t is suffixal: beside Avestan spənta- we have the comp. spən-yah- and the superl. spən-išta-, the abstract span-ah- (< *k'wenes-), etc., leading up to a verbal root (*k'wen-, cf. Latvian zero-grade svinu 'celebrate a holiday').
Yes, I fully agree and beg your pardon. Another *-t-suffixed form is Khotanese ysamaśśandai 'holy land'<*zam šu̥antaka. As for Old Indic, besides pan1- 'glorify, admire' a (related?) root pan2- 'sacred' may be considered:
(Vedic) arámatiH pánīyasī : (Avestian) Spənta- Ārmaiti- 'holy Armaiti';
(Vedic) soma pánipnatam, soma pányam-panyam 'soma the holy',
and may be derived from PIE *k'uen as well.
 
But some questions do exist. For instance:
- what about Old Indic śvāntás 'flourishing'?
- what about Old Indic śváñčate 'opens (itself)'< (?) *k'uenk-?
- what about PIE *k'uen-:k'un- 'dog'?
- is PIE *k'uen-to- rhyming with *suen-to- 'healthy' just casually?
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-. There are possible cognates for that, e.g. Germanic *xiw-i- '(fine) shape, beauty, colour' (English hue), Sanskrit śoNa 'bright-red', and possibly the Tocharian 'sun' word (A koM, B kauM, though W. Winter interprets them differently). Is THIS kind of relation what you meant? 
Yes, exactly.
Thank you for the explanation.
But also a hypothesis that *k'uen-to may be derived from root *k'eu-(H-):*k'u-aH- 'force, hero, souvereign'<'fill up, swell' can be considered.
 
Sergei