From: João Simões Lopes Filho
Message: 36831
Date: 2005-03-22
----- Original Message -----From: elmeras2000Sent: Tuesday, March 22, 2005 11:07 AMSubject: [tied] The 'lamb' word [Was: Re: Mi- and hi-conjugation in Germanic]
--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@......>
wrote:
> On 05-03-21 13:37, elmeras2000 wrote:
>
> > What do you mean when you say the presumed etymon *explains* the
> > form in an economic way? Surely a protoform with a long vowel
> > involving no change would be even more economic.
>
> It would, if there were a convincing explanation of the short
vowel in
> the other branches. Some of the evidence is dificult to evaluate,
but
> the key witness is Greek, where hypothetical *a:gWnos would not
have
> been affected by Osthoffian shortening.
That is exactly the point. How did you decide? Can you point to an
example of a word with a retained long vowel before /mn/ from *-gWn-
in Greek? I can't find any pertinent material one way or the other.
On the other hand, no extra cost
> is attached to the assumption of lengthening in the ancestor of
Slavic,
> Winter's Law being an independently established process.
Well, Winter's Law is precisely not an independently established
process in the environment we are talking about at the point to
which this debate has now progressed. The question still remains: If
Winter's Law has produced lengthening here, why was it blocked in
other examples?
> > And, in case that
> > matters, how unmotivated would a vrddhi form be in a word
denoting
> > the young of an animal? That looks to me like a functional
component
> > that could very well be signalled by a vrddhi structure; also
the
> > accented thematic vowel structure is fully compatible with
vrddhi.
>
> But see above on the implausibility of IE vrddhi already in pre-
BSl.
> *(H)a:gWnós (I mean "vrddhi" showing up as vowel length;
*//h2egWnó-//
> with a full vowel may of course represent the widespread type of
> formation with infixed *e). As for the possibility of Slavic
> morphological lengthening, *-eNt- neuters don't show it, cf. *tel-
eNt-
> 'calf', *s^c^en-eNt- 'whelp, cub'.
That is not necessarily relevant. I don't think there is any
morphological category that regularly shows vrddhi, and yet there
are isolated examples that do. But certainly it would be preferable
to avoid it and have only exhaustive explanations that stay away
from loop-holes.
Jens