From: Mate Kapovic
Message: 32306
Date: 2004-04-25
----- Original Message -----
From: "elmeras2000" <jer@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, April 25, 2004 12:06 AM
Subject: [tied] Slavic G. pl. (was: Re: Nominative Loss. A strengthened
theory?)
<Perhaps, but then fully, so that nothing remains of the old *system*
<which to nyou is so important to show which belonged where to begin
<with. I do not know of any IE language that distinguishes the
<gen.pl. form of thematic stems from that of the other stem classes
<(except by analogy as Lat. -o:rum, -a:rum).
<None of the languages quoted is fit to show what you make it. Latin
<has no long vowel at all before final -m (sim, amem, insulam,
<gen.pl. deum all have short vowel + -m). The same goes for Old
<Irish, cf. acc.sg. toil from *toleN < *tolam < *tola:m (nom.
<tol 'will'), and Hitt. -an is short also in te-e-kán from *d(h)ég^h-
<o:m 'earth'.
You have a point, these are also ambiguous. Also, Celtiberian -um in Gpl
(as opposed to -om in Asg) shows the *-o:m reflex.
<Meaning 'neck'?
Yes.
>> As for compensatory lengthening, it is not so compelling evidence.
Firstly,
>> in Slovene and Kajkavian neocircumflex doesn't always become from
comp.
>> lenthening.
<And what then is its origin here?
I already explained it. You have what looks like a metatony in a. p. c stems
(although it isn't really that) like Nsg *vor^g7 "enemy; devil" - Gpl
*vor~g7. And that is the cause for the same sort of thing to happen
analogically in a. p. a stems like Nsg *por´´g7 - Gsg *por^g7 "door-step".
>Sure, but if the ending is lost there is all the more reason to
>expect a compensation. And loss of a longer ending could be expected
>to produce more compensation than loss of a shorten ending. Why can
>there not have been degrees of reduced vowels that were both lost? I
>think you are leaving the ground of empiry here.
It seems to me you are looking for any kind of formal explanation so
even a long reduced vowels satisfies you no matter how improbable
phonetically it is.
Mate