I've been thinking more about a-colouring (uvularisation) and the
satem dialect area. Your arguement is persuasive, Piotr, and although
I'm disappointed that my pet mini-theory has exploded, I have to
agree that the satem shift developed at a late date. Boohoo. Them's
the crumbling cookies.
Anyways, then I got to thinkin' more about MidIE vocalism and the
talk about *o for stressed and *e for non-stressed. Now, the open
MidIE *a appears to cause the uvularisation of velars (*k + a -> *q)
and laryngeals. However, I've just realized that *a doesn't appear
in non-stressed syllables and that this a-colouring helps me prove
this (since *k and *q were once allophones).
Take our Uralic *ka"la"wa" for example. It relates to IE *g^loxus.
The initial phoneme is palatal *g^, isn't it? I've had trouble
finding whether this is the case on my searches on the net but one
site on Phrygian confirms this. If it's palatal, this means that
the correct reconstruction of the MidIE version should start with
*k:e- in order to explain the lack of a-colouring (or presence of
palatalization) of the velar *k:.
However, Uralic points clearly to *a at some remote stage of IE.
The most satisfying explanation at this point is to develop a new
and simpler rule on MidIE vocalism to explain all this:
Basically put, *a occurs *only* in stressed syllables, leaving the
schwa *& to unstressed syllables... However, both *& and *a can
exist in stressed syllables. The result of this rule is that MidIE
*a (LateIE *o) alternates with unstressed *& (LateIE *e) as we
see in the declension of words like *po:t "foot" or *wodr "water".
However, there are clearly words with stressed *& (IE *e), alternating
either with either nil-grade or a reinforced *e - this simple fact should
put monovocalism to bed.
Back to the reconstruction of MidIE's version of *g^lo:us, we are
then, by the above rule, forced to reconstruct *k:elaxwe (*e = schwa)
with accent on *a (the penultimate syllable) and the declensional
paradigm should look like this... (I've done away with writing MidIE
with syllabic liquids in order to explicitly show the vocalism):
*k:elaxwe (oblique)
*k:elaxwe-m (accusative)
*k:elexwe-se (genitive)
*k:elexwe-ta (ablative)
Please note the *a/*e ablaut caused by changing accent.
There are other conclusions that one can draw from this simple
rule of vocalism such as, for instance, that the unaccented
thematic *-o- seen in many IE nouns is truely a late invention
while *o/*e ablaut must be a very old pattern.
-------------------------------------------------
gLeNny gEe
...wEbDeVEr gOne bEsErK!
home:
http://glen_gordon.tripod.com
email:
glengordon01@...
-------------------------------------------------
>From: "Piotr Gasiorowski" <gpiotr@...>
>Reply-To: cybalist@yahoogroups.com
>To: <cybalist@yahoogroups.com>
>Subject: Re: [tied] Re: Satem shift
>Date: Mon, 23 Jul 2001 14:22:00 +0200
>
>(A) I have already given my reasons for regarding satemisation as an
>irreversible process. Quite apart from which, I don't see any compelling
>reasons for the grouping of Germanic with Balto-Slavic (or the Satemic
>cluster in general) in genetic terms. The affinities usually cited are
>either North European areal traits or trivial parallels that are too common
>cross-linguistically to indicate genetic relationship. The most interesting
>resemblances (such as *-m- vs. *-bH- endings and adjectival declensions)
>have been discussed on this list before.
>
>
>(B) Any variable, including differently regularised paradigms, can become a
>shibboleth provided that people are aware of it. In Irish English, the
>standard name of the letter H is "haitch", but in Protestant communities
>"aitch" is preferred to that Catholic deformation.
>
>Piotr
>
>
>
>----- Original Message -----
>From: tgpedersen@...
>To: cybalist@yahoogroups.com
>Sent: Monday, July 23, 2001 1:02 PM
>Subject: [tied] Re: Satem shift
>
>
>(A) It's nice to hear that I've produced something useful. Could that also
>serve as an explanation of the proposed re-centumisation of Germanic?
>
>
>(B) But I was also thinking also of the idea that shibboleths arise first
>within inflexion patterns which are then regularized differently within
>different sociological groups, and then become shibboleths...
_________________________________________________________________
Get your FREE download of MSN Explorer at
http://explorer.msn.com/intl.asp