Re: [tied] Re: Nominative: A hybrid view

From: Glen Gordon
Message: 22951
Date: 2003-06-10

I just want to make sure that I'm understood. I do
accept *e:/*e as being accent related just like
*e/zero and I do accept roots that contain long
*e: such as *hWre:g-. We are just in opposition
as to how *o/*e relates to the above, it seems.

Perhaps {gulp} Jens and I may even reach a
consensus sometime this year! Scary!


>The situation is neutral on this point. If o/e
>can be shown to work better and to produce more
>illumination of dark spots in the grammar, it
>should be preferred.

This kind of thinking is very much like adding
laryngeals to anything smelling of *o so that
everything can be reduced to one vowel *e. Of
course, this is a perfect example of simplicity
ad absurdum.

However, on examining Jens' idea, I can't help
but appreciate that we are thinking alike except
on one very, very subtle difference. Jens needs
to account for *o/*e as a product of length
differences caused by accentual changes. I feel
the need to account for *o/*e as a product of
an older "Vocalic Restriction" rule that
ultimately is still the product of accent (and
perhaps if reintrepreted slightly, length too).

The Mid IE Vocalic Restriction bars open central
*a from ever being an unstressed vowel. Only mid
schwa can be the unstressed vowel here. Stressed
*e and *a both become schwa when unstressed.

So this is still an accent based rule as yours is
and it might be intrepreted that *a, being more
open, requires slightly more time and effort to
enunciate than mid schwa. So for the purposes of
unifying ablaut to one single automatic process,
we might say that *& (> IE *e via Paradigmatic
Resistance) is a shorter version of stressed
MIE *a (> *o).

In that way, in a sense, we have the illusory holy
"triple-morae" that is so vital to Jens linguistic
faith, although it is really only double morae in
the end. That is, in Mid IE, we'd have three ablauts
unified by accent-based length changes: e:/*e,
*e/*& (> *e/*e via PR, or > *e/zero) and *a/*&
(> *o/*e via PR, or > *o/zero). My solution is
better because explaining *o/*e doesn't trip over
the explanations for *e/zero and *e:/*e. We don't
need the uneconomical phonemic triple length
contrasts, something that is very rare anyways.
I can only think of Lappish.

So there! Jens should now be satisfied because
everything is unified under length contrasts
caused by accent. I can only imagine what he
may object to next.


= gLeN

_________________________________________________________________
The new MSN 8: smart spam protection and 2 months FREE*
http://join.msn.com/?page=features/junkmail