From: Patrick Ryan
Message: 40291
Date: 2005-09-22
----- Original Message -----
From: "etherman23" <etherman23@...>
To: <cybalist@yahoogroups.com>
Sent: Wednesday, September 21, 2005 7:58 PM
Subject: Re: [tied] *kW- "?"
--- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...> wrote:
> > I disagree that /e/, /o/, and zero are allophones of *a, though.
>
> *a is just a label. It means (here) "the PIE ablaut vowel". Or "the set
> of vowels which descend from pre-PIE /a/". Just like English [-a-]
> really means /ä/ or /eI/ or /&/.
Sure, but I also believe in PIE /a/ so I'd rather write *e for /e/,
/o/, and zero.
***
Patrick:
That usage obscures rather than illuminates the facts.
I prefer *A (for Ablaut not /a/) to indicate the variable phoneme
(*e/*o/*Ø).
In my opinion, *(-,Y,W)a is better reserved for the preceding stage, which
set the conditions for *A with the introduction of strong expiratory and
tonal accent.
Of course, this still leaves some ambiguity because *(-,Y,W)a itslef
developed from *e, *a, and *o (at whatever height they may have been
realized).
***
***