Re: [tied] Abstractness (Was Re: [j] v. [i])

----- Original Message -----
From: Jens Elmegaard Rasmussen
To: cybalist@yahoogroups.com
Sent: Wednesday, June 04, 2003 9:41 PM
Subject: Re: [tied] Abstractness (Was Re: [j] v. [i])


> Well, that's funny, for /e, i, u/ by an analysis along these lines was
exactly what Kurylowicz and Benveniste made for IE - and that was rejected
on *typological* grounds. Is typological wisdom only a matter of
terminology?

No. I said /a, i, u/ was impeccable (and common). /e, i, u/ (with /e/ a
distinctively [+front] vowel, which is what makes the difference) is indeed
too odd. A quadrangular system such as /E, O, i, u/ (with a pair of low
vowels, one front, one back) would be acceptable, and the five-term
inventory /a, e, o, i, u/ is possibly the most common type of vowel system
on earth.

Piotr