On Wed, 4 Jun 2003, Piotr Gasiorowski wrote:
> [...] I'd still
> argue that the short-vowel inventory of Sanskrit is of a common triangular
> type (/a, i, u/). I can't believe that any phonetician would contest that.
Then what about an analyis of PIE in which all root vowels are (rightly or
wrongly) reduced to varieties of /e/, and /y/ and /w/ are realized as [i],
[u] when not adjacent to vowels, so that the language has /e/, /i/ and /u/
(because we choose to call /y/ and /w/ by their vocalic names)? Or /a/,
(i(, (u( for that matter, if we choose to call /e/ by a variant we like
better? Would *that* language be typologically permissible?
Jens