--- In
cybalist@yahoogroups.com, "Grzegorz Jagodzinski"
<grzegorj2000@...> wrote:
> 2) What is the basis for assumption that uvular sounds would cause
the shift
> *e > *a? And if you think /a/ was lower, please explain the
mechanism why
> uvular sounds should lower the following vowel. Compare your explanation
> with the facts from Turkish where we have kæ together with qA -
> uvularization causes back character but not lowering.
Do you mean Turkish? According to O'Connor, whereas one would expect
to have */A/ (or /V/ or /7/) to give a nice 2 by 2 by 2 grid of vowels
(high v. low, front v. back, rounded v. unrounded) for the language's
8 vowels, we actually have /a/ (i.e. it is pronounced [a]) as the only
vowel with the openmost aperture. He comments, 'Symmetry is very
frequent in vowel systems but it is not invariable, not is it an
absolute requirement, and one should beware of forcing a factitious
symmetry in order to keep the picture tidy.'.
Richard.