Re: The palatal sham :) (Re: [tied] Re: Albanian (1))

On Thu, 29 Jan 2004 09:01:08 +0000, P&G <petegray@...> wrote:

>> >But you have to admit that the evidence for PIE *a is very slim.
>>
>> No, I can't admit that. The evidence is quite solid.
>
>Can't we turn the argument on its head and suggest that PIE *a is in fact
>the only true phonemic vowel in PIE, since *i and*u can be analysed as
>allophones of *y and *w, and the ablaut vowel is phonetically conditioned
>and may disappear entirely?

At least in the examples of *ka- I was looking at yesterday, /a/ is fully
capable of disappearing itself in zero-grade forms [confirming that /a/ in
these cases is what would otherwise have been /e/ if not coloured by *k].

I also don't think that *i and *u can always be analyzed as allophones of
*y and *w. E.g. the 2sg. personal pronoun is *tú (~ *tú:), with full-grade
*u(:).

=======================
Miguel Carrasquer Vidal
mcv@...