Re: [tied] Re: Short and long vowels

----- Original Message -----

From: etherman23

To: cybalist@yahoogroups.com

Sent: Monday, July 18, 2005 1:41 PM

Subject: [tied] Re: Short and long vowels

--- In cybalist@yahoogroups.com, "Patrick Ryan" <proto-language@......>
wrote:
>
>   ----- Original Message -----
>   From: etherman23<mailto:etherman23@......>
>   To: cybalist@yahoogroups.com<mailto:cybalist@yahoogroups.com>
>   Sent: Sunday, July 17, 2005 3:52 PM
>   Subject: [tied] Re: Short and long vowels
>
>
>   --- In cybalist@yahoogroups.com<mailto:cybalist@yahoogroups.com>,
"Patrick Ryan" <proto-language@...<mailto:proto-language@...>...>
>   wrote:
>   > Zero-grade with the removal of one vowel from *dhe:- /dHee/, *sta:-
>   > /staa/, and *do:- /doo/ leaves *dhe-, *sta-, and *do-.
>
>   Now let's proceed to the next step. Why do non-Greek, non-IIr.
>   languages have a in all of these forms?
>
>   ***
>   Patrick:
>
>   I would be glad to try to address your question but how about some
specific and examples of the specifics?
>
>
>   ***

Well the specific examples would be the reflexes of dhe:, sta:, and
do:. Greek gives us e, a, and o in the zero grade, IIr. i, and all the
other languages a.

It seems that you want to explain this by a merger of e, a ,and o to a
(except in Greek). But then how do you explain those cases where Greek
e and o don't correspond to a in other languages (except IIr where it
corresponds to a(:) instead of i)?

***

Patrick:

 

What is problematical is that we do not know the exact phonological nature of the sound we conventionally write *H. Given the limitations of cuneiform, we do not know if Hittite <h> represented /x/ or simply /h/ or /ç/ or something related but different yet.

 

But I think it is a little easier to make informed speculation about what the value of *H had in the three major groupings you have delineated: for Greek, it probably had the value of simply /h/; it is fairly easy to understand that with Greek, *Vh . *V:, while, in the zero-grade form, the *V reemerged with quality intact.

 

I expect that in all non-Indo-Iranian/non-Greek predecessors, it was /h/ also. But that the /h/ maintained itself longer so that at a time when Greek had eliminated it in favor of lengthening the vowel, *Vh continued to be *Vh with the zero-grade form *h, which, avocalically, was voiced to /H/, i.e. voiced /h/, a sound that would easily develop into /a/.

 

You know my views on Indo-Iranian, but for completeness, I will reiterate them. In Indo-Iranian, apparently, /h/ was fronted to /ç/ so that early <a:> was, in reality, /aç/; the zero-grade would have been /ç/, but like /h/ above, was voiced to /y/-/i/. 

 

Now I am not sure I understand the very last part of your question. Could you give specific examples?

 

Also, please do not bring in Balto-Slavic accentuation. I simply am totally in the dark there, and I am hopeful that an explanation of how *H behaved in that branch is not absolutely necessary to make my hypotheses discussable.

 

As for Hittite, I have no reason to think it was an unspecified but unitary phoneme, simply whatever PIE *H yielded in Hittite.