From: Arnaud Fournet
Message: 61543
Date: 2008-11-11
----- Original Message -----
From: "Piotr Gasiorowski" <gpiotr@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, November 02, 2008 7:19 PM
Subject: Re: [tied] Re: [pieml] Labiovelars versus Palatals + Labiovelar
Approximant
>
> On 2008-11-02 18:52, Arnaud Fournet wrote:
>
>> The word *bhuH-o that has the same structure as *k^uh2-on behaves the
>> same : short u.
>
> Not before a consonant. For example, he aorist *(e-)bHuHt is reflected
> as Ved. ábHu:t etc. (_with_ laryngeal lengthening).
>
> Piotr
>
==============
Ok,
this is a stumbling block.
My point of view is :
1. There is a couple of roots where u is a fundamental vowel.
*puH1 "blow"
*k^uH2-on "dog"
*bhuH1 "grow"
2. One possibility was :
"H is lost between u and another vowel".
this is more or less refuted by the example of *bhuH-
Ok, this is a dead end.
3. Another possibility is (my new proposal)
"CuH was metathesized into CHu-".
Example :
*puH1 > pH1u
Greek phusa, Armenian phukh, Indic phut-
(But Basque buhatu, Uralic puj-)
In the case of *k^uH2-on
(Ample macro-comparative data)
the result is equivalent to *k^(H)w-on
Because you can't make *k^H from *k^
(Or do you ?? please inform me)
Next, in the case of bhuH1
The metathesis bHH1u was thwarted because *bh in *bhuH1 is already
aspirated.
In other words, the rule is
"CuH was metathesized into CHu-, unless C was already aspirated".
I let you try to refute my proposal
with other examples of CuH-
I think my new proposal works.
=> Your turn.
Looking forward to your answer.
Arnaud