Re: [pieml] Labiovelars versus Palatals + Labiovelar Approximant

From: Arnaud Fournet
Message: 61235
Date: 2008-11-02

----- Original Message -----
From: "Piotr Gasiorowski" <gpiotr@...>

>
> On 2008-11-02 14:52, Arnaud Fournet wrote:
>
>> Apart from *bhuH-o
>> Do you have another example of sequence uH2o ?
>> or even -uHx-o
>
> If you had read my responses with some care, you would have noticed one
> particularly relevant example (among those I have already given): PIE
> *h2juh3o:(n). It gives Skt. yúva: 'young (man)'. In the weak stem before
> inflectional and derivative suffixes beginning with a vowel (like
> gen.sg. *h2juh3n-os > yu:nas and feminine *h2juh3n-ih2 > yu:ni:) you can
> see laryngeal lengthening, but if a consonant follows and the stem-final
> *n becomes syllabic, the laryngeal begins the second syllable, there is
> no lengthening and a hiatus-filling -v- is inserted: loc.pl.
> *h2juh3n.-su > yuvasu, derived adj. *h2juh3n.-k^o- > yuvas'a-. All that
> is _very_ different from the behavious of *k^won-.
>
> Piotr
>
============
I'm sorry for asking the same question twice,
I remembered afterwards that you had already answered it.
I read your answers with much care (not some care).
Nevertheless, to go back to the issue,
the cases are not exactly identical.
In *k^uh2-on (according to me) *u is definitely a vowel and h2 is part of
the root, all this being based on pre-PIE etymologies.
In *h2juh3o:(n), it's unclear what -u- is (vowel/consonant), and in any case
there is a morpheme boundary *h2ju- // -h3o:(n).
and probably another morpheme boundary in *h2j- //-u-//- h3o:(n).
And to be frank, if -u- here in *h2juh3o:(n) is the same thing as in pol-u-
then it's from *w.
So this example *h2juh3o:(n) and *k^uh2-on are very much different
structurally so it's no wonder they don't behave the same.

The word *bhuH-o that has the same structure as *k^uh2-on behaves the same :
short u.

A.