Re: ka and k^a [was: [tied] *kW- "?"]

On Pon, rujan 26, 2005 1:25 am, Miguel Carrasquer reče:

> On Sun, 25 Sep 2005 15:54:34 -0700 (PDT), glen gordon
> <glengordon01@...> wrote:
>
>>Piotr:
>>> The underlying long vowel in this word could perhaps
>>> be dispensed with if the *a vocalism had no zero
>>> grade; then *Hna:s would owe its long vowel to the
>>> normal nom.sg. lengthening, and the weak stem *Hnas-
>>> etc. plus the total absence of *Hn.s- would
>>> represent the expected state of things.
>>
>>And what about:
>>
>> nom *h2neh2s (> *na:s)
>> acc *h2neh2sm (> *na:sm)
>> gen *h2nasos (> *nasos)
>>
>>With such a paradigm, the odd alternation would
>>cause some dialects to use weak stem *h2nas- for
>>all cases (hence short vowel) while in others,
>>the long vowel of the lost laryngeal would become
>>the standard. Does that work?
>
> No. Vocalized */&2/ does not equal */a/. The weak stem is
> nas- (not *nis-) in Old Indic, therefore there cannot have
> been a laryngeal between the /n/ and the /s/.

Just a quick thought - couldn't the a/a: in OI be secondary to a/a: in
cases like pa:t, pá:dam, padás or such? Wouldn't the pattern a:/i be a wee
bit too aberrant?

Mate

Mate