On Mon, 31 Mar 2008 01:12:35 +0200, Miguel Carrasquer Vidal
<
miguelc@...> wrote:
>On Sun, 30 Mar 2008 22:45:53 +0200, Miguel Carrasquer Vidal
><miguelc@...> wrote:
>
>>Short (it's aínumai, with recessive accent, as all Greek
>>verbs: if the /u/ had been long, it would have had to be
>>ainûmai [note that -mai counts as _short_, it's *-maj not
>>*-mai]).
>
>Sorry, my mistake. It's long. Only if the _last_ syllable is
>long is antepenultimate accent not allowed. Since the last
>syllable here is short (-maj), the penultimate can be long,
>and the stress can still fall on the antepenultimate (_must_
>fall, in a recessive form).
>
It was actually Sihler that confused me. Earlier today, I
was re-reading what he says on Cowgill's law (Greek o > u
between labial and resonant).
Among the examples, he lists the suffix -nu-, which he
derives from *-nh3- > *-no- > -nu-. That must be a short
vowel, and that stuck in my mind.
Upon closer reading now of the section on verbal morphology
in Sihler, I see that he derives -nu:-/-nu- (the vowel _is_
short in the plural: deíknu:mi ~ deíknumen) from a merger
between *-neu-/*-nu- and *-neh3-/*-nh3-. The expected forms
(*-neu- ~ -nu- and *-no:-/*-no- [or -nu- if you apply
Cowgill's law]), merged as -nu:-/-nu-.
The middle (e.g. aínumai, deíknumai) could have short or
long -nu-, I'm not sure.
Which brings me to me actual query: the other day I saw that
Pokorny gives the Gsg. of <nux> as <nu:któs>. How solid is
that?
=======================
Miguel Carrasquer Vidal
miguelc@...