On 2006-01-07 09:43, Sean Whalen wrote:
> Do you think that hiarós/iarós comes from metathesis
> is_H1ros > is_rH1os? If so is there any reason you
> use &1 not H1?
I used *&1 for PIE */h1/ in */ish1ro-/, a position in which it would
most probably constitute a syllable nucleus. This is what underlies Gk.
hierós (note Myc. i-je-ro-). As for the variants <(h)iarós> (Doric, New
Greek), <i:rós> (poetic Ionic) and even <ros> (Lesbian), they may be due
to various inner-Greek and pre-Greek special developments, but I doubt
if we have any right to regard any them as very old. Doric sporadically
has /a/ for Attic /e/, like Artamis for Artemis.
> How do you account for the retroflex n. if you don't
> reconstruct a retroflex series in PIE (it's not just
> after u/i (vi-sad- and vi-san.n.a- "sad"))?
It's <vis.an.n.a>, with "ruki" /s./ and distant assimilation of *n,
which is regular in Sanskrit in this environment (see Richard's reply).
The development of <ks.un.n.a> is the same.
Piotr