Re: [tied] The phonetic value of PIE *h3 and the 'drink' root.

You are right, Sergei, the "drink" word is the only evidence (as far as I
know) for the voiced character of h3, and there are so many problems with
that root that its evidence must treated with a great deal of caution!

It seems reasonable that PIE might have had a voiced laryngeal corresponding
to the unvoiced h1 and h2. I think this also persuades some people that h3
was the voiced equivalent of h2, but this is hardly evidence. Others (e.g.
Beekes) think that it was a labialised version of h2 (to explain the /o/
vocalism). H3 is in any case less common than h1 and h2, and the evidence
for it is not nearly as clear.

I won't go into other possible phonetic values of the laryngeals - there's
good discussion of it all in the archives - if you can find it.

Peter