Xavier Delamarre wrote:
> > Irish "bainis" (= wedding, wedding banquet) has a similar history.
> > OI "banais" is a compound of "ben/ban" (woman) + "feis" (see above).
> OI bainis 'marriage feast' is even a old common Celtic, and possibly
> Indo-European denomination. The comparandum is Breton banvez 'marriage
> feast, banquet'. Both from CCeltic *banowetsi- < IE *gwn.h2-o-wedh-ti-
> 'leading of the bride'. The verbal root is *wedh- 'to drive, to lead' and
> already in IE times 'to marry (of men)' : Lithuanian ved¨ 'lead, marry',
> Sanskrit vadh˙:- 'bride', etc. See P. Schrijver EC 34 (2000) 136 and Studies
> in Brit. Celt. Hist. Phonology 411.
I was following Vendryes here, as you probably guessed. He mentions
Breton "banvez", but derives the Irish, and by implication the Breton,
from a compounding with "feis". I will assume that Schrijver's etymology
is sound, so where does this leave "feis" (LEIA not having reached the
F's yet)? Is its apparent role here merely fortuitous? Could it still
derive from *wedh- via the semantic development: marry > cohabit with >
sleep the night with > spend the night ?