Donnchath in Rí scríbas:
> feamainn (seaweed) < OI "femmain" & Welsh "gwymon" < IE
> (the nasalized zero-grade of the root *weip-, to turn, temble
> vaccilate). Incidentally, the normal French word for seaweed,
> goémon, was borrowed from Breton "goumon" or "gwemon".
That's the etymology given by GPC, but would IE *-mp- really give -
mm- in Celtic? Are there other examples?
I can only think of one other Celtic word that has - probably - an IE *-
mp-, and that is Gaulish uimpi, Cymraeg gwymp "nice, beautiful".
Note, that PIE *p really ought to have disappeared in Celtic.
The IE root *wejp "LIV 612: in schwingende/zitternde Bewegung
geraten; to start moving oscillatingly/vibrantly" is basically not
attested with a nasal infix. Only Lithuanian vimpù, vìpti "den Mund
aufreissen, gaffen; to gape, to stand gaping" shows one, but is said
by its redactor Martin Kümmel to be a new formation.
> Can someone suggest the reconstructed Common Celtic form of the
The evidence of Irish femm, feamainn, femnach seems to point to
an original n-stem, which has been given up in all branches (Karin
Stüber unfortunately doesn't dicuss this word in her monography
on Celtic n-stems). The vocalism of the British form, the lacking
palatalisation of the m in feamainn and femnach, and the seemingly
"endingless" form in femm would point IMHO to an original
amphikinetic n-stem (as e.g. also in talam, talman).
The IC (Insular Celtic) preform would therefore be, I guess,
something like nom.sg. *wimmu:, gen. *wimmonos.
-mm- could result from -mb-, or more probably in this case from -
sm- ('coz we need an n at the of this cluster for the amphikinetic -
mon-suffix, like in talam < *telh2mon-), so we would arrive at *wis-
mon-. In the nom.sg. we would in PIE expect the full grade of the
root *wéjs-mo:(n), in the gen.sg the zero grade of root and suffix
*wis-mn-és, but this would have been levelled to zero grade in the
root and o-grade in the suffix throughout already in CC (cf. Karin
Stüber, The Historical Morphology of n-Stems in Celtic, p.11 ff. and
Thus we arrive at a root *wejs. LIV 613 gives three different roots of
the shape *wejs, of which the first two from the semantical and
distributional point of view are of interest for us: the first one means
"to sprout, to flourish", the second one "to flow". Both would
probably fit semantically. In the first case the word wejs-mon-
would mean "having flourishing; flourisher", in the second case
If my analysis is correct I can't say, but this would be the most
straightforward reconstruction (I just see, that Peter Schrijver has
the same analysis in "Studies in British Celtic Historical Phonology"
* Mag. David Stifter *
* Goldschlagstr.37/17 *
* 1150 Wien *
* ++43-(0)1-789 66 03 *
* [log in to unmask] *