At 7:14:27 PM on Friday, February 6, 2009, Richard
Wordingham wrote:
> --- In cybalist@yahoogroups.com, gwalstawd <gwalstawd@...>
> wrote:
>> I have trawled through the DIL looking for a possible
>> cognate but have failed. This may be due to my lack of
>> rigour.
> The word might be related to *k^el (Pokorny #889, pp552-3)
> 'thin shaft, stalk', but I can't see how that helps decide
> between a Graeco-Latin and a native origin.
The Old Welsh Juvencus glosses include the gloss 'culmos .i.
calamennou' (Loth, Vocabulaire vieux-breton, p.63; Falk,
Fick, & Torp, p.85). Loth makes this the plural of
*calaman, a derivative of *calam (my asterisks).
Both Irish and Sc.Gael. have <cuilc> 'reed' (not in the
DIL). MacBain and FFT both derive it from *kolki-, where
*kol- would be the same root seen with suffix *-mo- in Latin
<culmus>, OE <healm>, ON <halmr>, etc. Watkins (2000) takes
these to be from *k^olh2-mo- (Pokorny *k^ol&mo-s, p.612).
The form of the Welsh word is entirely consistent with
borrowing from Latin, and it seems to me that this becomes
even more probable if <cuilc> is from *k^olh2-.
Brian