Re: woad

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> W dniu 2012-03-14 01:25, dgkilday57 pisze:
>
> > See message #68217, which corrects my earlier posts. I regard the
> > /z/r/-less forms 'meed', 'woad', 'kien' as borrowed from NWB
> > *me:do:-, *waida-, *kaina- in which the */z/ between a vowel and a
> > voiced consonant was vocalized. OE <weard>, <werd> reflect inherited
> > Gmc. *wazDa- 'woad', and Go. <ouisdil> (etc.) reflects Gmc. *wizDila-
> > with /e/-grade, from PIE *wezdH-. In my opinion the Greek and Latin
> > words are unrelated to this.
>
> Just out of curiosity: what is the OE documentary evidence for <we(a)rd>
> 'woad'? I am only aware of a single glossary entry in
> Corpus/Epinal/Erfurt which reads <sandix: uueard>. As "sandix" seems to
> refer primarily to some kind of dye-yielding seaweed (rather than the
> woad as a plant), it is not certain that <uueard> is really a variant of
> <wa:d>; it could be a syncopated byform of <waroð> 'seaweed', for
> example. And of course Gothic *wizdil- is only an insecure
> reconstruction based on late mediaeval copies of Oribasius, and so not
> much of evidence.

I have no thorn/edh today, unfortunately. According to Bosworth-Toller <waroT>, <warT>, etc. (with short /a/) means 'shore, strand', while <wa:roT> (hapax) means 'seaweed', identical in sense to <waar>, <uaar>, <ua'r> etc. glossed 'alga'. I do not see how <uueard> could be a variant of <wa:roT>. The identity of <uueard> is discussed lucidly by O.B. Schlutter, Anglia 30 (N.F. 18):249-50 (1907).

Besides Aelfric's gloss <wa:d> 'sandyx', B-T give <wyrt oTTe wa:d> 'sandix', <wa:d> 'sandix', and <waad> 'fucus' (this last explains the confusion with 'seaweed'; <fucus> must have been understood as 'dyestuff' by the glossator). The OED cites early ME (ca. 1200) <wod> 'sandix' from Wright-Wuelcker. Moreover there is an OHG gloss <weit> 'sandix'. I see no reason to doubt that <uueard> meant the same thing as <wa:d>.

In message #68216 George pointed to the DWB, which cites a number of ML forms with -sd- besides the explicit "isatis quam Gothi visdilem vocant". Some of this material is also mentioned by Schlutter, e.g. "isatis .i. uuasdus unde tingunt persum". There can hardly be a question that we are dealing with Gmc. forms in *wazd-, *wezd-, or *wizd- meaning 'woad'. Schlutter draws the parallel to Nhmb. <meard> 'merces' (i.e. 'meed') against Go. <mizdo:>, which is obliquely noted by the DWB. Considering the distribution of 'meed' and 'woad' (also 'kien' against Russ. <sosna>), I regard borrowing from NWB to Ingvaeonic as the most plausible explanation of the WGmc forms without expected */r/ from */z/ in this position.

DGK