I've wondered about that bilis connection too. As for the
adjective, if bli'waz > Go. blius (cf. lasius beside the more normal
lasiws), it could easily have come to be treated as a u-stem, which
would explain dauthubljans.
On the other hand, OE bleo (the noun) is discussed in Campbell's Old
English Grammar, 578.5: bleo "has a diphthong due to early
contraction but is historically a ja- stem, Gmc. blija-." And in a
footnote: "distinguish bliwa- 'lead', ON bly', OHG blio, bli". He
sounds quite definite on this, but I don't know the full evidence.
There's also (maybe coincidentally) an OIc.: daudyfli, neut. 'dead
thing, carcass' - though Zoega divides this up daud- + yfli.
Not sure if that gets us anywhere...
--- In firstname.lastname@example.org, "thiudans" <thiudans@y...> wrote:
> As I am hand-correcting Bosworth-Toller for Sean Crist's project,
> I often imagine what the Go. form of a certain OE word might be,
> or if there's one that already exists. Translating an OE sentence
> happened upon the word (ge)-bleoh, -bleo' "coloured". A search
> in Gerhard Koebler's GW finds the entry *-bleis, "coloured"; but
> the etymological information is lacking. He suggest possible
> connexion to Lw. Lat. -bilis.
> This seems a bit questionable in light of the possibility of being
> cognate with the abovementioned Gmc. (OE) form. Falk Fick
> Torp (286) has under the entry *bli-1 scheinen (< PIE
> *blîwa- "Farbe" citing A-S and Fris. examples with meanings
> colour or coloured.
> Thus we may postulate a Germ. *bli'wam "colour" and *bli'waz
> "coloured" Go. -bleis (from 1 Cor. 4,9 in Cod. Ambr. A, the Acc.
> form "dauthubljans" death-hued.) could have been an error for
> wa-stem adjective (composited form) or an example of a wa-
> stem having gone over to ia-stem adjectives (i.e. long ja-stem,
> like wiltheis). Wright notes how in fact only "a very few
> of [the wa-class] are extant in Gothic".
> Admittedly, this has the minor problem of explaining how a
> single manuscript form "dauthubljans" could be a mistake for *
> dauthubleiwans. Ignoring the possibility of taking "-bljans" as a
> shortened final composite form in the code "-coloured", I would
> suggest here that *bleiws (already perhaps too close to *blews
> 'blue' or unfavorably finding itself in the paltry wa-class) had
> been reformed as *bleis after wiltheis etc.
> Koebler refers to Holthausen, which I don't have access to at the
> moment. I wonder if there is connexion to a root form with
> meaning "pale", thus "death-pale" rather than "death-coloured",
> cf. NHG bleich. This however might require a velar suffix ?*-blei-