u-stem, wa-stem, adjectives (thick, murk, etc.)

[llama_nom]

Also, I'd be interested to hear any opinions on potential cognates of
'thick' and 'murk'.  I'm not at all knowledgeable about PIE, but both
have cognates outside of Germanic; are theer any clues there that
could help to decide the declension?  Old English has ja/jo-stems:
'þicce' and 'mierce', but their Modern English reflexes show no sign
of palatisation.  In the case of the latter, at least, this might be
due to its being influenced by, or reintroduced by, the Norse cognate:
'myrkr', masc. ac. sg. 'myrkvan'.  Otherwise, the OED speculates that
the palatisation in English might have been blocked by an intervening
-w-.  I suppose an alternative could be that palatisation failed, if
it was an old u-stem, and non-palatised forms were generalised from
those parts of the declension with no following 'i', 'j'.  Koebler
reconstructs Go. *þiqus, on the basis of It. 'attecchire', OFrench
'tehir', the Germanic word, *þek(k)uz?, being from < PIE *tegu-.  So
I'm wondering which seems better: Go. *mairqeis (as OE mierce, OS
mirki), or Go. mairqus, given that little remains of the u-stems
outside of Gothic, and that many of the Gothic u-stems exist as
ja/jo-stems in OE?

LN


--- In gothic-l at yahoogroups.com, "llama_nom" <600cell at ...> wrote:
>
> 
> > *arus (adj. wa) "quick" < Gmc. *arwaz
> 
> Ah, yes.  Thanks for that one.  Any relation to *arwjo (Koebler: ohne
> sichere Etymologie)?  I'm not sure what the semantics of that would be
> though: more haste less speed...?  Koebler cites the personal names
> Arosinda, Aragunti, Armirus.  Medial vowels in Migration Era Germanic
> personal names recorded by Classical authors are notoriously variable.
>  Alternatively, could these be from 'ara' "eagle"?  How well are these
> attested as naming elements attested in the other Germanic languages,
> I wonder.  Incidentally, does anyone know what the loanword (or
> loanwords) is that Gamillscheg's *arþ- "Wohnsitz, Aufenthalt;
> residence, domicile" is based on.  All the other Germanic dialects, as
> far as I know, would be in keeping with PG *arduz, with -d-, but then
> we have Go. bloþ, gen. bloþis, so I suppose it's not impossible that
> Gothic was the odd one out here too.
> 
> > *aiws (Mu/i) "age" < Gmc. *aiwaz
> 
> It follows a mixed (w)a/i declension in Gothic, doesn't it?  dat. pl.
> 'aiwam', acc. pl. 'aiwins'.
> 
> > Hm. Not sure. Maybe I would do this: salus salwa-, falus falwa-...
> 
> Maybe...  That's another possibility I hadn't thought of.  Although
> such a variation isn't actually attested in the sparse remains of
> Gothic as written, and we have got masc. nom. sg. 'manwus', and no
> examples of such a variation after a consonant.  But there seem to be
> some conflicting views out there about, for example, 'manwus'.  Is it
> from *manwaz [
> http://www.dbnl.org/tekst/boer043sync01/boer043sync01_001.htm ], or
> *manuz [ http://us.share.geocities.com/iliria1/etymology1.html ]?  The
> alternation 'glaggwaba' : 'glaggwuba' suggests a sound change:
> unstressed 'a' > 'u' after 'w'.
> 
> > Unfortunately we never see *taihswa in a strong masculine nominative?
> 
> At least we can tell that it hasn't been assimilated to the u-
> (u-/ja-) declension, since there's no -j- intervening before the
> ending.  But maybe it's use as exclusively a weak adjective would have
> forestalled the assimilation to the u-stems that others of its
> declension may have undergone.  Is a Gothic origin posited for Italian
> 'salavo'?
> 
> > I would lean not toward unetymology because I have the information I
> > cannot avoid, nor pretend error or "casuality".
> 
> I'm not quite sure what you're saying here.  If you mean: when in
> doubt, reconstruct etymologically 'correct' or regular forms, rather
> than analogically altered forms, I'd tend to agree except where a
> regular pattern of analogical alteration is attested in the language
> (e.g. the treatment of neuter ez-stems as neuter a-stems: agis, sigis,
> hatis).
> 
> LN
> 
> 
> --- In gothic-l at yahoogroups.com, "thiudans" <thiudans@> wrote:
> >
> > Hm. Not sure. Maybe I would do this: salus salwa-, falus falwa-...
> > 
> > Orel has *salwaz and *falwaz in PGmc, but for shadow he has *skaduz,
> > *skadwaz. He has for skaus *skawaz, for few *fauhaz or *fahwaz; but
> > *manwuz is a (w)u-stem. There is gothic farwa- (ns. farus?)
> > appearance, appearing in Orel as *farhwo'. Of course, Gothic *badus
> > has only common Gmc. precursor *badwo'. Also there is
> > 
> > *aiws (Mu/i) "age" < Gmc. *aiwaz
> > *arus (adj. wa) "quick" < Gmc. *arwaz
> > 
> > Unfortunately we never see *taihswa in a strong masculine nominative?
> > 
> > 
> > 
> > I would lean not toward unetymology because I have the information I
> > cannot avoid, nor pretend error or "casuality".
> > 
> > 
> > 
> > --- In gothic-l at yahoogroups.com, "llama_nom" <600cell@> wrote:
> > >
> > > 
> > > 
> > > I'm very tentatively leaning towards *salwus, *falwus...
> > > 
> > > 
> > > --- In gothic-l at yahoogroups.com, "llama_nom" <600cell@> wrote:
> > > >
> > > > 
> > > > Does anyone have any thoughts on the best way to reconstruct
> > > > hypothetical Gothic cognates for English 'sallow' and 'fallow'?
>  Pure
> > > > u-stems: *salus (cp. skadus < *skadwaz)?  Wa-stems: *salws?  Or
> > > > u-stems from original wa-stems with -w- retained as part of the
> stem:
> > > > *salwus (cp. manwus)?  And is it significant that the surviving
> > > > wa-stem adjectives have long roots -- CVCC (triggws), CVCV
> (lasiws) --
> > > > or a monosyllabic root ending in a short vowel: CV (*faus, *qius,
> > > > *unskaus)?
> > > > 
> > > > LN
> > > >
> > >
> >
>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/gothic-l/attachments/20070330/62f60765/attachment.htm>