From: dgkilday57
Message: 70957
Date: 2013-02-19
>Of course I could be wrong, but I do not recall proposing *f > *xW, and I cannot follow your argument about it.
> --- In cybalist@yahoogroups.com, "dgkilday57" wrote:
> >
> > --- In cybalist@yahoogroups.com, Rick McCallister wrote:
> > >
> > > Could it have been Lusitanian? That could count for the P.
> >
> > Lusitanian has *ikko- from *h1ek^wo- 'horse', so I would expect its 'wolf' word to be *ulko-, *wolko-, or *lukko-, depending on the form it started with.
> >
> > Many years ago Miguel suggested that the tribal name Volcae meant 'Wolves'. That may well be, if they belonged to the Illyro-Lusitanian branch and used the *wl.kWo- form. They were associated with the Silva Hercy:nia, and if Proto-Celtic borrowed Volcan *Perku:nia: as *Ferku:nia:, that would explain the lack of *p...kW assimilation.
> >
> Why do you think so many words were borrowed? You use that excuse whenever any ev. doesn't fit your rules; don't you think you could just be wrong? f > xW would have hapened in Proto-Celtic long before this borrowing by one group in one place long after Celtic split up.
> > I do not find Sean's ad-hoc claim that Celtic did not assimilate *perkWu- to *kWerkWu- convincing,In Germanic we do have *kWu- > *ku-, *gWu- > *gu-, also *swu- > *su- (LG _sump_ against _swamp_, ablaut-variants) and *fwu- > *fu- (_funk_ < *fwunkka-), with analogical restoration in pronouns. Obviously this took place before root-/u/ and /o/ fell together in Gothic.
>
> It's not ad hoc. kW > k by u first, which is seen in Greek, too. Compare fairhwus = world, fairguni = mtn Got; showing that at least sometimes KW>K by u, apparently also after u in the pronouns ugkis vs. igqis . *kWuruz > kaurus shows it's fairly old (before ur>or , at least). Some analogy, assuming the change was regular, later occurred (qim-, qum-, etc.).
> >But *p...kW > *kW...kW is NOT very common.
> even though Bolelli proposed a similar makeshift to salvage Hercy:nia as pure Celtic (Cronologia relativa di alcuni fenomeni della fonetica celtica, Ricerche linguistiche 5:101-4, 1962). Labialization of a labiovelar is fairly widespread, but assimilation of a labial to a following labiovelar is rare, and it strains credulity to suppose that Celtic and Italic did this independently,
> >
> It doesn't have to be independent; if kW > k by u happened in one dia., that doesn't stop both from having p-kW > kW-kW a little later (though the rule isn't as simple as you say).
>
> It could still be independent. Compare Skt and Greek, both having variants of Ch-Ch > C-Ch, even though no other IE felt the need to do any such thing. Comparatively, kW > p in various env. is very common.
> >Also, this place-name is conceivably derived from fundus *Perpenna:nus, owned by an expatriate with the attested Etrusco-Latin name Perpenna, so I will not try to cite it as evidence.
> resulting in Itc. *kWerkWu- (Lat. _quercus_) against Ctc. *ferku- (or whatever Sean supposes). I find it much more plausible that Ctc. also had *kWerkWu-, leading to Gaul. *perpu-.
> >
> > I do not know if Perpignan, presumably from fundus *Perpinia:nus owned by a Gallo-Roman *Perpinius, can be cited as evidence. It would require a somewhat awkward Gallo-Latin 2nd-decl. *perpus, -i: 'native oak' vel sim. against expected 4th-decl. *perpus, -u:s.