Re: Lat. gladius and Sorothaptic

--- In cybalist@yahoogroups.com, "Tavi" <oalexandre@...> wrote:

>
> --- In cybalist@yahoogroups.com, "dgkilday57" <dgkilday57@> wrote:
> >
> > > Matasovic' thinks Latin gladius could be inherited, with *kl- > gl-
> as
> > > in glo:ria < *klowesja:. He reconstructs a Celtic protoform
> *kladiwo- on
> > > the basis of Old Irish claideb, with the Brittonic forms being
> loanwords
> > > from Goidelic. A direct loanword from Gaulish into Latin can be also
> be
> > > dismissed.
> >
> > Most Latin words continuing *kl- have cl-, so Matasovic' seems to be
> postulating an optional soundlaw, and we should always try to do better
> than that.
> >
> After reading his dictionary and having corresponded with him, I think
> cross-borrowing between Celtic and other IE languages isn't in
> Matasovic's theoric framework.

With the apparent exception of Germanic. This says something about the geography of his framework.

> > With characteristic modesty, I submit that Ligurian substrate theory
> may be capable of explaining the voiced onset of <gladius>, <glo:ria>,
> and the West Romance *gattu- 'cat' which largely prevailed outside of
> central Gaul. My working hypothesis is that Gallia Propria was
> conquered by P-Celts, originating in N Iberia, who passed through
> Aquitania and poured across the Garonne in search of greener pastures.
> The pre-Celtic IE-speakers in S and SE Gaul were principally Ligures;
> those in NW Gaul were principally Veneti. To the north, the Belgae had
> superposed themselves upon the Veneti and Ligures. The eastern Belgae
> in NE Gaul were Gallicized in pre-Roman times; the western Belgae in
> Belgica Propria were not.
> >
> IMHO, etymological proposals should be largely independent of homeland
> theories. BTW, you seem to have forgotten Etruscan and its northern
> cousin Rhaetic.

Etymological proposals and geography are inextricably intertwined. The model I was describing has to do with the Celticization of Gaul, and if I overlooked anyone it was the Volcae. The Etruscans had no significant impact on this process. I regard Rhaetic as Illyrioid, thus IE, not Etruscoid.

> > It is reasonable to suppose that here, to the west and south of the
> western Alps, Ligurian exerted a significant influence on the local
> Gaulish, which we might term Liguro-Gallic, just as we use the term
> Gallo-Latin to denote the Latin which was significantly influenced by
> Gaulish. Now, French <cabaret> and <cabriolet> (obviously not inherited
> the usual way from Latin) have been borrowed into Milanese as
> <gabar�> and <gabriol�>. French is noted for allowing very little
> aspiration with word-initial tenues. I hypothesize that the borrowing
> of Fr. /k/ as Milan. /g/ in word-initial position reflects a phonetic
> discrepancy going back to pre-Roman times. That is, central Gaulish
> dialects, which due to rapid conquest had undergone very little Ligurian
> (or Venetic) influence, allowed very little aspiration with word-initial
> tenues, particularly /k/. But Ligurian allowed considerable aspiration
> here, initial /k/ being sounded as [kH], so that Gaul. /k/, a pure [k]
> even initially, sounded more like /g/ than /k/ to Ligurian ears. Thus
> when the Ligures east of the Rh�ne and south of the Alps borrowed
> words from the first wave of Gaulish settlers, *k- became *g-, but of
> course native Ligurian words retained *k- (pronounced [kH-]). As more
> Gauls settled among these Ligures, effecting a slow conquest over
> several generations, their own Gaulish became Liguricized as
> Liguro-Gallic, with *k- in words corresponding to those in which Ligures
> had maintained native *k-, but *g- in words which Ligures had borrowed
> from Gaulish words with *k-.
> >
> Replacing Gaulish for Latin and Ligurian for Paleo-Basque wouldn't
> substantly alter the picture.

I do not understand that comment.

> > I believe the same mechanism can explain Latin <gladius> 'sword' and
> <glo:ria> 'glory' as borrowings from Liguro-Gallic *gladjos and
> *glowesja: (representing Gaulish words in *k- borrowed into Ligurian
> with *g-). The former hardly requires comment, since *kladjo- is
> otherwise attested in Celtic.
> >
> As I mentioned before, the Celtic form is *kladiwo-, not **kladjo-. I
> also see no evidence of it in Gaulish.

The extant Gaulish corpus is much smaller than that of Latin or Greek, and one should not expect to find every lexeme attested. The same could go for reflexes of *k^leu-.

DGK