Re: Ligurian

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

>
> > The presence of p is not against this if the common Celtic change
> > was p > pF , with most dia. pF > F > xW , etc.
>
> I beg your pardon?
>
> > Most Celtic shows p- > (h)- and -p- > -w-,
> > just like Armenian, and pt > xt but wtH in Armenian, which allows an
> > intermediate p>F>xW to be reconstructed.
>
> I see. But other stops also give /x/ before /t/, so this can't be conclusive.
>

Unlike the traditional model, I think the fact that Proto-Celtic *F *synchronically* correlates to /p/ in other IE languages doesn't necessarily imply that *diachronically* it evolved from a proto-phoneme *p.

Just as P-Celtic created a new phoneme /p/ from /kW/, similar processes could have happened earlier in different paleo-IE dialects. This is precisely why we've got *akWa: 'water' vs. *ap-/*ab-, etc.

Also according to macro-comparative evidence, the words 'oak' and '5' had an initial cluster *xW- ~ *XW- which was assimilated to kW in Q-Celtic and Latin but it gave p elsewhere (f in Germanic).

However, in the case of the toponym Hercynia (NEC *Xwy:rkkV 'tree, oak-tree') we could posit an evolution *XW- > *F > h- .  Anyway, this word is from some paleo-IE dialect spoken in Western Europe, and thus not from the Steppe dialect (i.e. "PIE").

To the skepticals, I'd propose they go to the closet and take out their Indiana Jones' uniform, because they will need it in order to find not the Lost Tribes of Israel but ehem the Celtic and Latin reflexes of *peh2-ur- 'fire'.

> But we've got actual evidence of p-Celtic loanwords into Goidelic where /p/ was adapted as /kW/.
>

Also early Latin loanwords like the ones quoted by Torsten had p > kW. Of course, this happened because Goidelic lacked /p/, so the foreign phoneme was replaced by the closest one in the native language.

> I should recognize your idea is interesting, but it still needs some
> working. For example, to explain the different outcomes of Celtic *w- in
> Goidelic (Irish f-) and Brythonic (Welsh gw-).
>

The shift [w] > [gW] is common in Romance languages, and it can be found in Germanic loanwords. Even today, many Romance speakers pronounce [w] as [gW ~ Gw].