Re: IE models (was: Ligurian)

--- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
<bhrihstlobhrouzghdhroy@...> wrote:

>
> > OEH has *akW-a: while Celtic was *ab-. They're from different
> > paleo-dialects.
>
> Everybody can see here at best to what extent a difference between
> models can grow. Facts are the existence of nouns from *ab- and with
> meaning 'water' in Celtic languages and of names from *akWa: for
> rivers in (approximately) Celtic lands.
>

Only that *akWa: isn't a Celtic word, much less Gaulish.

> A model (your one) starts from the axiomatic rejection of synonyms in
> the same language and concludes that *ab- and *akWa: must go back to
> different layers.
>

I'd rather say "axiomatic negation" of the existence of multiple layers.

> Another model (my one) starts from their factual
> co-existence and projects it back to the (IE) prehistory of the first
> historically known local layer (Gaulish).
>

Exaggerating a little, you could also devise a Proto-World or "Adamic"
language with all required lexemes, morphemes and "sound laws", and
derive all present-day languages from it, but I'm afraid this won't give
us any insight into human linguistic prehistory.