Re: [tied] Latin is a q-Dialect having p- from kW , PIE is simil

On 2007-05-14 01:46, alexandru_mg3 wrote:

> I. Pre-Germanic '5' was *pempe > Proto-Germanic *fimfi > Gothic
> fimf

Not all that LOOKS similar IS the same. In Italic and Celtic *penkWe >
*kWenkWe > Lat. quinque, OIr. cóic, but P-Celtic and P-Italic *pempe, as
in OWel. pimp. What seems to have happened in Germanic (where the
*p...kW > *kW...kW rule doesn't work) is the distant assimilation of
*kW/xW to *p/f (*penkW- > *fenxW- > *fimf- or perhaps pre-Gmc. *penkW- >
*pemp- > PGmc. *fimf-), see below.

> Pre-Germanic '4' was *petwores > Proto-Germanic *fithwor > Gothic
> fidwor
>
> So:
> Pre-Germanic *pempe
> p-Celtic was *pempe
> p-Italic was *pVmpe probably *pempe too
>
> and you are talking about a recent transformations *kW>*p or even
> *xW>*f for Germanic?

Yes, of course. And the change of Middle English x > f (spelt <gh>)
after rounded vowels, as in <cough, trough, laugh, enough, rough,
slough> is even more recent (and _proves_ that such a change can recur
independently).

> similarly : we have the same common forms for '4' *petwor(es)

You fail to explain why the change is regular in P-Italic and P-Celtic
but seems to be weakly conditioned in Germanic (by another labial in the
same word: *feDwo:r, *wulfaz, *fimf-, *twailiB-).

> II. Next because *wlpos is attested in Latin (lupus) too: wulfaz and
> lupus fitting perfectly => you cannot treat this as a pure
> coincidence : same evolution in two different PIE languages

Why can't I? Is there anything highly unnatural about *kW > *p?

> As result: the variants *penpe, *petwores, *wlpos didn't belong to
> the inner evolution of one PIE-language or of one of the dialects
> inside one PIE language: they all belong to Dialectal-PIE-times
>
> And if so, akWa/apa belong there too

Non sequitur, also because *h2ap- is found in languages that show no *kW
~ *p variation.