Re: [tied] ph3 > b PIE transformation?

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...>
wrote:

>
> On 2006-03-14 02:52, alexandru_mg3 wrote:
>
> > Is ph3 > b a PIE tranformation or it appears independently in
> > different languages?
>
> Judging from the wide distribution of *pí-ph3-e/o- > *píb(h3)e/o-

(from

> Skt. to Lat. to OIr.), it's a proto-thing.
> >
> > As an example we have
> >
> > *h2eph3-o-l- > Celtic *ablu => Old Irish ubull
> > *h2ph3-l-o- > Slavic *jabl&k& => OCS abl&ko
>
> This is somewhat suspect. There's no guarantee that all instances

of

> otherwise unmotivated *b go back to *ph3. It's quite possible that

*b

> was rare rather than non-existent in PIE.
>
> > see also
> > *h2ep-h3on- 'river'
> >
> > Question: could somebody help me with the meaning of the suffix

*-

> > h3on ? Also if it can be decomposed in h3o-n- or not
>
> There's a lot of evidence that formations with the "suffix" *-h3on-

were

> in fact compounds. Birgit Olsen has identified the second element

as

> *h3on(h2)-, from the verb root *//h3enh2-// meaning 'burden,

charge' (as

> in Lat. onus), so compounds with the corresponding deverbative

noun

> mean, approximately 'loads of X, a place with a lot of X' (if
> endocentric) or 'having loads of X, being burdened

with/responsible for

> X' (if exocentric).
>
> Piotr
>

Thanks, Piotr.

One additional question : Why Only h3 cause the voicing of p and
not also *w or *o or *h2 etc...?

Could we imagine an intermediary phase "pb"? like a Celtic *apbon >
*abon 'river' and also an intermediary stage *apb-lo for 'apple'
too..

This *pb could next explain why there is no lost of intervocalic b
in Romanian abure (Alb avull) > Dacian *ápbula > PIE *h2ep-h3(nh2??)-
lo

Thanks again,
Marius