[tied] Re: PIE *pel- 'to sell' < - > 'to buy'

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

>
> On 2006-06-13 08:15, Abdullah Konushevci wrote:
>
> > You have right, for as /n/, voiced apical nasal, gets /d/,

voiced

> > apical stop and, as well as /m/, voiced bilabial nasal, gets
> > homorganic sounds /b/, voiced bilabial stop.
> > In case of <bletë> 'bee' I think that reconstructed form

should be

> > *melit-teH2 (cf. Greek melissa, Attic melitta)
>
> In the Greek words we have -ss- ~ -tt- from *mélitja < *melit-ih2.

PIE

> *-tt- would have given -ss- in _all_ Greek dialects!
>
> > and thanks to iCC >
> > eCC we could reconstruct PAlb. *mHlitta:/*mlitta: with lost

of

> > pretonic /e/. But, in case of <mbles> 'marriage broker, go-

between'

> > we have evidence of prefix *H1en-.
> >
> > Konushevci
> >
> >
> > For -iCC- > -eCC- see also Latin loan:
> >
> > Lat. littera > Alb. <letër> 'paper';
> > Lat. missa > Alb. <meshë> 'mass, office';
> > Lat. sagitta > Alb. <shigjetë> 'arrow' with variant

<zhgjetë> 'id.' etc.

>
> Certainly. But in inherited words *-tt- developed into a proto-

Albanian

> affricate which has yielded modern /s/ (falling together with *-tj-

), so

> *melittah2, apart from lacking any Greek support, is questionable

on

> inner Albanian grounds. The most reasonable reconstruction seems

to be

> *mlit-ah2 (*melit- was an ablauting stem, so the loss of the *e

doesn't

> have to be blamed on the Proto-Albanians).
>
> Piotr

You are right : there wasn't a -tt- there...based on the fact that
PIE *tt gave finally s in PAlb.

So *meli-teh2 > *mli-ta: etc...

But an issue remains: why i > e? If we cannot explain this,
the 'loan-option' should be follow.

Marius