Re: [tied] Re: Albanian = Illyrian (1)

18-11-03 14:09, Abdullah Konushevci wrote:

> First of all, it can't be in any circumstances a Slavic loan, because
> rhotasism cease to function in Tosk dialect before Avars and Slavs
> invasion. Compare Sl. zakon > Alb. zakoni 'law', Slavic seno > Alb.
> sanë 'hey'.

There are different chronological layers of Slavic loans in Albanian,
and e.g. <zakon> is definitely a rather late borrowing, like other words
that have to do with legal or administrative matters. But I think you
are right: I can't think of any unquestionable Slavic loan showing
rhotacism in Tosk. To be sure, this may be an accidental attestation
gap, since the oldest loans from Slavic are few; in fact, we _may_ have
an authentic example of rhotacism in Labërija < *labanija with Slavic
metathesis (< *albanija). Still, the chances that the name Vlorë was
mediated by Slavic are not high.

> So, Geg Vlona, Tosk Vlora is by all means pre-Slavic
> place name. Indeed, there exist some kind of romanization of this
> toponym in Italian Valona, form borrowed also by Slavs. I express
> also my doubt about other place name Valbona, according to Jerichek
> from Lat. vallis bona 'good valley'.
> /au/ > /av/ or /af/ is also one outcome in Albanian: *aus- > afsh 'to
> do something with great pleasure, like shining'

This just can't be right in an inherited word. I can't tell you off hand
where <afsh> comes from, but it certainly doesn't descend from PIE *h2aus-.

> or in Latin loans:
> Lat. laus, -dis > Alb. lavd 'praise' or Lat. causa > Alb.
> kafshë 'cause, thing'.

These example show that you can get Alb. /v/ or /f/ from the Latin
diphthong /au/ (via Romance sources) -- something that supports the
hypothesis that <Vlorë> is the Albanisation of a Latinised (early
Romance) placename.

Piotr