11-11-03 05:52, alex wrote:
> if you see Rom. "une-" as reflex of Latin unu- , be your guess.
Romanian un- _is_ a reflex Of Lat. u:num < OLat. oino-, just like French
un and Italian or Spanish uno. The initial u- < Lat. u:- represents the
characteristically _Latin_ development of *oi.
> The nonrothacisation of Alb. "n" seems to be due the palatalisation of
> the "n", thus the palatalisation of "n" was before rothacism.
That's correct, but the <nj-> of <një> doesn't correspond tho the <-n->
of <un->. A trace of the latter can be seen in the _second_ nasal of
inflected <njân-> in in Geg. There's no rhotacism there, since rhotacism
is a Tosk innovation, and Tosk has generalised the shorter (uninflected)
form një < *njâ < *(ë)nje:n.
One final remark (I mean FINAL): pre-Albanian *u- (unlike other vowels)
was not lost even when unstressed, so <një> simply _can't_ derive from
anything like *un-. I wish you stopped your idle fantasising. It's
entirely sans rhyme or reason.
Piotr