11-11-03 00:10, Abdullah Konushevci wrote:
> Ok. I will agree with you if you are able to explain to me or to
> others: Why Latin <centrum> derives in Albanian <qendër> (Geg
> <qândër>, (t.) qëndër) 'center' and <centum> just not <quind>, but
> <qind>. Phonetically speakig, much convincing is Gall. or Bret.
> <cant> then Latin <centum>, taking into account that consonontal
> group <-en> derives regulary: (g.) -ân and (t.) -ën. See also Hemps
> explanation about numb one, (g.) njâ, (t.) një.
Sorry, "<quind>" was a typo. As for <qind>, virtually everybody
(including Hamp) derives it from <centum>. It can't be derived from
Celtic *kanto- (show us how it can, if you think otherwise); nor can it
be inherited (we'd get *that- from *k^m.tóm, or something still stranger
if the original place of stress had been preserved, as it was in the
other Albanian numerals, but certainly nothing like <qind>). The change
of -ent- > Mod.Alb. -ind- in loans from Latin is uncontroversial, cf.
parente- > prind. A nasal vowel could not develop here, since the loans
aren't old enough. The variable reflexes -end-/-ind- aren't particularly
problematic, since /e/ and /i/ can alternate in Albanian.
Piotr