I.
> Piotr wrote:
> I don't believe
> <gjuhë> has anything to do with Slavic *golsU; other etrymologies
> have been proposed.
For each word more then one derivation was proposed along the time.
We can never know exactly of course if was really related or not.
But semantic fit, phonetic fit. Derksen, Demiraj, Orel agree that
they are cognates.
But of course if there are arguments against *gols&, and pro- for
another derivation please post it.
II.
> Piotr wrote:
>As for <djethtë>, comparison with <gjashtë> 'six' shows
>that *-k^s- in complex clusters did not develop like intervocalic *-
>k^s-
>there must have been some not-too-clear phonetic similifications
> (loss of one of the consonants).
There is no complex PIE cluster CCC in <djathtë>, because is a recent
formation, Piotr. djathë is attested at BUZUKU (so with no -të). \
See Demiraj that consider djathë too. So we have an intervocalic k^s
in djathë: PIE *-k's- > PAlb c > Alb th
On the other hand gjashtë represents the phonetic cluster sek's-to
(I think that is not directly from PIE but doesn't matter here)
So PIE *k'st > Alb sht
Rule: tri-consonantic clusters that end in st lost there first element
PIE *Cst > Alb sht
So there is no original PIE *k's > Alb h (as you have proposed for
kohë) in Albanian. There is only ks > h.
Best Regards,
Marius