a96_aeu@... wrote:
> Hi!
>
> Why not deriving from PIE *sei-N- 'to be damp, to drip' like Alb.
> gjizë 'curd' is supposed to come from ?
>
> You have actually rebuilt gji as *sein-.
>
> Alvin
There should be maybe a problem. We seen here that the "gj" in albanian
can be from an "s", no matter if this is a PIE "s" or from Latin "s".
the problem is that Albanian and Romanian has some words in common which
should belong to the very old layer of the languages. And here is not
anymore the connection "s" > gj" in albanian or "s" > s/S in romanian.
the conection is Albanian /gj/ versus Romanian /ghi/
Examples
gjemb = ghimpe ( thron )
I am not quiet clear how the Albanian threatament is but for Roumanian I
have my ideea.
I observed that in the thracian space the "k" is maintained but I am not
sure which of them ( k, k', k'h)
I observed too that only g became palatalised but I am too not very sure
which of them ( g, g', g'h)
Therefore in Romanian I can have an "g" from one or more "g"-sets but
too an "z/j" from this g.
If this works for Albanian too then maybe the Albanian "gj" is the same
sound which has at origin an "s" or an "g"-set like in romanian.
For example: solt , Albanian 'gjellëz' romanian "sare" versus Albanian
gjemb and Rom. "ghimpe". If I should be right with my assumption that an
PIE *s gave an "gj" in Albanian but an PIE *g from "g"-set an "g" , then
we have in both languages the same situation. But so long I am not sure
it is still difficult to say if here is a Latin word or a independent
evolution and I dont forget that in Albanian is an another word for solt
and this is 'kripë' which seems to do not derive from the PIE *sal-
Alex