Re: [tied] *g'(h)- > d as aberrant outcome

From: elmeras2000
Message: 32531
Date: 2004-05-10

--- In cybalist@yahoogroups.com, "alex" <alxmoeller@...> wrote:

>
> If I won't see here Alb. "gjen" with the meaning "find" I will ask
you
> if you want to have examples of non-palatalised PIE *gh- in
Romanian.
> It seems a bit strange that until now none from the "Kenner" of
this
> subject did not show any examples. Are the established etymologies
in
> Albanian so unsure that one cannot show any PIE *gh- with its
reflex
> in Albanian?

There are examples of *gh > g, but none really good before front
vowels as far as I can see. I may have overlooked some from more
recent scholarship. Since gardh 'fence' may be a loanword (Slavic or
Germanic) and dergjem 'am sick' may have an original back vowel
following /gj/, I know only shteg 'path' off the top of my head.

I do not think it is really a matter of dispute that *gh yields /g/
before back vowels even of examples may be scarce. But is should be
treated with significant reservation that this applies also before
front vowels, unless a series of good examples can be produced.

Jens