--- In
cybalist@yahoogroups.com, Miguel Carrasquer Vidal <mcv@...> wrote:
> As an aside, there's also a "guru thing" in Armenian (not
> surprisingly, given the other evidence that labialization was a
> feature of Armenian phonology until relatively recently). Birgit
> Olsen gives the following examples:
>
> *gwr. > kur (for expected *kar):
> kerakur "food" < *gwerh3ogwr.-
> krawnk` "religion" < *kurawn- < *gwr.hm.no-
> *gwl. > kul (for expected *kal)
> klanem, (praet.) ekul "swallow" < *gwl.(h1)-
> *kwr. > or (for expected *k`ar)
> -ord (suffix) < *-kwr.t
> ordn "worm" < *kwr.(t)mn.(t)- (?)
> *kwl.h1 > (h)olo (for expected *k`ala)
> holow "rolling" < *kwlh1ti-
> oloRn "pea" < *kwl.h1-r-n-
> olor "twisting" < *kwl.h1-r-
>
> Note also the development *kwo- > *p`o- > (h)o- above, and also in a
> word like <o> "who?" (< *kwos).
>
Catching up on a few thing, I'd say:
There was no kW>p; only opt. qW>XW , since in Arm., as in Greek, xWr, > xWor , etc.
*kWr,mi- > *qWr,my- > *XWr,nYy- > *XWornYdY- > ordn (w reanalysis)
*kWekWLó+ > *qWoqWLó+ > *XWoLXWó+ > holow
r, > ur / etc. was the normal development of syllabic C after the environment-specific changes shared w Greek (mostly after/before x() ), as in -kul , durgn , *-r/n+ > *-ur/un+ , etc. It had nothing to do with CW , etc.
-ord shows t>d after r, but between V this is dependent on stress, such as:
*pYLut*ó+ > *pYóLt*u+ > *pYór.t.u+ > ortH -u- = calf/fawn Ar;
(w double dim. in
pórtax \ pórtis (f) = calf G;
)
more easily seen for rp , as in:
*t.er.p.t.r.ó+ \ *t.r.,p.t.r.ó+ >
terpós \ tarpós \ tárpe: = large wicker basket G;
tHarpH = large wicker fishing-basket, tHarb = wicker trellis-work Ar;
in which G shows stress-alt. and Ar the resultant C-alt.