--- Piotr Gasiorowski <
gpiotr@...> wrote:
> There is a rule, recently formulated and exemplified
> by Hinge, that
> initial laryngeals fail to vocalise in Greek before
> a heavy syllable
> containing a syllabic consonant. It solves a number
> of hitherto
> baffling cases like e.g. *h2n.h2-tj(a)h2 > ne:ssa
> 'duck'. If
> *h2wl.h1-n-o/ah2- had a syllable division between
> *h1 and *n, Hinge's
> rule works and no arbitrary metathesis is needed.
Also I think it's more likely that the first h2 in
"duck" dissimilates to zero. There is no failure in *
h1l.ngWHros > h1.ln.gWHros > elaphros. This is a
unique or "arbitrary" dissimilation not a rule since *
h1r.h1tmom > h1.rh1.tmom > eretmon.
I think h does not become syllabic before a syllabic
consonant unless there is r/l/n/m/h after the syllabic
that can become syllabic in its place. Actually, it
may be better described as: h becomes syllabic before
any sonorant C but then desyllabifies if the sonorant
is syllabic.
h>+syl/#_C+sonor
C+syl C+syl C+sonor C > C+syl C C+sonor+syl C
h>-syl/_C+syl
h1l.ngWHros h2n.gelyei
h1.l.ngWHros h2.n.gelyei
h1.ln.gWHros h2.n.gelyei
h1.ln.gWHros h2n.gelyei
elaphros angellei
n. > en after h1 in same syllable
n. > an after h2 in same syllable
n. > on after h3 in same syllable
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com