--- In cybalist@..., "Piotr Gasiorowski" <gpiotr@...> wrote:
> I think all these crunchy terms are expressive variants of (half-
onomatopoeic) roots like *skrembH- (cf. Slavic *skrob- ~ *xrob-).
*xormU is an interesting word. One feels tempted to connect it with
Sanskrit sra:ma- 'lame', which could in turn be the same as Germanic
lam- with a mobile *s (*slom-os). The only trouble is that this
attractive etymology would absolutely require an Iranian intermediary
(*slomo- > PIIr *srama- > Iranian *hrama- > Slavic *xromU) and I
don't think it's documented in Iranian. Of course an accidental
attestation gap is possible but the fact that we must assume it
weakens the etymology.
>
I see your point, but one can't explain all the instances of Slavic
*x- in anlaut via Iranian influence, and I'm aware of at least 5
correspondences between Slavic and mostly Baltic that match the
pattern Slavic *xr- ~ Baltic skr- (word-initially). One can explain
all of them as expressive/onomatopoeic (even *xrIbItU 'backbone'),
but *skr- > *xr- is not impossible phoneticcaly, because there's a
number of rather convincing examples of Slavic *x- < *sk- in anlaut.
I'm still looking for examples of Slavic *-xr- in inlaut (as to *-xn-
, here, as you hinted, we have examples like *sUxno,ti 'get dry').
Sergei