[tied] Re: Allofamy, allofams

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> On 2006-07-05 15:45, tgpedersen wrote:
>
> > I recall seeing a couple of names of creepi-crawlies which, oddly,
> > had we,- in Polish and a- or o- elsewhere. I'll look them up.
>
> If you mean <wa,z.> 'snake' (gen. <we,z.a>), the /v/ comes from the
> prothetic glide *w, obligatorily added before some rounded vowels, in
> particular nasal *oN, in the immediate ancestor of Polish (cf. Russ.
> uz^, without prothesis). The Slavic protoform was *oNz^I, which is

the

> same thing as Baltic *angis and Lat. anguis.

I got it from Wilczak: The Pre-Germanic Substrate. He posits a
substrate where *w- disappears before front vowel, thus
PIE *we:los > PGerm substr. *e:laz > German Aal, Eng eel, cf Baluchi
wal "worm" etc. This in contrast to the usual West European word Lat.
anguilla, OPruss. angurgis Pol. we,gorz from Lat. anguis "snake",
Polish wa,z. etc.
But it occurred to me that if was allofams *(w)ax- <> *(w)an, ->
*(w)a:- <> *(w)an,- the two words might be identical.
The /w/ in parenthesis since the set of allofams is derived by a
development *kWVnk- > *gWVnk- > *wVnk- > *Vnk-).

Is this prothetic glide rule exceptionless in its environment?


Torsten