Re: oldest places- and watername in Scandinavia

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

>
> On 2008-11-11 08:49, Brian M. Scott wrote:
>
> > >> [PG:] There could be no original Brugmannian lengthening in
> > >> such roots as *h2weks- (= *hweg-s-),
> >
> > > [TP:] That's a possible generalization of an s-aorist of *awg-
> > > which is a class VI verb.
> >
> > No, it's Class VII: *aukanã, *eauk, *eaukun, *aukanaz (ON
> > <auka>, <jók>, <jóku>, <aukinn>; Goth. <aukan>, <aíauk>,
> > <aíukum>, <aukans>).
>
> _And_, if it had an s-aorist in PIE (cf. Lat. auxi:), it was
> *h2e:ug-s-/*h2aug-s-.

Mea culpa. I took one look at ON auka, jók, jókum, aukinn and thought
it corresponded to class VI.

> How does one get the pattern of *faranã, *fo:r, *fo:run, *faranaz
> as a generalisation of that?

I hope you haven't understood what I proposed as that *aug- should
provide the pattern that Gmc. class VI is built on. I proposed that
that whole class VI was loaned from an ablaut-free a-language, ie. one
in which PPIE *a had not been split up as in PIE *e/*o/zero, so
a-language *parana, *pa:r, *pa:ront, *paranos >
PGmc. *farana, *fo:r, *fo:run, *faranaz

But I just got an idea. Suppose the aWg-/weg- was related to this
alternation:
http://tech.groups.yahoo.com/group/cybalist/message/61531


Torsten