Re: [tied] Rounding (was: Greek labiovelars)

--- In cybalist@yahoogroups.com, Sean Whalen <stlatos@...> wrote:

> > > I'm not giving up; initial w in P-Celtic can
> > cause
> > > rounding independent of the Irish types.
>
> > Well, if we ignore British Celtic *wor (and I don't
> > see how we can),
> > you still can't assume a change of initial *we- >
> > fo- (or *wi- > *fu-
> > for that matter) in Irish. What about fer < *werah <
> > *wiros, feithid
> > < *wet-e-, feidid < *wed-e- etc.? Or am I missing a
> > point?
>
> In Irish CW causes rounding, in Welsh w causes
> rounding (in some environments, possibly irregular);
> uper- moves through stages with both causes of
> rounding in my view.

But Ir. for? it has simple *w-, not *CW-.

>
> > > Sihler says final u causes rounding and drops;
> > > spelling daur indicating darW etc., and if so the
> > > roundings would be for the same reason at the same
> > > time.
> >
> > Yes, but not the same thing and certainly not
> > conditioned in the
> > same way.
>
> Are there any cases of final CW from initial
> kW/gW/gWH the newly rounded C's can be compared to?
>

Labiovelar rounding has only been demonstrated for stressed vowels,
as far as I know and only when the labiovelar precedes the vowel.

You don't get rounding in, e.g. ech 'horse' < *exWa- < *ekWo-. Or is
that not what you're saying?

u-infection affects stressed and unstressed vowels alike.

By the way, I hope the Cybalist readers aren't getting too fed up
with this thread - at least I think we're able to keep the subject
on track.

Anders