Re: a discussion on OIT

--- In cybalist@yahoogroups.com, "jsjonesmiami" <jsjonesmiami@...> wrote:

>
> --- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@> wrote:
> >
> > --- In cybalist@yahoogroups.com, "jsjonesmiami" <jsjonesmiami@>
> wrote:
> > >
> > > As for PIE dorsals, there are 3 views that I can recall.
> > >
> >
> > D) There were originally velars, which had allophones c^e/ko etc,
> > and labiovelars, which had allophones ke/kWo etc. Both satem and
> > kentum languages got rid of the allophone alternation in
> > paradigms, the satem languages by generalizing the former
> > allophone, the kentum ones by generalizing the latter.
>
> I thought there was at least one more view, but I didn't remember
> what it was. (I'm not sure that all phonologists would be happy
> about [k] being analyzed as an allophone of 2 phonemes. I could be
> wrong.

You are right. It's all my invention, and you are the first to comment
on it.

> But that's a matter of terminology.) So what happens outside of
> paradigms?

The 'other' allophone in each language which generalizes becomes
'shibbolethized', ie. the groups that still use it are stigmatized
socially and politically, and therefore it disappears also outside of
paradigms.
http://www.angelfire.com/rant/tgpedersen/Shibbolethisation.html
By tradition it's called hypercorrection.

> > I've tried to sketch how it could be done
> >
> http://www.angelfire.com/rant/tgpedersen/PIEstops/PIEstopsCurrent.html
> >
> > it's definitely not definitive. Please ignore the gunk at the
> > bottom.
> >
> >
>
> The link is just to a page with a big table and no explanation,
> definitive or not. I don't think I can make assumptions about how to
> interpret it.
>

It's been so long now since I wrote it that I have difficulty getting
it myself. The top table is my reconstruction of PIE, though.


Torsten