Re: Order of Some Indo-Iranian Sound Changes

> > I played with the idea that the original ppp morpheme was -dh-,
> > somehow related to the *dhe:- verb, and that PIE originally had an
> > endingless nom. like an accusative language should (onto which
> > later a nom. -s was slapped, after root lengthening, that would
> > make Szerémenyi lengthening much less awkward to formulate), and
> > that nom. -dh# > -t# which was generalized in some cases. I think
> > that took care of Bartholomae, but it's a long time ago.
> >
> >
> > Torsten
> >
>
> That's very interesting (the counter-argument to Bartholomae), but
> then how do you account for Avestan <duGDar> "daughter"?
>
> Andrew
>

I hadn't thought of that one. If I should tackle it the same way, I'd
have to somehow 'open up' the suffix to expose the consonant cluster
to a word boundary. I've toyed with the idea that the *-ter ending was
a locative, used in deference, with the verb in passive/impersonal,
also since PIE supposedly had moieties, so instead of eg. 'father
(*px-) says', something like 'on father's side, one says/it is said';
later the verb was normalized to active. That would make *-ter cognate
with 'there' etc, which, since it goes with a set of deictics in PIE
*t- must be composite.


Torsten