--- In
cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...>
wrote:
>
>
> I was wondering:
>
> if hysterodynamic nouns were actually composed of two nouns,
namely
> a combining form + a proterodynamic noun, we could merge
> proterodynamic and hysterodynamic into one type.
>
> ex. in the family relationshp nouns
>
> *pax-ter
should be *px-ter
>
> *-ter appears as a suffix in the most varied contexts, and as -tar
> ('inhabitant of') in Basque and Iberian. Was it once a noun?
>
>
I just read in Sihler that the combining form of compounds in
Sanskrit have lengthened grade. Might that render the -x- (h2) of
the 'family relations' suffix superfluous?.
Now suppose the correct general analysis of nouns is not
root + extension + ending
but
root + (extension + ending)
ie. the extension is (was once!) an independent word, then the
compound would have begun as two words with independent and equal
stress:
weak case
róot éxtention+ending
strong case
róot extension+énding
but once the two words merged to a compound, they had to choose
between one of the stresses,
either that of the combining form
weak case
róot+extension+ending
strong case
róot+extension+ending
or that of the second word
weak case
root+éxtension+ending
strong case
root+extension+énding
The first of these is the acrostatic pattern, the second one the
hysterodynamic one.
So if we assume that the language originally had the proterodynamic
pattern, we can construct from it both the acrostatic and the
hysterodynamic pattern. Nice.
Torsten