--- In
cybalist@yahoogroups.com, "Andrew Jarrette" <anjarrette@...> wrote:
>
> --- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@> wrote:
> >
> > Repeating a rule is not affirming it, and since he doesn't go
> > into the specifics I can't use his Pawnee parallel for anything
> > so I think I'll stick with my version until someone refutes it.
> >
> >
>
> Would you mind directing me to the posts where you explain your
> version of the development of *tt etc. in IE? (Or just summarizing
> it again?) I would love to see an alternative view. I've never
> understood why *tst should become /ss/ (in Gmc, Latin), or why an
> /s/ should be inserted.
Did you see this one?
It contains the basic idea
http://tech.groups.yahoo.com/group/cybalist/message/63349
There's a longer discussion in the archives, but I can't think of a
good search word.
The idea is connected to an idea I had that perhaps both alternatives
in the two shift-paradigms (kentum/satem and decem/taihun) were
present already in PIE as allophones in paradigms (cf. Polish or
Russian) and that kentum and satem languages got that way because they
regularized their paradigms with /k/ and /s/, respectively
(decem/taihun similarly) and by hypercorrection (shibboleth) the
preferred allophone spread also outside the paradigms.
Here is how I tried to experiment with how the allophones would look
in PIE.
http://www.angelfire.com/rant/tgpedersen/PIEstops/PIEstopsCurrent.html
Please ignore the gunk at the bottom. I can't seem to get rid of it.
As mentioned, some languages would go -þt- > -st-
As for -tt- > -ss- (typically ppp's):
PPP's end in -tó-, it is therefore a thematic adjective.
Sturtevant thought the tómos/tomós noun/adjective pairs came into
being as the adjective developed out of the end-stressed genitive (in
order that the genitive, now as adjective, could be inflected along
with the noun it governed, to mark they belonged together. The tómos
noun came into being now by generalizing root-stress for contrast to
the adjective. In other words, the tóms/tomós pair started as a single
non-thematic noun, nom. tom-s, gen. tom-ós. In the same manner ppp's,
in -tó-, started as non-thematic ´-t-s, -tó-, of which the former
became those Sanskr. nouns with agent suffix -t-, Latin sacer-do(t)s,
the latter became the classic ppp. Now, if by the above rule a root in
-t had gone
-þt- > -st-, then it would have
nom. ´-s-t-s, gen. -s-t-ós, becoming
nom. ´-s-s, gen. -s-t-ós
and the various languages now generalized the stem of one of the two
cases (-ss- or -st-) in their ppp paradigm
(cf. Latin o(s)s, Greek osteon).
Torsten