Something's been puzzling me:
The absolute chronology on p. 3 of the reference M. Kelkar was nice
enough to provide may be controversial, but the relative chronology
seems middle of the road. Now how come, a few nodes down the tree
Indo-Iranian splits off, satem, with 2nd palatalization, RUKI; then
Balto-Slavic splits off, satem, with 2nd palatalization, RUKI;
and the remainder Celtic, Germanic, Italic are all
kentum, no secondary palatalization, no RUKI?
Why did Indo-Iranian and Balto-Slavic follow the same path?
Personally I think this means that IE, at least at the node above the
split-off of Indo-Iranmian, must have been a language in which both
kentum and satem forms occurred within the same paradigm, depending
the following ablauting vowel and that satem and kentum languages
arose as a result of generalizing satem or kentum forms. The
alternative is to assume Indo-Iranian and Balto-Slavic just happened
by accident to follow the same path, which is too weird.
This is how I imagined velars varied in PIE
http://www.angelfire.com/rant/tgpedersen/PIEstops/PIEstops.html
(D voiced stop, Dh voiced aspirated stop, T unvoiced stop)
which is probably full of errors, but it's a start.
So
1) what does cybalist think of this solution, and
2) if no, what's the alternative explanation of the parallel
development of Indo-Iranian and Balto-Slavic?
Torsten