--- In
cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
> Another problem is that this presupposes reduplication in
> the stative, but the Hittite hi-conjugation (and isolated
> forms like *wóid-h2a) lacks it. How to explain the
> o/e-Ablaut in Hittite (or even outside Hittite: e.g. Lith.
> málti vs. Slav *melti)?
I think this is the core of the matter, so if other theories can
account for that they should be better. I think this reflects the IE
intensive, the Vedic type várvarti, várvrtati from IE *wr-wórt-ti,
*wér-wrt-nti. From *melH- 'grind' this gives *ml-mólH-ti, *mél-mlH-
nti. The Balto-Slavic continuations *ma:l-/*me:l- will be the direct
reflexes of the accented part of each allomorph, producing exactly
the paradigm *mólH-/*mélH- postulated by Jasanoff. What it takes to
derive that paradigm from a motivatable intensive structure is
dereduplication doing away with the unaccented part of each
alternant, i.e. deletion of the reduplication itself where the root
has full grade and original lack of accent can be sensibly assumed
for the reduplication, and deletion of the repetition of the root
material where the root is in the zero-grade and the accent was on
the reduplication. The process will be similar to that of Vedic
desideratives like bhíks.ate and s'íks.ati from *bhi-bhg-H1se/o- and
*k^i-k^k-H1se/o- which look as if the roots bhaj- and s'ak- have
just changed their root vowels to -i-.
It may be a matter of taste whether one considers such a development
possible for the hi-conjugation. I for one find it unsatisfactory
that a complete rewriting of the basic concepts of IE verbal (and
other) morphology is based on something which is only a matter of
taste. It takes more to get me to join the bandwagon. Conversely, if
common opinion was already Jasanoff-style, and a novel theory
introduced the reduplicated perfect and the s-aorist as fully
developed IE categories I believe I would have immediately switched
to that.
Jens