> >If Hittite is not as old as thought, that would make Indo-Iranian
> >perhaps the oldest language ...
This needs to be kept firmly in perspective. Hittite evidence mostly
overlaps with the earliest Greek evidence, although there is some Hittite
material from a few hundred years earlier. Indo-Iranian material is much
harder to date, for various reasons, but the best guesses seem to be that
both Avestan and Vedic overlap with the earliest Greek material as well.
This means that the three language groups should be thought of as rough
contemporaries, and we are wrong if we think that Hittite reflects an
earlier stage simply because it is attested from a slightly earlier period.
Any argument that Hittite (or Indo-Iranian) reflects an earlier stage of IE
must be based firmly on linguistic evidence, not the date of attestation, as
archaeology and literary analysis can only suggest the relatively trivial
difference of a few hundred years. So we must think of PIE producing two
parallel daughter groups: the Anatolian group with a fairly simple verb
system & the I-I / Greek group with a more complex verb system, the seeds of
which can also be found in other IE languages.
One aspect of the linguistic evidence is that the I-I / Greek group is
highly innovative. It should therefore not be a surprise if other groups
(such as Anatolian or Germanic) do not show this highly developed verb
system. The only surprise is that whereas the seeds of this system can be
traced in most other IE languages, there is no hint of some features in
Anatolian. But it is not clear if this is total loss (as opposed to partial
loss such as we find in Germanic or Celtic or Baltic), or original absence.
Peter