Horse, Indo-Iranians, nomads and origins?

Jo Kirkpatrick jkirk at MICRON.NET
Mon Mar 23 12:46:20 EST 1998


May I humbly inquire, only because there seems to be some illogic going
on here: considering nomadic cultures, how can one postulate a "more
compact source of origin"? I assume one can hypothesize a geographical
area which is traversed by a nomadic culture, but if that area is
extremely large, a vast expanse of teritory,  then what?  How does one
identify in this case a "compact source of origin", especially since
nomads leave a less concentrated archaeological record of their presence
than do settled people.

Cautiously yours,
J. Kirkpatrick

*********************
Edwin Bryant wrote:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>{cut]
 Kenoyer argues that nomads were constantly
criss-crossing this area to and fro resulting in a common language
family.
While this might work archaeologically, and is convenient in so far as
it bypasses the need for refuting either an 'into' or 'out of' India
model I suspect that most linguists would say that a language family
needs a more compact source of origin for its members to develop
significant shared morphological and lexical features (based on the
geographical history of other language families that are presently
known).   Regards,
> Edwin



More information about the INDOLOGY mailing list