Diffusion of Sanskrit

[joe at sfbooks.com]

Vidyanath Rao recently wrote, quoting me:

>> [...] in particular (Meadow 1991 below).  In the chapter which John
>> Lukacs worked on, there is an unequivocal assertion of significant
>> population changes in the centuries prior to 800 B.C.
>
>In `Indo-Aryans of Ancient South Asia', K. A. R. Kennedy writes (pp 49--54)
>``Evidence of demographic discontinuities is present in our study, but the
>first occurs between 6000 and 4500 B.C (a seperation between the Neolithic
>and Chalcolithic populations of Mehrgarh) and the second is >after 800 B.C.,
>the discontinuity being between the peoples of Harappa, Chalcolithic Mehrgarh
>and post-Harrapan Timargarha on the one hand and the late Bronze Age and
>early Iron Age inhabitants of Sarai Khola on the other. ''

I'm curious what the date is on the Kennedy reference.  This is partly
because calibration of radiocarbon dates in 1980s publications re South
Asian archaeology could be wildly inconsistent (one of the best
archaeologists in India at that time, Makkhan Lal, once spent an entire
paper mercilessly showing up opponents in a controversy by demonstrating
inconsistencies in their use of radiocarbon).  While Kennedy is unlikely to
be remotely as foolish as Lal's hapless opponents, there's a specific
problem with 800 BC in particular, which is that very close to that date,
the amount of C14 in the atmosphere started dropping.  Ordinary radiocarbon
dates for the period 800 - 400 BC, or slightly better ones for the periods
800 - 550 or 550 - 400, have highly unpredictable uncalibrated values.  I'd
need to know how the dates in Kennedy's paper were being calibrated to know
whether this effect (only demonstrated during the 1980s) was being properly
offset.

This is a very long way of saying that I agree wholeheartedly with your
fundamental point in this post, but in doing so, may well upset your
specific example.  "After" and "before" 800 BC can mean very similar things
depending on which calibration curve is being used!

One of my biggest uncertainties about the results of Lukacs et al in the
Harappa site report is precisely that I don't know what curve *they* were
using, though I'm inclined to assume it was a current one.

>This is the problem I have with much of the discussions on this topic.
>Whom do I believe? Did the population change (actually change in the
>cranioetric data) occur before or after 800 BC? (now add the fact that
>change in cranioetric data can occur due to causes other than population
>change).
>
>When I take the trouble to check some of the ``data'' presented, I come
>across troubling inconsistencies which the debaters want to ignore
>and fail to answer the questions. (which is why I am not participating
>in sci.archaeology debate. Usenet is awful in this regard.)

I'm going to have to come back to this matter of Usenet, because I'm
becoming convinced it's *precisely* the best place for debate on this
issue.  But I feel compelled to note, however ironically, that I've got
record of your participating fairly extensively (and intelligently) in just
that debate.  I'm glad you changed your mind.

Joe Bernstein