Truth and method in Indology, III

Jan E.M. Houben JHOUBEN at RULLET.LEIDENUNIV.NL
Sun May 31 21:01:05 UTC 1998


More than a month ago, Fri 20 Mar 1998  Edwin Bryant wrote:

>You are right that no serious scholar talks of invasions anymore (although
>see Allchin as late as 1993 in Possehl's "Harappan Civilization").
>However, we should be aware that many people (including scholars in many
>universities in India) do not have access to state-of-the-art material
[ . . . ]

This is a sympathetic position, but:
If Indian universities and institutions make serious efforts to get the latest
information and get trained in the best methods in medical and technical
disciplines and establish international exchanges etc. in order to acquire the
ability to make reliable decisions in these fields
(and to make atom bombs that WORK . . . ),
why should much lower standards be accepted if someone claims to make reliable
statements about the cultural and linguistic past of South Asia ? After all,
this past contains a rich 'cultural capital' (to speak with Bourdieu), and its
own potentially explosive material.

The quality of statements in this area depend on the quality of the methods and
that of the data, and the quality of the latter again depends on methods. Let
Indian scholars hold dear their cultural past, and let them challenge accepted
hypotheses: if their methods have become solid by international training and
exchange (one may think of e.g. Sukthankar and Dandekar having been partly
trained in Germany, Subrahmanya Iyer having studied in Paris, etc.) the results
of their research are on the long term more valuable for both Indian and
international scholarship. It is said that a rice-plant grows better if
replanted in another bed: something similar could very well apply to
academicians, including philologists and archeologists.

These are just some reflections on: how to proceed in a better way?

Greetings, JH





More information about the INDOLOGY mailing list