Paper short abstract:

Exploring traditional knowledge among Santals tribals and Tulu castes, I show how it is produced and sustained in the context of social situations. These transactions in knowledge allow remodeling of the corpus of knowledge, its coherence determined by cultural criteria for validity.

Paper long abstract:

A comparative perspective on human knowledge allows us to study a number of cultural worlds which people construct. Following F. Barth (2000) I will take as my point of departure that 'knowledge always has three faces: a substantive corpus of assertions, a range of media of representations and a social organization'.

Exploring medical and religious traditions of knowledge among the Santals, a 'tribal society' of Central India, and among a variety of Tulu-speaking castes of Karnatala, I shall show how these bodies of knowledge are produced in the vontext of the social situations they sustain. I shall argue that some of these social situations, such as therapeutic rituals, or religious and literary movements, stress the pattern of action that unfolds. These transactions in knowledge allow the remodeling of the corpus of knowledge through the agency of the priests and medicine-men whose performance is always negotiated. In these two regions, the coherence of medical and religious knowledge is determined by a variety of circumstances that generate the criteria for validity that govern knowledge. But knowledge is also affected by the constraints that arise from the agency in which knowledge is being cast, the medicine-man, the priest or the medium who are in a position of power. In these local traditions, the agents who convey the criteria of validity that govern knowledge are also criticized, as the systems of knowledge allow new forms of agency to emerge through cultural contacts and change.

We shall see, however, that while the Santals allow change in knowledge they stress conyinuity as a central value, essentializing Tradition, while the Tulus allow knowledge to benefit from hybrid contacts without curbing unexpected innovations. This difference might be related to the ritual framework which, in Tulunadu, is the main factor defining truth. From these two examples, I shall point to factors which make some forms of knowledge more 'catching' than others, as Sperber (1996) puts it.