Tuesday, 7 October 2003
Partial Knowledge
Daniel Andler, Université de Paris-Sorbonne (Paris IV)
& Ecole normale supérieure
12:05 pm, 817R Cathedral of Learning

Abstract: The knowledge which any given agent, community, or subagency, possesses at any given time is typically partial or incomplete, or so it appears in the usual situations where that knowledge is notionally immersible, possibly at the cost of some changes, in at least one larger or richer body. But is partial knowledge genuine knowledge? On the one hand, it seems that it had better be, if it is in fact pretty much all we've got, except perhaps in special cases. On the other hand, partiality may well affect the robustness of knowledge as traditionally construed. Nor is this a merely theoretical worry: some forms of relativism, for example, feed off the apparent brittleness of partial knowledge. Several further problems arise, in connection with the distributed character of scientific knowledge, commensurability of subspecialties, growth of scientific knowledge, etc. Also, one may wonder why partial knowledge fails to regularly lead to disaster, practical or theoretical, when it is applied or deployed. To recover some realistic and reasonable notion of working knowledge, we may need to reconsider some tacitly held models of partial knowledge and its relation to complete, or less partial, knowledge. One aim of the paper is to suggest that we should pay more attention to the partiality of knowledge and seek better models of the above relation. Another aim is to suggest that a promising place to look is the matrix of common skills which allows situated partial knowledge to be deployed. By focusing on partiality, we may eventually be led to view knowledge in a new light.