ABSTRACT
To begin, we need a baseline, which, in the history of science, is the genre known as history of ideas. In its canonical form this tries to understand scientific ideas as evolving under their own inner logic, though much mileage has recently been got by sociologizing and contextualizing the picture. Thus the sociology of scientific knowledge approach has sought to display the social interests that sustain the particular bodies of knowledge that one finds associated with particular groups; and, less theoretically, the historiographic avant garde would probably agree that the best history of ideas is contextualist, relating specific ideas to their historical, cultural, social, etc., contexts. The point I want to stress, though, is that all these variants share the view that science is, above all else, a knowledgeproducing enterprise; and that we should therefore put scientific knowledge at the centre of our historical accounts and arrange everything else around it.2 The first displacement I want to recommend is, then, to do away with this obsession with knowledge. A strange suggestion, I know, but one that might seem less strange if I put forward an alternative.
