On the 'Hows' and 'Whys' of Cultural Change (Response to Woolgar)

It is little more than a decade since the possibility of a thoroughgoing sociological analysis of scientific culture once more began to be taken seriously. But since that time developments have occurred with breathtaking speed. The view that scientific culture is constructed like any other is now well elaborated and exemplified. It is beginning to enter the third stage of William James' well-known theory of scientific research programmes: first it was denounced as absurd; then it was disdained as true, but only trivially-true; now it is recognized as so important that its opponents are starting to say that they themselves discovered it.1 The significant question today is how precisely the sociological study of all the various forms of culture should proceed: there is no longer any need to ask whether, in the case of science, it should proceed at all. The many disputes and differences of view in the sociology of science are now, for better or worse, simply those current in sociology generally. Fortunately, despite these disputes and differences there is a certain amount of common ground. Nobody actively involved in the sociology of knowledge seems to accept either rationalism or cultural determinism. It is everywhere agreed that cultural change can never be sufficiently accounted for as the product of 'reason' and 'experience' alone. And it is everywhere denied that the proper mode of use of a cultural artefact can be fixed in advance that future use of a concept is ever fixed by its 'meaning'; that future applications of a rule are ever inherent in the rule; that verbally expressed values or norms ever formally 'imply' this or that specific action; and so on. People, it is agreed, are not under the control of their discourse or their own verbal artefacts: the relationship is the other way round. Let me illustrate the extent of this common ground with the case of the next instance of correct application of a concept. The problem is what makes the next instance of a bird, a bird properly so called, or the next mental-patient, a mentalpatient, or the next electron, an electron? In general, for instance X and concept A, what fixes X as properly either A or not-A? Or, for conventionally-mutually-