Sir

We welcome the tone of Gottfried and Wilson's assessment of what they (not we) call ‘Edinburgh school’ sociology of scientific knowledge1. Their eschewal of the ad hominem attacks that have characterized too much of the ‘Science Wars’ debate is refreshing.

Three points of clarification are, however, in order. First, the goal of the sociology of knowledge, in our view, is the explanation of belief, not its evaluation. ‘Symmetrical’ analysis of the emergence, development and acceptance or rejection of bodies of knowledge does not involve the (absurd) claim that all beliefs are of equal merit. Rather, ‘symmetry’ implies that current evaluation of the truth or falsity of beliefs should not bias the empirical study of the processes through which knowledge develops. For the historian or sociologist studying nineteenth-century evolutionism, for example, both Darwinism and anti-Darwinism stand equally in need of explanation.

Our second point is a simple corollary of the first. Predictive success is, of course, a vital measure of the merit of bodies of knowledge, and an important cause of their acceptance or rejection. But later predictive success cannot be appealed to as a cause of earlier acceptance.

Third, we have always held that satisfactory analysis of the development of knowledge must be multicausal. Causal input from the real world and psychological and social processes are all involved. Scientific knowledge is not a collective fantasy devoid of relation to the real world, but neither is it a simple mirror of reality. The sociology of scientific knowledge stands or falls by the value of the best, empirical, case-studies of these complex questions. We are glad that these case-studies are attracting the attention, however sceptical, of distinguished natural scientists. We benefit from their scrutiny; perhaps they may learn something in return.