To paraphrase Marx, sociologists have only interpreted science; the point is to improve it. The Rational and the Social attempts both. It begins by sketching recent sociological approaches to science, notably the strong programme – Bloor's 'science of science' and Barnes's 'finitism' – and that of the 'anthropologists in the lab', Collins and Latour and Woolgar. The author argues that although sociological accounts are valuable in many respects, when morals are drawn about the structure and epistemology of science, they are badly flawed. In rejecting the sociological theory of science, it is not necessary to conclude that science develops without reference to the social. James Robert Brown argues for an alternative account. He proposes a novel way of viewing the history of science as a source of evidence for how to do good science and argues that the most important aspect of methodology is that it is comparative. Rival theories are evaluated by comparison and the contribution of the social to this process is inevitable and should be acknowledged. This is the challenge to science.