A central issue in the philosophy of the social sciences is the possibility of naturalism: whether disciplines such as sociology, anthropology, economics and psychology can be ‘scientific’ in broadly the same sense in which this term is applied to physics, chemistry, biology and so on. In the long history of debates about this issue, both naturalists and anti-naturalists have tended to accept a particular view of the natural sciences – the ‘positivist’ conception of science. But the challenges to this previously dominant position in the philosophy of science from around the 1960s made this shared assumption increasingly problematic. It was no longer clear what would be implied by the naturalist requirement that the social sciences should be modelled on the natural sciences. It also became necessary to reconsider the arguments previously employed by anti-naturalists, to see whether these held only on the assumption of a positivist conception of science. If so, a non-positivist naturalism might be defended: a methodological unity of the social and natural sciences based on some alternative to positivism. That this is possible has been argued by scientific realists in the social sciences, drawing on a particular alternative to positivism: the realist conception of science developed in the 1970s by Harré and others.
The critical realist conception of open and closed systems