ABSTRACT
I start from the assumption, which I had better make explicit, that the problem posed by the title of this paper is peculiar to the social, as opposed to the physical and biological, sciences. There is a sense in which these can, if you wish, be said to have a problem of ideology too. If, in the manner of T. S. Kuhn or, before Kuhn, of N. R. Hanson, 1 we think of a scientific theory less as a set of connected laws and more as one way among others of looking at the world, then the choice between rival theories is perhaps a little more like a choice between social and political philosophies than used to be supposed by positivist philosophies of science. But apart from the detailed criticism to which views of this type have themselves been subjected, 2 I take it for granted in any case that there is a difference between a set of connected scientific laws and a Weltanschauung; that scientific theories, however haphazard the process by which they are arrived at and however provisional their status at any given time, stand or fall by being publicly tested against potentially disconfirming evidence; and that scientific progress can roughly be described as a cumulative demonstration that specified sets of operationally definable terms are co-extensive. The construction of wide-ranging and well-tested theories is, to be sure, the common goal of the social sciences too. But the question which this still leaves open is whether the social sciences are at the same time ‘ideological’ in some way which makes them incompletely ‘scientific’ in the conventional sense.
