Each of the sciences, the physical, biological, social and behavioural, have emerged from philosophy in a process that began in the time of Euclid and Plato. These sciences have left a legacy to philosophy of problems that they have been unable to deal with, either as nascent or as mature disciplines. Some of these problems are common to all sciences, some restricted to one of the four general divisions mentioned above, and some of these philosophical problems bear on only one or another of the special sciences.
If the natural sciences have been of concern to philosophers longer than the social sciences, this is simply because the former are older disciplines. It is only in the last century that the social sciences have emerged as distinct subjects in their currently recognizable state. Some of the problems in the philosophy of social science are older than these disciplines, in part because these problems have their origins in nineteenth-century philosophy of history. Of course the full flowering of the philosophy of science dates from the emergence of the logical positivists in the 1920s. Although the logical positivists’ philosophy of science has often been accused of being satisfied with a one-sided diet of physics, in fact their interest in the social sciences was at least as great as their interest in physical science. Indeed, as the pre-eminent arena for the application of prescriptions drawn from the study of physics, social science always held a place of special importance for philosophers of science.
Even those who reject the role of prescription from the philosophy of physics, cannot deny the relevance of epistemology and metaphysics for the social sciences. Scientific change may be the result of many factors, only some of them cognitive. However, scientific advance is driven by the interaction of data and theory. Data controls the theories we adopt and the direction in which we refine them. Theory directs and constrains both the sort of experiments that are done to collect data and the apparatus with which they are undertaken: research design is driven by theory, and so is methodological prescription. But what drives research design in disciplines that are only in their infancy, or in which for some other reason, there is a theoretical vacuum? In the absence of theory how does the scientist decide on what the discipline is trying to explain, what its standards of explanatory adequacy are, and what counts as the data that will help decide between theories? In such cases there are only two things scientists have to go on: successful theories and methods in other disciplines which are thought to be relevant to the nascent discipline, and the epistemology and metaphysics which underwrites the relevance of these theories and methods. This makes philosophy of special importance to the social sciences. The role of philosophy in guiding research in a theoretical vacuum makes the most fundamental question of the philosophy of science whether the social sciences can, do, or should employ to a greater or lesser degree the same methods as those of the natural sciences? Note that this question presupposes that we have already accurately identified the methods of natural science. If we have not yet done so, the question becomes largely academic. For many philosophers of social science the question of what the methods of natural science are was long answered by the logical positivist philosophy of physical science. And the increasing adoption of such methods by empirical, mathematical, and experimental social scientists raised a second central question for philosophers: why had these methods so apparently successful in natural science been apparently far less successful when self-consciously adapted to the research agendas of the several social sciences?
One traditional answer begins with the assumption that human behaviour or action and its consequences are simply not amenable to scientific study, because they are the results of free will, or less radically, because the significant kinds or categories into which social events must be classed are unique in a way that makes non-trivial general theories about them impossible. These answers immediately raise some of the most difficult problems of metaphysics and epistemology: the nature of the mind, the thesis of determinism, and the analysis of causation. Even less radical explanations for the differences between social and natural sciences raise these fundamental questions of philosophy.
Once the consensus on the adequacy of a positivist philosophy of natural science gave way in the late 1960s, these central questions of the philosophy of social science became far more difficult ones to answer. Not only was the benchmark of what counts as science lost, but the measure of progress became so obscure that it was no longer uncontroversial to claim that the social sciences’ rate of progress was any different from that of natural science.
Constructive Methods in Economics