A Realist Philosophy of Social Science, ExpIanation and Understanding (review)

This book, which complements the author's superb History and Philosophy of the Social Sciences (1987), is a beautifully clear and readable defense of realist philosophy of science, as developed by Rom Harré, Roy Bhaskar and others; this approach has been extremely influential, I believe deservedly, in social science in the United Kingdom, Scandinavia and elsewhere. Realism understood in this common-sense way, according to which science is about describing the structures and mechanisms of the world, offers the social sciences the prospect of overcoming the opposition between interpretive understanding and causal explanation which has dogged social science since its beginnings in the late 19^th and early 20^th centuries.

The realist strategy is to reconceptualize explanation – not in terms of general laws ("covering laws" in the language of mid-20^th century positivism), but in terms of the powers and tendencies of things, including people. To describe my curtains as inflammable or my polemical writing as inflammatory is to point to a tendency or liability which may or may not be realized, depending on other conditions of ignition or readership. Social scientific laws, which had been justifiably accused of being either trivially true or subject to innumerable exceptions, could be rehabilitated as statements of tendencies which might or might not be fulfilled. This conception also brought together the explanation of human action and of other processes: the reasons for quitting smoking, in the face of addiction and habit, could be incorporated into a causal explanation of successful or unsuccessful attempts to do so.

The originality of this book is that Manicas draws out a little-noticed implication of this approach, namely that the very relation between explanation, often presented as the holy grail of the social sciences, and understanding needs to be rethought, and that explanation has been over rated. Many explanations in the social sciences, as in natural science, take the form of postulating possible mechanisms to explain observed or presumed facts, such as the tendency of objects to fall to the ground, of Protestants to be economically innovative in early modern Europe, or of Protestants to be more suicidal than Catholics. To understand such effects has little to do with testability and prediction in the Laplacean sense and much to do with enriching our understanding of the kinds of mechanisms in play in complex processes. As Collingwood (53) states in The Idea of History, "When [the historian] knows what happened, he already knows why it happened." Understanding social processes as processes of a particular type is also a form of explanation; the terms "understanding" and "explanation" are often interchangeable, depending on the context.(123).

The central part of the book is concerned with working out the implications of this approach, notably for the relations between agents and social structures. Micro-social mechanisms can often provide the foundations of larger-scale processes. The relation between history and the (other) social sciences is also much closer than has often been assumed. Chapter 5 and two of the substantial appendixes are concerned with these issues, as raised by the superb work of people like Barrington Moore, Charles Tilly and Theda Skocpol working on the interface between history and sociology. For me, as a long-standing realist who shares Manicas' scepticism about the cult of explanation, these sections of the book were the most stimulating. Further appendixes are devoted to the limitations of multiple regression and neo-classical economics. The final substantive chapter is devoted to theories of markets as social mechanisms, from Adam Smith in the 18^th century to Charles Smith in the 20^th and 21^st.

The basic message of the book is summarized by Manicas in a quotation from a 1996 paper by William Sewell which ends Appendix B (Comparison, Mill's methods and narrative):

"Sociology's epic quest for social laws is illusory, whether the search is for timeless truth about all societies, ineluctable trends of more limited historical epochs, or inductively derived laws of certain classes...