Robert C. Bishop: The Philosophy of the Social Sciences: An Introduction.

Robert C. Bishop

The Philosophy of the Social Sciences: An Introduction.

New York: Continuum 2007.

Pp. 416.

US$120.00 (cloth ISBN-13: 978-0-8264-8952-4); Cdn$44.95/US$29.95 (paper ISBN-13: 978-0-8264-8953-1).

Jon Elster

Explaining Social Behaviour: More Nuts and Bolts for the Social Sciences.

New York: Cambridge University Press. 2007.

Pp. 493.

US$85.00 (cloth ISBN-13: 978-0-521-77179-5); US$27.99 (paper ISBN-13: 978-0-521-77744-5).

Why do you refrain from picking your nose when a train is passing by, despite the fact that no one else is on the platform and you know no one on the train, and will see none of them again? Why do we tip taxi drivers in one-off encounters, knowing we'll never see them again? Such prima facie puzzling behavior is the subject matter of the social sciences. Many deep philosophical issues (e.g., concerning rationality, action, beliefs, desires, etc.) are raised in trying to answer even such seemingly silly and trivial puzzles as those just posed.

Elster and Bishop have written two very different books on the social sciences--in terms of their approach, at least. Bishop's is more obviously philosophical than Elster's (the latter's book contains the charming examples I began with). Bishop's is a textbook. Elster's is more akin to a manifesto, perhaps even a credo--an 'Elster manual' if you like! In fact, Elster's is not really a philosophy book at all, though it steps on the toes of many philosophical issues. He defends, through a series of sparkling examples, the view that social phenomena are reducible to individual psychology. Bishop defends a similarly subjectivist view according to which one cannot gain a value-neutral 'outsider's perspective free from all ethical commitments' in the social sciences (356). They are not so very different, then, in terms of their agenda. Bishop is by far the more radical and ambitious, though: he seeks to break down the barrier between individualism and holism. Elster retains this barrier and firmly ensconces himself on the individualist side.

Elster has written many books on social phenomena and behavior in a variety of contexts. He pioneered the use of game theory and other concepts and tools from economic theory to political science, for example. His work is characterized throughout by a devout adherence to methodological individualism: the view that social phenomena are explainable only by reference to the behaviour of individuals (their psychological states, for example). Rational-choice theory is another mainstay of Elster's work. The two frameworks are not unconnected, of course: rational choice theory is often thought to be underpinned by methodological individualism. His work is also characterized by an adept marshaling of classic literature to make contemporary points. All of these aspects feature heavily in the present book; though he is here, as he admits (given the very many examples of irrational social behavior), rather more skeptical of rational-choice explanations of action (164).

The book is in five parts. In the first part Elster presents his own view on explanation and mechanisms. The subsequent parts apply this view to various puzzling cases: 'The Mind' (with chapters covering motivations, self-interest and altruism, myopia and foresight, beliefs, and emotions); 'Action' (caused by the desires and beliefs of agents--an analysis based on rational choice theory and problem cases for rational choice); 'Lessons from the Natural Sciences'; and (Social) 'Interactions' (between agents, based on game theory). The mind is clearly fundamental to action and the interaction of people, and so to social behavior: Elster unpacks the mind using 'introspection and folk psychology' (67). This 'methodology' is called upon throughout the book.

Explanations of the various ideas Elster is trying to get across are often carried out by utilizing classic literature--Montaigne, Proust, and Tocqueville figure highly--and proverbs, i.e., folksy wisdom. This is certainly enjoyable and widens one's knowledge. There are indeed many truths to be found in the works of the great authors; that is to be expected--and is surely why they are great!--but nice quotations, however apposite, are no substitute for real argument. Of course, that's Elster's point: this is a 'nuts and bolts' account, building up social behavior in whatever way is possible. It is not a formal treatise. The formalists will, therefore, not find much to whet their appetites here: however, they will find some exceptionally clear examples on which to hang their beloved formalism.

Given his individualism, Elster is against functionalist explanation in the social sciences, i.e., explaining social phenomena by appealing to their outcomes or consequences (1, 5). Elster also claims--not argues: a recurring practice!--that every explanation must specify a causal mechanism, i.e. some causal regularity such as one billiard ball hitting a stationary one and the stationary one's moving off. The discussion omits vast swathes of recent work on causation and explanation. What of accounts of explanation that involve unification or the best balance of systematization and simplicity? These can surely play a role in the social sciences? There will be some explanations that cohere better with well-established knowledge. In this sense, Darwinian-type explanations of social phenomena provide a promising account; more so, perhaps, than Elster's causal/mechanistic account. The treatment of causation is even worse: it simply assumes a regularity account, with no defense or mention of other possibilities. Yet this account plays a central role in his notion of mechanism.

Also absent is a considerable amount of recent work on social behavior: Steve Durlauf and William Brock's 'social interactions models' for example. There has also been much work, including work by Kenneth Arrow, on the modeling of social phenomena under the assumption of zero rationality. This is anathema to Elster, yet it fits the stylized social facts very well, and thus surely demands some discussion. The omissions are, unfortunately, relevant and numerous. Alternatives that aren't omitted are briskly dismissed, with virtually no reasons. He dismisses a dynamical systems model of an arms race with no real argument whatsoever (460).

What bothers me most about Elster's book is the insistence that social science cannot explain or predict. It is perfectly true that individual behavior is a problem. There is, as Elster points out, 'complexity and instability' in human behavior (467). This is true of most systems involving large numbers of interacting parts. But the aggregate behavior of such systems is often predictable, in a coarse-grained sense. One can at least often predict distributions of values of properties over the individuals (just not which individual gets which value). Elster argues that even if this is possible, it does not deliver any explanatory gain for it neglects 'context'. The aggregation necessarily involves abstracting such details as the desires and beliefs of individuals. For Elster, of course, these are the true explanatory factors. But this begs the question. If we want to explain social behavior, then a possible strategy is to model the social system as an individual in its own right (a complex system), with its own properties. Elster gives no argument to the contrary.

Whereas Elster adopts a hard-line methodological individualism, and eschews holism, Bishop is suspicious of such clean distinctions construed as mutually exclusive and exhaustive positions: a little of both is Bishop's conclusion. The book is organized into four parts. The first part is, naturally, background material. Since this is intended for social scientists and philosophers, key ideas from both fields are introduced in the first part. Part 2 focuses on value-ladenness in social research, and finds it in a wide variety of situations. This is then (in Part 3) applied to a number of cases where decision and action are involved: psychology, rational choice, political science, and economics. Part 4 deals with some traditional philosophy of science issues using the earlier material.

Bishop's book is a very well executed textbook, perfectly suited for undergraduate courses or, if supplemented with additional material, graduate courses. It is geared towards students all the way through--i.e., rather than being a textbook written to impress one's colleagues! There are useful 'further study' questions at the end of each chapter that are both perfect for self-study and for use by instructors (to base tutorials around, for example). Recommended readings, pitched at the right level, also follow each chapter (though there should have been more and varied suggestions, in my opinion). It describes itself as 'the definitive companion', and I think it really is. But it does have its problems.

Like Elster's book, it misses much that is of contemporary relevance. I suppose Bishop has a possible excuse: the desire for clarity and coherence. The contemporary debates require a good grounding in the classic foundational problems, and that is just what Bishop gives. By no means does he sit on the fence though: this is not an impartial introduction. There is an almost evangelical zeal running through the chapters, giving the book a coherence not often found in textbooks. The glue is a belief that the attempt to force social phenomena into the mould of the natural sciences creates problems: the findings of social sciences are inextricably value-laden. Bishop argues that social science should reject the natural science template and the models of explanation and prediction that are associated with it and forge its own style. However, at times I would have preferred it if he had kept the book more agenda-free.

Like Elster, Bishop does not think that the social sciences are like the natural sciences. He does not think that the methods and theories of natural science should be applied to social phenomena: such a move serves to 'distort the social world we are trying to understand' (8). Bishop takes one of the central philosophical problems of social science to be the question, 'Are social sciences and their subject matter different as a matter of degree or as a matter of kind?' (27). A major part of the problem I have with both books is the lumping together of all social sciences as if they face identical problems and ought to be dealt with in the same way. I see no reason to assume this is so. Economic behavior might well demand a wholly different conceptual and methodological basis than psychology or anthropology. Indeed, there do seem to me to be essential differences between these: economics (financial economics, certainly) involves vast numbers of interacting agents, while anthropology does not. Given the numbers of interacting agents one can make a case that financial economics is closer (in some sense) to (statistical) physics than anthropology is. One can then test this hypothesis (the statistical physics analogy) by inspecting the time series data and looking for such things as positive auto-correlation and non-Gaussian distributions of changes. The case against the use of natural science methods in social science has not been made in either book.

To sum up, as background reading on the concepts and methods of social science for philosophy undergraduates, Elster's book would make a good first choice: it contains many interesting puzzles and examples, and excellent elementary discussions of the major concepts of the social sciences. (Indeed, I think lecturers on a variety of philosophical topics, such as political philosophy and ethics, would find the book a treasure trove of suitable and interesting case-studies and examples.) It is certainly a book that gets you thinking hard, which has to be a good thing. As a philosophical text, however, it is sadly lacking and should only be used as a supplement to a more pedagogically responsible text like Bishop's. I shall certainly use Bishop's book for courses on the philosophy of social sciences. However, though a thoroughly superb textbook, given its lopsidedness, Bishop's book would best be used alongside a book defending an opposing view, such as Harold Kincaid's Philosophical Foundations of the Social Sciences (Cambridge 1995).

Dean Rickles

University of Sydney

Article Details

Printer friendly Cite/link Email Feedback
Author:	Rickles, Dean
Publication:	Philosophy in Review
Date:	Jun 1, 2008
Words:	1929
Previous Article:	Francisco Benzoni: Ecological Ethics and the Human Soul: Aquinas, Whitehead and the Metaphysics of Value.
Next Article:	Andrew Brook and Kathleen Akins, eds.: Cognition and the Brain: The Philosophy and Neuroscience Movement.