K. Brad Wray’s new book is an excellent overview of the scientific realism debate, as well as a development of the state-of-the-art. Wray, whose views seem most strongly influenced by Bas van Fraassen and Thomas Kuhn, develops crucial aspects of the debate, such as the argument from underconsideration and the ability of anti-realism to explain the success of science. This book is clearly written, tightly argued, and well researched. I recommend it highly to all philosophers and students of philosophy interested in scientific theory change or the scientific realism debate.
Resisting Scientific Realism is divided into two parts. In Part 1, ‘Against Realism’, Wray reviews a number of key arguments against scientific realism, including the arguments from underdetermination and underconsideration, as well as several versions of the pessimistic induction. This involves evaluating realist replies to these anti-realist arguments, and so there are chapters directed against Ludwig Fahrbach’s (, ) and Michael Devitt’s () rejoinders to the pessimistic induction, and against realist appeals to inference to the best explanation and the epistemic privilege of current scientists.
Part 1 begins with a chapter dedicated to the history of the Copernican Revolution. Later chapters of the book refer back to this history, which, along with the history of the periodic table presented in Chapter 7, serve as the main case studies supporting the claims of the book. The historical elements of the Copernican Revolution do not go much beyond Kuhn’s () book on the same subject, but the history is put to good use as an illustration of some of the key anti-realist claims of the book, for example, the assertion that false elements of rejected theories were often instrumental to their predictive success.
Wray proceeds to evaluate the central challenges to realism: underdetermination, underconsideration, and historical inductions. The chapter on the argument from underconsideration—Wray’s phrase for what van Fraassen calls the problem of the bad lot—is one of the highlights of the book. This argument grants that scientists are reliable at determining which extant scientific theories in a given domain are best (that is, which theories best exemplify the empirical and theoretical virtues) and contends we have no reason to think that the true theory is among the extant theories. Therefore, we have no justification for thinking that our best theories are true. Wray musters a serious challenge to Peter Lipton’s () critique of this argument. One compelling theme of his response is that because of the comparative nature of scientific theory selection, even if realists are able to establish that the theoretical virtues are truth-tropic, they have yet to show that we are justified in believing that our best theories are likely to be true, rather than simply more likely to be true than their competitors. Finally, the chapter on the pessimistic induction contains some excellent disambiguations of different forms of the argument and a good overview of the state of the debate. Wray’s close and subtle readings of Larry Laudan’s () and Hilary Putnam’s () versions of the pessimistic induction clarify important aspects of these arguments that have often been misunderstood.
The remainder of Part 1 criticizes realist responses to the above anti-realist challenges, with reflections on radical theory change (including a helpful account of Kuhn’s () analysis based on lexical changes) and further exploration of the theoretical virtues. Another important chapter, Chapter 6, criticizes recent arguments by Fahrbach (, ) and Devitt () against versions of the pessimistic induction. A theme of this criticism, and of Part 1 as a whole (see, for example, Chapter 4, on epistemic privilege), is that today’s scientists are in the same boat as their predecessors. They know the theories they accept are more instrumentally accurate and based on more precise observations than those held by previous scientists; at the same time, they know previous scientists were in this very same position with regards to the theories held by still earlier scientists. Yet the best theories of their predecessors have been revised radically or rejected outright; consequently, today’s scientists should not believe that their best theories are true. Today’s scientists do not have grounds to state that they, unlike previous scientists, are epistemically privileged, such that the theories they accept will not be radically revised or rejected in the future.
Part 2, ‘Strengthening Anti-realism’, evaluates the most common argument against anti-realism, the realist’s no miracles argument, and seeks to defend and expand a popular anti-realist response, the selectionist explanation of the success of science. The extended discussion and elaboration of this explanation is another important contribution to the literature.
The no miracles argument contends that the only plausible explanation of the instrumental success of science is that scientific theories are approximately true. Wray defends van Fraassen’s () contention that the instrumental success of science is explained fully by the fact that scientists discard unsuccessful theories. We no more need an additional explanation for why today’s best scientific theories are instrumentally successful than we need an additional explanation for why today’s non-threatened species are successful at reproduction, other than that non-reproducing species die out. Chapters 9, 10, and 11 elaborate and defend this selectionist explanation. A major component of Wray’s defence is that the anti-realist’s explanation of the success of science, unlike the realist’s, is able to explain why false theories have been successful and why inconsistent competing theories can both be successful. Structural and selective realists (for example, Worrall  and Psillos , respectively) try to explain the success of false theories by appealing to veridical aspects of false theories. Against this manoeuvre, Wray gives a number of examples where it seems that non-veridical aspects of false theories are invoked in the explanation of the theories’ successes. If these examples are legitimate, it seems realists must somehow work out how to acknowledge that theories can be successful for reasons other than being true.
The book concludes with a short synthesis of the main points, preceded by an intriguing chapter on the role played by scientists’ changing interests over time. There are a number of interesting ideas here. In particular, Wray is correct to emphasize that scientific theories are partial representations. This should be considered in the scientific realism debate, and I found myself wishing that this chapter had pushed the exploration further. Wray observes that ‘a theory that effectively served the interests of scientists at one time is apt to seem inadequate at some later time, when scientists have different research interests’ (p. 187). He aspires to show that sometimes scientists reject a previously held theory not because they come to deem it false, but because ‘their research interests shift to such an extent that the long-accepted theory proves to be inadequate’ (p. 198). He cites the fact that scientists were once, but are no longer, interested in the question of whether planets are self-illuminating (p. 187), and Newton’s development of a novel interest in the mass of planets (p. 191). But these changes in research interests were in part due to advances in our understanding of the world: we came to know planets are not self-illuminating and that their masses are related to their motions, so we lost interest in theories that supposed otherwise. Clearer examples of scientists discarding theories for reasons other than their falsity would have been more helpful to Wray’s case. I look forward to further development of this topic, especially the role that scientists’ interests play in determining which variables can be left out of their models (p. 199).
In my opinion, there are several other areas of the book that would have benefitted from greater development. First, Wray announces that rather than being primarily concerned with the question of whether we have grounds for believing that our best theories are approximately true, he will focus on the question of whether our best theories will likely be overthrown by theories ‘that make significantly different ontological assumptions’ (p. 1). This is an interesting direction to take. It would have been helpful to hear more about what Wray takes to be the pay-off of this shift in focus, especially as he often defaults to the standard focus of the debate. Wray frequently refers to scientists being ‘realists’ about various aspects of their theories. For example, he says that Kepler and Galileo were ‘realists about Copernican cosmology’ (p. 25). In this context, Wray clearly means that they believed, for example, that the Earth revolves around the Sun, and not vice versa. Kepler and Galileo’s realism is described by Wray with the standard focus on the approximate truth of the theory they accepted.
In addition, this example shows that Wray’s understanding of the distinction between observable and unobservable entities departs from van Fraassen’s. For van Fraassen (), scientists who accepted the heliocentric theory as empirically adequate would be in complete agreement with scientists who accepted the heliocentric theory as true. The reason for this is that the empirical adequacy of the theory relates to its observable substructures, which for van Fraassen are identical to the theory itself. Even though the revolution of the Earth around the Sun was not directly observable to Kepler and Galileo, it was observable in van Fraassen’s sense, because there is a perspective from which humans can observe this phenomenon with unaided senses. Wray, who is focused on the kind of transient underdetermination investigated by Larry Sklar () and Kyle Stanford (), seems to conceive of observability as a transient property that changes as our observational repertoire expands. I raise this point not to object to the adoption of a transient notion of observability, but to say it would have been useful to hear something about how Wray understands observability, and what the implications might be for his anti-realism if observability is taken to be transient.
In sum, Resisting Scientific Realism exhibits excellent scholarship and philosophical creativity. It is a valuable contribution to the literature on scientific anti-realism, containing important arguments with which realists must contend, and directing the scientific realism debate towards exciting new (or unfairly neglected) topics. The writing is admirably clear and elegant. If you are interested in scientific anti-realism or scientific theory change, you should read this book.
Samuel Ruhmkorff Department of Social Studies Bard College at Simon’s Rock Great Barrington, MA, USA firstname.lastname@example.org
Devitt, M. : ‘Are Unconceived Alternatives a Problem for Scientific Realism?’, Journal for General Philosophy of Science, 42, pp. 285–93.
Fahrbach, L. : ‘How the Growth of Science Ends Theory Change’, Synthese, 180, pp. 139–55.
Fahrbach, L. : ‘Scientific Revolutions and the Explosion of Scientific Evidence’, Synthese, 194, pp. 5039–72.
Kuhn, T. : The Copernican Revolution: Planetary Astronomy in the Development of Western Thought, Cambridge, MA: Harvard University Press.
Kuhn, T. : ‘What Are Scientific Revolutions?’, in J. Conant and J. Haugeland (eds), The Road Since Structure: Philosophical Essays, 1970-1993, with an Autobiographical Interview, Chicago, IL: University of Chicago Press, pp. 13–32.
Laudan, L. : ‘A Confutation of Convergent Realism’, Philosophy of Science, 48, pp. 19–49.
Lipton, P. : ‘Is the Best Good Enough?’, Proceedings of the Aristotelian Society, 93, pp. 89–104.
Psillos, S. : Scientific Realism: How Science Tracks Truth, London: Routledge.
Putnam, H. : Meaning and the Moral Sciences, London: Routledge and Kegan Paul.
Sklar, L. : ‘Do Unborn Hypotheses Have Rights?’, Pacific Philosophical Quarterly, 62, pp. 17–29.
Stanford, P. : Exceeding Our Grasp: Science, History, and the Problem of Unconceived Alternatives, Oxford: Oxford University Press.
Van Fraassen, B. : The Scientific Image, Oxford: Clarendon Press.
Van Fraassen, B. : Laws and Symmetry, Oxford: Clarendon Press.
Worrall, J. : ‘Structural Realism: The Best of Both Worlds?’, Dialectica, 43, pp. 99–124.