In the 1990s incensed scientists began responding to what they saw as ideologically motivated attacks emanating from left wing camps in the humanities and social sciences. Their ripostes included Higher Superstition by Paul Gross and Norman Levitt (Reference Gross and Levitt1994) and Fashionable Nonsense by Alan Sokal and Paul Bricmont (Reference Sokal and Bricmont1998). These works entertainingly exposed the arrogance and ignorance of the science bashers. The authors also delighted in deflating the obscurantist babble that passes for prose among many science critics.Footnote ¹

Still, for all their insight and entertainment value, these scientific polemics lacked philosophical depth. There really are some deeper issues lurking in the murky depths of social constructivist and postmodernist science critique. The scientific polemicists sometimes write as if any question about science being merely a ‘reality driven’ enterprise must be simply perverse—like asking whether you fall down when you jump out a window. Yet, the palmy days of positivism and knee-jerk scientism did not end because of the wild claims or turgid prose of the new wave of science detractors. As every reader of Philosophy of Science knows, there are serious questions, asked by serious scholars, about the nature of scientific rationality. Global skepticism about science is not justified, but neither is an attitude of automatic deference. What we need is a guide through the debates about scientific rationality, a guide written by someone who emphatically rejects the cynicism of the new science critics, but with the sophistication to deal fairly with the substantive philosophical issues. I think that Susan Haack's Defending Science—Within Reason is by far the best such book yet.

Haack deserves her reputation as a passionate moderate. Those who pursue a middle way are often scorned as fence sitters, but as she shows, in this and other works, truth often does take a middle path. Specifically, she opposes both what she calls the ‘New Cynicism’ and the ‘Old Deferentialism’—gratuitous contempt for science on the one hand, and equally unjustifiable adulation on the other. She sees both attitudes as based on unrealistic expectations about science:

Of course, this is not a novel thesis, nor is her reiterated recommendation that the rationality of science be understood as continuous with the rational practices of other areas of inquiry and of common sense. For professional philosophers of science, Haack is treading well-trodden ground. However, the so-called ‘science wars’ of the 1990s expanded the debate over scientific rationality beyond the confines of academic philosophy of science. Haack is speaking to the wider audience that took an interest in those debates. Also, Haack's aim is not to startle us with new analyses and concepts, but to put our feet back on the ground. When debate becomes polarized, as it has in the ‘science wars’, the temptation is to rush violently from one extreme to the other. Haack is here to remind us, emphatically, that we have all along possessed the tools for an understanding of science that accounts for its successes, while guarding against unreasonable expectations.

She calls her view ‘Critical Common-Sensism’. The key is to see science as embodying the epistemic virtues and sorts of spontaneous practices that go with any serious human inquiry. As Haack notes, anybody who honestly investigates any empirical question, even the most mundane, has to have a procedure for checking hypotheses and methods for safeguarding against error and bias. So scientific inquiry is methodologically continuous with everyday inquiry, though, of course, science is carried out at a much more sophisticated level. The extent to which hypotheses in the natural sciences are more rigorously constrained than everyday hypotheses is not due to scientists’ possession of any exclusive ‘Scientific Method’. Rather, the precision with which scientific hypotheses and their predictions are expressed allows for stricter testing. Also, scientists often enjoy contingent advantages in the quantity and quality of data available and the reliability of the tests, instruments, and methods (little ‘m’), they employ. She illustrates these points with the familiar story of Watson and Crick, which she relates (warts and all) while still showing that, in the end, the double helix model was strongly warranted.

Actually, Haack's treatment of the ‘naturalness’ of scientific rationality needs to be somewhat more detailed and nuanced. Robin Dunbar's The Trouble with Science (Reference Dunbar1995) notes numerous instances of ‘scientific’ methods of hypothesis testing being practiced in ‘primitive’ cultures, as well as by small children, and even animals. Such examples strongly support Haack's claim that science is regimented common sense and that scientific methods are refinements of natural and spontaneous cognitive practices. However, Dunbar also notes that the famous studies by Kahneman and Tversky, and the results of many administrations of the Wason Selection Task, all show unequivocally that many of the intellectual tools essential for science do not come naturally and are learned only with considerable difficulty (if at all). The most general aspects of scientific rationality may arise naturally, but the particular tools and techniques do not. In fact, we seem strongly inclined to think irrationally in some ways—as, for instance, the pervasive tendency to ignore base rates in our intuitive estimates of probability. So, Haack's case would have been strengthened by a somewhat deeper analysis of the ways that science is natural and the ways that it is not.

Haack is fond of quoting the saying of Percy Bridgman that asserts scientific method is nothing more than “doing one's damnedest with one's mind, no holds barred.” This is pithy, but not very helpful. Haack's most useful analogy compares the process of scientific discovery to solving a crossword puzzle. Nature provides the clues, and the answers we have already filled in represent our presumed background knowledge. Solving the puzzle is not an algorithmic procedure, but an effort involving trial and error, inspired guesswork, intuition, and the occasional ‘Aha!’ insight. The reasoning goes back and forth: Clues suggest answers, but sometimes, when a clue suggests an answer that will not fit with what we have already written in, we try to reinterpret the clue. Support for any particular entry comes both from the clue and from the fit with other entries. However, all answers are tentative and we often find that we have to go back and erase earlier entries because they conflict with what we now think that we know. As analogies go, I think this is a useful one, and Haack develops it in greater detail in her earlier Evidence and Inquiry (Reference Haack1993).

The upshot is that science fumbles, stumbles, and sometimes forges ahead. As with any human enterprise monotonic progress is not to be expected, but neither is extreme pessimism and cynicism justified. Consider the claim that observation is too theory-laden to serve as an objective anchor for scientific claims. Haack (Reference Haack2003, 125) admits that observation depends on theory in various ways, and as such is always fallible and revisable. Yet, she rejects the radical epistemological conclusion that scientific claims are therefore always relative or subjective. She argues, as philosophers such as Scheffler (Reference Scheffler1982) and Newton-Smith (Reference Newton-Smith1981) have also argued, that the theory dependence of observation does not mean that it cannot be a strong and objective constraint on theory. However, she does oppose drawing too sharp a distinction between the observable and the unobservable—a distinction she thinks not robust enough to support Bas van Fraassen's constructive empiricism (Reference Haack2003, 137–138). Also, she finds puzzling Ronald Giere's claims that we can have ‘science without laws’ and ‘realism without truth’. She comments dryly, “[Giere] also tells us that he doesn't, strictly speaking, mean that science could really do without laws, or without truth. I agree with the part where he takes it back” (Reference Haack2003, 138).

Her disagreements with philosophers such as van Fraassen and Giere are polite and restrained, unlike her criticisms of sociologists of science such as David Bloor, Steven Shapin, Harry Collins, Bruno Latour, and Steve Woolgar. Instead of directly attacking the sociologists’ theses about science, which has been done by a number of others (see Parsons Reference Parsons2001; Koertge Reference Koertge1998, and Brown Reference Brown1989), she traces the history of their skeptical claims. She notes that these claims began big and bold, and were intended to strike at the very heart of ‘rationalist’ philosophies of science. Sweeping claims were made about how class interests, politics, or ideologies determine theory choice in science, or how ‘reality’ is just whatever scientists agree that it is. Such claims soon sank under the weight of their sheer implausibility and the failure of their proponents to provide adequate supporting arguments. Also, sociologists of science could not escape the problem of reflexivity: the ineluctable fatuity of treating all knowledge claims as social constructs—except for the opinions of social constructivists. The consequence is that the once bold claims have petered out into hackneyed truisms about the fallibility of scientific knowledge (see Hull Reference Hull1995). I think Haack's treatment is devastating and should serve as a tombstone for the whole program of radical sociology of science.

After addressing general concerns about scientific rationality, Haack moves on to particular bones of contention, such as science and the law, and science and religion. I found her treatment of the latter especially interesting. Since Andrew Dickson White's A History of the Warfare of Science and Theology in Christendom was published in 1896, the ‘warfare’ theory of the relation between science and religion has been thoroughly debunked (see Lindberg and Numbers Reference Lindberg and Numbers1986). Many recent writers have pursued the irenic vision that there is no real conflict between science and religion. For instance, Stephen Jay Gould (Reference Gould1998) argued that science and religion are ‘non-overlapping magisteria’; that is, science and religion do not conflict because they deal with distinct spheres—science with fact and religion with value.

Haack rejects such views as naïve and overly sanguine. Science and religion may not be in a state of perpetual warfare, but Haack contends that there are various areas of actual or potential conflict. For instance, science is committed at least to methodological naturalism, since it rejects any appeal to supernatural, otherworldly, or spiritual forces (Reference Haack2003, 266). Insofar as religion might insist that some things have a supernatural provenance (life, the mind, or the universe, perhaps) there is a potential for conflict between religious and scientific explanations. More fundamentally, says Haack, “Science is not primarily a body of belief, but a federation of kinds of inquiry” (Reference Haack2003, 266). She says that religion is not primarily a kind of inquiry but a body of belief, and the sort of faith needed to sustain such belief has no place in science (Reference Haack2003, 267). Theology is a form of inquiry, but its appeal to religious experience and allegedly revealed texts makes it discontinuous with ordinary empirical inquiry, unlike science (Reference Haack2003, 267). Her conclusion is refreshingly old-fashioned and non-conciliatory: “As I see it, religion and science really are profoundly at odds on all the dimensions I have distinguished; and science really is, on all those dimensions, far and away the more admirable enterprise” (Reference Haack2003, 267).

A few years ago, Haack's astringency might have been regarded as a return to discredited stereotypes of both science and religion. At that time, it might have appeared that the only religious attacks on science were instigated by marginalized fundamentalist groups such as the young-earth creationists. Since such groups had no standing in academe, it might have seemed that respectable religion had no conflict with science, and that it could be only regressive to go back to the rhetoric of the ‘warfare’ view. But creationism can no longer be relegated to the fringes. With the rise of ‘intelligent design theory’ and the vocal support of that theory by some very distinguished professional philosophers (see, e.g., Plantinga Reference Plantinga and Pennock2001), the religious critique of natural science is no longer easy to dismiss. As Phillip Kitcher puts it, “Creationism is no longer hick, but chic” (Kitcher Reference Kitcher and Pennock2001, 258). So, in this reviewer's opinion, it is quite appropriate for Haack to note that characteristic goals, standards, and claims of science and religion might really clash, and to express unabashed support for science when they do.

Because Haack's book touches on so many topics, it inevitably reads as something of a manifesto, with the details of some arguments left for other times and places. However, manifestos are written as the Zeitgeist calls for them. In recent years, science bashing has become the avocation, if not the preoccupation, of ideologues of both the left and the right (see Parsons Reference Parsons2003). The ‘science wars’ were far more serious than the typical academic in-house quarrels. These debates raised vital questions about the future of science and technology in our intellectual culture. When the stakes are high and the rhetoric heated, a passionate but cogently argued manifesto can be heard as the voice of reason. This is how I think that Haack's Defending Science should be heard.