Issues in Science and Technology, Online A Publication of National Academy of Sciences National Academy of Engineering Institute of Medicine University of Texas at Dallas
  • Home
  • Back Issues
  • Search


Book Review

In defense of science

Fashionable Nonsense: Postmodern Intellectuals' Abuse of Science, Alan Sokal and Jean Bricmont. New York: Picador, 1998, 300 pp.

Daniel Barbiero

Since the publication in 1994 of Higher Superstition: The Academic Left and Its Quarrels with Science by Paul R. Gross and Norman Levitt, a debate has opened up between critics of science, mostly (though not exclusively) associated with the various postmodernist ideologies in academic humanities departments, and defenders of science, mostly (though again, not exclusively) drawn from the ranks of science itself. The point at issue is nothing less than the legitimacy of science's claims to reliable knowledge.

Summarizing the postmodern critique of science is difficult, since there are seemingly as many postmodern critiques of science--and indeed, postmodernisms--as there are postmodern critics of science. But if a common denominator is to be identified, it would be a form of radical epistemological relativism holding that truth claims are relative to languages or social formations and thus have no basis in objective fact. Undergirding postmodern radical relativism, more often than not, is a misunderstanding of the products and processes of science.

Where does such misunderstanding come from? At least in part from some well-known figures in the humanities, as physicists Alan Sokal and Jean Bricmont show. Wielding a generous helping of quotations and a rigorous analysis, Sokol and Bricmont examine how several influential French thinkers have misunderstood and misused concepts from science and the history and philosophy of science. As the authors explain, their book grew out of Sokol's well-known parody "Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity" (reprinted in the book as an appendix). The parody, which Sokal deliberately filled with pseudoscientific nonsense "justified" by liberal citations of leading postmodern theorists and fellow travelers, was published unwittingly as a serious article by a leading cultural studies journal. By presenting in book form an expanded sampling and analysis of the kind of theorizing lampooned in the parody, the authors set out to explain in lay terms exactly why such writings are absurd.

Sokol and Bricmont devote individual chapters to exegeses of work by psychoanalyst Jacques Lacan, literary critic Julia Kristeva, feminist theorist Luce Irigaray, sociologists Bruno Latour and Jean Baudrillard, philosophers Gilles Deleuze and Felix Guattari, and architect Paul Virilio. The majority of these writers came to prominence in the 1970s in France, where their work continues to exert a profound influence on French intellectual life. In the past two decades, that influence has extended outward, particularly to the United States.

Sokal and Bricmont focus only on the treatment of science and the use of scientific concepts. They offer lengthy excerpts of each writer's work in order to be certain that nothing is taken out of context. Much of the material quoted makes for dismal reading. Lacan's arbitrary, often incoherent applications of mathematical formulae and concepts to psychoanalysis come across as fraudulent, as does his former student Irigaray's attempt to critique solid and fluid mechanics with gender-derived metaphors. Kristeva's effort to relate poetic language to set theory founders in error and irrelevance. Virilio's and Baudrillard's tendency to dig out of a grab bag of scientific terms seems to serve no purpose other than to mystify.

In many of the cases described by Sokal and Bricmont, technical terms are taken out of their original context and are made to serve purposes for which they are largely irrelevant. In other cases, technical and nontechnical meanings of terms such as "linear" and "chaos" are mixed promiscuously, without any attempt to make explicit the connection between the two uses or the relevance of the technical terms to the larger argument. A good example of this can be found in a Deleuze and Guattari pronouncement on chaos: "Chaos is defined not so much by its disorder as by the infinite speed with which every form taking shape in it vanishes. It is a void that is not a nothing but a virtual, containing all possible particles and drawing out all possible forms, which spring up only to disappear immediately, without consistency or reference, without consequence. Chaos is an infinite speed of birth and disappearance."

As Sokal and Bricmont point out, Deleuze and Guattari are not using "chaos" in connection with nonlinear dynamical systems--although they do so later in the same work without calling attention to or explaining the reason for this sudden shift in usage. In this passage they conflate a description of quantum field theory with a description of the properties of a supercooled liquid, both of which they drew from a popular science work by Ilya Prigogine and Isabelle Stengers. To be sure, Deleuze and Guattari's uses of chaos here, as well as other authors' uses of other technical terms, often seem to be metaphorical or analogical. But as Sokal and Bricmont note, metaphors and analogies ordinarily employ familiar concepts to illuminate unfamiliar ones; not the opposite, which is the case with these writers.

The situation becomes only more confused when a Lacan, a Latour, or a Baudrillard protests that a given use of a scientific theory or concept should be taken literally. In an essay on the end of history, for example, Baudrillard claims that chaos theory calls for a reversal of the relationship between cause and effect, that it has something to do with Benveniste's (discredited) hypothesis on the memory of water, and that it somehow demonstrates that history will be deflected from its end. Taken literally, the first claim is mistaken, and the other two claims are irrelevant simply because chaos theory has no real connection to the points Baudrillard wishes to make. And yet Baudrillard's use of chaos theory is not intended as a metaphorical illustration of his theory of the end of history but rather as a basis from which it can be derived.

The heart of the debate

It would be easy enough for Sokol and Bricmont simply to ridicule the scientific ignorance they expose and leave it at that. To their credit they don't. Their purpose instead is to offer a substantive critique of the positions such lack of understanding can produce. Their major target is epistemological relativism, which they discuss in the first of two "intermezzos." Over the course of analyzing work by historians and philosophers of science Karl Popper, W. V. Quine, Thomas Kuhn, Paul Feyerabend, David Bloor, and Latour, they show how concepts from the history and philosophy of science have been misused in order to license what might be called a naïve antirealism embodying extreme relativist claims regarding the nature of knowledge and objectivity. Although the majority of the philosophers Sokol and Bricmont cover are neither postmodernists nor likely to agree with much that is postmodernist, their work has often been cited by postmodernist critics of science, particularly in the English-speaking world. The attraction appears to lie in what Sokol and Bricmont characterize as the partial failure of Popper et al. to formalize the scientific method. If this couldn't be done, critics of science contend, then it must follow that scientific rationality itself is merely a myth. Of particular interest to radical relativists are the notions that theories are as a rule underdetermined by evidence, that they arise and are evaluated relative to a conceptual framework or paradigm, and that science is a social practice. But do these notions really warrant a radical relativism?

Consider the relativist argument from underdetermination. Because the evidence does not unambiguously select one theory over another, it is claimed that theories are virtually if not actually unconstrained by the natural world and that theory selection is therefore a function of nonempirical social factors. But although it is true that theories are underdetermined by facts, in that a finite set of facts may be consistent with an infinite number of theories, it is also true that a theory's being consistent with the facts isn't the same as its being confirmed by the facts. For the latter to be the case, the facts would have to obtain in the context of a successful test of the theory. In addition, theories come with auxiliary hypotheses attached, so that even two theories equally entailed by the same set of evidence may be distinguished by the relative soundness of their respective auxiliary hypotheses. In cases where no convincing theory is available, scientists may suspend judgment, but this doesn't mean that theory choice is a virtual leap in the dark; it just means that in this particular case the evidence is insufficient to support any of the theories that have been proposed.

The argument from framework relativism, derived rightly or wrongly from Kuhn and frequently encountered among postmodern writings, holds that since our descriptions of the world are necessarily relative to the conceptual repertoires through which we have access to the world and to the conditions under which our descriptions are formulated, then our knowledge has no necessary validity for others outside of the contexts in which our descriptions are formulated. Consequently, we cannot look to objective factors to evaluate knowledge; on the contrary, the very notion of objectivity must be abandoned. But this is obviously untrue; the speed of light, for instance, is the same no matter which conceptual framework one holds to. It may be that a framework lacks a concept for the speed of light, but that says something only about the framework, not about the speed of light. It simply isn't the case that because a description can be made relative only to a given linguistic, social, or historical context, the facts described are valid relative only to that context. In the end, features of the world either will or will not bear out the descriptions we make of them.

The assertion that science is a social practice driven by interests, purposes, and concerns has become a dogma among postmodern writers, particularly those engaged in social and cultural studies of science. If the claim is simply that science does not take place in a vacuum, that it is situated within particular cultures at particular points in their histories, then it is a reasonable if not banal one. The false step, which many postmodern critics of science seem more than willing to take, is to go from the generally accepted observation that science is not simply a matter of the mechanical application of general topic-neutral laws of derivation to bare facts, to the false conclusion that there therefore is no fact of the matter to scientific knowledge outside of social determination. Socially negotiable interests can influence the terms in which we will describe something, the choice of what it is we will describe, or what questions we will attempt to answer in generating the description. But ultimately we are still left with the crucial question: Is our description an adequate one, given what we want to know?

In the end, does it matter that a handful of academic humanists pretend to a scientific competence they do not have? To the extent that the positions they espouse remain influential, yes. As Sokol and Bricmont point out, the abandonment of clear rigorous thinking and expression can have only negative consequences not only for humanities departments but for the culture at large. But beyond this, the acceptance of radical relativism and the rejection of the goal of objectivity can lead to worse abuses in which ideological standards replace standards of sound reasoning as the criteria against which assertions will be judged valid. As philosopher Paul Boghossian remarks in a thoughtful consideration of the Sokol hoax, that way leads to completely socially determined science and to such destructive madness as Lysenkoism in the Soviet Union and "Aryan science" in Germany. Against this dangerous prospect, Sokol and Bricmont offer an uncompromising defense of science and reason.