Review of Intellectual Impostures

by Val Dusek

[To appear in Metascience 9.3]

Sokal, Alan, and Jean Bricmont.
Intellectual Impostures: Postmodern Philosophers' Abuse of Science.
London: Profile Books, 1998.

Sokal and Bricmont in their exposé of allegedly meaningless statements about science by recent French philosophers take errors of particular applications of philosophical ideas to science as refutations of the whole general framework utilized. They also seem to think that taking snippets out of context is sufficient to expose the "fashionable nonsense." In the early twentieth century, British analytic philosophers such as Bertrand Russell and A. N. Whitehead did the same with Hegel on mathematics. After deciding not to bother to read Hegel because of distaste for what he wrote about mathematics, Whitehead was later surprised to learn that his own relational process philosophy resembled that of Hegel in various respects.

Sokal and Bricmont, like a number of other physicist and mathematician science warriors, strive to maintain a view of science that preserves the attitudes of the past century by reinterpreting the apparently unsettling developments of twentieth century science. They wish to reassure non-scientists that chaos theory and quantum mechanics have not radically changed to nature of the universe presented by science. They debunk claims that twentieth century science has undermined determinism or the independence of the observer from the observed.

During the first half of the twentieth century, many leading theorists of modern physics were also philosophers and humanistic scholars. Werner Heisenberg first learned of atoms, not in a physics text, but from reading Plato's Timaeus. He claimed that later reading of the same work (in Greek) for relaxation during lunch break had some influence on his conception of uncertainty in physical reality. Schrödinger took his lab notes in classical Greek and wrote as did Heisenberg about the Presocratic philosophers in his search for a way to understand subatomic reality.

After W.W.II, with the congealing of the official interpretation of quantum theory, and the rise of big science, philistinism took over. Feynman, a leading genius of the period, despised philosophy, though he often misrepresented the positions of the philosophers he ridiculed. His proudly dismissive attitude toward philosophers was linked with unconscious personifications of Nature and an implicit philosophy of a plurality of causes in mechanics resembling that (he would be horrified to hear) of Aristotle. One perceptive reviewer of Feynman's anecdotes says that someone familiar only with the beauty of Feynman's physics papers, would, on reading his books of humorous anecdotes, react like Antonio Salieri on his first encounter with Mozart. Physicist and science essayist Jeremy Bernstein, in reaction to the mention of the influence of Hindu thought on Schrödinger's later writings, replied simply "Yogic, Schmogic," and claimed recently that only an historian of physics would have any interest in reading Niels Bohr, the creator of the standard, "Copenhagen interpretation" of quantum mechanics. Einstein and Heisenberg read the philosopher Kant as teenagers, and the Kantian strain in Bohr and Heisenberg is an alien realm to late twentieth century Anglo-American physicists.

The changed situation in twentieth century philosophy is similar. In the 1920s, not only Henri Bergson, but also Whitehead, and George Herbert Mead with their "objective relativism," strove mightily to grapple with the general philosophical consequences of Einstein's relativity theory (whatever one may think of their particular conclusions). Today in Anglo-American philosophy the philosophers of science discuss such issues, but usually without attempting in any way to discuss their implications for culture or for patterns of thought in general, saying, with W. V. O. Quine, that "Philosophy of science is philosophy enough." On the other hand, most general analytical social philosophers don't even try to grapple with the consequences of contemporary science and math for our worldview, and often uncritically and tacitly presuppose older, flawed interpretations. (One source of the "overdetermined" support for Sokal and Bricmont's position, besides neo-conservatives denouncing "political correctness" and traditional literary critics angry at French theory, is the community of analytical philosophers who reject continental philosophy.) Would Sokal and Bricmont be happier if general philosophers were to ignore science totally? Would they admire the Oxford ordinary language philosophers of the 1950s who sneeringly ignored both science and politics as irrelevant to "ordinary language" (the supposed font of all wisdom)? Would they agree with neo-conservative Straussian Alan Bloom that science has no relevance to human life? They ought to welcome the sometimes fumbling attempts of recent philosophers to make sense of science as a cultural phenomenon and to speculate about the cosmological metaphysics that science reveals.

Alan Sokal calls the founders of quantum mechanics Bohr and Heisenberg, "vulgarizers (in both senses)."(p. 255 fn 14) Odd that they should have invented the things of which they are mere vulgarizers. Similarly, Sokal and Bricmont sharply distinguish Einstein's "pedagogy" (discussed by Latour, and explicitly called pedagogy by him, contra Sokal and Bricmont) and Einstein's real theory (p. 116). What they ignore is that the early Einstein (who credited Hume and Mach for inspiration, and Mach in turn credited Berkeley) really did approach relativity in terms of the thought experiments concerning possible measurements by conscious observers that he describes in his popular book. Logical positivism and Percy Bridgman's operationalism were inspired precisely by this approach of Einstein.

Sokal and Bricmont likewise are distrustful of philosophical claims concerning implications of chaos theory (Ch. 7, pp. 125-136 ). Sokal satirizes and with Bricmont certainly exhibits and exposes some confused and misleading statements about non-linearity and chaos. Yet not all such extrapolations from chaos theory are solely the product of mathematical ignorance. One wonders what they think of André Lichnerowicz, one of the great mathematicians our time, lending his name and authority to a collaborative work that, although it does not, like Prigogine or Bohm, use the word postmodernism, ranges afield into speculative applications of chaos theory to biology, economics and philosophy, mentioning suspect, supposedly anti-scientific figures such as Bergson, Tielhard d' Chardin, Freud, and Foucault.

In the original French version of their book, Sokal and Bricmont discuss Bergson's misunderstandings of Einstein and then trace what they consider the sad history of French philosophers praising Bergson. Bergson probably suffered from writing too well and deceptively simply. This made him extraordinarily popular, that led to his soon being dismissed by "serious philosophers." Part of Bergson's loss of respect in the English speaking world is due to Bertrand Russell's History of Western Philosophy (with a quotation from which Sokal and Bricmont's chapter on Bergson begins) that portrays Bergson's intuitionism as proto-Nazi, when in fact Bergson died from illness contracted while waiting on a bread line in occupied France after he refused the Nazis' offer to give him special treatment as an "honorary Aryan."

Before dismissing Bergson as a fool, and his philosophy of the intuition of time and of the fundamental reality of process as nonsense, one needs to separate several issues. Bergson indeed made mistakes (pointed out by Einstein himself) in arguing about special relativity theory. Bergson himself recognized his lack of expertise in physics, and refused to allow further editions of his work to appear during the thirties. Do these mistakes mean that Bergson's views on time ought to be dismissed, or that his philosophical claims about time have no value? Bergson was not, contrary to Bricmont's opinion, trying to "refute" Einstein. Bergson rejected Newtonian absolute space and he accepted the demise of the classical aether, unlike a number of reactionary philosophical holdouts against relativity theory. Several physicists, such as de Broglie, Watanabe and Costa de Beauregard have seen value in Bergson's ideas in relation to wave mechanics and thermodynamics, despite his particular errors in relativity theory. Two major mathematicians whom Sokal and Bricmont cannot accuse of ignorance of mathematics made sympathetic use of Bergson. Norbert Wiener, in his Cybernetics, opened with a discussion of "Newtonian vs. Bergsonian time," and A. N. Whitehead incorporated parts of Bergson's philosophy of process into his own interpretation of relativity theory.

The spirit of Bergson's earlier writings contradicts the letter of his unfortunate sally into relativity theory. Bergson's own, earlier Matter and Memory contradicted the later denial of multiple temporal rhythms in his discussion of relativity theory. Milic Capek points out in Bergson's emphasis on the difference between time and space and his denial of absolutely separate material particles fits well with much of relativity theory and quantum mechanics, but that Bergson's own treatment of time in reaction to Einstein mistakenly treated Minkowski's diagram as a dreaded "spatialization of time" similar to that of classical treatments of time as a fourth dimension in d'Alembert and others (a mistake shared, by the way, by some of the physicist defenders of Einstein's theory as portraying a "block universe" without genuine change).

At the center of the debate between Bergson and Einstein was the "twin paradox." If rapid travel shows down time, then a twin sent into space at high speed would return younger than the twin who remained on earth. Yet if velocity is relative, should not the twin on earth be younger than the space traveler, since, relative to the space traveler, the twin on earth receded and then approached at high speed? Even if Bergson's claims about the twin paradox are confused, that's not to say that the twin paradox is totally cleared up. When physicist Herbert Dingle argued in Nature for the genuine paradoxicality of the twin paradox, a number of physicists indignantly claimed the solution was clear and simple, but gave "obvious solutions" inconsistent with one another. Marder edited a whole book of such "obvious solutions" to the twin paradox some of which are mutually incompatible.

One point that Sokal and Bricmont miss making because of their fragmented approach to out-of-context quotes is the suspicious resemblance of Latour's "third observer" in his account of special relativity theory to Bergson's account (possibly via Deleuze's book on Bergson). Latour rarely gives reference to the sources of his ideas, preferring to appear to have created them out of whole cloth. He claims that in Einstein's special relativity theory there is a third observer who is describing the two observers mentioned in the exposition. Bergson makes a similar move in claiming that there is a unitary time subsuming the relative times of the two observers in Einstein. Sometimes, as in Latour's account, this is the time of the third observer subsuming the other two. Latour and his critics, as well as his physicist defender David Mermin confuse the issue of the number of physical "observers" needed in special relativity with the philosophical question of whether in thinking about some topic we are also thinking about ourselves thinking about it. (This latter issue leads to the old paradoxical claim that one cannot imagine oneself dead or unconscious, because one is imagining oneself consciously imagining oneself dead or unconscious. ) The third observer is not one of the observers in the physical system, but is this self-conscious theorist or reader thinking about the other two physical observers. Both Latour, Sokal and Bricmont, and Mermin treat this transcendental conscious observer as if it were an actual, physical observer located somewhere in the physical space-time being described.

Ironically, David Bohm, whose deterministic quantum theory is admired by many of the physicist science warriors, has speculated in manners strangely similar to Bergson concerning notions that there might be other ways to think about or make models of physical processes other than the classical ones. This has been elaborated on by Capek, and the suggestions resemble in many respects Bohm's suggestions about trying new imaginary models and rejecting Bohr and Heisenberg's claim that we are trapped conceptually in classical models -- that supposedly prevents us from thinking directly about quantum reality.

One philosopher who makes use of Bergson's ideas concerning time and process is Gilles Deleuze. Sokal and Bricmont seem to be particularly annoyed at Deleuze because he was excessively praised by Michel Foucault. (Deleuze and Guattari shared many interests such as anti-psychiatry and rejection of unitary systems, were both gay, interested in sadomasochism, and evidently took drugs together).. Much of the rancor in the science warriors' attack on the postmodernists seems to be jealousy at their undeserved fame. In Bricmont's case, the perhaps undeserved Nobel Prize and excessive fame of the popular writings of Ilya Prigogine, a fellow Belgian physicist who occasionally mentions Bergson and postmodernism, seem to stoke the fires of his resentment. Sokal and Bricmont should feel less of this now that they, through the Sokal hoax and this book, themselves have achieved worldwide fame.

Sokal and Bricmont, like many of the uncritical epigones of Deleuze interested primarily in gay liberation, anti-psychiatry movements, focus on Deleuze's work with the psychiatrist and political activist Felix Guattari. Deleuze collaborated in his later life on a number of wild and unbuttoned books with his buddy. Sokal and Bricmont treat the two together in their critique, but have the harshest words for a passage by Guattari alone, with which they conclude as the ultimate in nonsense. Certainly Guattari, a lifelong rebel (whose early support of the Algerian independence and of reform of authoritarian mental institutions was admirable) rebelled even against the revolutionary sects he joined. His raging against the Oedipus complex seems to betray a major one of his own. Guattari was much wilder and sloppier in his writing than Deleuze, and the latter permitted much looser and free-associative formulations in joint productions written with his companion. However, Deleuze also wrote some seven academic books on various philosophers, such as Leibniz, Spinoza, Kant and Nietzsche, that Sokal and Bricmont do not discuss. For instance, Deleuze's book on Leibniz, The Fold, contains references to topology (the mathematics of continuity), the use of which by postmodernists Sokal and Bricmont descry. Since Leibniz was an inventor of both the calculus and analysis situs (precursor of topology) and made the principle of continuity central to his philosophy, these references are not guilty of the irrelevance of which Sokal and Bricmont accuse Deleuze's other references to mathematics.

One claim that Sokal and Bricmont make throughout their work is that if the authors they criticize and expose are using scientific metaphors to illustrate their philosophical, psychological or literary ideas, these would not be illuminating to an audience ignorant of science. They suggest these scientific or mathematical examples are simply added to impress the scientifically illiterate literateurs. This may be the case with some of the phrases of Kristeva and Lacan. However, another use of scientific and mathematical concepts in philosophy is as models for metaphysical speculation. Since much of our thinking is based on images and spatial diagrams (following Kant but pace Hegel, Wittgenstein, and others), the precise, worked-out structures of mathematics and physics can suggest metaphysical models. Here the mathematical models are not window-dressing to impress the ignorant, but sources of admittedly vaguer metaphysical extrapolations. Deleuze, in a manner similar to (though nowhere as ably done as) Whitehead, mathematical structures are used as models for metaphysical ideas. Sokal and Bricmont do not totally reject philosophical thinking or even metaphysics, as they present some philosophy of science in order to set aside skepticism and to argue against relativism and subjectivism.

Sokal and Bricmont do comment on two of Deleuze's serious works. Bricmont also mentions, in an open letter concerning the dropping of Bergson from the English edition of the book that Bergson's influence on Deleuze shows the relevance of the former. Evidently Anglo-American analytic philosophers convinced Sokal and Bricmont to ignore Bergson in the English edition, though several English books on Bergson have recently appeared. Ironically, two of the passages in Deleuze that they ridicule assert that relativity theory, measurement in quantum theory, and information in statistical mechanics should not be interpreted subjectively.(pp. 14-150). This agrees with Sokal and Bricmont's own position, but they do not note this. It would spoil the fun.

Sokal and Bricmont hold up for ridicule selective passages in Deleuze's Difference and Repetition concerning the differential calculus. They quote long passages, followed by the remark that the passage is meaningless or nonsense (pp. 151-155). They claim that the problems of the calculus were solved by Cauchy in the early nineteenth century. (They even claim that the problems "were solved by the work of d'Alembert around 1760," (p. 151) though d'Alembert did not clarify in terms of inequalities or explicitly apply the limit concept that he advocates in the Encyclopedia.) They claim the status of the infinitesimals in the derivative is no longer worth bothering about, as it has been replaced by the limit.

Sokal and Bricmont's comments on Deleuze on the calculus resemble Bertrand Russell's comments on Zeno's paradoxes of motion. Russell claimed that the nineteenth century theory of real numbers and Weierstrass's "static theory of the variable" solves Zeno's paradoxes (and makes irrelevant the reflections on them of process philosophers like Bergson). But some later analytic philosophers noted that showing that mathematics is internally consistent hardly solves the physical version of Zeno's paradoxes. Unless one is willing to say that the mathematical structure (of all the real numbers) is physically existent, or one says that the mathematical formalism is all we need and that questions of physical reality should be rejected (a position that a scientific realist would have to reject) then there is still a physical problem of motion and infinitesimal processes, and the question of whether an infinite number of acts can be performed in a finite time. Similarly, Sokal and Bricmont, claim that the question of the status of the infinitesimal is eliminated by the limit notion. Sokal and Bricmont claim that Cauchy solved the problems of the status of infinitesimals with the concept of the limit and criticize Deleuze for puzzling over the status of differentials. If, indeed, the only consistent way to present derivatives were by reducing them to limits, this would be true. That is, if the infinitesimal has been reduced to a meaningless notational component of a ratio that is really a limit, then puzzling over the status of the infinitesimal in isolation is made obsolete. However Abraham Robinson's non-standard analysis (and Lawvere's less well known category theory approach) has shown how one can make direct mathematical sense of infinitesimal quantities without resorting to the replacement of their ratios by limits, and eliminating the individual differentials.

Deleuze seems to borrow some of his discussion from Hegel. Similar criticism to that of Sokal and Bricmont has been made of Hegel., claiming that Cauchy's formalization of the concept of limit has made all such discussion otiose. However some are beginning to reexamine Hegel's writing on the calculus with less dismissive attitudes than had Whitehead and Russell.

Marx also wrote philosophical discussions of the calculus. Edmund Wilson,, consulted a mathematician, who told him that Marx's comments on the calculus were worthless, and Wilson duly reported this. Some Marxist mathematicians, on the other hand, have defended the value of Marx's remarks on the calculus, even claiming he arrived at results similar to Cauchy. Marx's side-kick Friedrich Engels wrote far worse stuff concerning elementary algebraic operations and the dialectic. Would leftist Sokal move from a similar discussion of Marx and Engels on mathematics to discrediting Marx's insights about capitalism as Intellectual Impostures moves from Lacan's, Irigaray's or Kristeva's mathematical errors to question their honesty?

Sokal and Bricmont skip a number of linking passages in Deleuze's discussion, that treat in great detail writings of various mathematicians and philosophers. These include early nineteenth century figures such as the mathematician Wronski (a mathematician with whose Wronskian matrix they are undoubtedly familiar, but whose mysticism probably embarrasses them) and the philosopher Salomon Maimon. In one of the passages that they do quote, they omit by means of ellipsis the reference to Maimon and Wronski's philosophical approaches to the calculus, that would help make sense of some of the "nonsense" of the passage. Deleuze does not simply discuss the early nineteenth century debates on the "metaphysics of the calculus," but also uses twentieth century philosophers of mathematics, such as Albert Lautman who wrote in the 1930s and Jules Vuillemin, a contemporary analytic philosopher. Lautman, whose conception of mathematical problems Deleuze uses, had a correspondence with great logician Jacques Herbrand and the philosopher of mathematics Jean Cavaillès, and was praised in a commemorative volume by the mathematician Jean Dieudonné, suggesting that his understanding of logic and mathematics was taken seriously by his peers. Several French philosophers of mathematics were inspired to attempt to build on Lautman's approach to a logic of mathematical problems and interpretations because of Deleuze's lectures.

Obviously Deleuze is no mathematical virtuoso, but his treatment of the issues of the calculus is far more detailed, informed and serious than Sokal and Bricmont let on. For instance Sokal and Bricmont note in footnotes that some of Deleuze's errors are shared by Hegel, such as a dated treatment of functions in terms of Taylor series, but they neglect to note that Deleuze himself, in discussing Hegel mentions that he is well aware that the series approach to the calculus has been replaced in modern writers.

According to the standard account of the Sokal hoax, scientifically illiterate literary critics and sociologists have been bamboozled by the pretentious claims of French postmodernists concerning science. Ironically those same benighted scientific illiterates now have to take on faith the words of physicists Sokal and Bricmont concerning the errors of the French theorists. In some cases the great unwashed masses of humanists and social scientists may be misled again.

Val Dusek
Department of Philosophy
University of New Hampshire
Durham, NH 03824