Fashionable Nonsense [A book review]
Post of the Month: November 1998
by Richard Harter
Fashionable Nonsense, Postmodern Intellectuals' Abuse of Science, Alan Sokal and Jean Bricmont, Picador (St. Martins), 1998, ISBN 0-312-19545-1, 300pp, hardcover.
This is the English version of Impostures Intellectuelles, originally published in French in France 1997. Elsewhere I have commented on Richard Dawkins' extended review that appeared in Nature.
This work was inspired by the notorious Sokal hoax, an article consisting of a collection of physics gibberish liberally salted with quotes from sundry post modernist authors, which ran in the "Science Wars" issue of the journal Social Text. In the course of constructing the hoax article Sokal collected quite a number of quotations from "post modernist" authors, mostly French intellectuals.
In FN (Fashionable Nonsense) Sokal and Bricmont (both Physicists) take a sardonic look at what they call the abuse of Science by said intellectuals. In the course of doing so they raise issues about attitudes towards Science in American academia. Their "expose" has been enthusiastically greeted and enthusiastically damned. The back cover has the following quote from Barbara Ehrenreich:
"Take the most hallowed names in current French theoretical thinking, divide by one of the sharpest and most irreverent minds in America, multiply by a half dozen examples, render in good clear English - and you have a thoroughly hilarious romp through the postmodernist academy. Two years ago, Sokal struck a devastating blow against intellectual obscurantism with his famous Social Text parody, and Fashionable Nonsense delivers the perfect coup de grace."
The book contains an introduction, dissections of Jacques Lacan, Julia Kristeva, Luce Irigaray, Bruno Latour, Jean Baudrillard, Gilles Deleuze and Felix Guatari, Paul Virilo, two intermezzos - "Epistemic Relativism in the Philosophy of Science" and "Chaos theory and 'Postmodern Science'" -, a chapter on abuse of Godel's theorems, a summary epilogue, and three appendices - the original hoax article, some explanatory comments of the jokes, and an afterword. [The above paragraph is laid out more neatly in the table of contents.]
In the introduction Sokal and Bricmont present counters to objections which they feel that the reader might raise. They consider the following objections:
What is not considered is the possibility that they (the authors) might simply have been wrong, either because they have completely missed the point or because they have confused jargon from another discipline with particular usage in science. They also make no real allowance for usage that reflects popularizations of Science rather than Science itself.
Popularizations of Science may be a cure worse than the disease. There are a number of fields of Science which are badly popularized and are quite fashionable - they include Godel's theorems, relativity, quantum mechanics (particularly the uncertainty principle), chaos theory, and catastrophy theory. As such, these fields (often bastardized) are part of the intellectual culture.
In considering these fragments of text we have to consider (more precisely, we don't have to - and some won't - but we should IMO) the possible context, how much of it is an ill-digested bit of pop science, how much is a fundamental misunderstanding, and the extent to which errors are intrinsic to the argument. There is a great tendency in FN to condemn out of hand anything that is not by the numbers, i.e., anything where the usage is not precise (or is not carefully delimited by definition.) Quite often, little effort seems to have been made to determine the sense of what is being said; in consequence there are a number of misreadings in FN of quoted passages.
As a note, FN has picked up the deplorable postmodern vice of speaking of "texts" and "discourses", terms which are not wrong as such, but which are prefatory to postmodernist linguistic abuses.
Most of the remainder of this article consists of chapter notes. It is not particularly well organized for which my apologies. I pick a fair number of nits; this should not be construed. On the whole the critiques offered are cogent and to the point. Since the book is jointly authored I refer throughout to FN rather than presuming that one author or the author is the originator of cited passages. There is a final summary at the end.
A fair amount of space is dedicated to Lacan who has, I am given to understand, quite a reputation as a psychoanalytic theorist. Judging from the cited passages Lacan is a premier example of the abusing Mathematics. It is not just that he uses mathematical terms and imagery - he misuses them and does so both badly and blatantly. Some of the passages may be jokes, e.g., the "derivation" of the phallus as the square root of minus one. It might seem that he is using terms such as topology and compact space as analogous structures but he regularly denies that he is doing so.
FN notes Lacan abuses mathematical logic less than other fields. Oddly enough it is here that the critique is weakest - FN misreads some of the quoted passage, misses some of the errors, and gives wrong counter-examples.
The chapter on Julia Kristeva is devoted to her early work in which she makes some fanciful sorties into mathematical logic much in the style of a mad poet. The object of her investigation is to ground poetic language in set theory; the result is neither mathematics nor poetry.
There is a misreading of the passage quoted on pages 39-40; it is clear from the text that Kristeva is using to mean finiteness, 1 to mean aleph-1 (countable infinities) and 2 as aleph-2 (the power of the continuum assuming the continuum hypothesis.) This is a confusing (and IMO silly) thing to do but the text is clear.
That said, Kristeva's usage in the passage quoted on pp 41-42 is very bad. She is confusing "next larger" and "outside the sequence". FN comments on the passage quoted on pp 42-43 "These paragraphs are meaningless ...". FN errs; the passage is meaningful but it is wrong.
On p45 FN makes the point that texts are finite in length and that the set of all possible texts is a countable infinity. FN remarks "It is hard to see how the continuum hypothesis, which concerns nondenumerable infinite sets, could have any application in linguistics." Au contraire, it is easy to see the thought - the rationale is that the meaning of a text cannot be fully established by any text or finite sequence of texts, ergo (!!) the "meaning" of a text is an infinite text (countably infinite). A given finite length text has, on this interpretation, a nondenumerable number of potential "meanings".
Kristeva's musings on Marx and set theory beggar description.
In summary, Kristeva has the germ of an interesting (but probably unsound) approach; she lacks the mathematical competence to undertake what she attempts.
This chapter is a survey of modern notions in the philosophy of science with a particular emphasis on those authors, e.g. Kuhn and Feyerbrand, who are regularly quoted in support of epistic relativism in the philosophy of science. The discussion is competent and interesting. FN makes the interesting observation that Kuhn, et al, write in a manner that lends itself to a dual interpretation, one moderate and one radical, the latter being seized upon by the "postmodernists".
Luce Irigaray writing's deal with a wide variety of topics, e.g. psychoanalysis, linguistics, and the philosophy of science. At the beginning of the chapter FN quotes her as follows:
Every piece of knowledge is produced by subjects in a historical context. Even if that knowledge aims to be objective, even it its techniques are designed to ensure objectivity, science always displays certain choices, certain exclusions, and these are particularly determined by the sex of the scholars involved.
This is an interesting and important thesis, one worthy of consideration in depth. Savor that paragraph - it is a small island of reason and sanity. The rest of the material quoted from Irigaray runs thick and deep and is filled with effluent. Perhaps Irigaray has written something worth writing but the stuff that FN dredges up is truly awful.
A real objection to Irigaray and those who follow in her footsteps is that her maunderings mask by obfuscation real questions, e.g., the "masculinity" of physics, the real tendency to treat the questions that one can answer as the only meaningful questions, and the effect of preferred metaphors.
The sociologist, Bruno Latour, is well known for his work, Science in Action. The chapter on Latour concentrates on a less well known article, a semiotic analysis of the theory of relativity. It demonstrates fairly convincingly that Latour doesn't understand the theory of relativity and, as a consequence, his analysis is a mishmash of error.
Towards the end of the chapter FN presents the following passage from Latour:
First, the opinions of scientists about science studies are not of much importance. Scientists are the informants for our investigations of science, not our judges. The vision we develop of science does not have to resemble what scientists think about science.
The last two sentences are plausible; the first, however, lays the groundwork for a fundamental error that dogs the strong programme and science studies, to wit the belief that one can understand social actions and psychology while remaining ignorant of the full environment of the subjects (scientists) being studied.
On p136 the following passage appears in FN about fractal dimensionality and catastrophe theory:
Like all scientific advances, they have provided new tools and focussed attention on new problems. But they have in no way called into question traditional scientific epistemology.
This is an overstatement, at least when one speaks of "all scientific advances". Science operates in part by the questions it does not consider and by restricting what are admissable as valid lines of reasoning. These restrictions change over time. Thus, arguments based on "innate tendencies" were once acceptable whereas statistical arguments were not. Science not only increases the stock of reliable knowledge over time; it also improves its methods for discerning reasonable knowledge.
In general the discussion of chaos theory (which is not a single subject) is not the happiest. There are outright errors. Thus on p174 FN gets the logistic equation wrong. It is given as
(1) dx/dt = r*x*(1-x)
whereas the equation actually is
(2) x(t+1) = r*x(t)*(1-x(t))
To be fair, FN attributes (1) to Verhulst (1838) and that may indeed be the equation Verhulst introduced. However (1) is not the logistic equation studied in biology and is not the equation whose behaviour was famously studied by May, et al. The difference is critical: Equation (1) is a differential equation; the solution, x(t), is a smooth function (the sigmoid function) in time. Equation (2) is a difference equation; the function may converge to a fixed value, oscillate between multiple values, or be chaotic, depending on the value of r.
There are other minor infelicities which are not, perhaps, to the point. The major point that is made, and it is made clearly, is there that there is a great deal of popular confusion about determinism, causality, predictability, and solvability and between the notions of linearity in mathematics and linear thought.
FN quotes a lengthy passage from Baudrillard and concludes with the remark:
The last paragraph is Baudrillardian par excellance. One would be hard pressed not to notice the high density of scientific and pseudo-scientific terminology - inserted in sentences that are, as far as we can make out, devoid of meaning.
This is, one supposes, correct: there is indeed a high density of said terminology and as far as the authors of FN can make out, devoid of meaning. In truth, Baudrillard does use a lot of terminology that is either techno-babble or which makes it eminently clear that he has a less than perfect grasp of theories he is appealing to. (I was particularly charmed by the notion that a fixed point attractor is a strange attractor!) However matters are not so simple.
It is fairly clear (or so it seems to me) that Baudrillard has some grasp of the relevant aspects of chaos theory and has applied them more or less correctly to his chosen topic, "the end of history". A discussion of this passage in detail is beyond the scope of this article; I may discuss it elsewhere.
The final paragraph of this section begins:
In summary, one finds in Baudrillard's works a profusion of scientific terms, used with total disregard for their meaning and, above all, in a context where they are manifestly irrelevant. Whether or not one interprets them as metaphors, it is hard to see what role they could play, except to give an appearance of profundity to trite observations about sociology or history....
The first sentence is clearly true; Baudrillard's "science" reads like star-trek fan fiction written by an eight grader. (This is unfair - the eight grader will produce pseudoscience with more verisimilitude of style.) The second sentence, however, reflects what I feel (perhaps unfairly) is a fundamental bias in FN. Despite the protestations in the introduction and elsewhere that the authors are not passing judgement on the non-scientific content of the authors they consider they do pass judgement. The judgement is: If the science is bad, the work is bad; mis-users of science are an intellectual imposters, using babble to cover up the lack of depth in their thought.
In the beginning of the chapter FN remarks of them, "In our opinion, the most plausible explanation is that these authors possess a vast but very superficial erudition, which they put on display in their writings." This is both profoundly right and profoundly wrong. The authors in question (and indeed many of the authors discussed) have a vast erudition of varied depth. The depth, however, lies in an entire philosophic tradition, a large body of writings, complete with jargon, forms of expression, and indirect references to issues considered at depth by prior authors. The shallowness lies in their understanding of science, its jargon and viewpoints. The commentary in FN at times is an excellent example of Kuhn's description of people with different paradigms talking past each other.
Thus in pp 155-158 there is an extended passage from Deleuze which, at first sight to anyone with a scientific background (or a background in analytic philosophy) is nothing but meaningless gibberish. Yet, on a careful reading, there is a clear and distinct thread of meaning. The problem is that Deleuze is using words and concepts that are clear to him and to someone familiar with the tradition that he is writing from but which are quite obscure to anyone not familiar with that tradition.
On the other hand pp 159-166 has a passage in which Deleuze considers ancient difficulties in the foundations of Calculus which is quite revealing. He makes a jaw dropping error which, FN notes, is a repetition of an error by Hegel. What is going on here is that Deleuze is considering the philosophic problem of infinitesimals. When calculus as a subject was young this was a live issue which was discussed communally by philosophers and mathematicians. Over time two traditions developed. As FN notes, the issue was resolved in the mathematical community by Cauchy's theory of limits, et al, and the Calculus was placed on a rigorous foundation. As FN does not note, Cauchy's treatment does not solve the philosophic problem but rather it eliminates the issue from needing to be considered in Mathematics.
Be that as it may it would seem that Deleuze and perhaps many of the other philosophers in continental philosophy are dangerously insular with respect to science.
Pp 166-168 has a passage from Guatarri; my notes read "Eeek!!". I see no reason to reread it.
Virillo purportedly writes on the philosophy of speed. It is hard to take seriously a philosopher of speed (except, perhaps, the chemical kind) who confuses acceleration and speed.
In this chapter FN has fun with Debray who propounds the following:
The secret takes the form of a logical law, an extension of Goedel's theorem: here can be no organized system without closure and no system can be closed by elements internal to that system alone.
There seem to have been a few details missing in my copy of Goedel. The quotation from Badiou on p181 is priceless.
The epilog is a summary discussion. It is frankly speculative and is an interesting take on American "postmodernism". In its way, though, it highlights an essential weakness of the book. The authors say in the preface that they have a dual agenda. One is to expose the "fashionable nonsense" in the European "deep thinkers". This they pursue with great vigor throughout the course of the book. The second objective is:
A second target of our book is epistemic relativism, namely the idea - which, at least when expressed explicitly, is much more widespread in the English-speaking world than in France - that modern science is nothing more than a "myth", a "narration" or a "social construction" among many others.
It is explicitly recognized that the connection between the two targets is indirect, being principally that the jargon of French Literary Theory is fashionable and that an entire brigade of French camels are poking their noses up under the edges of the tent of postmodernism.
The weakness of the book is that it never comes to grips with the second target except in the most general of terms nor does it really connect the two targets.
The appendices consist of the text of the original Sokal hoax, and two articles of comment. The hoax has discussed ad nauseum and needs no comment from me.
The book is interesting; some of the quoted material is hilarious - at least it is if one has a modicum of scientific literacy. One is left with the impression that the authors are a bit too arrogant, a bit too ready to insist on the literal use of their preferred jargon, and a bit too literal in their reading of the passages that they quote. On the other hand (surely someone has remarked that a one handed philosopher would be a boon) their critiques of their chosen targets are on the mark.
I get the impression that the relationship of French Literary Theorists and Science is much like the attempts of Westerners to assimilate Eastern religion. The ideas are partially absorbed and woefully misunderstood; the result is something strange and wonderful.
As a counterpoint to this book I will commend Science Wars. This is a collection of the articles that appeared in the issue of Social Text in which the Sokal hoax article appeared, plus a number of additional articles. The articles are written in clear English (no fusty academic obscurantism). Social Text is the premier journal (so I am given to understand) of Cultural Studies. You may judge for yourself how apropos the critiques of Fashionable Nonsense and Higher Superstition are.
Richard Harter, firstname.lastname@example.org,
The Concord Research Institute
URL = http://www.tiac.net/users/cri, phone = 1-978-369-3911
If you can laugh at something it can't hurt you.
It can kill you but it can't hurt you.
First posted 18 November 1998
Home Page | Browse | Search | Feedback | Links