by Alan Sokal
Science and technology have long been the subject of philosophical and political debates: on nuclear weapons and nuclear energy, the human genome project, sociobiology, and many other subjects. But these debates in no way constitute a ‘science war’. Indeed, many different reasonable positions in these debates are advocated by scientists and non-scientists alike, using scientific and ethical arguments that can be rationally evaluated by all the people involved, whatever their profession.
Unfortunately, some recent developments may lead one to fear that something completely different is going on. For example, researchers in the social sciences can legitimately feel threatened by the idea that neurophysiology and sociobiology will replace their disciplines. Similarly, people working in the natural sciences may feel under attack when Feyerabend calls science a ‘particular superstition’241 or when some currents in the sociology of science give the impression of placing astronomy and astrology on the same footing.242
In order to alleviate these fears, it is worth distinguishing between the claims made for research programmes, which tend to be grandiose, and the actual accomplishments, which are generally rather modest. The basic principles of chemistry are today entirely based on quantum mechanics, hence on physics; and yet, chemistry as an autonomous discipline has not disappeared (even if some parts of it have gotten closer to physics). Likewise, if one day the biological bases of our behavior were sufficiently well understood to serve as a foundation for the study of human beings, there would be no reason to fear that the disciplines we now call ‘social sciences’ would somehow disappear or become mere branches of biology.243 In a similar way, scientists have nothing to fear from a realistic historical and sociological view of the scientific enterprise, provided that a certain number of epistemological confusions are avoided.244
Let us therefore put aside the ‘science war’, and see what kind of lessons can be drawn from the texts cited in this book concerning the relationship between the natural and the human sciences.245
1. It’s a good idea to know what one is talking about
Anyone who insists on speaking about the natural sciences – and nobody is forced to do so – needs to be well informed and to avoid making arbitrary statements about the sciences or their epistemology. This may seem obvious, but as the texts gathered in this book demonstrate, it is all too often ignored, even (or especially) by renowned intellectuals.
Obviously, it is legitimate to think philosophically about the content of the natural sciences. Many concepts used by scientists – such as the notions of law, explanation and causality – contain hidden ambiguities, and philosophical reflection can help to clarify the ideas. But, in order to address these subjects meaningfully, one has to understand the relevant scientific theories at a rather deep and inevitably technical level;246 a vague understanding, at the level of popularizations, won’t suffice.
2. Not all that is obscure is necessarily profound
There is a huge difference between discourses that are difficult because of the inherent nature of their subject and those whose vacuity or banality is carefully hidden behind deliberately obscure prose. (This problem is by no means specific to the humanities or social sciences; many articles in physics and mathematics use a language more complicated than is strictly necessary.) Of course, it is not always easy to determine which kind of difficulty one is facing; and those who are accused of using obscure jargon frequently reply that the natural sciences also use a technical language that can be mastered only after many years of study. Nevertheless, it seems to us that there are some criteria that can be used to help distinguish between the two sorts of difficulty. First, when the difficulty is genuine, it is usually possible to explain in simple terms, at some rudimentary level, what phenomena the theory is examining, what are its main results and what are the strongest arguments in its favour.247 For example, although neither of us has any training in biology, we are able to follow, at some basic level, developments in that field by reading good popular or semi-popular books. Second, in these cases there is a clear path – possibly a long one – that will lead to a deeper knowledge of the subject. By contrast, some obscure discourses give the impression that the reader is being asked to make a qualitative jump, or to undergo an experience similar to a revelation, in order to understand them.248 Again, one cannot help being reminded of the emperor’s new clothes.249
3. Science is not a ‘text’
The natural sciences are not a mere reservoir of metaphors ready to be used in the human sciences. Non-scientists may be tempted to isolate from a scientific theory some general ‘themes’ that can be summarized in few words such as ‘uncertainty’, ‘discontinuity’, ‘chaos’ or ‘nonlinearity’ and then analysed in a purely verbal manner. But scientific theories are not like novels; in a scientific context these words have specific meanings, which differ in subtle but crucial ways from their everyday meanings, and which can only be understood within a complex web of theory and experiment. If one uses them only as metaphors, one is easily led to nonsensical conclusions.250
4. Don’t ape the natural sciences
The social sciences have their own problems and their own methods; they are not obliged to follow each ‘paradigm shift’ (be it real or imaginary) in physics or biology. For example, although the laws of physics at the atomic level are expressed today in a probabilistic language, deterministic theories can nevertheless be valid (to a very good approximation) at other levels, for example in fluid mechanics or even possibly (and yet more approximately) for certain social or economic phenomena. Conversely, even if the fundamental physical laws were perfectly deterministic, our ignorance would force us to introduce a great number of probabilistic models in order to study phenomena at other levels, like gases or societies. Besides, even if one adopts a reductionist philosophical attitude, one is by no means obliged to pursue reductionism as a methodological prescription.251 In practice, there are so many orders of magnitude separating atoms from fluids, brains or societies that vastly different models and methods are quite naturally employed in each realm, and establishing a link between these levels of analysis is not necessarily the most urgent task. In other words, the type of approach in each domain of research should depend upon the specific phenomena under investigation. Psychologists, for example, do not need to invoke quantum mechanics to maintain that in their field ‘the observer affects the observed’; this is a truism, irrespective of the behaviour of electrons or atoms.
Moreover, there are so many phenomena, even in physics, that are imperfectly understood, at least for the time being, that there is no reason to try to imitate the natural sciences when dealing with complex human problems. It is perfectly legitimate to turn to intuition or literature in order to obtain some kind of non-scientific understanding of those aspects of human experience that cannot, at least at present, be tackled more rigourously.
5. Be wary of argument from authority
If the human sciences want to benefit from the undeniable successes of the natural sciences, they need not do so by directly extrapolating technical scientific concepts. Instead, they could get some inspiration from the best of the natural sciences’ methodological principles, starting with this one: to evaluate the validity of a proposition on the basis of the facts and reasoning supporting it, without regard to the personal qualities or social status of its advocates or detractors.
This is, of course, only a principle; it is far from universally honoured in practice, even in the natural sciences. Scientists are, after all, human beings and are not immune to fashion or to the adulation of geniuses. Nevertheless, we have inherited from the ‘epistemology of the Enlightenment’ a totally justified mistrust toward the exegesis of sacred texts (and texts that are not religious in the traditional sense may very well fulfil that role) as well as toward argument from authority.
We met in Paris a student who, after having brilliantly finished his undergraduate studies in physics, began reading philosophy and in particular Deleuze. He was trying
to tackle Difference and Repetition. Having read the mathematical excerpts examined here (pp. 151–155), he admitted he couldn’t see what Deleuze was driving at. Nevertheless, Deleuze’s reputation for profundity was so strong that he hesitated to draw the natural conclusion: that if someone like himself, who had studied calculus for several years, was unable to understand these texts, allegedly about calculus, it was probably because they didn’t make much sense. It seems to us that this example should have encouraged the student to analyse more critically the rest of Deleuze’s writings.
6. Specific scepticism should not be confused with radical scepticism
It is important to distinguish carefully between two different types of critiques of the sciences: those that are opposed to a particular theory and are based on specific arguments, and those that repeat in one form or another the traditional arguments of radical scepticism. The former critiques can be interesting but can also be refuted, while the latter are irrefutable but uninteresting (because of their universality). And it is crucial not to mix the two sorts of arguments: for if one wants to contribute to science, be it natural or social, one must abandon radical doubts concerning the viability of logic or the possibility of knowing the world through observation and/or experiment. Of course, one can always have doubts about a specific theory. But general sceptical arguments put forward to support those doubts are irrelevant, precisely because of their generality.
7. Ambiguity as subterfuge
We have seen in this book numerous ambiguous texts that can be interpreted in two different ways: as an assertion that is true but relatively banal, or as one that is radical but manifestly false. And we cannot help thinking that, in many cases, these ambiguities are deliberate. Indeed, they offer a great advantage in intellectual battles: the radical interpretation can serve to attract relatively inexperienced listeners or readers; and if the absurdity of this version is exposed, the author can always defend himself by claiming to have been misunderstood, and retreat to the innocuous interpretation.
How did we get here?
In the debates that followed the publication of the Social Text parody, we were often asked: How and why did the intellectual trends that you are criticizing develop? This is a very complicated question belonging to the history and sociology of ideas, to which we certainly do not claim to have a definitive answer. We would like, rather, to put forward some possible answers, while emphasizing both their conjectural nature and their incompleteness (there are undoubtedly other elements that we have underestimated or missed entirely). Moreover, as always in this kind of complex social phenomenon, there is a mixture of very diverse causes. In this section we shall limit ourselves to the intellectual sources of postmodernism and relativism, leaving the political aspects for the next section.
1. Neglect of the empirical
For a long time, it has been fashionable to denounce ‘empiricism’; and if that word denotes an allegedly fixed method for extracting theories from facts, we can only agree. Scientific activity has always involved a complex interplay between observation and theory, and scientists have known that for a long time.252 So-called ‘empiricist’ science is a caricature belonging to bad schoolbooks.
Nevertheless, our theories about the physical or social world need to be justified in one way or another; and if one eschews apriorism, argument from authority and reference to ‘sacred’ texts, there is not much left besides the systematic test of theory by observations and experiments. One need not be a strict Popperian to realize that any theory must be supported, at least indirectly, by empirical evidence in order to be taken seriously.
Some of the texts cited in this book completely disregard the empirical aspect of science and concentrate exclusively on language and theoretical formalism. They give the impression that a discourse becomes ‘scientific’ as soon as it seems superficially coherent, even if it is never subjected to empirical tests. Or, worse, that it is sufficient to throw mathematical formulae at problems in order to make progress.
2. Scientism in the social sciences
This point may seem bizarre: isn’t scientism the sin of physicists and biologists who seek to reduce everything to matter in motion, natural selection and DNA? Yes and no. Let us define ‘scientism’, for the purposes of this discussion, as the illusion that simplistic but supposedly ‘objective’ or ‘scientific’ methods will allow us to solve very complex problems (other definitions are certainly possible). The difficulty that constantly arises when one succumbs to such illusions is that important parts of reality are forgotten simply because they fail to fit within the framework that was posed a priori. Sadly, examples of scientism are abundant in the social sciences: one can cite, among others, certain currents within quantitative sociology, neoclassical economics, behaviourism, psychoanalysis and Marxism.253 Often what happens is that people start with a set of ideas having some validity in a given domain and, instead of trying to test them and refine them, they extrapolate them unreasonably.
Unfortunately, scientism has often been confused – by its supporters as well as by its detractors – with the scientific attitude itself. As a result, the entirely justified reaction against scientism in the social sciences has sometimes led to an equally unjustified reaction against science as such – and this on the part of both the ex-partisans and ex-opponents of the old scientisms. For example, in France after May 1968, the reaction against the scientism of certain rather dogmatic strains of structuralism and Marxism was one factor (among many others) that led to the emergence of postmodernism (the ‘incredulity toward metanarratives’, to quote Lyotard’s famous catchword).254 A similar evolution occurred, in the 1990s, among some intellectuals in the former Communist countries: for instance, the Czech president Václav Havel wrote that
The fall of Communism can be regarded as a sign that modern thought – based on the premise that the world is objectively knowable, and that the knowledge so obtained can be absolutely generalized – has come to a final crisis.
(Havel 1992, p. E15)
(One wonders why a renowned thinker such as Havel is incapable of making the elementary distinction between science and the Communist regimes’ unjustified claim to possess a ‘scientific’ theory of human history.)
When one combines neglect of the empirical side with a good deal of scientistic dogmatism, one can be led into the worst lucubrations, of which we have seen all too many examples. But one can alternatively fall into a sort of discouragement: since such and such (simplistic) method, to which one had dogmatically adhered, does not work, therefore nothing works, all knowledge is impossible or subjective, etc. And so one passes easily from the climate of the 1960s and 1970s to postmodernism. But it is based on a misidentification of the source of the problem.
One recent avatar of scientism in the social sciences is, paradoxically, the ‘strong programme’ in the sociology of science. To try to explain the content of scientific theories without taking into account, even in part, the rationality of scientific activity is to eliminate a priori an element of reality and, it seems to us, to deprive oneself of any possibility of effectively understanding the problem. To be sure, every scientific study must make simplifications and approximations; and the approach of the ‘strong programme’ would be legitimate if its advocates were to provide empirical or logical arguments showing that the neglected aspects are indeed of marginal importance for understanding the phenomena in question. But no such arguments are given; the principle is posed a priori. In reality, the strong programme is trying to make a virtue of (apparent) necessity: since it is difficult for sociologists to study the internal rationality of the natural sciences, it is declared ‘scientific’ to ignore it. It is like trying to complete a puzzle when one knows that half the pieces are missing.
We believe that the scientific attitude, understood very broadly – as a respect for the clarity and logical coherence of theories, and for the confrontation of theories with empirical evidence – is as relevant in the social sciences as it is in the natural
sciences. But one must be very prudent toward claims of scientificity in the social sciences; this holds also (or even especially) for the currently dominant trends in economics, sociology and psychology. The problems addressed by the social sciences are extremely complex, and the empirical evidence supporting their theories is often quite weak.
3. The prestige of the natural sciences
There is no doubt that the natural sciences enjoy an enormous prestige, even among their detractors, because of their theoretical and practical successes. Scientists sometimes abuse this prestige by displaying an unjustified feeling of superiority. Moreover, well-known scientists, in their popular writings, often put forward speculative ideas as if they were well established, or extrapolate their results far beyond the domain where they have been verified. Finally, there is a damaging tendency – exacerbated, no doubt, by the demands of marketing – to see a ‘radical conceptual revolution’ in each innovation. All these factors combined give the educated public a distorted view of scientific activity.
But it would be demeaning to philosophers, psychologists and sociologists to suggest that they are defenceless in the face of such scientists, and that the abuses exposed in this book are somehow unavoidable. It is obvious that no one, and in particular no scientist, forced Deleuze or Lacan to write as they do. One can perfectly well be a psychologist or a philosopher and either speak about the natural sciences knowing what one is talking about, or else not speak about them and concentrate on other things.