Against Method

Paul Feyerabend‘s Against Method (originally published in 1975) is another one of those books I have been meaning to read for years, but never got around to before now. Feyerabend (1924-1994) was a philosopher of science, famous (or notorious) for his “epistemological anarchism,” his insistence that “the only principle” that can be justified for scientific research is that “anything goes.” I’ve turned to him now, partly out of my interest in science studies, and partly because I’m supposed to give a talk in a few months at a symposium on “foundations and methods in the humanities,” a task I am finding difficult because I have no belief in foundations, and little use for methodologies.
Feyerabend critiques and rejects the attempt — by philosophers of science, primarily, but also by popular apologists for science, and sometimes by scientists themselves — to establish norms and criteria to govern the way science works, and to establish what practices and results are valid for scientific research. Feyerabend’s particular target is Karl Popper’s doctrine of “falsification,” but more generally he opposes any a priori attempt to legislate what can and cannot be done in science.
Feyerabend’s argument is partly “deconstructive” (by which I mean he showed how rationalist arguments were necessarily internally inconsistent and incoherent — though he does not seem to have much use for Derridean deconstruction as a philosophy), and partly historical and sociological. He argues that actual scientific practice did not, does not, and indeed cannot, make use of the rationalist norms that philosophers of science, and ideologists of science, have proclaimed. He analyzes Galileo’s defense of heliocentrism at great length, and shows that Galileo’s arguments were riddled with non sequiturs, loose analogies, ad hoc assumptions, self-contradictory and easily falsifiable assertions, rhetorical grandstanding, and so on. The point is not to undermine Galileo, or to assert that there are no grounds for choosing between seeing the earth and the sun as the center. Rather, Feyerabend wants to show that such (disreputable) strategies were strictly necessary; without them, Copernicus and Galileo never could have overthrown the earth-centered view, which had both the theoretical knowledge and the “common sense” of their time, as well as the authority of the Church, on their side. It was not a matter of a “more accurate” theory displacing a less accurate one; but rather, a radical shift of paradigms, one which could only be accomplished by violently disrupting both accepted truths and accepted procedures. It is only in the hundreds of years after Galileo convinced the world of the heliocentric theory, that the empirical evidence backing up the theory was generated and catalogued.
Feyerabend is drawing, of course, on Thomas Kuhn’s work on “paradigm shifts,” but he is pushing it in a much more radical direction than Kuhn would accept. Kuhn distinguishes between “normal science,” when generally accepted research programs and paradigms are in place, and rationalistic criteria do in fact function, and times of crisis, when paradigms break down under the weight of accumulating anomalies, thus forcing scientists to cast about for a new paradigm. For Feyerabend, however, there is no “normal science.” There was no crisis, or weight of anomalies, that forced Copernicus and then Galileo to cast about for a new astronomical paradigm; it would be more to the point to say that Galileo deliberately and artificially created a crisis, in order to undermine a paradigm that was generally accepted and that worked well, and put in its place a new paradigm that he supported more out of passion and intuition than out of anything like solid empirical evidence. Because “facts” are never independent of social contexts and theoretical assumptions, Galileo could not have provoked a shift in the theoretical assumptions of his time merely by appealing to what were understood then as the “facts.”
Such an argument was quite shocking in 1975. It has become much less so in the years since, as rhetorical theorists, sociologists, and others in “science studies” have studied in great depth the way science actually works, and have contested many other instances of (capital-S) Science and (capital-R) Reason on historical and sociological grounds.
There remains a subtle but important difference, however, between Feyerabend and more recent science studies historians and thinkers like Bruno Latour, Stephen Shapin, Steve Fuller, and many others. Feyerabend justifies his “epistemological anarchism” on the ground that it is necessary for the actual, successful practice of science, and indeed for the “progress” of science — though he explicitly refuses (page 18) to define what he means by “progress.” What this means is that Feyerabend opposes methodological norms and fixed principles of validation largely on pragmatic grounds : which I do not think is quite true of Latour et al. Where Latour sees a long process of negotiation, and a “settlement,” between Pasteur and the bacilli he was studying, Feyerabend doesn’t see Galileo (or Einstein, for that matter) in engaging in any such process vis-a-vis the earth, or the sun, or the universe. Instead, he sees them as blithely ignoring rules of evidence and of verification or falsification, in order to impose radically new perspectives (less upon the world than upon their own cultures). Galileo’s and Einstein’s justification is that their proposals indeed worked, and were accepted; this is what separates them from crackpots, though no criteria existed that could have assured these successes in advance.
What I don’t see enough of in contemporary science studies — though one finds it in Deleuze and Guattari, in Isabelle Stengers, and in the work of my friend Richard Doyle — is Feyerabend’s sense of the kinship between scientific and aesthetic creativity, in that both are engaged in creating the very criteria according to which they will be judged.
More generally, Feyerabend, like Latour and other more recent science studies thinkers, is deeply concerned with democracy, and with the way that the imperialism of Big Science threatens democracy by trying to decree that its Way is the Only Way. Indeed, one probably sees more of this threat today — in the “science wars” that reached a flash point in the mid 1990s, but that are still smouldering, in the popularization of science, and in the pronouncements of biologists like Richard Dawkins and Edward O. Wilson, and physicists like Steven Weinberg and Alan Sokal — than one did when Feyerabend was writing Against Method. But Feyerabend wisely refuses to get lost (as I fear Latour does at times) in the attempt to propose an alternative “settlement” or “constitution” to the one that Big Science has proclaimed for itself. Feyerabend’s genial anarchism, pluralism, and “relativism” (a term he accepts, but only in certain carefully outlined contexts) simply precludes the need for any single alternative account, such as the one Latour struggles to provide. Finally, for Feyerabend, there is no such thing as Science; we should rather speak of the sciences, as a multitude of often conflicting and contradictory practices, none of which can pretend to ultimate authority, and all of which have to be judged and dealt with according to a range of needs, interests, and contexts.
Pluralism is often derided as wishy-washy, wimpy, “soft,” unwilling to take a stand. None of this is true of Feyerabend’s pluralism, though I am not sure how much of his exemption from such charges is due to the rigor of his arguments, and how much to the charm of his rhetorical style — he’s an engaging, inviting, and unaffected writer, able to be clear and focused without becoming simplistic, and able to argue complexly without becoming abstruse. Of course, the attempt to separate logical rigor from stylistic effects is precisely the sort of pseudo-rational distinction that Feyerabend is continually warning us against.