Abstract

In this commemorative article the significance of Paul Feyerabend's work for philosophy of science in general is reviewed. Its unifying perspective is identified as the fight against any possible constraint on imagination (i.e. on the capacity of generating alternatives). This alternative-maximizing search was already central in Feyerabend's 'pre-anarchistic' studies. In fact, I claim that the really significant theses and arguments, as far as the intrinsic debate within the philosophy of science is concerned, were present in these earlier studies (criticism of the conditions of consistency and meaning invariance, pragmatic theory of observation, incommensurability thesis ...). The context of the 'historical turn' is sketched in which these arguments were developed as aresponse to current views on science. But these arguments do possess an independent normative significance. As Feyerabend construes them however, this independence is uncertain, because it is the value of proliferating imaginative alternatives as such which partly conditionalizes the significance of these arguments (as it conditionalizes the very value of science itself). In this way the philosophy of the critical imagination could eventually turn the critical values against the philosophy of rational criticism itself. And indeed, (1)a certain combination of the tenets concerning consistency and meaning variance undermines the idea of crucial test, and (2) in Against Method xht panegyric of the imagination (couched in a new narrative of the history of science) is used as the main weapon to override any methodological account. Philosophy of science is now convicted of what is the only deadly sin — the murder of imagination. But science itself is also accused of the same crime: it has become not merely an ideology — which is nothing to worry about — but a totalitarian ideology, monopolizing epistemic and educational claims. I argue that, although the reintroduction within philosophy of science of such questions belonging to a philosophy of culture is called for as a complement for intrinsic approaches, the Feyerabendean way of treating those questions makes it impossible to state anything about the specificity of science within culture. On the other hand two 'master ideas' still seem apt to stimulate further discussion: (1) the critique of the consistency condition, backed by the attack on the thesis of the 'relative autonomy of facts', and (2) the defence of meaning variance, backed by the pragmatic theory of observation, and leading in turn to the idea of incommensurability. At the same time this idea of incommensurability could also play a role in a philosophy of the relation between science and culture (which remains to be developed)