Yet, despite these strictures, from The Birth of Tragedy onward, where he contrasts the shallow optimism of science to an alternative Dionysiac insight into the nature of things, Nietzsche will repeatedly oppose a vision of ultimate reality to accepted truths. Indeed, in The Birth of Tragedy he employs the Kantian concept of the noumenal to illustrate precisely this opposition: 'The contrast of this authentic nature-truth and the lies of culture which present themselves as the sole reality is similar to that between the eternal core of things, the thing-in-itself, and the entire world of appearances.' ¹⁶ In general, Nietzsche's critique of metaphysics, and his denial of the ability of philosophy to establish epistemological criteria, drives him towards an idealism which argues that the structures of knowledge are entirely constitutive of the object, while his insistence that all consciousness should comprehend itself as perspectival pushes him back towards a reinstatement of the distinction between appearance and reality.

I would argue that a similar dilemma, encapsulated in Nietzsche's dictum that 'Knowledge and Becoming exclude one another', ¹⁷ pervades the work of those post-structuralist thinkers who have been most directly influenced by Nietzschean schemas. We have already examined how Lyotard's motif of the libidinal band, which fuses a Freudian-inspired theory of cathexis with the doctrine of the Eternal Return, makes possible a denunciation of all theoretical discourses as 'apparatuses for the fixation and draining away of intensity'. ¹⁸ Lyotard, however, is too conscientious -- and too restless -- a figure to be satisfied for long with the monistic metaphysics of libido on which Économie Libidinale relied. It can be no accident that, shortly after the publication of this work, he began to set off in a new direction, replacing the description of forms of discourse as 'dipositifs pulsionels' with the less ontologically loaded notion of 'language-games', borrowed from Wittgenstein. In Lyotard's case, the attempt to develop a critique of objectifying theory from the standpoint of an ontology of flux represents an explicit, but only temporary, phase of his thought. With Foucault, however, the tension which this attempt implies is both a more covert, but also a more persistent, feature of his work. It is already apparent in Madness and Civilization, where Foucault wishes to develop a critique of the objectifying and alienating nature of modern psychiatric treatment and its theorizations, while also being sensitive to the difficulty of appealing to the 'rudimentary movements of an experience' which would be 'madness itself'. ¹⁹ In The Archaeology of Knowledge Foucault renounces this approach: 'We are not trying to reconstitute what madness itself might be . . . in the form in which it was later organized (translated, deformed, travestied, perhaps even repressed) by discourses, and the oblique, often twisted play of their

-53-