I need more time to work through Graham Harman’s critique of certain aspects of my Whitehead book. But I think I can give a quick answer to his PS about Kant.
Graham quotes, against my reading of Whitehead as a kind of post-Kantian, Whitehead’s own assertion that Process and Reality involves “a recurrence to pre-Kantian modes of thought” (page xi). Throughout my book I am referring, instead, to a passage where Whitehead instead credits Kant as “the great philosopher who first, fully and explicitly, introduced into philosophy the conception of an act of experience as a constructive functioning” (page 156). This particular way of relating to Kant is crucial to my whole reading of Whitehead; it is why I position Whitehead as a kind of (radical) post-Kantian. Ultimately, which of these two sides of Whitehead’s attitude towards Kant one wishes to cite is a tactical decision. But in any case, I think that my Kant/Whitehead juxtaposition is more than what Harman dismisses as a “bats and birds both fly” argument.
I certainly do take Graham’s point when he says: “So, why do I choose to portray Kant as an enemy rather than an ally? Largely because of how Kant has been appropriated.” From an anti-correlationist position, Kant is indeed the source of the problem diagnosed by both Harman and Meillassoux, and it makes sense for them to move against him.
My own positioning of Kant as an ally has a different genealogy. It really comes out of the implicit “Kant vs Hegel” faultline in French philosophy. My very first book was on Blanchot and Bataille, and a big part of my argument (inherited from my initial advisor, Joseph Libertson) was that they were reacting against Hegel (whom they encountered via Kojeve) by disassembling the whole (Kojevian more than actually Hegelian, perhaps) “labor of the negative” — hence Bataille emphasizes “negativitÃ© sans emploi,” and Blanchot “desoeuvrement” — both of which mean that negativity cannot be put to work, cannot perform a labor. Negativity is weak, and not productive. I saw a link here between the Bataille/Blanchot critique of negativity and Kant’s emphasis on limits: even though Bataille and Blanchot never themselves say anything about Kant. But Foucault definitely positions Bataille in relation to Kant (rather than to Hegel) for this reason in his crucial early article on Bataille, “A Preface to Transgression.”
Subsequently, thisconfiguration seemed to me to be the key to Deleuze’s hatred of the dialectic, and to his presentation of Nietzsche as the thinker who “stood Kant on his feet” in a manner analogous to how Marx stood Hegel on his feet. Deleuze separates productivity entirely from negativity. Similar anti-dialectical stances in Foucault and even in Derrida are grounded in this sort of argument, as well. From there I came to a sense that one could read the second half of the First Critique (the Transcendental Dialectic) as, in effect, Kant’s rejoinder-in-advance to Hegel’s critique of him in the Encyclopedia Logic. And this, in turn, takes on considerable relevance today, since Hegel’s critique of Kant is such a centerpiece of everything that Zizek does. I never managed to work this out in a form that I was satisfied enough with to publish — but it definitely stands behind why I found Kant of such importance in talking about Whitehead and Deleuze.
To try to put this more concisely: Kant’s importance is in saying that there are limits to the pretensions of thought to determine the cosmos. Hegel and Zizek argue that any limit to thought is illusory, since it is turns out to be thought itself that is positing such a limit. I think that Kant’s Transcendental Dialectic is best read as a rejoinder in advance to this sort of argument. It is in this particular sense that I argue for reading Kant as (surprisingly, perhaps) an ally of some of the speculative realist arguments against unlimited correlationism; rather than seeing this part of Kant’s philosophy as being — as Deleuze sometimes implies — a policing action against speculation. Of course it is both, but my reading of Kant in Without Criteria is designed to bring out some of the often overlooked “minor” aspects of Kant — which is something that gets me in trouble with more orthodox Kantians,and more generally with normativists, as much as it does with speculative realists who see Kant as the enemy.