Paul was writing his last post at the same time as I was writing mine--what's even more interesting is we were almost writing about the same things: what motivates Meillassoux, ultimately, to redescribe or "revise" (as he says at one point) the history of philosophy. My allusion to Derrida, picking up off of Paul's, was to suggest that while for him (Derrida), this derailing of the consistency of correlationism would be in order to politicize the latter (in a philosophical/extra-philosophical act which is nothing less than tarrying with the aporias through which politics/history is inscribed in philosophy), it is still unclear to me what Meillassoux's derailing has to do with politics. Therefore, from the Derridian perspective (certainly an extreme one), it is an eminently philosophical act of rewriting the history of philosophy that we see here (and there's nothing wrong with that of course). But occasionally that only makes things more confusing, especially as Meillassoux certainly has some sort of "state of things" in mind that he is engaging, and which I think is certainly also political--as much as it is religious. This might simply mean that Meillassoux has a different conception of politics/history in general and its relation to philosophy (such that the history of philosophy doesn't always reappear primarily as a political or even historical issue), but I think what we're also recognizing here (and I think Evan registers it too) is that something seems very (intra-)philosophic about this attempt. That's not a fault at all (it has become even more easy to call philosophy into question by talking about it in this Derridian way--but I want to do here what it allows us, which is to really stick with what is weird about philosophic operations and that history which even the Deleuzian maneuver of calling it, precisely, "Oedipal," might dispense with too quickly). In fact Meillassoux's effort is puzzling precisely because he wants to bring back something of this properly philosophic character, and it is "the state of things" that actually isn't quite ready to accommodate it.
And yet that state of things precisely is ready to accommodate this, in certain ways, because the narrative he gives is so compelling and hits home what many feel to be problematic right now. But I think your sense, Paul, that this narrative has to almost return in spirit and in content to the tumultuous era immediately after Kant--and I'm actually quite inclined to agree with you (and I think many many people in the reviews of the book I am reading are registering something similar in talking about Meillassoux's "Hegelianism")--makes it clear that philosophy has been accommodating this state of things for some time. In short, if Meillassoux is confronting us with the "repressed" of recent philosophy, which is some manner has to involve proper philosophizing itself--and I think both our accounts verge on saying something like this (even though Heidegger wasn't quite fond of psychoanalysis, I take your comment saying there's something Heideggerian about the way things proceed to in fact gesture towards this repression)--then you are hinting that there is another repression (secondary repression--which Freud called, precisely, revision): repressed in the repression is also that this was philosophically problematic once before. I think this is a fascinating response to what is Meillassoux's general contention, it seems: that correlationism was "weak" when it was adopted, then proceeded to get "stronger," as I sketched last time. Wasn't it "strong" in some sense also when it was adopted? Not in the sense that it completely involved de-absolutizing the absolute, or absolutizing the correlation (the precise sense in which Meillassoux means "strong"), but in some other way we--or indeed the post-Kantians themselves--might articulate? And doesn't Meillassoux somewhat concede this point in saying that "every philosophy which disavows naive realism has become a variant of correlationism?" (5) What appears to be an expansive claim is also the recognition that, if we properly see that certain philosophies to not wholly "disavow" naive realism, they can be seen to at least be dealing with, if not escaping, correlationism? This of course with the caveat that a sort of general blase acceptance of naive realism can be fundamentally constitutive for many of the philosophies that assert the inaccessibility of the in-itself. As Meillassoux says, this is actually what is involved in the postmodern's indifference in the face of science, which is a sort of acceptance that actually makes thought recuse itself. What you might be saying, in some deep sense, is that pulling apart this sort of blase acceptance from a more significant, though seemingly inapparent, affirmation of naive realism (or simply realism), is what is precisely possible now: it is this that the postmodern situation, or the intensification of correlationism, provokes us to do.
But now I want to move on to my promised explication of that first correlationist "rejoinder" to the problem of the arche-fossil. It intrigues me because it treats of space--which is why Brassier will come to it in his (well articulated) criticisms of the problematization itself of the fossil and the ancestral--and for that reason I think it probably intrigues you, too, Paul. The argument proceeds by identifying the problem of the arche-fossil with what is indeed a variant of a "familiar and inconsequential anti-idealist argument" (18): the old one most easily applied to Berkeley, where if no one is around to see a tree, and esse est percipi, the tree ceases to exist unless we have some God that keeps it in being. In other words, the rejoinder would proceed by "trivializing the problem," reducing it to this one (18).
Now, remember what the arche-fossil is supposed to do: it is supposed to be a "material substrate" that indicates to scientists the existence of an ancestral reality, or one that is anterior to every recognized form of life on earth (10). What this is to open up--and the linkages between the two might become problematic--is that there is a being which is not just not given to thought, but also remains completely outside any way that it could be given to thought, though (and this is what is crucial) it still can be thought (precisely by thought going outside of itself). What this opens up is the truly dazzling (from the transcendental perspective, 27) possibility that science thinks not only these beings, but also the emergence of the givenness itself--in other words can account for both.
The rejoinder cuts off the latter question about emergence completely (or approaches it in a weird way) by saying that what is not given to thought is precisely what phenomenology accounts for. "The lacunary nature of the given has never been a problem for correlationism. One only has to think of Husserl's 'givenness-by-adumbrations:' a cube is never perceived according to all its faces at once; it always retains something non-given at the heart of its givenness" (19). being partial to phenomenology, of course, this is what I have always loved about it: I can't get enough of houses with five unperceived sides (or six if the roof is normal) towards which I nevertheless constantly orient myself. But the rejoinder can go even further: "Generally speaking, even the most elementary theory of perception will insist on the fact that the sensible apprehension of an object always occurs against the backdrop of the un-apprehended, whether it be with regard to the object's spatiality or its temporality" (19). What has happened, Meillassoux points out, is that we have confused the being-not-given of the fossil with something commensurable with not-givenness in precisely a correlationist sense: but what is not-given in an ancestral sense--in the case of the arche-fossil--is supposed to be different than the merely "lacunary" within the otherwise given. This is why the argument boils down to a rejection of a mere anti-idealist challenge, when the challenge brought by the fossil is actually much deeper: a conflation of the ancestral with the un-witnessed, which of course (the correlationist says) does not cease to be when it is not perceived. Perception itself (in a phenomenological account or even the most basic and sensible one) involves precisely accounting for the being of what is unwitnessed.
But then Meillassoux takes another step in his reasoning--the one that really interests me. So far, I've presented his account just in relation to the general givenness of a being: the conflation of the unwitnessed with the ancestral is with respect to the nature of the givenness in question. But Meillassoux goes on to say that this mistaken correlationist rejoinder involves some effort to correct what is seen as the privileging of "temporal seniority" in the challenge of the fossil (18). The conflation with the unwitnessed with the non-givenness of the fossil proceeds by asserting that "spatial distance would raise exactly the same difficulty… An event occurring in an immensely distant galaxy, beyond the reach of every possible observation, would in effect provide the spatial analogue for the event occurring prior to terrestrial life" (18). Supposedly, the conflation would then involve trying to argue from this distant event, saying that "'distance' and 'ancientness' are both vague" (18): "above all, we would immediately notice that the question of the relative proximity the object under consideration becomes irrelevant to the force of the argument once the scope of the latter has been extended to space" (19). We don't know where what is proximate begins and where what is distant ends, and the same can be said about the recent and the ancestral. The conflation proceeds, in other words, by seeing what is not-ancestral as relative to our position (thus the appearance of "recent," when the ancestral qua ancestral has no opposing term of this sort)--and this proceeds to make the problem for correlationism an intra-correlationist problem. But again, why this has to proceed so thoroughly via space is interesting. It seems to be because to think of the spatial event beyond our observation, this would mean conceding that givenness as such may be in existence, when the whole point of the fossil is to think beyond the given but the emergence of the given itself. The latter cannot be thought--Meillassoux says--if we can't think that it may occur while there is no givenness existing. Certainly this is correct, and it is what keeps Meillassoux away from any pseudo-idealist philosophy, but I just keep thinking about why time remains the way that we have to think this:
The reason why the traditional objection from the un-witnessed occurrence--it being a matter of indifference whether the latter is spatial or temporal--poses no danger to correlationism is because this objection bears upon an event occurring when there is already givenness. Indeed, this is precisely why the objection can be spatial as well as temporal. For when I speak of an event that is distant in space, this event cannot but be contemporaneous with the conscious presently envisaging it. Consequently, an objection bearing on something that is unperceived in space necessarily invokes an event and a consciousness which are considered as synchronic. This is why the event that is un-witnessed in space is essential recuperable as one mode of lacunary givenness among others--it is recuperable as an in-apparent given which does not endanger the logic of correlation. But the ancestral does not designate an absence in the given, and for givenness, but rather an absence of givenness as such. And this is precisely what the example of the spatially unperceived remains incapable of capturing--only a specific type of temporal reality is capable of capturing it (20-21).
Again, nothing wrong here, just curious about that last statement. It isn't that I can imagine any space that would be capable of capturing it: the spatially ungiven. Rather, what is odd is that the absence of givenness itself is actually not quite so temporal a thing, leading us to wonder why we have to go to such temporal lengths to capture it. For, as Meillassoux says,
Though ancestrality is a temporal notion, its definition does not invoke distance in time, but rather anteriority in time (20).
There's nothing wrong with this, again, since what we're doing is thinking through precisely how the fossil exists in space and time (as intra-worldly) so as to disrupt the correlation. But it just brings up the specificity of the fossil and what it is doing in the argument: it is supposed to call correlationism in question--to be something for which it cannot account. But, like Brassier (and like Harman, in a different way), we can then begin to wonder whether it is perhaps more than sufficient to do the job.