I am hesitant to write this post because I believe that metaphysical and epistemological issues should be grounded in metaphysical and epistemological reasons, rather than normative reasons. Indeed, this was one of my main critiques of Deleuzian scholarship in Difference and Givenness: That it was too often conflating normative considerations based on a particular politics with grounds for endorsing Deleuze’s ontology. However, given the aridness of me and Alexei’s recent debate over the grounds of mathematics vis a vis correlationism (Alexei) and realism (me), I thought it might be worthwhile to say a word or two as to just why such a debate is important. It is not unlikely that Alexei– and he’s said nothing that would indicate this to me, I’m just speculating –gets worked up about this debate for very similar reasons to the ones that work me up. Moreover, it’s not unlikely that Alexei’s reasons for getting worked up about such an arid issue– despite claiming he wasn’t worked up… he certainly spilled a lot of ink for someone not worked up –are not simply about the soundness of a particular argument or philosophical theory. However, I can only speak for myself.

One of my reasons for getting so worked up about correlationism has nothing to do with flaws I see in the position (and I’ve become increasingly convinced that it is a deeply flawed position), but rather with a certain skepticism it introduces into philosophy and, more particularly, science. Speaking in a very crude way, the correlationist claims that we can only ever know things as they appear for us, not as they are in-themselves. This, in and of itself, is not the skeptical consequence of correlationism. The philosophical pay-off of making the correlationist move is that you get a rejoinder to the Humean skeptic concerning our inability to ground inductive reasoning. That is, because the structure of mind is a universal and a priori structure, the apparent subjectivism of correlationism (that we can only speak of things for-us, not in-themselves) yields an empirical realism. Appearances are, according to the correlationist, law-like in how they appear. In other words, the subjectivism of correlationism is not a private subjectivism restricted to individuals, but rather phenomena are universal and objective. The correlationist is thus able to defeat Hume’s skepticism.

The price the correlationist pays for this defeat of Humean skepticism is that our knowledge is restricted to phenomena or appearances, not things-as-they-are-in-themselves. It could be, the correlationist claims, that the things themselves, independent of the structuring activity of our minds, do not follow these laws at all. But none of this really matters, because insofar as the structures of mind are transcendental and universal, phenomena will always be structured accordingly and science and mathematics will be well grounded. Additionally, the correlationist argues that there are limits to this knowledge. As Kant so pithily puts it, “concepts without intuitions are empty, intuitions without thoughts (concepts) are blind.” Pure a priori categories of the understanding only find their legitimate employment in relation to sensibility. This sensibility, of course, can be of either the empirical variant (what we must receive through experience) or the a priori variant (the pure forms of space and time that mind imposes on the world), but nonetheless concepts employed independent of sensibility or intuition are illegitimate. Consequently, one meritorious consequence of the correlationist move towards finitude (our restriction to sensibility), is that we can no longer claim knowledge of things beyond the scope of our experience. In a single stroke Kant is thus able to banish rational psychology (attempts to demonstrate the immortality of the soul, for example) and natural theology (attempts to rationally demonstrate the existence of God), from the domain of what can be known as these things exceed the limits of our intuition (and here I’m simplifying quite a bit as to how Kant goes about demonstrating this as the arguments of the parlogisms are quite different than those of the ideas).

read on!

The shift towards antirealism or correlationism can thus be quite attractive from the standpoint of those who are fed up with theological arguments and wish to set them outside of philosophy. Kant famously said that he wished to limit reason so as to make room for faith, but his position is equally congenial to an atheism that wishes to foreclose any attempts to demonstrate the existence of God, the nature of the soul, etc.

However, and this is part of the reason I get worked up by the realism/anti-realism debate, the correlationists restriction of knowledge to appearances or phenomena, while, in a certain sense, “saving science”, nonetheless opens the door to another form of religious skepticism. Take, for example, the contemporary debate over evolution. The religious skeptic might concede the correlationist argument, claiming that at the level of phenomena or how things appear for-us, evolution is the only plausible conclusion. Nonetheless– and here I’m indebted to Meillassoux’s analysis in After Finitude –the religious skeptic can still point out that this knowledge is restricted to appearances, and that the level of things-in-themselves the world could be organized in a completely different way, along creationist lines. “Since we cannot know things-in-themselves,” the religious skeptic reasons, “there is no reason to conclude that things are as they appear.” Consequently, the correlationist move still leaves open wiggle room for faith trumping what our experimental investigation of nature tells us. If the religious skeptic is committed to revelation as an article of faith (not knowledge), then he will feel warranted in rejecting the findings of these sciences (not that we would ever convince the religious skeptic anyway, but perhaps those in the audience viewing the debate). We find exactly this line of argument in vulgar form among those who argue (with frightening frequency in the states) that either a) Satan put fossils in the earth to mislead us, or b) that God created the world in such a way so as to appear to work along evolutionary terms so as to test our faith (apparently God is an ego-maniac that needs our faith like a vampire needs our blood, according to those who believe such things). But we also find very sophisticated forms of this argument. It is not a surprise, for example, that the heirs of correlationism have, in many instances, shifted towards a theological discourse: Nancy, Derrida, Marion, even Henry in his own way.

I thus think that lurking in the background of the realist/anti-realist debate is this central issue. The realist move– if possible today –attempts to bite the bullet and argue that occasionally we discover a bit of the real and that this real is not just phenomena for-us, but is how things are in-themselves. That is, these things are as they are regardless of whether we know them or experience them, and regardless of whether or not anyone exists. In making this claim it refuses the conflation of the epistemological (the for-us) with the ontological (the in-itself), arguing that claims about beings (the ontic) cannot be reduced to claims about what things are for-us. In part, I think this is what was at stake in the recent arid debate about the status of mathematical entities and whether or not there are mathematical entities that cannot be constructed in intuition. If it is so important to defend the existence of entities that cannot be constructed in intuition– which is entirely different from the claim that they cannot be known –then this is because what is being defended is the position that these entities aren’t simply what they are for-us, but exist as they are in their own right regardless of whether or not anyone exists to know them.

However, it is important to note that this realism would not be a naive realism. It is not being asserted that things exist in-themselves as we perceive them or that we posses an immediate relation to mind-independent objects as they are in-themselves. Following a line of argument advanced by Nick over at Accursed Share, it could turn out that things such as trees, tables, rocks, etc., do not exist as real objects, but instead exist in a completely different way at the level of real and mind-independent being. That’s an issue for ontology to work out. These real objects, rather, are only arrived at through a laborious and careful process in the development of knowledge where many theories are tried out, many motives for pursuing that knowledge are operative, and many of these theories and concepts turn out over time to be mistaken (as demonstrated experimentally, where the real gets to add its two cents with respect to our constructions or models).

It is also worth noting that this realism does not foreclose the possibility of theology or God as in the case of the correlationist move. It could turn out that God exists or is real. However, if things such as evolution describe the real, if things such as contemporary cosmology, geology, subatomic physics, and psychology, turn out to be true, it would also be the case that this God that exists is very different than the anthropomorphic God we find in the revelation of sacred texts throughout the world religions and that we should side with what our investigations into nature show us to be true, rather than what the revelation of some sacred text shows us to be true. From the correlationist standpoint this assertion cannot but be nonsense, as revelation is a matter of faith, whereas science is a matter of how things appear for-us. Thus, as Galileo put it in a less than heroic moment, his claims about planetary motion were not real but were useful fictions that aided in calculation.

As I said at the beginning of this post, I am hesitant to write all of this. If I am hesitant to write all of this, then this is because the desire to refute the religious skeptic is not a legitimate reason for endorsing a particular ontology, nor a reason for dismissing a particular epistemology. If we are intellectually honest, the reasons for endorsing or dismissing a particular ontology or epistemology must themselves be ontological or epistemological in character. Wishing, desiring, does not make something so, and this is the problem we find among ontological and epistemological arguments that are normatively driven such as in the case of those Deleuzians who seem to think they can dismiss Kant because he is a “state-thinker”. Poppycock! If Kant is to be dismissed, then this can only be on the grounds of 1) there being significant flaws in his position, and 2) through offering an alternative. However, while wishing does not make something so, this does not undermine the fact that these supposedly arid and remote issues have consequences that reverberate far beyond what they’re immediately about. I strongly suspect that both Alexei and I agree on the truth of the maths or the sciences, regardless of the fact that he’s an antirealist or correlationist and I’m a realist. However, the realist and anti-realist positions nonetheless have consequences that go beyond these rarified matters.