Correlationism

Alexei has written a very valuable response to my post responding to his diary over at Perverse Egalitarianism.

In a previous post, Alexei had objected to my Ontic Principle on the grounds that some differences are more important than other differences. I had responded by arguing that this conflates normative and epistemic issues with ontological issues. Additionally, I pointed out that the Ontic Principle only states that for anything that is, that thing both is a difference and makes a difference. In short, to be is to differ. I have never suggested that all differences are equal. Indeed, as my Principle of Reality states, the degree of reality or power of an entity is a ratio of the extensiveness of the differences it produces. Thus, while by the Principle of Translation or Latour’s Principle there is no difference that does not make a difference, there nonetheless are degrees of power among differences where one or more differences can overdetermine the rest.

Before responded to the passage quoted above, I would first like to thank Alexei for the tone and thoughtfulness of his response above. While Alexei certainly disagrees with my position, he disagrees without being disagreeable. Rather than suggesting that I have failed to understand Kant and should go back to the text, and rather than suggesting that I am equivalent to the disgruntled first-year student that simply has a violent reaction to the difficulty and foreignness of philosophical texts, thereby wishing to simply dismiss them like the horse swatting its tail at an irritating fly, Alexei instead opts to clearly state his own position and argument, showing why he believes that object-oriented philosophy is doomed to failure. As I remarked in responding to another interlocutor here at Larval Subjects, I generally think one is on the losing side of an argument when they begin from the premise that their opponent is ignorant. This is not to suggest that misinterpretations of philosophers do not often take place– indeed, it’s likely that development in philosophy is the result of philosophers misinterpreting one another –but when one makes a charge of misinterpretation, the specifics of this misinterpretation should be clearly detailed and laid out so that those involved in the discussion can get back to the philosophical issues at hand.

Alexei has approached discussion in the right way. Rather than treating the issue as a textual dispute over whether or not Kant has been understood, he instead states his correlationist commitments as a philosophical (not interpretive) claim and proceeded to show both why he believes we must both begin with correlationism and why it poses a significant challenge for realism. I wish I were myself better at emulating Alexei’s approach to discussion. Like Kevin Klein’s ridiculous character in A Fish Called Wanda, I become really enraged when I experience someone as calling me stupid– “don’t call me stupid!” –or suggesting that I am ignorant. It’s something that I need to work on. In my view, charges that another has not understood or that they have misinterpreted something tend to be counter-productive– even when true –because they implicitly send this message and thereby generate polemic. I strongly believe that one of the most serious diseases infecting how Continental philosophy is practiced in the United States lies in a tendency to perceive all philosophical disputes with a philosopher as a failure to correctly interpret that philosopher. Interpretation is a tremendously important activity and skill, but we really need to get over this sort of knee-jerk reaction.

read on!

Hopefully, then, we can set aside the question of whether Kant and the correlationists are being interpreted correctly. As Alexei elegantly writes in response to my post,

I don’t actually think I’m conflating ontological and normative categories. I take myself to be saying, like a good correlationist should, that ‘Ontological’ is a concept, which is delineated through other concepts, and that these concepts are intelligible as only in virtue of their normative character. More technically put, the normative structure of thought is the sufficient condition for being able to think about ontology. My anxiousness about object-oriented philosophy (and the foreshortened versions of Kant we’ve been seeing) is really a function, I think, of the fact that I have no idea what it would mean to think ontologically, without relation to the normative. For that seems to mean not only (1) trying to climb out of your own skin to think about the world, but (2) trying to think this non-subjective world without concepts. Quite simply put, then, Ontology is only ever conceptual, and concepts are first and foremost normative animals.

This seems to me to be the biggest stumbling block in uderstanding Object Oriented philosohy. For no matter how complex and nuanced our concepts may be, no matter how differential, they are always simplifications of something or others (Adorno’s critique of the Begriff, etc etc).

This is precisely what I mean by correlationism or philosophies of access. I take it that there are variants of correlationism as well. Thus, Alexei here appears to endorse that Kantian version of correlationism, where the world can only ever be thought through concepts. A Foucaultian version of correlationism would place the emphasis on epistemes and power. A Gadamerian form of correlationism would place the accent on our thrownness in history. A Derridean version of correlationism would place the emphasis on our embeddedness in language and texts. A Luhmannian correlationist would place the emphasis on our situatedness within operationally closed autopoietic systems. Berger and Luckmann would place the emphasis on the social and the social construction of reality. And so on. All of these are variants in one form or another of correlationism. All of these versions of correlationism have their merits and reveal different things.

By contrast, this is not what I take correlationism to be: I do not take correlationism to be the Berkeleyian thesis that esse est percipi or that categories, power, society, history, texts, systems, etc., create existence. Just as Alexei puts it, I take correlationism to be the thesis that our relationship to the world is always mediated by some agency (concepts, systems, history, power, texts, language, etc.), that prevents us from ever knowing what the world is in itself. Correlationism is thus an epistemic thesis pertaining to how we know, what we know, and what the limits of our knowledge are. Correlationism doesn’t deny the existence of a world independent of humans, but simply points out that our relationship to this world is always mediated in some way or another. The battle of the “gigantomachia” among correlationists thus revolves around what the primary mediating agency is. Is it concepts? Is it history? Is it economics? Is it society? Is it language? Is it intentionality? Etc.

With Alexei’s elegant formulation of correlationism, I hope that we can, at last, lay the question of interpretation of this or that philosopher to rest and instead discuss the merits of these various positions. In other words, Alexei presents the root claim that Kant is making such that, from there, discussion of all the intricacies of his system aren’t all that relevant or important to the philosophical dispute underway. Moreover, I believe that correlationism in its many variants is a powerful and compelling argument. Indeed, it is a position or argument that I have accepted for a good deal of my philosophical career.

Keeping in mind, then, that I have repeatedly emphasized that I am still working through these issues myself and do not yet have arguments against correlationism that I find entirely convincing, what then are my problems with correlationism?

First, following Roy Bhaskar and his early transcendental realism, I think correlationism is guilty of the epistemic fallacy. That is, correlationism is guilty of collapsing ontological questions into epistemic questions, or of working from the premise that ontological questions can be completely reduced to epistemic questions. Because of this, correlationism is perpetually beset by a sort of internal contradiction (with the exception of Hegel who is a sort of “absolute correlationist”). Correlationism, with the exception of Hegel, simultaneously says that there is existence that is not dependent on humans, and that we can only speak of being as it is for us (for the reasons Alexei nicely outlines). However, the ontological question is not the question of what being is for us, but the question of what being is. Certainly the correlationist can minimally acknowledge that this is a real problem and that the issue of what being is cannot be reduced to the question of what being is for us? Such an acknowledgment does not, of course, prohibit the correlationist from arguing, as does Alexei, that nonetheless we cannot “climb out of our own skin”. Yet, following Hegel, in acknowledging that there is something other than what being is for us, it seems to me that the correlationist has, already, in part, crawled out of his own skin.

Second, and following from this first problem, I object to the way in which the correlationist privileges the human-world relation, the subject-object relation, over and above all other relations. There’s a way in which, for the correlationist, the human functions like the empty set in set theory. Readers will recall that, according to set theory, the empty set is included in all sets precisely because every collection necessarily includes nothing. Moreover, because nothing is nothing, it follows that it is always exactly the same empty set that is included in each and every set (and here, Dominic, I know I’m butchering things), for how could there be more than one empty sets without them being something? It is precisely this universal inclusion that allows the set theoretician to weave all sets from the empty set without having to assume the existence of any particular given thing (this is really the brilliance of Badiou’s ontology as a rejoinder to Heidegger’s critique of metaphysics). Well, for the correlationist the subject functions in this way. For the correlationist the subject is necessarily included in each and every relation because, while we may recognize that there is a world that exists independently of the subject, nonetheless we can only speak of this world as it is for us. While it may be epistemically true that insofar as we speak of an entity we can only speak of it in terms of how we relate to it or how it is for us, the ontological issue of what entities are both independent of us and in relation to each other (object-object) relations nonetheless remains. In other words, we need to avoid collapsing the ontological into the epistemic; and, more importantly, the correlationist already makes this move in claiming the existence of a world independent of us and that is not for us.

Following Meillassoux, I think this universal inclusion of the subject or human in discussion of any and all relations causes significant problems when dealing with something like the Arche-Fossil and tends to generate skepticism. Because scientific statements about things that pre-exist humans refer to things that are purported to exist in a way not dependent on humans in any way, the correlationist necessarily seems committed to the thesis that 1) the scientist is dogmatic and doesn’t recognize that he’s only speaking of being for us, and 2) that we are not authorized to speak of anything not involving the human. This, I think, spells ruin for scientific practice, despite Kant’s defense of science to the contrary. In this connection I’m inclined to argue that scientific practice– in many respects –is organized in precisely such a way as to de-suture objects from both ourselves and relations to other objects. That is, the scientist is at pains to construct an artificial environment that disconnects the object from those relations that mediate it so that the object or mechanism might itself speak. In this way the scientist is able to trigger or evoke the action of the object. I realize this argument is rather weak as Alexei will come back and respond that categories are still mediating our relationship to the object; however, it seems to me that the great accomplishment of contemporary science has been to reveal a world of objects in sharp opposition to our conceptual mediations.

A third objection I have to correlationism more directly pertains to Kant, as well as philosophers like Žižek. Expressed in a manner that might make me sound Hegelian, I think Kant’s account of the subject-world relation, as well as Žižek’s account of the language-world relation, are both too one-sided. In the case of Kant, the lion’s share of the activity comes from the mind. In the case of Žižek, the lion’s share of the activity comes from language or the symbolic. However, as my Principle of Translation asserts– and I’m perfectly fine with conceding at this point that these principles are dogmatic –there is no transportation that does not involve translation. As I’ve often expressed in the past, these ways of treating the human-world relation strike me as being premised on an untenable form/matter dichotomy arising out of technological models of how matter is given form through the imposition of a mould on a matter as in the case of making bricks out of clay. The matter is treated as passive and formless, while the mould is treated as active and structuring. Thus, in Kant, we have the categories or concepts as the structuring principle that is active– and indeed Kant repeatedly insists that the faculty of Understanding is characterized by spontaneity and activity –and we have the manifold of intuition as a sort of passive matter upon which form is imposed. The manifold of intuition contributes very little of its own.

As I have argued or indicated, I simply don’t think this is a very tenable model. Rather than presupposing form that is then imposed upon a passive matter, I require a dynamic interplay of singularities belonging to a field involving all local elements, leading to the genesis or production of form. In other words, in my view, form is not there at the beginning, but is a result of the interactions among these singularities. Moreover, at the level of human cognition and relations to the world I think my thesis accords with the best findings in neurology, developmental psychology, cognitive psychology, etc. However, if it is the case that the genesis of form is the result of a play of singularities– singularities pertaining to my mind, my body, the objects of my environment, others, etc. –then correlationism, at least in its Kantian formulation, is dead in the water. Why? Because we can no longer speak of one agency as contributing the lion’s share of difference through its activity, but rather must now speak of a pre-individual or a pre-personal field as described by Deleuze or Whitehead. The subject here is not an origin or a beginning point, but is a result.

Now it is notable, here, that I have two options left open to me. On the one hand, if the model I just quickly outlined (please follow the links) holds, then I can either opt for a radicalization of correlationism where I arrive at some variant of Absolute Correlationism, or I can shift towards a realism. Absolute Correlationism, as exemplified by Hegel, could possibly accord with my position insofar as Hegel demonstrates the identity of identity and difference, the identity of substance and subject, etc. Under Hegel’s view we can’t speak of one thing, subject, and another thing, world, but rather must see them in constant relation to one another such that there is nothing other than these relations. Since my own position holds that not all things are related and that therefore there are no totalities, I’m instead left with the option of a realism. Again, I’m happy to confess that I have a long way to go in developing this.

Finally, fourth, as I argued yesterday, I think that correlationism’s claims to be critical are highly inflated. The force of the correlationist position arises from the claim of having taken something into account that has been ignored by the dogmatic philosopher; namely, reflexivity or the role that our own conceptual mediations play in our relationship to the world. The dogmatist would thus be someone who believes they have a direct relationship to the world, to objects, without the mediation of concepts. First, I wonder if this isn’t a caricature of Pre-Critical philosophies. Spinoza, for example, constantly emphasizes the role that imagination plays in our relation to the world, while also asserting the reality of objects independent of mind. Moreover, it seems to me that the critical turn is dependent on Hume’s particular discussion of causation as its premise. Has Hume characterized causality correctly in treating it as a succession of events and, epistemically, a relation between impressions? Second, as I argued in my post “First Draw a Distinction”, correlationist philosophies harbor a dogmatism within themselves as they are always premised on a distinction as a condition for indication that they cannot themselves ground, and of which they are not even aware as this distinction perpetually recedes into the background or the unconscious. Yet if that’s the case, it seems to me that the entire distinction between the critical and the dogmatic breaks down.

I confess that I’m uncomfortable with this argument because far from being an overturning of correlationism, it seems to be a universalization of correlationism. That is, if my argument from distinction holds true, it follows that all relations to the world are dependent on distinction such that a world independent of these distinctions is impossible to know. To make matters worse, because these distinctions are abyssal or perpetually recede or withdraw, we fall into a skepticism where all positions are equally dogmatic. Consequently, not only would this position spell ruin for any realist orientation of thought, but it would also generate a universal dogmatism. At any rate, I thank Alexei once again for his terrific response to my post.