The Limits of ‘Correlationism’

The Sea of Ice

Graham mentioned that he liked my distinction between structuralism (read broadly) and other forms of correlationism. I’ve felt, probably since first reading After Finitude, that Meillassoux’s categories tell a very small story and that that story should be broadened to include new categories or designators. In some ways this has meant broadening ‘correlationism’ but it also means looking at how contemporary thinkers designate their contemporaries. Graham’s work for instance is very good on this, as he lays out clearly why his position is different from materialism (of many types), the two forms of correlationism, empiricism, occasionalism, relationism, etc, etc. He writes a story with many characters (17 apparently, if Treatise on Objects will contain his own system against 16 others), which I really appreciate. I suppose my own efforts stem from this desire to tell a big story, to have many characters rather than a strict argument against one single position. In my paper in Claremont for instance, there were several such systems at work, with some overlap: eliminative materialism, eliminative idealism, weak and strong correlationism, speculative materialism, OOO, transcendental realism, and neo-vitalism. The two forms of eliminativism could perhaps (in the case of eliminative idealism at least) be subsumed under the heading of correlationism, as I mentioned yesterday of structuralism.

As for structuralism as a variety of idealism, I don’t think this is quite right. I don’t know if the category ‘idealism’ makes sense anymore after we have chosen to adopt ‘correlationism.’ As Graham has said before, initially Meillassoux thought himself to be arguing against idealism, but he realized that there were important differences between ‘weak’ and ‘strong’ forms of correlationism, both of which would be, under more classical terminology, ‘idealist’ in the sense of being anti-realist. But the important difference has to do with things-in-themselves, and not simply the principle of correlation. Since structuralism as I have defined it maintains the possibility of novelty, of things being other than what they are, and of things existing outside of thought (however traumatic this transition from unthought to thought may be), they are not really classifiable under Meillassoux’s schema. Nor are they classically idealist. If we take perhaps the two extreme forms of idealism to be those of, on the one hand, Berkeley, and on the other, Fichte, then this becomes clear. For Berkeley, there are no things in themselves because everything is always being perceived by God. So while we could say he is an idealist in the sense of all things being mental, he is also an anti-realist in the sense that nothing exists apart from being perceived. For Fichte, we have the relation of Subject and Object boiling down to the fact that the Object is always created by the Subject as a form of opposition. The Object only has being from the Subject, who creates it in order to strive against it (in the form of absolute freedom and the striving for Justice). He is therefore both an idealist (the Object is created entirely by the mind of the Transcendental Subject) and an anti-realist (the Object has no reality apart from this relation to human mind). There also really are no things-in-themselves for Fichte for this reason, as the Object is always excluded from Subjectivity in order to be further integrated and included. So both extreme examples of idealism share the fact that they are strong correlationists (there are no things-in-themselves) as well as the fact that they are anti-realist (all things are reducible to mental processes or Mind generally).

Structuralism, which I should point out I am entirely against, maintains along with the weak correlationist that there are things-in-themselves, but claims we not only imagine them as a possibility, we encounter them as trauma. We also do not cause them to exist, as they have some disorganized being outside of thought, existing whether we like it or not. So the structuralist is neither a correlationist in the weak or strong sense (things-in-themselves actually do exist and we know them in some traumatic way), nor are they a strict anti-realist (Nature exists in opposition to Culture, even if it is a swirling tumult of chaos).

I have also proposed a lineage of “Transcendental Realism” in both my Claremont talk as well as another essay, where I maintain that there is a lineage after Kant that takes Kantianism (and critical philosophy generally) seriously, while also maintaining that there is more to things than our ideas of things. I locate this tradition with the rejection of Fichte by his star students, Schelling and Novalis, and see it as the ground of Romantic philosophy broadly understood to include Schopenhauer, Fechner, Nietzsche, von Hartmann, etc. It’s also a tradition which takes Spinoza very seriously, as well as aesthetics and mysticism. This is because they represent a group that knew that the logical consequence of Enlightenment thought was the reduction of the real to the rational and that this isn’t the case. They accept that there are things-in-themselves and that we have some vague knowledge of them through non-cognitive means, like sensation, imagination, intuition, etc. This is precisely the critique of Kant that Schopenhauer makes, that we actually know something of the in-itself because we are able to grasp the in-itself in us intuitively. They are also realists in the sense that things exist prior to human thought and will exist after we are gone. This is because they accept the metaphysical unconscious, that conscious perception not only is not the ground of reality, but is really only a very small part of the cosmos.

I think this gives some idea, at least in part, of the limitations of ‘correlationism’ as a category. While it is useful and telling of certain figures, it is in no way the whole story. We simply need to add more characters to the story of contemporary philosophy.



Filed under Uncategorized

6 responses to “The Limits of ‘Correlationism’

  1. One reason I like your expansive classificatory scheme is its nose for the subtle differences that get lost if we have only one label; and this actually salvages “correlationism” as useful too. I’ve been saying for a while now that this word is (or at least, risks becoming) the new logocentrism, i.e., swear word. This doesn’t make it completely useless, since there are indeed family resemblances and certain characteristic excesses we can lump together under this rubric; but calling someone a correlationist mustn’t substitute for actually engaging their arguments. It is not enough (if I may suggest a reductio) to spot the correlationist moment in Kant, or Schopenhauer, or whoever, and forgo reading them any further; just as if we could see how Leibniz or Descartes was a “metaphysician”, so we needn’t bother with them. I know this strikes one as silly, but I really did hear people talk this way in the wake of Derrida 20 years ago. (Which of course would have made Derrida blanche.)

    As you note, its to his credit that Meillassoux already saw that correlationism per se won’t do; hence his strong and weak versions of it. Played “to the end,” these turn, respectively, into Meillassoux himself and into Harman.

    But of course, the dangerous thing about all such labels is the risk of freezing thought. I remain (like Harman) uncertain myself whether strong correlationism can be consistently distinguished from idealism, but this isn’t because I think that the proper Venn diagram would show idealism as the larger and subsuming set; it’s because our discourses aren’t wholly stable, and they flow into each other. Even Marx has his idealist moments. My sense is that the more you distinguish different possible positions, as you do here, the more fine-tuned and careful your own analyses can be. But if we take over anyone else’s schema wholesale, we risk missing the fluidity of thinking. One good antidote to this (not without its own risks, but very salutary nonetheless) is a close eye for genealogy, such as you put forward here.

    One question: when you talk “Transcendental realism,” do you mean to point to Wolfendale’s project of the same name? For I take it his genealogy is somewhat different from the one you lay out– though he too takes Kant very seriously.

  2. Pingback: and another post by Austin « Object-Oriented Philosophy

  3. “Transcendental realism” is a very broad term. I know Pete uses it as well, but I think we both understand it as the logical consequence of metaphysical speculation that takes Kant seriously. In the canon I have outlined, it is used explicitly by von Hartmann to describe his own work (as well as his readings of both Hegel and Schopenhauer) and is also evident in Schelling’s late period (where it is related to what he called “Positive Philosophy” in opposition to the purely “Negative” [though necessary] thought of Critical Philosophy).

  4. Hello Michael,
    Would it be possible for me to read a copy of your recent paper delivered in Claremont? I’m also very interested in your project of pursuing the metaphysical foundations of psychoanalysis, and your upcoming book on Freud (how is that coming along?),this is close to the terrain I’m working on myself at the moment. When will your essay be published in the Zizek journal, sounds very interesting?

    Yours Jan

  5. inthesaltmine

    Hi Michael, this is my first week blogging in the SR/OOO area. I’ve referenced your article in my recent post. I’m attempting to carry out a Nietzschean “forgetting” of transcendental realism, by starting the dance of moving “with Novalis beyond Novalis”.

    Here you go:

  6. I appreciate your type of analysis and distinction here. I wonder if you would could give me your insight into my essays. ‘The Significant Event’, which carries forth through its parts in posts, considers OOO and correlational ism and the ‘speculative’ in general.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s