No, this is not a Kant bashing diary.
Given the flurry of writing today it’s probably fairly evident that I’m trying to avoid grading. Despite my antipathy to Kant and transcendental idealism, I do find his thought endlessly fascinating and replete with brilliant and devious arguments. It was thus with great pleasure that I got to recently explore the Prolegomena once again with my students. And as we worked through the Prolegomena I found myself particularly struck by the logic and structure of the a priori categories which Kant introduces in the transcendental analytic of the Critique of Pure Reason. In particular, I found myself fascinated by the manner in which every third category is a combination of the preceding two categories. For the object-oriented ontologist the categories falling under quantity and relation are particularly important.
To be clear, I am not endorsing Kant’s specific theorization of the categories (i.e., that they are a priori structures of the mind). As a realist I am, of course, committed to the thesis that attributes like being a substance belong to the things-themselves, not the mind regarding objects (viz., they are primary, rather than secondary, qualities). Nonetheless, there is a great deal of interest in these categories, despite the short shrift he gives to their elaboration (does Kant somewhere treat them in detail in his lectures?).
read on!
Take the categories falling under the heading of quantity. We first find unity and plurality. Unity pertains to a being that is one, that is that being, whereas plurality refers to a collection of objects (a collection of “units”). The key point to draw from the category of plurality is that apart from being collected together, the units of a plurality share no internal relations with one another beyond being collected together. In this respect, pluralities are unstructured sets. It is this that allows us to distinguish between the category of plurality and totality. If there is a difference between a plurality and a totality, it lies in the fact that there are no relations of dependency among the elements of the former category, while in the latter category the unities falling under the totality stand in relations of dependency with one another. It is in this sense that the third category is a combination of the first two categories. Totalities are pluralities that are unities.
Where a plurality is a mere aggregate, a totality is an assemblage. Thus, for example, were I to deviously sneak into your yard in the middle of the night and take apart your car, you would have a plurality or a mere aggregate of parts. The car itself falls under the category of unity. But note– and here is part of the reason the car is of interest for the object-oriented ontologist –the car is also a totality. That is, it is composed of parts forming an assemblage that have entered into internal relations with one another. However, what’s really interesting here is that the category of totality does not efface the parts, for a totality is a unity of a plurality. The parts of the totality remain autonomous objects in their own right, with their own powers or potentials, while nonetheless being harnessed in an assemblage. According to OOO every object is a unity that is also a totality.
The case is similar with respect to the categories falling under relation. As listed in the Prolegomena, we here find the category of substance, causality, and community. A substance, of course, is an individual object that is independent of any other object. Causality refers to one object affecting another object and is therefore unilateral. Causality is here treated as one object affecting another object. Like the category of totality, with the category of community we get a combination of substance and causality. How is this to be understood? In the Critique of Pure Reason Kant lists community as reciprocity between agent and patient. Thus where the category of causality is unilateral the category of community is bilateral. Rather than one way causality, we here get “n-way” causality. Causality is operative in the relation between a match and a piece of paper.
Community is operative in the relation between the moon and the earth, the sun and the planets, the organs in the body, in so on. In other words, communal relations (in the metaphysical sense) are relations in which each substance is simultaneously an agent and a patient with respect to all of the other substances. If there is a difference between totalities and communities (though they overlap), it lies in the difference that in the former the unities have merely formed a unity with one another, while in a community the substances are simultaneously affecting one another and being affected by one another. However, like the category of totality the point not to be missed is that communities are not the erasure of substances, but are rather a combination of causality and substance. Substances qua substances continue to exist within communities. And like the category of totality, it can be said within the framework of onticology (I’m not sure this would hold for all variants of OOO), every object is simultaneously a substance and a community. That is, every object is a substance composed of substances that are simultaneously affecting and being affected by one another, producing the generative powers of the object.
One of the trends of the last hundred years of philosophy has been to think questions of ontology in terms of categories of totality and community. This “episteme” takes an endless variety of forms (Heidegger’s worldhood, Whitehead’s “organism”, Hegel’s system, Wittgenstein’s “language games”, the structuralist’s structures, etc), but generally we can refer to the core thesis behind each of these positions as that of “relationism”, where relations are held to hold ontological priority over parts. Now, the corresponding gesture of relationism is the erasure of substances by their relations. Within a relationist framework substances are their relations. This line of thought is a variant of what I have called the “hegemonic fallacy” and what Harman has referred to as the undermining of objects. The hegemonic fallacy consists in treating one type of difference as making all the difference. Thus, for example, Barthes, in a relationist moment, treats relational and signifying difference as making all the difference in fashion, ignoring whatever difference different fabrics might contribute. The object is here undermined or erased in these signifying differences, functioning as nothing more than a support for these relations (if even that), and becoming a mere epiphenomenon of the relational network constituted by signifying difference.
What we get in these relationist applications of the categories of totality and community are a sort of “amphiboly” in the Kantian sense. The reason relationist employments of the categories of totality and community are unsound is that they apply an illicit mereology where plurality is erased under unity in the category of totality, and where substance is erased under the category of causality in the category of community. Yet within totalities and communities pluralities don’t cease to exist in the category of totality, nor do substances cease to exist within communities. Indeed, the very concepts of totality and community become incoherent with the erasure of plurality and community. Every totality swarms with plurality and every community swarms with substances. And if this swarm that is in excess of the unification of plurality and substances is to be thought, objects, in principle, must be thought as 1) independent of the totalities to which they belong, and 2) totalities and communities must be thought as harboring a χώρα, an excess, that always threatens to undo the higher level object, the totality or community, from within.
Larval Subjects . 
March 25, 2010 at 5:17 am
I’m anything but a Kant expert. (I hate reading Kant myself) But isn’t this a broader manifestation of the old thesis, anti-thesis, synthesis that gets so misleadingly attributed to Hegel but which is right out of Kant? Certainly in his antimonies.
Peirce gets his categories right out of Kant, reducting them to pure 1stness, 2cdness and 3rdness. While he’s a huge Hegel critic at first by the end of his life his thought has come to have many Hegelian parallels.
Getting back to your earlier point it seems like what Kant’s examples of third categories might best exemplify is mediation in general. Of course that’s the Peircean in me coming out.
March 25, 2010 at 5:40 am
There’s definitely something of the thesis, anti-thesis, and synthesis triad at work in Kant’s sorting of the categories. I think a tremendous difference between Hegel and Kant here is that for Hegel the third category is the truth of the other two. Not so in Kant. In other words, for Kant it is not that the first two categories are incomplete or partial, requiring the third category to complete them. As a consequence, there’s not really a dialectical movement here in the Hegelian sense where the partial categories (I’m forgetting my Hegel speak at the moment) pass over of their own accord to the new category. Thus, for example, the idea of a unity that is independent, autonomous, and outside of a totality is thoroughly unthinkable for Hegel. Not so, I think, in the Kantian sorting and certainly not so for OOO. This is a major difference between OOO and anything like what Hegel proposes. Substances (individual objects) do not find their “truth” in totalities and communities such that they are inseparable from totalities and communities:
1) A substance (X) can be quite indifferent (and “unaware”) of how a higher-order object (Y) “makes use” of it in forming itself as a substance or totality. Thus, for example, a worker (X) might be largely unaware that she plays a role in the constitution of a particular capitalist network of production and distribution (Y). Y “uses” X to sustain itself, but X is also independent of Y (which isn’t to say X isn’t influenced by Y) moving in directions quite different than Y. This is why, for example, markets can quite literally be outside capitalist modes of production, while nonetheless providing all sorts of input unbeknownst to themselves for a capitalist network.
2) A substance (X) can actively be at odds with the totality (Y) that attempts to integrate it even while being a vital feature of that totality. A cancer, for example, can be thought of a rebellious organ. Similarly, a social totality (Y) might make use of the militant activity of an activist group (X) to sustain itself as an organized totality. Thus, for example, the Obama administration (Y) is now using the terrorist activity of certain Tea Party members (X) as a way of shoring up itself as a totality and strengthening its ongoing existence. The aims of X and Y are quite different here and while X, for the moment at least, is a vital component in the “totalization” of Y, X is nonetheless independent of Y.
3) Substances can simultaneously belong to different discrete totalities and communities and in entirely different ways.
4) Substances (X) belonging to a higher order totality or community (Y) can themselves contain other substances or lower order substances (Z) that a) do not themselves belong to the higher order substance (X), and 2) that elicit any of the first three differences listed.
I’m sure there are many other weird formal relations I haven’t thought of here. The point is that nowhere do we get a sort of totality of totalities or community of communities where substances are all gathered up in a system where they become something like mere organs in a body.
March 26, 2010 at 5:48 pm
[…] Against the Hegemonic Fallacy (and here) […]