My feeling about the use of the word « correlation » changed as a result of reading the excellent article by Katerina Kolozova on Laruelle’s non-marxism and on the need for monstrously radical concepts. It confirmed my feeling that the OOOxians’ concept of “correlationism” is a part of the disastrous impoverishment of thought propagated by their philosophy, and that one should not cede the semantic terrain to them but make full and open use of this potentially fruitful concept.
I was intrigued by the fact that Alexander Galloway’s texts hesitate between a restricted use of the term “correlation” as in Meillassoux’s work (implicating a relation between subject and object, or between thinking and being) and generally affected with a negative valence as constituting an aporia to be overcome or sidestepped, and a more general use of the word as in Laruelle’s work and affected with a more variable valence. Unfortunately Galloway does not consistently make use of the more extended notion and lapses most often into the OOOxian acception of the term despite advancing an argument tending to dismantle the problematic supporting that acception. Further, he calls the more extended notion, when he talks of the homology between object-oriented ontologies and the structure of npost-fordist capitalism, “secondary correlation”, thus giving rhetorical primacy to the OOO acception.
I must admit that up to now I have tended to avoid the word “correlation” due to its imbrication in what for me is a naive and impoverished problematic. So I was initially a little non-plussed by Kolozova’s liberal but unfamiliar use of the word, with a positive valence. I saw that she was using “correlation” to designate a relation of thinking to the real “without the pretention that Thought in its constitution could ever be the direct reflection of the Real and vice versa” (2). Intrigued by this use of the word, I turned to Kolozova’s contribution to the book LARUELLE AND NON-PHILOSOPHY, another excellent article entitled “Theories of the Immanent Rebellion: Non-Marxism and Non-Christianity”. In this text I found the same more extended use of the term and a very useful explanatory note that I reproduce here in full:
“Laruelle, let us note, uses the term “correlation” in a different sense – it is a relation which is not “relationist”, one that remains in the One, one that merely correlates with the Real without mirroring it, within the gesture of relative constitution of both terms. So Meillassoux’s “correlationism” corresponds to the non-philosophical notion of the relative mutual constitution of the Real and the Transcendental, i.e., of Philosophy’s Unity (of the Two) or auto-reflectivity” (p223, footnote 16).
I think that this philosophical gesture has an almost “Wittgensteinian” effect of dissolving the OOOxian distorted usage of the term and bringing it back to a more “ordinary”, even if non-standard, use.
I still do not like the notion of correlation, but if we are going to use it I would say that Alexander Galloway’s so-called “secondary” correlation between neo-liberalism and OOO is in fact primary. On this view, there is no disjunction between politics and ontology, and attempts at imposing such a disjunction are themselves politically contaminated. The political is not just one side of a disjunction, it is constitutive of the World.
Badiou accepts the ontology/politics correlation as posing a problem to his own system, but claims to get round it via the Event and Eternal Truths. If we do not accept that solution then we are stuck with the problem of not wanting to reflect or reinforce in our enunciative structures what we purport to critique at the level of enunciated content. Ontology exists under the condition of politics.
So for Badiou, we are presented with a binary choice: eiither his miracle-endowed materialist dialectic or resigned democratic materialism. This does not exhaust the possibilities for Feyerabend, who probably has more in common with Laruelle than one would think at first sight. According to Feyerabend there is a third possibility: “mysticism with arguments” (« Is argument without a purpose? No, it is not; it accompanies us on our journey without tying it to a fixed road »). This is very close to Laruelle’s gnostic theorems.
AGENT SWARM 
This is an intriguing post, Terrence, but I’m missing a sense of what your think is wrong with the notion of correlation. It seems to me that Meillassoux’s coinage captures what is common a range of post-Kantian philosophies. Correlationisms claim that the concept of reality must be explicated in terms of our means of thinking or representing it. So Husserl’s claim that a perceptual object is an open horizon of possible experiences is correlationist. Putnam’s claim that what objects there are depend on what linguistic practices we adopt is similarly correlationist. I can see that the position that Laruelle is getting at might not be a correlation in this sense (I don’t know his work) but that doesn’t mean that correlationism isn’t an identifiable position that unites much of phenomenology as well as many pragmatist and post-analytic positions.
J’aimeJ’aime
I think that the term « correlation » does not represent a conceptual innovation, but designates what used to be called the problematic of the subject. No interesting post-Husserlian or post-positivist figure is entrapped in that problematic. Extending the notion of the subject to language, power, etc to include Foucault, Deleuze, Wittgenstein, Popper etc is diluting it to the point of vacuity, an empty epithet that one can apply to anyone one happens to dislike.
J’aimeJ’aime
Laruelle does not dispense with correlation if only to capitalize where philosophy has gone wrong. Correlationism is a fodder to philosophical decisionism. But he escapes the correlationist hang-up via his dualysis where philosophical correlationism as it is hitherto practiced becomes a material for non-philosophical analysis. At a point where Laruelle arrives at non-standard philosophy, a new status of non-philosophy that comes after, let us say, an effective work of dualysis, he rescues correlationism from its fossilized status in philosophy and exposes correlationism’s unilateral function, namely, it is just another objectification of the Real.
J’aimeJ’aime
Well, I wasn’t claiming that the notion of the correlation constituted an innovation but that it describes something. I’m not sure why distinction between subjectivity and language is significant here. Kant argues that objects are constituted by categories, but categories are rules for synthesizing a manifold not Cartesian representations. Putnam’s internal realism argues that objects are constituted by linguistic practices. Husserl’s noematic content is not linguistic – for sure – but it is has a similar structure. Foucault is hard to pin down, but in archaeological mode he seems to treat objectivity in terms of our practices for deciding what is an object. Deleuze seems to be an anti-essentialist realist not a correlationist. As far as I can tell Wittgenstein’s position seems to lead us towards something like Putnam’s internal realism. I realize that there are nuances here. Some correlationisms are linguistic, while others are not. Some hold bivalence, while others do not. Derrida’s an interesting case here. He is often counted as an anti-realist and as a correlationist. But I don’t think he leaves enough of language or subjectivity to correlate to.
J’aimeJ’aime
I don’t think even Kant is a correlationist, as his notion of the « understanding » is a structural one and is not that of a personal subject. I see no reason to regard internal realism as correlationist, when it was elaborated to deal with the sorts of worries that the correlologists (those who believe in the notion of correlation) raise post festum, and Harman’s notion of the non-absolute reality of the scientific object, in THE THIRD TABLE, is in fact parasitic on internal realism, but is a dumbed down version. Questioning objectivity of objects and reconfiguring our notion of it is not in itself correlationist. Deleuze explains quite clearly that even in the first Foucault of the archeological period language does not constitute the materiality of objects, the visiblities do not coincide with the enunciatables. Also the archive is a material element, it does not constitute itself. Once « correlation » is extended to anything that has a similar structure it loses any definite meaning, as anything and everything can be called « correlational ».
J’aimeJ’aime
Ping : CORRELATIONISM, PLURALISM, AND REDUCTIONISM: Kolozova, Brassier, Wolfendale and Harman | AGENT SWARM