There would seem to be a substantial difference between Graham Harman, for whom correlationism is the philosophy of “human access”, and Quentin Meillassoux, for whom correlationism is any attempt (ever since Parmenides) to say that thought and being are the same.

For Harman, to reject correlationism is to be able to “speak in the same way of the relation between humans and what they see and that between hailstones and tar” (Prince of Networks 124). But for Meillassoux, the reject correlationism is to be able to “think a world that can dispense with thought, a world that is essentially unaffected by whether or not anyone thinks it” (After Finitude 116).

But these are not the same. Harman is saying that all ntities, and all relations among entities, must be given the same ontological status – as opposed to specially privileging relations between human subjects and objects in the world, as nearly all post-Kantian philosophy has done. This means rejecting the model in which human thought actively shapes an otherwise inchoate object-world; but it says nothing about thought per se.

Meillassoux, however, insists that the only way out of the correlationist circle is to think being absolutely devoid of thought.

This makes a difference from the point of view of panpsychism, which does not privilege the human subject in the least, but which insists that thought is immanent to all entities. Though Harman is reserved on the question of panpsychism, his object-oriented philosophy does not disallow it. [Cf. his article in Skrbina’s Mind That Abides collection]. The relation between hailstones and tar may involve thought, just as the relation between human beings and what they see (tar or hailstones, say) involves thought. This is in effect what Whitehead says: every act of prehension has a mental or conceptual pole, as well as a physical pole.

Whereas Meillassoux, because he wants to conceive the absence of thought, and because he insists on the scientific mathematization of reality as the one and only way in which this thinking the absence of thought is achieved, would accept neither Whitehead’s panpsychism, nor the model of interactions that Harman gives us.

[What still needs more consideration here is how Harman’s anti-relationalism affects this question. Harman says both 1) that relations between tar and hailstones are on the same ontological level as (epistemological) relations between an observing human subject and tar as an object; and 2) that relations themselves are inessential. I accept the first of these points, but reject the second.