Levi Bryant interestingly links his thesis (and Graham Harman’s) about objects being withdrawn from relation, or from contact with other objects — with autopoiesis, making the connection via Katherine Hayles’ How We Became Posthuman — here.

This is useful to me, precisely because it links together two arguments that I tend to dislike: withdrawn objects on the one hand, and autopoiesis on the other. Both of these arguments seem to me to produce a kind of solipsism: for Harman and Bryant, the object is firewalled off from contact with any other objects (this argument actually goes back to Leibniz’s monads, as Bryant notes); and for Luhmann and Varela and other advocates of autopoiesis,

systems are informationally closed –radically alters the idea of the informational feedback loop, for the loop no longer functions to connect a system to its environment. In the autopoietic view, no information crosses the boundary separating a system from its environment. We do not see a world “out there” that exists apart from us. Rather, we see only what our systemic organization allows us to see. [Bryant quoting Hayles]

Now, what I am saying here is not (yet) an argument — it is just a statement of the position I want to argue for. BUT — it seems to me that the whole excitement of “speculative realism” or the rejection of correlationism, and of object-oriented ontology in particular, is precisely that it brings us to recognize a plethora of things (objects, entities, occasions, processes) that ARE “out there” and that “exist apart from us.” To emphasize autopoietic closure is to return us to epistemology instead of ontology, to make the alleged impossibility of knowing (anything about) what is out there take precedence over the ontological affirmation of the reality of objects. The supposed withdrawal of objects from one another is only an epistemological matter, not an ontological one.

This is why I favor Whitehead’s and Tarde’s position over that of Harman and Bryant (and ultimately Leibniz — for this is precisely the point where Tarde and Whitehead revise Leibniz). Objects are not isolated from each other. They are continually bumping and jostling against each other, and affecting each other. An object’s knowledge of other objects is indeed merely perspectival and partial (in this sense I agree with Harman’s and Bryant’s claim); but the way objects affect one another cannot be reduced to their knowledge of one another. Harman says that when fire burns cotton, the fire doesn’t engage with all the properties of the cotton. I would say rather that the fire, in destroying the cotton, does indeed engage with the cotton in its totality; I would revise Harman’s claim to say that, although the fire affects the cotton utterly, it doesn’t “know” all the properties of the cotton. I think we need to insist upon the distinction that the fire affects more of the cotton than it “knows” — although I am putting “know” in scare quotes, I actually think that it might be better to use the word quite literally. The fire only has limited epistemological access to the cotton, but ontologically the fire and the cotton are in total contact and interaction. Of course, this leads me down the path to panpsychism — but, as I have already written, at this point I am entirely willing to go there.

PS: Harman’s reply to this is here.