Pete Wolfendale at Deontologistics has put up a pdf of a paper in which he calls the bluff of anyone who thinks the only options are naive realism, idealism, correlationism, and speculative realism. It's a beautiful piece of work and I urge people to read it carefully. He's also, in an earlier post, provided a handy inventory of his engagements with Harman's work, and some of this thinking shows up in the paper as well.
I am most impressed by Wolfendale's construction of not one but two alternatives to correlationism that don't succumb (he claims) to Meillassoux's arguments-- a deflationary realism, strongly informed by Robert Brandom (Wolfendale is one of a small online chorus that has been urging people to read Brandom for a long while now), which sees (as I understand Wolfendale) the world and thought to be structurally dependent on each other, in the sense that it does not make sense to say that one understands the one without understanding the other. However, they are not dependent on each other in a different sense. Not everything that is true of the world is true of thought; but one cannot understand the one without understanding the other.
These two senses Pete distinguishes (again, following Brandom) as sense-dependence and reference-dependence, respectively. Obviously there is more than a touch of Frege in this distinction, and this is important later on, for Pete is going to offer an account of how we make and dispute claims that reads, to my mind, as a primarily Quinean main course judisciouly salted with Wittgenstein; behind all of this lies Frege. (Wolfendale’s account also has more than a hint of pragmatism).
For Wolfendale’s deflationary realist, the structure of the world and the structure of thought are reciprocally sense-dependent, but not reference dependent. Thus while one must understand either in terms of the other, it remains possible to say truly that the structure of the world is X without saying that the structure of thought is X, and vice-versa. (In an email before I posted this, Pete ventured took issue with this way of phrasing it; and I hope I have got him right now.) The deflationist takes the structure of the world to be reflected by the structure of thought. This isomorphism is, for instance, what is found (as I read him) in the Tractatus. Every fact about the structure of the world corresponds to a fact about the structure of thought. The fact that the world is composed of objects and properties corresponds to the fact that thought has a subject-predicate structure. (I think that this holds whether we take a substance/quality sort of metaphysics, or an actant/power sort, as does Levi Bryant, since the question is not the precise relation between object and property, but the distinction). However, the deflationist can also hold that a world could exist without thought, because this isomorphism between thinking and world is not thought (as in classical idealism) in ontological terms as an identity of Being and thought. Still, there is nothing to the structure of the world that is not to be found (in some sense) in the structure of thought.
I hope Wolfendale will correct me if I have mis-stated his account. However, deflationary realism is not where he stops--he's only getting started. Pete notes that these definitional terms of dependence and independence allow for what he calls a thin concept of reality; but the disputes we tend to have both in philosophy and ordinary life all tend to depend upon a thick notion of reality such as we seem to have intuitively (even though it is infamously difficult to articulate). When I first read this paper I got snagged here; I thought he was using “thick” and “thin” in a way more or less informed by the anthropological use of these terms, which Clifford Geertz drew out from Gilbert Ryle; a usage in which they tend to merge with "objective" description (=thin) and first-person evaluation (=thick). I still think this hangs in the background in a sense (I’ll try to explain why towards the end here), but this is not the primary sense Wolfendale needs as he goes on to show an alternative to deflationism, which he calls Transcendental Realism. This stance dispenses with the reciprocality of sense-dependence between world and thought; that is, one needs to understand the structure of thought to understand the structure of the world, but no longer vice-versa.
Deflationists, Pete wrote me in explanation, don’t take the word 'real' to be doing any genuine work;
For them, there's no real difference between saying 'numbers exist' and 'numbers really exist'. Their notion of 'real' is thus a deflated, or 'thin' one. The classical debates all need the notion of 'real' to do genuine work in order for them to make sense, precisely because it is this notion which articulates the difference between the various positions, e.g. platonism ('numbers really exist') and nominalism ('numbers don't really exist').So the deflationists’ critique succeeds against classical realism (and indeed idealism) only insofar as we concede that there is no good ‘thick’ definition of 'real' which can do justice to these differences. This is one reason why so many analytic and pragmatist attacks on the philosophical debates can dissolve them into “pseudo-problems.” This, by the way, they do share with Meillassoux’s “correlationists.” But this is not sufficient to establish that they fall back into correlationism.
However, Wolfendale is unsatisfied with deflationism, and for good reason. He wants to do justice to the common-sensical experience that we have a good intuitive grasp of a 'thick' notion of real, one that makes sense of these debates. Adequatley defining such a 'thick' notion, and in non-ontological terms (as we can only understand what ontology is in terms of reality), showing how this would work, and drawing out its consequences (pointing to ramifications in fields from philosophy of mathematics to ethics), occupies the bulk of his paper.
Wolfendale’s account combines considerable care with ambition, and refreshingly moves quite at ease between the analytic and the continental, but never self-consciously. He has learned well from Brandom’s ability to see Hegel and Quine (for instance) on the same page, even as he criticizes what he sees as Brandom’s deflationism. (I should add that I defer to Pete in matters about Brandom as he is obviously more informed about him than am I.) I am struck by his recourse to the practices of revisability and how this chimes with the need to be able to account for the possibility of falsehood. For Wolfendale, genuine description (as opposed to interpretation) is possible if and only if we can give complete authority to the world over whether what we say is true or false. (I read this as Popperian at first, though Pete claims to feel much closer to Lakatos--“a strange hybrid of Popper and Hegel,” he wrote). I.e., real description is possible, precisely because misdescription is possible; because one can be wrong.
This resonates with a passage from Plato (Sophist 241) I recently cited as germane to the reading of both Harman and Bryant:
testing the strength of the philosophy of our father Parmenides, proving by force that in a certain sense, non-being is, and being is not...[for] unless these words are either disproved or accepted, no one who speaks about false words, or false opinion—whether images or likenesses or imitations or appearances—about the arts which have to do with them, can ever help being forced to contradict himself and make himself ridiculous.Funny how it always comes back to Parmenides and "being and thinking are the same."
I have to leave out a tremendous amount of detail here, obviously, but I want to mention one important strain in Wolfendale’s paper, which I find personally refreshing, but also important. This is his rehabilitation of Kant. This is not a reactionary circling of wagons—no one online has engaged more with object-oriented philosophy in a spirit of non-hostile criticism than has Wolfendale, as one can see from his extensive posts on the subject. Rather, it is a sense of going to Kant to re-read him again in the wake of the coup that Heidegger staged against the neo-Kantianism of Cohen, Cassirer and Rickert. I like especially Wolfendale’s use of Wittgenstein and Nietzsche, including them in the lineage that descends from Kant. I have long thought that Wittgenstein and Kant are making similar moves: both are in effect turning the liability of thought’s limitation into an asset that becomes the very condition for the possibility of insight. As to Nietzsche, his reaction against Kant was so famously antipathetic that it must have concealed some too-close-for-comfort recognition.
One cannot help but notice that the anti-Kant animus of speculative realism provokes not only arguments but passions. "Am not!" and "Are so!" start to fly. (I assume I need not name names here.) I do not say there is something wrong with this; what it suggests, though, is that this is where the rubber meets the road. When people start to get riled, we are touching upon values. This is why I maintain that the “thick” of reality description Wolfendale wants to articulate does indeed have something to do with evaluations. Harman has argues that correlationism has been the default position for the past hundred years. Is it just coincidental that this has coincided with the famous, all-too-infamous “analytic-continental divide”? What’s right with Wolfendale’s paper however is that he doesn’t think of himself as “bridging” this gap; he just draws upon the material that’s there. But above all, he’s pointing the light at a place where we need to look.