On “theory” in Graham Harman’s theory of everything

There’s an interesting new technical point in Graham Harman’s most recent book, Object-Oriented Ontology: A New Theory of Everything (part of the revamped series of Pelican introductions). “Technical” unfortunately means I’m jumping right in without much background explanation.
Harman - a new theory of everything
Here’s a preview: In a section of his new book, Harman devotes special care to “theory,” which in his philosophy refers to the tension between sensual objects and real qualities (SO-RQ). He frames his discussion as OOO’s response to the problem of good and bad knowledge, and explains it by inverting another object-quality tension, “allure” (RO-SQ), the aesthetic tension that has so far gotten the most love. He associates theory with the idea of a paradigm, meaning the background structure of intentionality that renders things visible to us. I think a subtle but important aspect of this description misses the mark within an object-oriented framework. I’ll also discuss a related issue. Readers like Jon Cogburn have favored allure as the sole agent of indirect contact with the real in OOO. Yet I think we have good reason to reconsider this view, and it seems Harman believes this as well…

* * *

The Pelican overview of OOO does its job very well. Harman x-rays the bones of his metaphysics, aesthetics, social theory/historiography, and ideas about politics and formalism in ethics and aesthetics. And he gets it done with some of his best rhetorical panache. Here’s a memorable example from the beginning: Instead of diving right into his famous (or ought-to-be-famous) concepts of overmining and undermining, he acknowledges the big question many readers will have. Why can’t physics provide a theory of everything? Surely philosophy has nothing to say, since it’s just a decadent and slightly embarrassing has-been.

To shed light on what a unified theory of physics fails to account for, Harman uses what he has elsewhere called the “hyperbolic method.” He imagines a version of his object of critique without any circumstantial accidents holding it back from complete dominance.

Imagine that good reasons are eventually produced for preferring one variant of string theory to the others, and that someone also comes up with a brilliant experimental arrangement that confirms the truth of the theory. In this case not only would our fictitious scientist Browne be right to call string theory ‘the only game in town’, since it would now be something more than just one game among others; but string theory would have become textbook science, learned by students everywhere as a basic fact about our world, much like Einstein’s theory of gravity or the periodic table of chemical elements. My claim is that even under this optimal scenario of maximum scientific triumph, string theory would still not be a ‘theory of everything’. [25]

The teaser works, right? The actual content after this point should be familiar to Harman’s regular readers. Even a triumphant string theory will make ontological assumptions that prevent it from accounting for everything. For instance, it will assume that everything can be explained by the relations of smaller, simpler physical stuff. But that leaves out the emergence of non-physical realities, such as corporations. Objects at all different scales have features not found in their components. If those properties are predictable, fine, but that doesn’t diminish Harman’s point. An object such as a marriage “does not require mysterious results” to be emergent. It only requires that “the married couple has joint features not found in either of the individuals in isolation.” The triumphant string theory also has nothing to say about fictitious entities or about anything that cannot be stated in literal prose terms. Throughout this opener, the spirit of the hyperbolic method leads Harman to anticipate objections in their strongest form, and this goes a long way to getting readers on his side.

The book has newer material, too, mainly in the last third. Chapter 6, for instance, gives an overview of OOO’s allies, “fellow travelers” (Bennett, Garcia), and adopters in architecture. These parts are meant to demonstrate Harman’s ideas in action, not to push OOO itself forward. A great example is the architecture section. In an eye-opening look at OOO in the field, Harman echoes a conceptual distinction he made in Weird Realism between a “cubist” aesthetic (SO-SQ) and an “allusive” or noumenal aesthetic (RO-SQ). He suggests Mark Foster Gage is more interested in the former, and Tom Wiscombe in the latter. To me this raises tricky but interesting questions about “applied” philosophy in the arts.

The exciting new conceptual moment for me – and the subject of this blog post – comes about halfway through, at the end of Chapter 3. After explaining OOO’s central notion of indirect causation, Harman raises a new topic: Knowledge. “Wait,” you say. “Doesn’t he treat knowledge as the ‘duomining’ reduction of an object to its parts and effects?” Yes, but here he wants to replace his polemical use of the term with one that is compatible with the indirect cognition he favors. The important thing for readers to keep in mind is that instead of critiquing “scientistic” ontologies (see the thought experiment, above), he now wants to give an object-oriented description of knowledge production.

* * *

A couple things jump out at me about this section. To start, I was reminded of Jon Cogburn, one of Harman’s sharpest readers. Cogburn claims in his book Garcian Meditations that unlike allure, theory fails to resolve the “OOO Paradox.” This refers to OOO’s portrayal of reality in terms of a strong gap between reality and its portrayal (the sensual). Rationalists like Adrian Johnston would claim OOO fails to sidestep the OOO Paradox. Harman responds in Chapter 2 that Johnston presents a false choice. He assumes “we either ‘capture’ a thing in terms of literal categories, concepts, predicates and properties, or else we are left with nothing better than ‘facile mysticism’ and ‘negative theology.’” Harman’s alternative is an indirect, allusive access to reality. Cogburn buys this. He just doesn’t buy that theory offers it. Only allure does. The reason, according to Cogburn, is that theory is a case of duomining, explaining things by what can be mapped and measured:

The way Harman describes [Husserl’s] eidetic reduction strongly suggests that he would consider the scientific attempt to map and measure various of nature’s capacities as an instance of ‘theory’ [SO-RQ]. But if this is right, and if it were the case that metaphysics were also theoretical, Harman would straightforwardly face the OOO Paradox. [GM 73]

Insofar as theory reverse-engineers real qualities and OOO claims to be an instance of theory, it faces the contradiction of trying to map what can’t be mapped. The solution for Cogburn is simply to ignore theory and turn instead to allure.

If metaphysical explanation were properly characterized as a species of allure, rather than theory, then there would be no reason to think that explanation of a Totality must proceed in terms of entities outside of that Totality. One of the things creative artists do is make sense of things in new ways. Part of how someone writing a narrative history gets their new perspective across is by arranging things already at hand in novel ways. Metaphysics then would be a bit Janus-faced, using language in common to over- and undermining [read: literal, Johnston-approved] explanations but to a fundamentally different purpose…Is philosophical wisdom really analogous to what is delivered to someone who responds properly to aesthetically valuable experiences? [GM 75. Cogburn calls this a ‘tantalizing possibility.’]

According to Cogburn, any indirect access must be a form of allure (RO-SQ). This even goes for real qualities. He explicitly claims as much in an important footnote:

Harman does at one point [in Quadruple Object, p. 28] say that ‘allure’ is used when we gain epistemic access to real qualities…Real qualities are themselves objects with a fourfold structure. We approach them qua real objects via allure. [GM 88, n. 30]

Sure enough, the page in Quadruple Object that Cogburn cites is about RQ—yet I don’t see any mention of allure. If anything, it is agnostic about how access to RQ works. Harman states there that real qualities are necessary for a sensual object to exist, and they are withdrawn from the intellect no less than the senses (sorry Husserl). For that reason, RQ “can only be approached indirectly by way of allusion.” OK, but does allusion here count as allure? Is there only one mechanism for alluding to the real? Are real qualities approached qua real objects?

In the Pelican book Harman still seems agnostic on the question. Let us consider each side. Certainly, he wants to answer “no.” In a nutshell, “OOO is completely opposed to the idea of knowledge as direct access to the real…We also know that knowledge cannot be metaphorical in character.” And a bit later: “Whatever the difference between better and worse medicine may be, it cannot consist in good medicine having a more accurate picture of reality than bad medicine does: for between any picture and the reality it depicts, the gulf is absolute.” This looks like he disagrees with Cogburn.

cogburn diagram

Yet he also writes that “knowledge, by contrast [with allure], must ascribe genuine qualities to the entities it knows.” And later: “knowledge cannot be metaphorical but must be literal, in the sense that the object of knowledge [SO] must be reducible to its qualities in such a way that it amounts to nothing but a bundle of components or effects,” the latter of which happen to be mysterious real qualities. The key difference between dogmatic and object-oriented knowledge is not that the former is Humean (bundle-oriented) and the latter is object-oriented. Both are Humean! The difference is that object-oriented knowledge is a bundle of mysteries. Is this sufficient for OOO to pass the OOO Paradox by instantiating its description of theory?

Given the above, Cogburn can still assert that theory ultimately has to work like allure. In this alternative to “normal science,” object-oriented scientists would approach their craft aesthetically. That isn’t such a crazy idea. Many a metaphor have guided the working assumptions of scientists. In any case, let’s assume SO-RQ refers to a bundle of mysterious qualities. What do we get when we specifically turn our attention to real qualities? More objects! Hence theory would be an undermining species of allure (replacing SO with a bundle of real objects). That’s really Cogburn’s point, I think. [He’s supported by Quadruple Object, 131.]

And we should keep in mind another possible connection between allure and science. Harman used to be less precise about the ontology of knowledge, but he clearly held that the underlying paradigm transforms in sudden leaps by way of allure. Take this passage from Guerrilla Metaphysics (p. 152):

In more recent times, [allure] is also the best way to interpret Thomas Kuhn’s notorious “paradigms” in the history of science. A paradigm is not an arbitrary principle constructed by a social community in a contingent time and place and imposed by the power of the mob, but rather the rule of a unified scientific object beyond all nail-filing arguments and contradictory evidence and public cataloguing of its traits. When the electron is introduced or phlogiston abandoned as sheer fantasy, the regime of objects has shifted, even when the jury remains out as to the details. “Normal science,” like normal perception, tidies up our lists of known properties and fixes previous inconsistencies in our map of things, but does nothing to shift the underlying field of objects that are accepted as real. Allure, with its severing of objects and qualities, is the paradigm shift of the senses.

Harman notes in the Pelican book that the paradigm (Kuhn) or scientific research program (Lakatos) has a “hard core,” which he associates with theory (SO-RQ), not allure. But I think he still would treat a paradigm shift as an instance of allure, and that sounds fine by me. Nevertheless, there may be an intermediate kind of transformation that Harman is not accounting for here, and if so, then he would be wrong to claim that “we can look on dispassionately at a riddle of knowledge” [183].*

How does theory work? It is the inversion of allure, so let’s review allure first. From 2014 onward, Harman has explained allure in “theatrical” terms, meaning the real object (call it the beholder) adopts the sensual qualities of the alluring object. Here’s the Pelican book’s diagram. It only looks complicated because it describes two steps – cancellation and substitution – rather than one:

allure diagram

Step one: Cancel the SO. “By assigning improbable but not impossible new sensual qualities to the sensual object – such as the metaphorical ‘wine-dark sea’ rather than the literal ‘dark blue sea’ – the sensual object ‘sea’ is cancelled (hence the crossing out of SO above), being unable to uphold such unusual qualities. A mysterious real object is needed to do the job.”

Step two: Substitute the RO. “But since sea as real object withdraws inaccessibly from the scene (hence the exclamation point! on the uppermost RO above), the sensual qualities of the metaphor are supported instead by the only RO that is not withdrawn from the situation: I myself, a real experiencer of the metaphor.”

There is no workaround in OOO to make a real sea available, no Heideggerian unveiling. From the surprise of the sea having wine qualities we only experience its seductive withdrawal (step one). So we replace the missing RO as “thespian beings” and perform its qualities, though not necessarily in the literal volitional sense (step two). Indirect experience of the real in OOO means roleplaying the unreal. And it marks the production of a new object, an attachment. “Rather than burrowing downward towards the reality of the sea, the metaphor builds a new theatrical sea-reality upward, on top of the sensual qualities of the wine” [181]. Aesthetic experience is “unjustified true belief.” Unjustified because it does not work by literal analogy, and true because it produces a new real object.

* * * 

In the Pelican book, Harman wants to show that theory uniquely points to the indescribable through the inverse of allure, or “justified untrue belief.” I think this phrase misleads more than it enlightens. Start with “untrue.” Theory (SO-RQ) obviously cannot be true in the sense of producing a new object; that’s what allure does. Nor can theory be untrue in the sense of translating but not mirroring the real, because that describes sensual qualities rather than real ones. No: either theory is true in a different way than allure is true, or untrue in an unspecified way that is not the opposite of allure’s truth. Now consider “justified.” When Harman calls allure unjustified, he means that aesthetic belief is not based in literal comparison or a “prosaic statement of identity.” Theory is also not a prosaic statement of identity, yet in a way that differs from allure. And if we call theory justified it is not by comparison with allure.

My criticism of Harman’s wordplay sounds harsher than I mean it, because (here comes my own wordplay) I think he was justified deep down to go in that direction. The philosophical underpinnings for it can be found in how he inverts the structure of allure to get at how theory works, a move that I find to basically be true.

theory diagram

Step one: Cancel the SQ. “The evident qualities of a thing are too shallow to provide us with genuine knowledge (hence the crossing out of SQ above).”

Step two: Substitute the RQ. “OOO holds that real qualities – no less than real objects – withdraw from both sensual and intellectual experience: hence the exclamation point! on the uppermost RQ above. For this reason, the sensual object SO can combine only with the substitute RQ that I myself as the knower bring to the table.” Those real qualities are “the unnoticed background assumptions that make [the SO] visible to us.”

Harman spends little time on step one. For my own part, I intuitively accept that “once we are focused on gaining knowledge about the sun, its sensual qualities are cancelled as being no longer enough.” A nagging question, though: What makes the evident qualities of a thing seem too shallow? What is the ontological feature of our attachment to the sun that causes us to focus on gaining knowledge about it?

Let’s skip the question for now. Theory refers to a sensual object that is “made up of elusive real qualities that are somewhat known…without any claim to access them directly or exhaustively” [187]. And those real qualities come from the beholder. (Notice he claims SO is made up of RQ, not in tension with them…) When the object-oriented doctor makes his diagnosis, he does not just robotically compile data and compare it with reference definitions like his dogmatic colleagues do. He recognizes a medical disorder, and he works to narrow down its precise nature. “SO can combine only with the substitute RQ,” the tacit expertise of the doctor. Or take the experienced stock analyst at a hedge fund who recognizes a supposed value company as a dud and wants to test his hunch with a closer study of the company’s accounting practices. In each case, we produce literal knowledge in accordance with “the unnoticed background assumptions that make it visible to us.” Harman associates this ontological structure with Kuhn’s paradigms, Lakatos’s research programs, and McLuhan’s background mediums.

One thing about this model of theory has been bugging me, though, and I’ve already hinted at it. In Quadruple Object (Chapter 7), Harman distinguishes between fission and fusion. When a sensual object’s link with its qualities is at stake, the quality pole splits from it. And when a real object’s link with its qualities is at stake, it fuses with the quality pole: “Whereas sensual objects are conjoined with their qualities in advance, such that fission between the poles is required, the real object is absent from the sensual field; hence, real object and sensual qualities will meet only when fused.” Allure as described above certainly fits the bill. The substitute RO fuses with SQ.

What about theory? Remember, according to Quadruple Object, SO separates from RQ because in their banal form they are already linked. “The word theory can serve as our term for the fission that splits a unified sensual object from the real qualities it needs in order to be what it is.” Yet in the Pelican book Harman indicates that SO combines with the substitute RQ; it is made up of the substitute RQ. His examples imply that theory enacts this fusion. I could be missing something – I usually am – but right now I think a more consistent interpretation of theory would start by observing that SO already involves unnoticed background assumptions prior to fission. Insofar as the object fits my selective form and I take it to be what it seems to be, my RQ conform to it and there is no theory to speak of. A child’s suspicion that something is inside the plastic Easter Egg or that an adult has hidden the Passover afikoman somewhere is not so much an instance of theory as “simulation,” or SO-SQ, the literal playing out of a known object’s qualities. Theory occurs when this unstated background becomes at issue in some way. For instance, if the child suddenly realizes the location of the afikoman had been secretly given away to a favored sibling, its significance for the child changes in disturbing ways. From this perspective, theory is when the form of my pre-existing attachment to SO transforms.

The above interpretation reminds me of another line from Quadruple Object: “we need fission accompanied by fusion.” Here’s a quick ‘n dirty shot at gathering three tensions under one roof. Allure (RO-SQ) names an indirect relation between real objects. For the beholder, it produces an attachment with a sensual object that is in tension with its qualities. This attachment-object was not there before, or else it would have already split from its qualities. (See Harman’s comments on “buffering” or “black noise” in his early writings.) In other words, allure produces a new sensual object that is difficult or “at issue” to some extent. It is unjustified to the extent that I “accept the mystery,” but true as a reality in itself. Next, the SO endures in my experience as I wrap my intentionality around its qualities (SO-SQ). And insofar as my form and that of the object do not cohere, its real qualities indirectly become at issue via my own real deportment (SO-RQ). This is called theory in OOO, and certain techniques, like the hyperbolic method described at the top of this post, are custom-tailored to encourage it.

Let us accept that theory involves “a form of expertise that is somehow better than non-expertise, even though it cannot consist of correct literal paraphrase of whatever subject matter concerns it” [186-87]. But to be more precise, theory occurs when this expertise is made at issue. It is justified in the sense of being retroactively “justifiable” or rendered no longer at issue, but the knowledge produced along the way is untrue in that it merely translates the real.** Theory tests allure, removing irrelevant debris that was either itself the duplicitous cause of allure or a dusty hindrance to the attachment.

Theory is different than what is meant by an expert’s “working theory” or paradigm. A chair designer dealing with the human butt does not design swords or flamethrowers. With her working theory, she simply conforms to her object. If object-oriented theory occurs, it is because the butt has become a vexing target of concern rather than a set of assumptions she enacts. The object-oriented scientist is a bit like that designer, constantly testing how well her uncertain beliefs conform to the cosmos.

* [July 11 note] Speaking to Harman’s point that boredom does not destroy the object of knowledge: If the object being studied is just a boring specimen in a theoretical experience (SO-RQ), then it is in the “SQ position” of the proper object of knowledge. That is, it’s just part of the background medium or paradigm the way a physical book is when I’m encountering ideas. My more basic point is that I do not see how SO can be a means to an end in OOO. My mind does not always encounter the same objects that my body does. (I do not mean this in an ontological sense; it’s just an observation about the content of my experience.)

** [July 11 note] Following what I said earlier, we could say theory is unjustified because it does not uncover the real, and true because it transforms the real.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s