Graham Harman’s book on Dante (Pt 2/3)

Following Part 1.

In the past, Harman has straightforwardly associated his model of causation with aesthetics. He shakes up that position a bit in Dante’s Broken Hammer. Though he continues to criticize anthropocentric metaphysics, he is skeptical of an art without humans and an ethics without things. To reconcile these views with OOO, he whittles down ethics and aesthetics to more specific (if still indefinite) regions of the world. As he teases in chapter 3, “the surprise of a broken hammer is not quite an aesthetic experience, [and yet] the broken hammer’s rift between the withdrawn hammer-object and the obtrusively visible hammer-qualities does stand at the gateway of aesthetics.” Here is an overview of the subject matter of the book.

The main philosophical topics in Dante’s Broken Hammer are: (1) attachment; (2) anti-formalist ethics (Kant’s two senses of autonomy); (3) charges of panpsychism and correlationism; and (4) aesthetic mimesis.

(1) Attachment
Harman argues (against formalism) that the basic unit in ethics and aesthetics is a composite entity rather than an isolated rational being or object of experience. Ultimately, attachment refers to how the real object in an intentional/sensual relation (RO-SO) manages to contact something real, and thus produce a third object that remains autonomous from either of them. Aesthetic attachment affects the real object through its mimesis of the sensual object. This idea does have roots in Harman’s earlier work. One of the weird but intriguing claims in the 2005 book Guerilla Metaphysics is that the real object in an attachment is passive (ch. 11; he doesn’t yet use “attachment”). When we say a spectator contacts something, what actually happens is that the spectator invites contact by the other object, like a host proudly letting in a salon of storytellers and musicians. In my understanding of DBH, ethics and aesthetics are not two different examples of the same metaphysical tension, but refer to two different metaphysical moments of an attachment.

(2) Ethics: Good Autonomy, Bad Autonomy. (Chapter 2)
Harman’s argument against formalism is not new for him, though his polemical use of the term is. My favorite concise version appears in his Quentin Meillassoux book (1st ed., p. 132). He observes that Kant’s career-defining notion of finitude – autonomy in DBH – conflates two distinct meanings: a human/world dichotomy (bad) and generic finitude (good).

(This reminds me of some comments by Peter Wolfendale in his article against OOO titled “The Noumenon’s New Clothes.” Wolfendale treats the notion of a fundamental gap between thought and causation as a self-evident truth. Since it is apparently “really contentious” to ascribe finitude to anything other than the limits of thought, any need to address Harman’s analysis of Kantian finitude vanishes. A formalist taxonomy, by contrast, offers both the triumphant certainty of knowledge and the humbling aggregate of what we don’t yet know, like Carl Sagan’s famous pale blue dot: “when [an encountering object] is understood as a knowing subject, the extrinsic explanation of excess has traditionally taken the name of finitude… [which] needn’t be interpreted in terms of some common qualitative excess, but could be seen as a disparate quantitative excess.”)

Back to Harman. So there’s bad autonomy and good autonomy. Kant’s own preference for “abstracting from all objects of our interest” is anathema to his own philosophy and to the viewpoints of others, such as Dante, Max Scheler, and Alphonso Lingis. Harman quotes Scheler on how love is instead a composite of two things:

In our account love [is] thus always the primal act by which a being, without ceasing to be this one delimited being, abandons itself, in order to share and participate in another being…This participation is such that the two in no way become real parts of one another.

At this point Harman only emphasizes the first sentence, the Latourian claim that ethical units are composites. Overall, Harman agrees with Dante and Scheler that this composite involves “a relation of love between one entity and another.”

Kant’s good sense of autonomy plays an important role, though. We learn in the next section that the attachment as a whole “does not exist for any heteronomous purpose but as an end in itself.” We should note a further consequence of this point. The second half of the Scheler quote – “the two [entities] in no way become real parts of one another” – implies that love causes a being not only to abandon itself in its attachments, but also to respect the boundaries of its beloved things. Love mingles without merging, and digs without destroying. Harman usually describes the object-oriented sense of autonomy in terms of the entire composite, rather than its parts: “The ethical must be self-contained, an end in itself, or it is merely a means to some other thing.” Yet I also take this to mean that an ethical comportment regards the sensual object itself (or rather, the presumed reality behind it) as an end in itself.

Harman notes that the conflict between different rankings of worthiness helps to explain rebellion or hate. In the same way, it seems that the respect for the independent reality of the beloved object provides object-oriented ethical theory with a source for good and bad conservative impulses (not to mention good and bad liberal impulses).

(3) Panpsychism and Correlationism.
Before going further, I should summarize how Harman handles a couple charges against his philosophy. The first is panpsychism. He has repeatedly noted in the past that his translation-based model of relation is not psyche unless one interprets “psyche” in an unusually broad way. I’ve always found this convincing. Harman further sums up the cost of ascribing panpsychism to OOO:

The basic unit of ethics, aesthetics, and metaphysics is…a composite in which the “subject” term need not be human. To call this “panpsychism” is to assume that we know what a psyche really is, that humans alone should be allowed to have it, and that the relation between real and sensual objects must already be called a psyche. But all three assumptions are false. (ch. 4)

If we treat DBH and Immaterialism as a combined attempt to articulate a basic psyche within Harman’s overall theory of vicarious relation, then the panpsychism epithet holds less water than ever before. That “If” points to the question, though, of whether Harman still interprets psyche as existing on the interior of a higher entity, like he did five or six years ago. Does he now distinguish psyche from other kinds of interior existence? The third item in his list of false assumptions indicates that RO-SO need not already be called psyche, which could be taken to mean that it is false to assume the link with a sensual object is psyche unless it indirectly contacts something real. But then again, Harman claims in an article on vicarious causation (2007) that sensual contact is itself contingent on the fusion of two real entities: “[A]lthough intentionality seems to be a relation between me and the sensual pine tree, this is merely its interior. The intention itself results only from the unexplained vicarious fusion of me with the real pine tree, or with whatever engenders my deluded belief that I perceive one.” In any case, my provisional answer is that “ethics” and “aesthetics” each still apply in distinct ways to asteroids, people, tarps, music, and bees, but only in a suitably general sense – hence the scare quotes.

This brings me to the second criticism. Since Harman claims that ethical and aesthetic attachments are typically grasped in terms of the human-world variety, he now has to double his vigilance against the charge that his own theory falls into correlationism. His main theoretical point is that correlationism is a species of formalism, because it takes the human-world division to be ontologically primitive:

For at bottom the correlationist accepts the modern taxonomy between subject and object, and merely wants to show that these two domains are not as separate as some people think. (ch. 4)

Any human-object pairing ought to reflect a more general ontological structure, rather than have its own special features inflated into such a framework. And prematurely emphasizing the human/world dichotomy is not just an abstract problem. For instance, it can influence the way disciplines divide their labor. A formalist taxonomy is supposed to save science from vitalism and philosophy from scientism, but in effect it finds “no other way of speaking about non-humans than through science.” Another way of putting this is that the term “formalism” is a warning about begging the question: like objects in mirrors being closer than they appear, conclusions about disciplinary propriety are sometimes closer to your presuppositions than they appear.

By contrast, Harman holds that real-sensual composites need not be limited to isolated human subjects or pieces of the non-human world, even if particular domains like art and science typically involve humans or non-humans. Furthermore, correlationist standpoints tend to underemphasize or ignore how my stance toward the world partly defines my own executant actuality (consider the practically standard view that I am “co-constituted” by my relations), and how each object in a compound entity has a non-relational reserve that can surprise others. Of course we must keep in mind that for Harman this surplus is not a storage depot of untapped capacity or formless material. It is a formed surplus of other objects not currently involved in the attachment. These contiguous non-relational props on the interior of each object do not yet make a difference – but they might!

I do think the correlationist charge raises at least one question worth taking seriously. Does Dante’s Broken Hammer adequately distinguish Harman’s specifically human ethics and aesthetics from his metaphysics? According to my current understanding, I would say “no, but in an illuminating way.” Harman does not try to set humans apart from pigs, flowers, whiskey, or operating systems.* Instead of defining humans typologically, Harman describes two different modes of human being, and these modes seem to reach beyond specifically human ethics and aesthetics.

But as Harman is always keen to point out, this cannot be taken to mean that his philosophy treats zebras and novels about zebra commandos the same way. In fact, it entails quite the opposite. Just because they are all objects with qualities does not mean they are all the same: the ontological task is to dig into particular things and unlock their inner strangeness. As he writes in all italics: “individual content is always more than its surface content.

[* Well, he does alert us to some possibilities in a rhetorical elaboration of how formalism leads us astray: “We are historical, think and speak at a high level, and undergo the repression of thoughts into the unconscious… While all of this is true, it does not justify basing ontology on a taxonomical dualism…” And in a wonderful passage somewhere in Guerilla Metaphysics, he implies that humans are not more critical than other animals, but more object-oriented.]

So how do we describe the interplay between amorous ethics and alluring aesthetics? Harman suggestively paraphrases one of Dante’s ideas, which he also quotes multiple times (from Canto XVIII in Purgatorio):

Awakened by the image, the soul can turn steadfastly towards the object with love, and then the nature of that object unites with the soul through beauty. In short, love and beauty offer a path to the real that does not fall short, as does perception or knowledge, of the inner substantial forms of things.

(4) Aesthetic Mimesis. (chapter 3)
In the arts, the formalist typically “treats the artwork as self-contained,” independent of its historical conditions and context. That seems object-oriented at first, but for Harman it also risks the bad kind of formalism by deciding in advance what belongs in art. Indeed, art critics such as Clement Greenberg and Michael Fried ultimately agree with Kant that aesthetics must be rooted “either in the subject or in the object.” The key difference is that Kant chooses one side, and Greenberg and Fried the other. Kant picks “the deeper conditions of experience that make human agreement and certainty possible.” His attempt to devise universal criteria for human experience thus reflects a desire for “an objectivity that nonetheless makes no pretension to grasp the things as they really are.”

By contrast, Greenberg and Fried emphasize the object side, “uncontaminated by human subjectivity.” For them, art objects exist apart from humans (despite their use of terms like “experiences”), sometimes in tension with a unitary background. Greenberg emphasizes the “flat, continuous, finitely bounded surface” of the painter’s canvas, and Fried outright dismisses the notion that humans are theatrical ingredients in art, rather than just external perceivers and judges. Cleanth Brooks joins them with his views on literature – see the last third of Weird Realism – and from what I’ve seen, many music scholars have similar attitudes. Though misguided as an ideal, I have noticed that formalism has an affinity with a very useful idea, namely, that a certain dissonance between spectator and art object is necessary for aesthetic contact to occur.

And Harman has agreed with that last point since the beginning of his career. A successful metaphor, for instance, “cannot be either too convincing or too unconvincing.” Some similarity between the two objects enables contact. But in order to be aesthetically resonant, their connection can’t just amount to their shared qualities. The metaphorical figure affects the ground without displacing it or symmetrically allying with it (see Immaterialism). As Harman writes in DBH: “This trivial similarity between them provides an alibi for a significantly more farfetched but also more beautiful merger.” Harman calls beautiful mergers allure. In his updated aesthetic theory, he calibrates the concept to account for an ostensibly special kind of allure that involves a beholder of causation. In short, the spectator mimetically takes on the role of the real object. (I should reiterate that Harman’s aesthetic theory reaches further than the fine arts. Since he leaves the details of that problem up in the air, though, let’s stick to the arts.)

Harman first gave a preview of where he is headed in his 2014 article, “Materialism is Not the Solution.” The description of mimesis in Dante’s Broken Hammer is even more memorable. The way I picture it, Harman must have started with the question of how artistic effects like metaphor can be said to make their objects “more real.” In other words, how does launching imaginary flame qualities into orbit around an imaginary cypress tree bring me into contact with the executant reality of a cypress? It is not enough to claim that metaphor merely simulates causation. From the OOO viewpoint, compound aesthetic units of spectator-plus-object must be real. Aesthetic attachment cannot just refer to our attachments with images, but to some kind of indirect contact with reality. Otherwise, we fall into the trap of explaining aesthetic contents in terms of a fundamental difference between human representation and everything else.

Harman finds his answer in the way a mask transforms its wearer and a theatrical character transforms its actor. When a metaphorical flame fuses with an imagined cypress tree, they actually fuse with me in the role of cypress. Similarly, when music entices me into its depths, it comes to life by uniting with me as a surrogate for its medium.

It is not that [the poet] López Picó or his reader produce an imitation of the cypress, but that they become the imitation of the cypress, just as actors become the characters they play.

Formalism proposes a division between a productive rational human agent and inert cypress image. But object-oriented mimesis refers to the formation of an emergent reality – me as cypress – which exceeds its two components of real human and sensual cypress image.

I have claimed that Harman’s ethical and aesthetic theories are metaphysical rather than specifically human (or sentient or whatever). But I still think the ethical/aesthetic division teaches us something about humans, since the general metaphysical tensions to which they refer do not just disappear or become meaningless for particular beings. I will try to elaborate on some of that in my next post.

Link to Part 3.


Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s