Both Jensen Suther and Timothy Bewes usefully treat my arguments against them as opportunities to elaborate on and clarify how they think literary criticism works with respect to intention: Suther by way of a contrast with what he thinks of as my resolutely but mistakenly “quietist” account; Bewes by an analysis of what he takes to be my apparently “ever-expanding” but actually “ever-diminishing” notion of intention. In a certain sense, they’re making the same point or, at least, expressing the same frustration. Suther discreetly regrets the fact that my rejection of the ambition to alter anybody’s critical practice has limited the intentionalist impact on literary criticism. Bewes low-key celebrates the fact that, as he now understands it, my increasingly capacious understanding of how intention works encourages me to allow most of the critical claims he thought “Against Theory”-style arguments opposed.

Indeed, at a certain point, Bewes declares himself in agreement with me. As long as the intention “need not designate an event in the mind of the artist” and can be identified, in Nicholas Brown’s formulation, with the “immanent purposiveness of the work,” he’s “on board.” And since a crucial thing about considering the question of intention as a question of action is that it begins with what you were doing (the work) rather than what you were thinking, maybe we really are in agreement.

Happily, however, we’re not. Bewes’ commitment to the (Deleuzian) “cinematic apparatus”—replacing “the unity posited by a human action together with its motivations, causes, and intentions” with “a nondetermined, acentered and decentered perception” [i.e., the one “located in the cinema apparatus itself”1]—involves the opposite of immanent purposiveness. It only achieves immanence by getting rid of purposiveness. The point of the wave poem was that you saw the marks on the beach as a poem because you saw in them the act of writing. But the point of Deleuzian immanence is “the rupture in the very relation between the image and action,”2 that is, between what you see and any action. The achievement of the cinema apparatus is not that it accomplishes the feat of embodying the intention in the work but that it accomplishes the feat of keeping the intention out of the work. A practical problem with the wave-poem example was the objection that once you’d seen it as a poem, you couldn’t unsee it as a poem. The cinematic apparatus is a way to deal with that problem. Instead of imagining the marks on the sand as seen by a person, it imagines them seen by the camera—which has never read Wordsworth, doesn’t know English, doesn’t know what writing is. But where the “Against Theory” argument is that, seeing the work this way, we see there is no work, Bewes thinks we here confront what in the work is un- or non-intentional. We see, as he puts it in talking about Forster’s Howards End, the “thought” that can be “attributed to the work (as opposed to its author).”3

How the cinema apparatus—which doesn’t have motives, beliefs, or intentions—has thoughts is unclear. But in Bewes’ practice, the cinema apparatus is less prominent. Howards End, for example, is hardly an object of pure Deleuzian opticality (it’s in English, and the cinema apparatus doesn’t know English), and Forster’s intentions are said to “contribute” to but not “determine” its meaning (the cinema apparatus doesn’t mean anything). Contributing but not determining is a more usual and perhaps more common sensical view of the role played by intention in meaning, but it’s just as far from the one I am defending, which is that to see Howards End as written in English is already to make its separation from Forster (“the work as opposed to the author”) impossible. (This is what’s helpful about the cinema apparatus; it makes visible what you actually need to imagine that separation.) It’s the inseparability from the author that makes intention identical to the meaning and not a contributor to it. And it’s the same inseparability that makes it the object of interpretation and not, in Jensen Suther’s terms, a norm of interpretation.

Suther’s exemplary norm is that of friendship, itself a Forsterian choice (“If I had to choose between betraying my country and my friend, I hope I’d have the guts to …) and more like Howards End than like the cinema apparatus because the identity of meaning and intention is unsettled by the author’s relation to his own act. He imagines a “mean girl” and a “soft boy,” both of whom pay “ambiguous compliments” to their “friend”—let’s say, “You’re looking better.” But where the mean girl is just being mean, the soft boy “is committed to being kind but is later told he was actually mean.” The interpretive question is how to understand “You’re looking better.”

In literary theory, one answer has been that the semantic and syntactic rules of the language he was speaking determine what his utterance meant. But the problem of interpretation here is not exactly that we don’t know the meaning of the words. And even if it were, that wouldn’t give us an alternative to his intention since to identify the language he was speaking is to identify the syntactic and semantic rules he was following, and the minute you’re interested in the rules he was following, you’ve already interested yourself in what he meant by the sounds he uttered. In other words, the rules of the language matter only because they are the rules he’s using, and the injunction to pay attention to the rules he was using is not an alternative to intentionalism.

To get a real alternative, we need to care about what his words mean in some other language, which sounds implausible but can be made less so if we say what matters is what they mean to us. Perhaps, to take a famous example, by “vegetable” Marvell just meant growing whereas what we hear in it is an allusion to a kind of organic but impersonal emotion. But this still doesn’t give us an alternative to intentionalism. For one thing, it involves recourse to what someone meant (here in the hypothetical form of what we would mean if we used that word). And, for another, replacing what we think the speaker meant with what it means to us actually eliminates the idea that we’re trying to understand the text, as we can see from the fact that it eliminates the normativity that marks the possibility of misunderstanding. If the crucial thing is that you hear “You look better” as supportive and I hear it as critical, the sentence itself is being treated as a Rorschach test, not a speech act. And it’s the same if you hear “growing affection” in vegetable love and I hear “erotic cabbage”; the question of which of us is right doesn’t apply—right about what?

This is what it means to claim that the intended meaning is the non-optional object of understanding. It’s non-optional in the sense that if you’re reading (or hearing) something as a text (or speech act), you’re already treating it as meaning what its author meant by it.

So, how do we understand “You’re looking better today”? Said by the mean girl, it’s an allusion to how bad you’ve been looking and maybe a suggestion that you still don’t look so great (she said better, not good), and she knows it. Said by the soft boy, it’s what? Suther says, “The soft boy is committed to being kind but is later told he was actually mean.” When he said, “You’re looking better,” he meant to be supportive but later he’s told (and presumably comes to believe) that he wasn’t. How might this happen? It might be that he uncomplicatedly meant to be supportive but that his friend misunderstood him. So, the soft boy failed to do what he intended (he failed to cheer his friend up), but his intended meaning is left unaltered. What we have here is a failure to communicate, not a failure to live up to your commitments. It’s the soft boy who understands himself as being kind but is in fact being mean who interests Suther. What does he mean by “You’re looking better”?

This is obviously more complicated, and the complication is built into Suther’s description of his behavior as “self-deceived kindness manifest as meanness.” The self-deceived part means not just that that he thinks he’s being kind when in fact he’s being mean but that he thinks he’s being kind when he knows he’s being mean. And how exactly that works is a question which can’t be answered by saying that he’s failing to “live up” to the kindliness “standard to which he had committed” himself since that takes away the complication—despite his commitment to being nice, he had on this occasion been mean. But, even if we hang on to the self-deception—we say not just that he was mean instead of nice but that he knew he was being mean—the identification of what he said with what he meant is unchanged. He didn’t, when he spoke, understand himself as being mean, but now that he does, he realizes that his previous interpretation of his own behavior (what I said was nice) is mistaken and he now correctly understands what his speech act meant. The “objective intention” (he realizes he meant to be mean) is just a better account of what the “subjective purpose” was. Neither is more or less subjective or more or less objective. They’re both subjective in the sense that they’re both what the speaker meant; they’re both objective in the sense that no speaker can mean anything except in the language he’s speaking.4

And reading is just trying to understand what the speaker/author meant. Suther accepts this but redescribes the object of interpretation (the author’s intention) as a “norm” (the “norm is intentionality”) on the model of the norms of friendship. But trying to understand what someone meant is not like trying to be a friend. When the soft boy fails at being a friend, it’s because he wasn’t in those moments of failure really trying—he was “committed to being kind” but wasn’t, as he subsequently comes to realize, acting on his commitment. But when you fail to understand what the author meant, it’s not because, despite your commitment to “recovering intention,” you were actually doing something else. There’s nothing else to do! Insofar as the author’s intention is the object of rather than a norm for interpretation, misunderstandings are mistaken accounts of that object, not failures to look for it in the right way. In other words, there’s no such thing as being “answerable to the task of recovering intention” because there’s no such thing as not being answerable to the task of recovering intention. Just to see the text as a text is to commit yourself to recovering intention.

Which is why “commitment” and “recovering” are the wrong words. Being committed to coming up with the author’s intention is just reading. So, what then should we make of claims for what is “legible” in texts “irrespective” of their authors’ intentions? How, for example, should we understand Toni Morrison’s account of how different she thinks major works of American literature look (“irrespective of an author’s intentions or a critic’s interpretive schemas,” as Bewes glosses her thought) when considered from the perspective of “a dark, abiding, signing Africanist presence” that hovers “at the margins of the literary imagination.” Morrison says that “Ernest Hemingway, who wrote so compellingly about what it was to be a white male American, could not help folding into his enterprise of American fiction its Africanist properties.”5 The “could not help” suggests that maybe he didn’t want to and didn’t mean to but nonetheless did. This would be soft boy Hemingway, not thinking of himself as racist but racist nonetheless. Of course, that’s a counterfactual Hemingway–actual Hemingway had been committed to the uses of racialization ever since the description of Robert Cohn’s nose being “flattened” and thus “improved” in a boxing match at Princeton that begins The Sun Also Rises. But soft boy Hemingway might understand his first paragraph as criticizing not Cohn but the people who made Cohn feel “race-conscious.” So, we could say of To Have and Have Not (the text Morrison focuses on) that it acknowledges and deploys [as in, “Here we see Africanism used as a fundamental fictional technique by which to establish character”6] the dark Africanist presence even though Hemingway didn’t intend to.

How might this work? Hemingway’s description of the black man who’s part of Harry’s crew (he has a name, but he’s usually called “the nigger”) is pretty unequivocally racist. Is it unintentionally racist? It’s more like casually racist, but the casual racism might be understood to reveal something Hemingway didn’t intend to reveal. The book isn’t intended to be about race, but using a black man to get at what it is intended to be about (manhood), it shows how central race and racism are even when they don’t seem to be. Why did he make that crew member black? Why did he use the black man’s less-than-manly behavior as the foil to Harry’s manly behavior? Hemingway couldn’t help but fold “Africanist properties” into his fiction because the main interest of his fiction was itself constructed in relation to those Africanist properties. So, the unintentional part is that Hemingway didn’t mean for the book to be racist or about race—he meant it to be about manhood—but the book nevertheless makes race central because his idea of manhood is inextricably linked to his ideas about race.

So (assuming something like this is true), does this give us an account of To Have and Have Not “irrespective” of its “author’s intentions” or its “critic’s interpretive schemas”? For sure not irrespective of the “critic’s interpretive schemas.” Like many American liberals, Morrison believed American society to be constructed around some version of and some set of relations to white supremacy, and her reading of Hemingway is dependent on this belief (aka interpretive schema). But if we see it as showing the centrality of race in a way that he didn’t see it, aren’t we at least reading it “for dimensions of the work that its author was not aware of or would not have wanted us to pay attention to”? Isn’t that going beyond the author’s intentions?

Well, there are things in almost every utterance that the author is not aware of. If I make an argument that I believe to be true when it’s actually false, the reader who sees it’s false is seeing something that I was not aware of but not finding some unintended meaning in it. The fact that I was intending to say something true but in fact said something false does not alter the meaning of what I said. (The question of whether what I said was true is beyond my intention but it isn’t some meaning beyond my intention.) If we understood race in Hemingway to work like this, we would want to say that he didn’t mean to be racist but that we can see what he couldn’t and that what he wrote was in fact racist.

But we also see the limitations of this way of understanding it. The explanation of what’s racist about it—he used a racial epithet that even then was controversial, he made the black man inferior, he made that inferiority parallel to and revelatory of the white man’s superiority—all involve descriptions of what he meant to do. So, he did mean to be racist, and our judgment that the text is racist actually is an interpretation of it precisely because it’s an account of what he meant to do. He might be unaware of the racism; he might not even have had the concept of racism. (Interestingly, an n-gram of “racism” shows the word first getting traction around 1940; To Have and Have Not was published in 1937.) But our judgment that what he wrote was racist does not depend on him having the concept of racism. And, yes, a Hemingway who lived to see that concept come into play would very likely not have wanted us to pay attention to that aspect of his writing. But not because he hadn’t meant it.

Bewes correctly predicts that I will have no problem accepting that some things the author was unaware of (e.g., that he was providing evidence of his racism) could figure in the text’s meaning, although the way he puts it is that I will find some “codicil” enabling me to retrofit the original identification of meaning and intention.7 But, as we’ve just seen, you don’t need to revise that identification to see how some things are legible in a text without being part of its meaning and how other things that may not look intended nonetheless are. What he really finds exasperating—and here we’re back to the similarity between him and Suther—is that despite what seems to him all the tough talk about intention, “Against Theory”-style intentionalism neither stops you from doing most of the things he wants nor helps you to do what Suther wants. So, what does it do?

It tells us that the meaning of a text is the function of the semantic and syntactic rules of a language only insofar as the language is being used by the writer. The question is not what “vegetable” means in English or even in seventeenth-century English but what Marvell used it to mean. It tells us that the meaning of the text does not change over time and that it doesn’t depend on the context in which it was read. It tells us (pace Suther) that intentionalism cannot be used to reproach de Man’s reading of The Triumph of Life for failing to distinguish between the question of “whether the poem can bear meaning” and the “question of what meaning it was meant to bear” since de Man’s defense is that raising the question of whether a poem can bear meaning is the meaning it was meant to bear. Intentionalism shows you what’s wrong with de Man’s theory of interpretation, not what’s wrong with his interpretations.

Is that enough? I’m not sure whether trying to show people that their methodological debates are empty represents “a quietist retreat from methodological debate,” but it is true that I think there are higher stakes elsewhere. In one of the replies to our critics, Steve Knapp and I urged theorists who might feel themselves rendered irrelevant by the end of theory to take up the study of its history instead and, although that recommendation was meant in part as a taunt, I now think it was pretty good advice. In 1967, Barthes had famously identified the death of the author with the birth of the reader, but the idea that the meaning of the text was more generally open-ended and that authorial intention was best understood as an effort to control meaning (as Bewes says, to assert “sovereignty” over the work) was widespread and not just in literary theory. In some art criticism, the idea of the artist controlling the meaning of the work was identified (pejoratively) with Modernism and the idea of the artist giving up that control (refusing to fix the meaning of the work was the influential way Barthes put it) with the postmodern. And more recently, Jacques Rancière has attacked the “stultification” he associates with the idea that “what will be perceived, felt, understood” by the spectator is what the artist has “put into” the work and has praised the “emancipated spectator” who is not forced into the “passive” position of seeing what the artist wants to “make her” see.8

But how are we supposed to understand the difference between writing something and trying to make someone see something or between understanding something and being made to see something? In Barthes’ terms, the problem here is that you can no more try to fix the meaning of your work than you can try to leave it unfixed. How would you do it? Talking about the tendency to interiorize intention by identifying it with the “will” to do something (rather than with doing it), Anscombe observes that “people sometimes say that one can get one’s arm to move by an act of will but not a matchbox” since staring at the matchbox and willing it to move doesn’t work.9 But, as she points out, staring at your arm and willing it to move doesn’t work either. There’s nothing that willing your arm to move adds to moving your arm; there’s nothing that trying to control or fix the meaning of your utterance adds to just saying it. Soft Hemingway means whatever he means; there’s no additional act he can perform of fixing or unfixing that meaning.

And once he’s meant whatever he’s meant, there’s nothing anyone else can do about it. But there is an interesting historical question of why this whole picture of meaning with all the debates it has engendered became so influential in the last half of the twentieth century. On the one hand, we see the intention imagined as something anterior to the act (what’s going on inside your head, like willing); on the other hand (actually the same hand) we see the act imagined as something caused by but not defined by the intention. This simultaneous interior and exteriorization is not only a mistake; it’s a symptomatic one, prefigured in Ludwig von Mise’s claim that every action is a “risky speculation” and powerfully embodied in all the ways that people have tried to make a politics (or at least an ethics) out of claiming or (in art and literature more usually) refusing sovereignty or empowering the reader or celebrating the indeterminacy that results from the redistribution of action into cause and effect.

Rancière’s celebration of the text (or performance or photograph) as a thing “that is owned by no one, whose meaning is owned by no one” is instructive here. The claim that the text means what its author intends it is not a claim that the author owns its meaning any more than the claim that the author doesn’t own the meaning involves it meaning something else. Writing something involves asserting ownership even less than it involves fixing the meaning of what you’ve written. But once you conceive of the intention as the cause of the text rather than its content, you make it available for all the dramas of interpretation (does its meaning belong to the author? To the reader? To no one?) that Rancière renders as explicitly political. The idea that the meaning belongs to no one gives us what he calls a “new scene of equality.”10 But, of course, it does so only by depicting someone trying to understand what someone else meant (i.e., reading) as a scene of inequality. The problem here is not that the question of intention is being made political; it’s that it can’t be made political but is nevertheless treated as if it were. So, during a period in which the distribution of actual property has reached levels of inequality that are almost inconceivable, we’re urged to seek equality by getting rid of property rights that actually are inconceivable. In other words, there is no politics of intentionality, but there is a political economy of the theory of intention. And producing an account and critique of that theory probably is more worthwhile than fretting about theorists who hope to make literary criticism more respectable by finding a method for it.