Skip to content Skip to sidebar Skip to footer

Intention, Interpretation, and the Balance of Theory

Then the recent major painting which Fried describes as objects of presentness would be painting’s latest effort to maintain its conviction in its own power to establish connection with reality—by permitting us presentness to ourselves, apart from which there is no hope for a world.

–Stanley Cavell, The World Viewed: Reflections on the Ontology of Film (1971), as cited by Michael Fried in Four Honest Outlaws: Sala, Ray, Marioni, Gordon (2011)




I want to begin by noting that this is not the first time around for the proposition —that “meaning just is what the author intended”—that we have been asked to explore.  Its initial outing was in Critical Inquiry in 1982, in an article by Steven Knapp and Walter Benn Michaels with the deliberately provocative title “Against Theory.”  That article, various critical responses, and further responses by both the authors and Stanley Fish, whose 1980 book Is There a Text in This Class? The Authority of Interpretive Communities significantly shaped Knapp and Michaels’s way of putting their issue, were subsequently collected as a book entitled Against Theory:  Literary Studies and the New Pragmatism.1 It’s worth noting for future reference that the next issue of Critical Inquiry included the well-known exchange on modernism between Michael Fried and T. J. Clark—in 1982 Fish, Fried, and Michaels were all at Johns Hopkins and in frequent conversation, and Michaels already had a well-established interest in photography, all of which has some bearing on how these issues emerge for us now.

The argument of the Knapp-Michaels essay is straightforward:  that “theory” emerges for literary study out of the attempt to establish a relation, positive or negative, between two terms—author’s intention and meaning that do not in fact have any relation because they are simply identical.  Full recognition of this identity necessarily entails an end to any imagination of a general standpoint that might govern our interpretive practices or to any claim to knowledge standing over and against our beliefs and capable of, as it were, correcting them.

I have, for reasons that I think will become apparent, no interest in actually engaging this argument, but there are very good reasons to be broadly sympathetic to it.  Some of these come out in considering what kind of argument it is:  it is itself necessarily not advancing a theory; it has more nearly the form of what we sometimes call a salutary reminder, and so its primary interest is in showing something of how we have managed to get wrong way round within our practice, how we find ourselves making claims we can neither mean nor believe.  In this the essay is comparable to Hans-Georg Gadamer’s Truth and Method, which arrives at its claims about interpretation only from within a history of interpretation and does so precisely in order to set itself against the coupling of truth and method as we would find it in a theory of interpretation.  That the Knapp-Michaels essay does not engage figures like Gadamer is arguably a significant feature of it; it is, in a sense, content to work with strawmen—although strawmen that the profession very obligingly offered up in the persons, most notably, of E. D. Hirsch and Paul de Man, a figure that continues to loom large in Michaels’s more recent arguments.2 Arguments of this kind typically entail not refuting your opponents on theoretical grounds but demonstrating that the underlying fact palpably confounds their own efforts to stand against it, and that is indeed the basic shape the exchanges published with the essay take.  More broadly, arguments of this kind invite some further exploration of how we can indeed come to stand somehow deeply wrongly within our practices and so invite also some further diagnosis of how or why we might have come to do so.  By and large, these possibilities are left very much in the background in 1982; Michaels’s more recent work does better by the diagnostic job, but continues, I think, to do less well by the mere fact of our being able to get turned around, and more particularly does not take on the thought that this possibility might itself be constitutive of our practices in ways that need acknowledgment.

I’ve said that there good reasons to be sympathetic to the “Against Theory” argument; the main one, which I’m not going to dwell on, is that it is pretty much right—right enough that arguing against it is mostly silly.  But there are, I think, also some good reasons not to sign on to its terms:  it exacts some considerable costs—one has to surrender a number of otherwise useful distinctions—as between language and speech acts and between meaning and signification–that one may be rightly loath to give up.  More generally, I am always happy to discover ways in which my language can be made larger, and I am willing to accept technical refinements of my language if they genuinely render the world more articulate, although I am always happier if I can find ways to recapture such vocabulary as syntax, but I find myself exceedingly wary of any attempt to make my language smaller.

One has also to accept a hard construal of the word “theory”—one that is perfectly correct but does not fully meet actual usage and so does not fully meet the practices at issue, and one may find some of the trading around the term “theory” suspect:  if we take theory in its proper, hard sense, it does not normally have consequences for practice: theory in this hard sense is above all testable; its consequences are themselves theoretical and likewise testable.  Very few people in 1982 are doing theory in this sense and so very few feel themselves actually addressed by the argument.  When theory does have a relation to practice, the relation is not one of consequence but of fit or adjustment—we understand the forces in play in theory but we do not know how this will work out in some more particular concrete or material situation.  Knapp and Michaels’s definition of theory as “the attempt to govern interpretations of particular texts by appealing to an account of interpretation in general” (Knapp and Michaels, 723) is an uneasy hybrid that doesn’t sit comfortably with the harder construal of theory essential elsewhere in the argument.  Fish addresses aspects of this well in his own response, but it still leads me to suspect that “theory” is not in fact the best name for the actual target of the Knapp-Michaels intervention. I think the real target comes more fully into view in their response to the essay’s critics, where they order the various aspects of that response under the headings “epistemology,” “method,” and “the profession.”  It’s a certain unholy alliance of these things they would like to defang, a goal I fully share—but also one that brings the question of diagnosis into full prominence and which leads me to Michaels’s more recent renewal of the argument in his book The Shape of the Signifier.

However, before doing that I want to look very quickly at two examples, one taken from “Against Theory” and one of my own.  “Against Theory” centrally offers a thought experiment involving your coming across a stretch of beach where you discover a set of “squiggles” in the sand that turn out to spell out a stanza of Wordsworth.  Knapp and Michaels then work through the differences between dismissing this as a mere and meaningless accident of the tides and taking it as meaningful and the different ways you might come to the one position or the other, the point being that in taking it as meaningful one automatically takes it as, however problematically, authored (Knapp and Michaels, 727-29).  I have no quarrel with this—none I want to pursue anyway—but it does seem to me that even having dismissed the sand-formation as mere and meaningless accident, I will still find it legible, and if my monolinguistic French friend asks me what it says or what it means (the difference here is unimportant), I will unhesitatingly paraphrase or translate (this difference too seems unimportant) it for her.  If the question is put specifically in terms of “meaning,” rather than “saying,” I may, if I’m careful, say “It would mean . . .” rather than “It means . . .” and that care is definitely of the Knapp-Michaels kind and is by no means dismissible.  But all I want out of this extension of their example is the subjunctive proper to the bit of nonsense’s continuing legibility.

Diego Velázquez, Los Borrachos (1628-29)

The second instance is a bit more elaborate and less imaginative.  It’s this painting, “The Drunkards,” as we call it, for no good reason.  I’m entirely an amateur, if that, of Velázquez, so I hope not to have to argue over this stuff seriously.  But my understanding is that pretty much from the moment of its completion we have not known what it meant. Steven Orso has recently made what seems to me a pretty convincing case that Velázquez intended it as an allegory of Spanish kingship drawing upon and, in effect, attempting to revitalize several existing legends about the emergence of a unified Spain out of its legendary conquest by Bacchus, thus making it an image of good government.3 So let’s say we now know what the author intended and so we also know, in decent detail, what the painting’s elements are and why they are disposed as they are.  The painting itself, it seems to me, remains as opaque as ever, which is to say that author’s intention and meaning appear sundered in it.  I cannot find my way to its undeniable authority along this path, and following Knapp and Michaels, I should call the painting “meaningless.”  Since I don’t mean to argue with them, I’m OK with that—even as I feel that being chivvied in this direction somehow threatens to lead me away from its actual continuing hold on me. Trying to reassert this about it may bring me to say that the author’s intention was only secondarily to mean and was primarily to make a painting—something in which Velázquez evidently felt he had succeeded, whatever he may have or have not made of the outcome of this particular moment in his early struggles with the business of painterly meaning. Some part of my response here must surely have something to do with what Manet later made of this painting, which seems integral in ways I may not be able to pin down, to something I still want to call “the painting’s meaning.” As I try to make this out I may find myself hesitating among several possibilities:  that Manet simply took advantage of the earlier painting’s meaninglessness; that he was in some way actively interested in the palpable discontinuity within the painting between artist’s intention and unrealized meaning; that his own painting stands as a reading of Velázquez’s, where reading means something distinct from but not without relation to interpretation.  The choices among these options presumably have to get played out more fully in any attempt at real description of the Manet.  One of the things that shows up here is the difference between a field in which interpretation insists as a or even the central possibility wholly apart from any descriptive project, and art history as a field in which the claims of description can potentially outrun those of interpretation.  I am among those—not simply reducible to “formalists”—who do indeed take the central job of art history to be description rather than interpretation.  Another thing I am saying that needs explicit remarking is that the very particular important thing at stake underneath the often confused and confusing theoretical bluster of literary study in the 1970s is the emergence of something called “a reading” as distinct from, but entirely opposed to, an interpretation.


Edouard Manet, The Old Musician (1862)

The larger point of these brief examples is not to prove Knapp and Michaels wrong—the examples don’t do that and there’s no reason why they should.  It is simply to show that we have perfectly good and highly various uses of words like “intention” and “meaning” and “legibility,” and so to suggest that we should prefer to keep our usage both unsettled and careful.  Ordinary language serves us well here, although only as well as we use it.  If we are to have an argument about these things, Knapp and Michaels should, I suppose, win; we shouldn’t want to have the argument.




As I’ve noted, Michaels takes up and extends the argument of “Against Theory” in his 2004 book The Shape of the Signifier: 1967 to the End of History.  The extension is twofold:  Argumentatively, Michaels moves to connect the argument about intention with a critique of identity and more particularly with a critique of the impact of identity politics on our accounts of our objects.  “The argument, in miniature, is that if you think the intention of the author is what counts, then you don’t think the subject position of the reader matters, but if you don’t think the intention of the author counts, then the subject position of the reader will be the only thing that matters.”4  In terms of scope, the book moves substantially beyond positions in literary study and notably takes on a range of recent works of literary and visual art as well as versions, at least, of the claim to “postmodernism.”  As with “Against Theory,” I feel a good bit of sympathy with this extended argument—while Michaels does rather downplays the ways in which identity emerged as an issue precisely out of direct political concerns for equality, I share his belief that identity alone cannot sustain a politics and instead functions as a substitute for it, and I share also his general sense that the relation between views about interpretation and views about politics is intimate in roughly the way he claims.  But I also continue to have trouble with his more particular ways of putting the question.  Chasing down these disagreements, like fully working through the question of theory and politics that is certainly the book’s leading issue, would go well beyond such time as we have, so I want to focus on one particular issue that I think lies near the heart of what keeps throwing Michaels’s position off balance.

The choice the argument in miniature gives us is stark and in this is true to the book’s argument in large:  we have before us in any given instance either an interpretation, good, bad, or indifferent, or we have something that is simply not an interpretation but an account of the reader and the reader’s situation.  There’s no room in between these two for the kind of thing that I’ve suggested is properly called “a reading” and which should be taken as one of the primary outcomes or discoveries or acquisitions of what we have come, however unhappily, to call “theory.”

This term “reading” needs some pausing over:  It is the one that will be favored by those who do in fact produce just the thing Michaels is attacking, and in this usage Michaels will say, rightly, that it is doing the job of granting an account the dignity we normally give to interpretation while nonetheless holding it apart from the standards we take to define that practice.  Used this way, it is implicitly, and surprisingly often explicitly, qualified as “my reading,” where that “my,” adequately theoretically glossed,  is also its justification, spelling out the terms of its exemptions from our standard ways of gauging the success of an interpretation.  I have no interest in defending this usage—I am as opposed to it as Michaels is; it is a prime instance of what it means to say that we have come to stand frequently wrong way round within our practice.  The thing about “my reading”—one of the things anyway—is that I don’t have to believe in it the way I presumably must believe in my interpretation (which is, as it were, only incidentally mine; it rests no claims there) and if I do happen to believe in it, the terms through which I have arrived at it don’t give me any way of understanding that belief.  It is, I am tempted to say, objectively cynical, and it is a worry to Michaels, as it is to me, that objective cynicism of this sort has turned out to be an extraordinarily viable professional commodity.  If the stakes were not of this order, this discussion would not be worth having at all.

Michaels doesn’t have my worry about wresting the notion of “a reading” away from this appropriation of it.  For him, there is on the one side, interpretation, and, on the other, something else.  Once you’ve given up on what the author intended, meaning has no place to rest except in your situation; as he puts it,

Everything that is there must also count—the table the pages are on, the room the table is in, the way the pages, the table, and the room make you feel.  Why? Because all these things are part of your experience of the pages and once we abjure interest in what the author intended . . . we have no principled reason not to count everything that’s part of our experience as part of the work.  And, of course . . . where you stand will be a little different from where I stand, what you feel will feel different from what I feel, who you are is not who I am. (Michaels, 11)

Some of you, I’m sure, are way ahead of me now.  This “everything that is there must also count” is familiar, and you will be not be surprised when I say that the “1967” that figures in his book’s subtitle is there to explicitly key the argument to the appearance of Michael Fried’s “Art and Objecthood,” a text with which Michaels takes his argument to be continuous—a view Fried himself has endorsed.  In Minimalism, as in what I’ll simply call bad criticism, the meaning of the work gives way to the experience of a situation. “Postmodernism” for Michaels comes down to the mutual agreement of artist or author and viewer or reader that this is indeed how it should be: the job of art, such as it is, is simply to provide the occasion for an experience. When artist and critic agree in this, we are, at best, nowhere; that this is indeed where we too often are in artworld and academy alike needs saying.

Michaels reading of “Art and Objecthood”—and the closely related “Shape as Form”—is woven into the complex fabric of a central chapter that also engages at various points with work by Robert Smithson, Cindy Sherman, James Welling, Paul de Man, and Kim Stanley Robinson, among others.  I am in what follows pulling out just one thread to examine, although I do take it be a key thread.

Michaels locates the central argument of “Art and Objecthood” in the essay’s fifth section, where Fried takes up Tony Smith’s account of his night drive on the unfinished Jersey Turnpike.  This is where I locate the center of the argument as well, and I am going to take the passage to be sufficiently well known to most of you that I am not going to review it in any detail.  Michaels in the course of his chapter quotes twice what is arguably the most important sentence in the essay—one written by Smith, not Fried, which reads,  “There is no way to frame it, you just have to experience it.”5  It’s clear that Fried finds this sentence deeply problematic—finds it, in fact, unacceptable.  The time since 1967 is the time of the sentence’s historical success in spite of Fried’s objection; the question for us is exactly what that objection is.

Michaels puts it this way:  “In Fried’s account of Minimalism, the object exists on its own all right; what depends on the beholder is only the experience.  But of course the experience is everything—it is the experience instead of the object that Minimalism values” (Michaels, 89).  Or again:  “The opposition between what can be framed and what can only be experienced is foundational” (Michaels, 90).  This is, Michaels says, “why Fried will invoke ‘the concept of  ‘meaning’ as against experience on the [Modernist work’s] behalf” (Michaels, 90).  “Fried’s fundamental commitment in these essays is,” he writes, “not to distinguishing kinds of experience (between, say, interest and conviction) but to distinguishing between those objects to which our experience is relevant and those to which it isn’t” (Michaels, 88).

All of this seems to me wrong.  While it is right enough to say that Fried is not interested in distinguishing kinds of experience, he is in fact making an argument that is wholly about experience.  That is, Tony Smith evidently believes or imagines that experience is simply what befalls one, thus essentially lost to articulation and publicness, while Fried clearly thinks that such an imagination of experience is in fact a betrayal of it that will ensure one’s not having one’s experience at all.  Fried evidently holds that experience is realized—thus had—in and as its articulation and nowhere else.  Smith’s problem is that he both lays claim to an experience and refuses the having of it, and the problem with his art is that it is the monumentalization of that bind, demanding of its viewer an experience that it also withholds, thus locking the beholder into his or her self with no outlet into anything other than the formlessness and senselessness of a situation.  “Art and Objecthood” makes this argument by being itself a sustained exercise in the description of its objects and Fried’s experience of them; Fried takes it that description and experience cannot be separated, and so that his job—what he asks the reader’s assent in, is to give an account of an experience that is not his alone but of the object.  That there is no other argument to the essay than that has then to be weighed against Michaels’s claim that on Fried’s account “the question of what [works] are cannot be answered by a description either of them or of our experience of them” (Michaels, 105).  My claim for reading is that it is an essentially descriptive practice bound to our experience of the work insofar as it is in fact “of the object,” with all the difficulty the ambiguity of that genitive entails.  I’ll just suggest that what such a description brings to light is something that has more nearly the shape of a thought than of a meaning.

In saying this, I’m not trying to defeat Michaels’s arguments about intention and interpretation but to shift the ground closer to the underlying condition they mean to address, one in which experience is not opposed to meaning but in which the question is about our capacity for experience and the ways in which we can find ourselves baffled within it—how it is that we can find ourselves wrong way round in our practices.  In philosophy, it is, I think, Stanley Cavell who has most continuously and compellingly worked through these kinds of questions—he’s also of course an important figure in the background of “Art and Objecthood” and, it may now seem, conspicuous by his absence in Michaels’s writing.  He’s also the figure outside the Continental tradition so strongly associated with “theory” who has been most insistent about the specificity and distinctness of “reading” in relation to interpretation. This, I think, is the possibility placed before us by “theory,” and which we have regularly refused.  We are evidently all too willing to remain baffled, and the construal of theory as method is the mechanism by which we maintain ourselves in that state.



I’ll conclude with just two remarks in, so to speak, the present tense, neither of which amounts to a conclusion.

The first is that it has not escaped my notice that I’ve raised questions about the reading or interpretation of a particular text and done so in a context in which the author has himself played an active role.  I don’t know how useful that is or isn’t, but it does strike me that one might well want to say that the argument, such as it is, is not about the author’s intention but more nearly within that intention, and that this may indeed be the case with most serious arguments of this kind.  It may also be worth saying that with this kind of text one can, and frequently will, say things like, “Fried—or Kant or whoever—must mean x or y.”  I don’t know that this “must” figures in our ways of talking about works of art—it seems to me an interesting question, touching on our sense of the relevance of such terms as thought or argument to our understanding of artistic work.

The second is that I have spoken both of a certain unholy alliance of epistemology, method, and professionalism and of an interest we evidently now have in being or remaining baffled in our experience.  Meetings like this play an important role in renewing and cementing that alliance, so it’s hard to think anything said in this session is likely to make much of a difference.  In 1982 Knapp and Michaels suggested that a good first step, for which they held out no particular hope, would be to stop teaching “theory,” that is (I take it), stop teaching it as method. In so doing, they were, of course, speaking “against theory.”  I am, in my way, speaking “for theory,” but that does not mean theirs was not—and is not still—a good idea.  The question before us is, on my understanding, not about theory but about our stake in it.


1.  Steven Knapp and Walter Benn Michaels, “Against Theory,” Critical Inquiry 8.4 (Summer 1982): 723-42; Stanley Fish, Is There a Text in This Class? The Authority of Interpretive Communities (Cambridge, Mass.: Harvard University Press, 1980); and W.J.T. Mitchell, ed., Against Theory: Literary Studies and the New Pragmatism (Chicago: University of Chicago Press, 1985).
2.  As Charles Palermo has reminded me, Knapp and Michaels do engage Gadamer (and others) at some length in “Against Theory 2:  Hermeneutics and Deconstruction,” Critical Inquiry 14.1 (Autumn 1987): 49-68.  Taking up the particular arguments advanced in that essay would entail moving considerably beyond the terms of the present paper without shifting the fundamental argument.
3.  Steven N. Orso, Velázquez, Los Borrachos, and Painting at the Court of Philip IV (Cambridge, U.K.: Cambridge University Press, 1993).
4.  Walter Benn Michaels, The Shape of the Signifier: 1967 to the End of History (Princeton, N.J.: Princeton University Press, 2004), 11.
5.  Tony Smith, cited in Michael Fried, “Art and Objecthood” in Art and Objecthood: Essays and Reviews (Chicago and London: University of Chicago Press, 1998), 158; originally published in Artforum 5 (June 1967): 12-23.