8     Irreconcilable Differences

Felicitous falsehoods admit of objectivity in art as well as in science. Because aesthetics is often considered the province of subjectivity, in this chapter I take up the role of reasons in aesthetics. I argue that aesthetic judgments are procedurally objective. This procedural objectivity does not ensure consensus, even in the long run. This raises the question whether, or to what extent, consensus should in general be an epistemic desideratum.

The Role of Reasons

Jim likes caramel ice cream; Kim dislikes it. Jim might try to talk Kim into trying it again, but he can offer her no reason to think that caramel ice cream tastes good. He can, of course, say that he likes it. He can add that lots of other people do too. These are sociological facts, and Kim, we may assume, has no quarrel with them. She may recognize that she is in a gustatory minority. But such facts do not get anywhere near the heart of the matter. They give her no reason to think that caramel ice cream tastes good. There is no such reason. Some people like it; some people do not. That’s the end of it. There is no room for argument because there is nothing to argue about. For a divergence in attitudes to be a genuine disagreement, it seems, there must be some matter of fact about which parties hold opposing views. Whatever their differences with respect to caramel ice cream, Kim and Jim do not disagree. Reasons have no purchase. What we call tasting good is entirely subjective.

Things are different in factual realms. There are genuine disagreements about how many Olympic medals Jesse Owens won, whether Neanderthals buried their dead, whether electrons have exact locations, whether there is milk in the fridge. Opinions may diverge; evidence may be sparse or inaccessible; but in genuine disputes, opinions are supposed to be responsive to reasons.

As we have seen, reasons are not mere expressions of personal conviction; they are considerations that should weigh with others (Korsgaard, 1996b). Ordinarily, a speaker can responsibly adduce as reasons only considerations that she accepts.1 These are her reasons. Her goal in adducing them is to convince or persuade her auditors of something she is committed to. Reasons thus provide answers to an audience’s (often implicit) question: why should we think that? In effect, a provider of reasons is saying: these are the factors that weigh with me; they should also weigh with you. A wholly satisfactory resolution to a disagreement consists of conclusive reasons.

A speaker’s considering her reasons conclusive is, of course, not enough. Her audience may find (and be right to find) her reasons too weak to be convincing. The audience may have defeaters in their doxastic systems that undermine the significance, relevance, or adequacy of the considerations she adduces. She may have overestimated their strength or bearing on the case. Moreover, not all reasons even purport to be conclusive. A speaker may adduce reasons that are, and are mutually acknowledged to be, relatively weak. They are at best indicative. There is, she says, some reason to believe that the butler stole the spoons or that the bacterium is resistant to antibiotics, but we need more evidence or better reasons before we consider the case closed.

In paradigmatically factual cases, there tends to be widespread agreement about what counts as a reason and, skeptical worries aside, about what would count as a compelling reason, if not a conclusive one.2 Maybe the evidence about Neanderthals is too sparse; maybe it always will be. Maybe the models, methods, or mechanisms needed to determine whether an electron is a sharp-edged particle or a cloud have not been developed; maybe they never will be. Still, what it would take to resolve paradigmatically factual disagreements generally is clear, or can be made clear. Where there is a genuine disagreement, reasons have a purchase. On this picture, the point of adducing reasons is to resolve a dispute by establishing once and for all which side is right. That reasons function to settle factual disputes is plain. A critical question will turn out to be whether this is their sole or primary function.

So far, we’ve surveyed purely subjective cases that are unresponsive to reasons, where there are differences in taste but no real disagreements, and factual cases where differences of opinion are genuine disagreements that are in principle resolvable by appeal to public, intersubjectively shared reasons. Discussions about art seem to want to have it both ways. Aesthetic responses are grounded in personal sensory and emotional experience. They seem clearly to be matters of taste. That being so, it might seem, what holds for ice cream should hold for art. Some people like serial music; others loathe it. Some are revolted by disturbational art; others are intrigued. These are sociological truths about subjective responses. We can dispute the sociology if we like, but not the responses. The difficulty is that when it comes to the arts, we do not exercise the open-minded tolerance we show to those with different tastes in ice cream. As Hume (2007) recognizes, we think some people are better judges of art than others. As Kant (1987) insists, judgments about art make a claim to universality. ‘I like caramel ice cream’ carries no implication that you should like it too. ‘The Nightwatch is masterful’ carries the implication that you should think so too. Of course, you might not. It seems then that with respect to the arts, there is something to disagree about. Kant and Hume focus on aesthetic assessment—in particular on judgments about whether an item is beautiful or sublime. But the same blend of subjectivity and putative universality is characteristic of interpretations of works of art. So I take their points to apply to interpretation as well.

Although there is something right about Kant’s claim, its scope has to be restricted. If a critic says that a work is beautiful or banal or original or inept; that it is an expression of the moral bankruptcy of bourgeois society or a commentary on the plight of the working class or a reaction against the stultifying restrictions of the diatonic scale, she implicates that other suitably sensitive, suitably knowledgeable audiences should find it so. In saying that Schönberg’s music is beautiful, or that it is a limiting case of tonality, she does not implicate that even deaf people should agree. Nor does she implicate that teenagers who listen only to heavy metal music should share her assessment. Rather, she implicates that all suitably experienced, suitably sensitive listeners to classical music should agree with her. She is in effect appealing to the other members of her specific realm of epistemic ends.

The critic’s implicature invites a question: why should we, the members of her realm of ends, think that? The critic obligingly supplies reasons to support her assessment or interpretation of the work. The problem is that another equally knowledgeable and sensitive critic takes and defends a diverging view of the work. He too claims restricted universality for his judgments, appeals to the same restricted audience, and supplies reasons to support his judgment. A dearth of reasons is not the problem. What makes controversies about the arts philosophically curious is that they seem intractable and interminable. Each party can supply reasons, but the reasons are and seem fated to be inconclusive.

To be sure, paradigmatically factual disputes may be irresolvable. But typically we can say what it would take to resolve them. Or if we cannot, we at least consider our failure to be an outstanding problem for the field. When Joe disagrees with Flo about whether there is milk in the fridge, it is perfectly clear what would settle the issue: open the fridge and look. Being in the grocery store, they are in no position to do that, which is why the issue remains unsettled. Nevertheless, they know exactly what would resolve their disagreement. When paleontologists argue about whether Neanderthals buried their dead, they know what it would take to settle the issue positively, at least. Find something that is incontrovertibly an undisturbed grave that contains remains that are incontrovertibly Neanderthal. Then find another one at a second site, thereby indicating that the original discovery was not a fluke. Negative facts are harder to establish. But indirect evidence about Neanderthal cultural practices and comparisons with other hominid cultures might strongly indicate that they did or that they did not bury their dead. A balance of indirect evidence pertaining to cultural development might settle the matter. Maybe it wouldn’t. We may never know. The evidence required to answer the question might not be available. Still, parties on both sides can agree about what direct or indirect evidence would resolve the dispute.

Aesthetic Disputes

Not so in aesthetics.3 There, disputes seem to be interminable. Roger Fry (1952, 13) maintains that the key to understanding Le Compotier is to appreciate how Cézanne constitutes mass and volume out of subtle gradations of color. Clement Greenberg insists that the key is to recognize how, by distorting shape, Cézanne emphasizes “the ineluctable flatness of the picture plane” (1966, 103).4 Each goes on to explain how and why the features he points to make sense of the work. In so doing, each marginalizes features the other points to. And each does so by the simple expedient of showing that good sense can be made of the work without adverting to those features. We have no idea what would carry the day. What sort of reason, if it could be found, would settle the issue between them?

Both Hume (2007) and Goodman (1978) provide resources for explaining the apparent intractability of aesthetic controversies. Hume believes that aesthetic judgments are procedurally objective. But making an aesthetic judgment is an extraordinarily delicate matter. The requisite procedure is hard to perform. It requires high levels of discrimination, good sense, and imaginative acuity. The judgment must be made in propitious circumstances, where the observation of the item being judged is unimpeded and the observer is sufficiently experienced, and is currently free of prejudice and sensory impairment. Making an accurate aesthetic judgment is on this view rather like doing an extraordinarily delicate scientific experiment, such as those conducted deep underground to block the myriad, barely detectable surface-level vibrations that would interfere with accurate measurement. In short, although there is a fact of the matter as to whether a work of art is beautiful, Hume believes, making an accurate aesthetic judgment is exceedingly difficult. The judgment itself is highly refined, and it is hard to tell in any given case whether the conditions for making it correctly have been satisfied. It is, therefore, not surprising that even sensitive, knowledgeable critics disagree. Still, there is an objectively right or wrong answer; and (skeptical worries aside) in principle a suitably situated, suitably sensitive critic could supply conclusive reasons for her judgment. Where there is disagreement, at least one of the critics is impaired.

Goodman intimates otherwise. Given the symptoms of the aesthetic, divergences in interpretation and assessment are to be expected. “Where we can never determine precisely just which symbol of a system we have or whether we have the same one on a different occasion, where the referent is so elusive that properly fitting a symbol to it requires endless care, where more rather than fewer features of the symbol count, where the symbol is an instance of properties it symbolizes, and may perform many interrelated simple and complex referential functions” (Goodman, 1978, 69), even viewers who satisfy Hume’s exacting standards for being competent, impartial judges who are judging in propitious circumstances, are apt to diverge in their interpretations and assessments of a work.

Critical discussions bear this out. Fry and Greenberg discuss the individual dabs of paint that make up Le Compotier. Fry notes that they are parallel to one another; that they are nearly rectangular; that there is no attempt to blend one into the next; that they slant from right to left; and that they do so regardless of the contours of the shape being depicted (1952, 45). As he interprets the work, these are significant features of the dabs of paint. Greenberg ignores the specifics, but takes the dabs to constitute “a mosaic of brush strokes that call attention to the picture plane” (1961, 52). On Greenberg’s reading, the fact that the dabs are discrete units like mosaic tiles is syntactically significant. But the shape and direction of the dabs, and the indifference of their shape and direction to the shape of the contours they figure in, are secondary at best. Fry takes the direction of the brush strokes to function aesthetically; Greenberg evidently does not. Greenberg takes the exemplification of the flatness of the picture plane to function aesthetically. Fry apparently does not.

An interpretation of a work is a constellation of considerations that bear on the work and seek to illuminate or make sense of it. These considerations, either individually or collectively, supply reasons that constitute a particular reading and perhaps assessment of the work. So, for example, Greenberg’s claim that the paint dabs are like mosaic tiles supports his point about the centrality of the flatness of the picture plane, because mosaic tiles not only are flat but achieve their effects at least in part by means of their flatness.

The diverging interpretations are both viable. Each makes sense of the work it pertains to and highlights features that, once identified, strike us as significant. There is no plausible way to say that any of them is obviously in error. According to Goodman (1968), we should expect divergence of critical opinions not (or not just) because interpreting and assessing a work of art is hard, but because a certain indeterminacy of interpretation follows from the structure of the symbols that constitute a work of art and the symbol systems they belong to. On his account, the works themselves admit of multiple, divergent, but equally tenable interpretations.

It might seem that this analysis consigns aesthetic differences to the ice cream realm.5 I said that parties to a genuine disagreement have to be talking about the same thing. If Goodman is right, a single work of art can be construed as any of several symbols, and bears an interpretation because of the symbol it is. There seems then to be nothing—no thing—to disagree about. Construed as one symbol, it has property p. Construed as another symbol, it lacks property p. Attempting to disagree would be like trying to argue with a Frenchman about whether ‘chat’ refers to a conversation or to a cat.

Things are not so simple. For works of art are, or are realized in, objects, inscriptions, and/or events—objects, inscriptions, and/or events that symbolize, to be sure, but still objects, inscriptions, and/or events. An object, inscription, or event is something about which we can disagree. So it is possible to disagree about how best to interpret the object that is a painting by Cézanne, or the inscription that is a play by Shakespeare, or the motion of a body in space that is a dance by Martha Graham. It is even in principle possible to disagree with a Frenchman about whether an inscription of ‘chat’ refers to a conversation or to a cat. The question is: what is to be gained by doing so? Why bother?

If such controversies can be settled by adjudication, the answer is obvious. We seek an objective answer to a specific question. To get one, we need to determine which interpretation is best supported by reasons. This is how Stadler reads the dispute between Fry and Greenberg. Assuming that if one is right the other is wrong, she finds for Greenberg, saying that his interpretation enables her to appreciate “the gawky ellipse of the compotier’s opening, flatter below, arched above”; Fry’s does not (Stadler, 1982, 202). Her approach reflects a familiar view of the point of argument. As Nozick puts it, “The terminology is coercive: arguments are powerful, and best when they are knockdown, arguments force you to a conclusion. Perhaps philosophers need arguments so powerful that they set up reverberations in the brain: if a person refuses to accept a conclusion he dies” (1981, 4). Nozick acknowledges not only that arguments are never that strong, but also that there is something ridiculous about the aspiration, even if we set aside the macabre idea that the goal of argument is to kill the opposition. If someone doesn’t mind being (or being called) irrational, he can disregard the force of argument entirely. Still, the idea that reasons should compel is familiar, and many think it right.

It is implausible in aesthetics. Even if we concede that Greenberg’s interpretation makes better sense of the odd geometry of the compotier, he has not given anything like a knockdown argument for his interpretation of the painting. There are aspects of the painting—the massiness of the apples, for example, for which Fry seems to give a better account. To adjudicate between the two would seem to require delineating and justifying the delineation of all the aesthetically important aspects of the work, establishing a grading scale to evaluate their relative importance, then grading the two interpretations of the work on that scale. That, of course, would require a third interpretation that identified the important properties, and justified the weighting of their relative importance. Its adequacy would have to be judged in turn.

Reason Giving

We seem embarked on a regress, with no end in sight. Do we have to take the first step? That depends on the purpose of reason giving. Why do we go in for it? If our goal is to convince others of the truth of p by the deployment of epistemically sound means, then both the point of reason giving and its value are plain. Reason giving transmits the grounds for epistemic entitlement across the community. It transfers justification for a belief. In discussions about art, however, such a goal seems spurious. We expect serious aesthetic debates to be irresolvable. But if there is no reasonable expectation of defeating the relevant alternatives, which would be required to convey epistemic entitlement to one’s claim, isn’t the enterprise idle? Is aesthetic interpretation akin to the project of squaring the circle? Why should we even start if we recognize that we’ll never succeed?

It might seem that we can evade this conundrum by shifting the terms of the debate. Rather than the disagreement between Fry and Greenberg being a disagreement about whether the key to Le Compotier is the way Cézanne constitutes mass out of color or the way he emphasizes the flatness of the picture plane, it is a disagreement about whether the work is best seen as one where mass is composed out of color, or one where the emphasis lies in the flatness of the picture plane. In that case there is, arguably at least, a fact at the heart of the dispute: how is the picture best seen? Then reasons perform their standard function of supporting or undermining a statement of fact. If this is so, then our mistake was that we mistook what the debate was about.

There is something attractive about this move. We need a basis for dismissing unacceptable interpretations. Some, we want to insist, are simply false. Under no remotely plausible interpretation is Le Compotier a portrait of the Pope. Because such an interpretation simply does not engage with its object, it is easily discredited. The problem comes with interpretations that contain no falsehoods, but are inappropriate or inadequate. They may be uninformative or unenlightening. They may focus on unimportant features or fail to account for important ones. An interpretation of Le Compotier that said only that it is a picture of a bowl of fruit is inadequate. It sheds virtually no light on the work. An interpretation that faults the work for failing to address the destruction of the rain forests is misguided. There is no reason to think that it ought to address that issue. If we insist that a correct interpretation says how the work is best seen, we have grounds for dismissing such interpretations. They simply fall short of the mark.

But we go too fast in assuming that the choice is between saying that there is a unique best interpretation and saying that all interpretations are on a par. The omitted alternative is to say that there may be—and indeed often are—multiple equally acceptable interpretations and a vast number of unacceptable ones. It might seem that this points to an easy way out of our conundrum. If multiple interpretations of a work are acceptable, it is plausible to consider them all true. Then we can simply conjoin them. If p is true and q is true, then p & q is true. So if ‘Le Compotier underscores the ineluctable flatness of the picture plane’ is true and ‘Le Compotier manifests the constitution of pictorial mass out of color’ is true, then ‘Le Compotier underscores the ineluctable flatness of the picture plane and Le Compotier manifests the constitution of pictorial mass out of color’ is true. There’s no disagreement. The two critics are just stating different, but mutually consistent truths.

Although logically impeccable, this is unhelpful. For even if the result of conjoining Fry’s and Greenberg’s interpretations is a truth about the painting, it is not a good interpretation of the work. The two interpretations pull in different directions. An interpretation contends that its object exemplifies certain features. It draws attention to those features and highlights their significance. The reasons it advances make the case that we should find it worthwhile to see the work as exemplifying them. Different interpretations highlight different aspects of the work, drawing attention to them. They provide a way to read the work as exemplifying those features.

Exemplification is selective. Exemplifying some features of an item requires marginalizing or downplaying others. In interpreting Le Compotier, Greenberg downplays the significance of features of the work that do not bear on the flatness of the picture plane. In so doing, he takes a risk. If features he overlooks are too significant to be ignored, his omissions will tell against the adequacy of his interpretation. Under his interpretation, Cézanne’s creating volume and mass out of color is relatively insignificant, if not off the table entirely. Has he omitted something too significant to ignore? Any given interpretation involves a balancing act. What is gained and what is lost by choosing to highlight or omit certain features?

If there is exactly one correct interpretation, the stakes are high. But they are nonnegligible in any case. The goal of an interpretation is to make sense of a work; if something salient is omitted or marginalized, the goal may not be met. Still, the idea that a single interpretation should do justice to every significant aspect of a work is unduly ambitious. Just as no literal point of view can disclose every visible aspect of an object, no single interpretation can accommodate every seemingly significant aspect of a work. Features that one interpretation marginalizes may be highlighted by another. Fry’s interpretation presents a different perspective from Greenberg’s, taking the flatness of the picture plane to be relatively insignificant and the role of gradations in color in constituting pictorial mass and volume to be central. We should doubt that there is a uniquely best interpretation of a work because every interpretation perforce omits or downplays aspects of the work that another interpretation may show to be significant.

Something similar occurs in science. Different models make manifest different features of the target. Diverging models afford different perspectives on the same reality. And it is no surprise that different perspectives reveal different aspects of that reality. There is no optimal model for the same reason that there is no optimal perspective. Every perspective, in revealing some things, inevitably occludes others. Science and art are not so far apart as they might seem.

The role of reasons in aesthetics is not primarily to convince one’s audience that the interpretation they constitute is the one true interpretation of the work or that it is the single best way to read the work. Nor is the role of reasons in justifying a scientific model to convince one’s audience that the model provides the single best way of representing the phenomenon. It is to highlight features and display their significance. In neither case ought the reasons purport to exclude alternative viable perspectives. Fry and Greenberg articulate reasons that lead us to discern different features of Le Compotier and enable us to discern different aspects of reality via our engagement with the work. In proffering reasons, each in effect invites us to read the work in a particular way. Each intimates that doing so will prove rewarding. Their intimations are tested by our responses. Do we find the interpretations enlightening? Do they illuminate aspects of the work, its connections to other works, its connections to the world beyond the work that we otherwise find bewildering or obscure? Do they enable us to make sense of the work? As I see it, an interpretation is closer to an invitation than to a thesis.

Our reflections on aesthetic disagreements show that differences of opinion should not automatically be construed as adversarial. In adducing reasons to support her opinion, the critic should not be thought of as seeking to best her rivals in a winner-take-all competition. Rather, as Dewey (1916) urges, we should treat those whose interpretations diverge from ours as resources for extending our epistemic range. They evidently see something in the work that we as yet do not. If we seriously entertain their point of view, we may come to discern something of value. This is not to say that every perspective will be rewarding. Some, like that of the Holocaust denier’s, can be quickly dismissed. But by accessing the opinions of others and the reasons for those opinions, epistemic agents gain access to insights and perspectives that they would otherwise miss.

In the last few pages, I have focused on aesthetics, because aesthetics is a realm where reason giving and disagreement coexist. It is therefore a good place to look for a role for reason giving other than to win an argument. We have found one. Opposing opinions can make different aspects of their shared object salient. When these are backed by publicly available, publicly assessable reasons, they can make a case for the epistemic value of considering those features important. Opposing opinions backed by reasons can thus enhance our understanding of their object. There is no reason to think that the arts are the only venue where this occurs. Indeed, the highlighting of relevant features is the way resolvable disagreements often get resolved. And it is why we find advancement in understanding in fields like paleontology where in principle resolvable disagreements may never be resolved. We understand more about the Neanderthals and about our understanding of them when we appreciate the force of the arguments for and against the claim that they buried their dead.

Practices that involve the giving and accepting of reasons, I suggest, are procedurally objective. A conclusion that emerges from a chain of reasons gains its epistemic status from the process that led to it. It is credible to the extent that those reasons in that epistemic environment back it up. It might seem that this notion of objectivity sets the bar too low. If even art criticism qualifies as objective, then, one might think, objectivity doesn’t count for much. I think this is wrong. A good interpretation shows us how to look at a work and how to look at the world through a lens that the work provides. It thereby has the potential to enhance our understanding. The interpretation itself is subject to dispute. Reasons can be adduced for and against it. Perhaps the reasons are not Nozick’s knockdown reasons. But it is not obvious that the measure of an argument’s objectivity should be the power of its punch. If we are going to take the admissibility of incompatible alternatives to discredit the claim to objectivity of criticism, we seem forced to discredit the claim to objectivity of the sciences that admit alternative incompatible models. It seems preferable to recognize that any field that admits of reason giving and sets standards for the appropriateness, acceptability, and weight of reasons is one that admits of objectivity. Only then can questions arise about how strong the reasons are or need to be.

Notes

1.  There are exceptions. If a speaker knows that her interlocutor holds a position that she does not, she can responsibly say, ‘Given your commitments, p is a reason for you. Since I do not share those commitments, it is not a reason I share. But you should take it seriously’.

2.  Here I use the term ‘paradigmatically factual’ for cases that are uncontroversially factual because I want, at least for now, to leave open the question of whether aesthetic judgments are or are not judgments of fact.

3.  My focus on well-known critics may suggest that I am committed to something like Dickie’s (2007) contention that the art world determines what counts as art and what counts as good art. This is not so. I am concerned with the ways that and reasons why people who engage seriously with works of art provide reasons for their claims. I appeal to a dispute between well-known critics because they are sufficiently focused and articulate that the points they raise readily strike us as reasons. The same considerations arise in discussions on blogs devoted to heavy metal music or Game of Thrones.

4.  Caveat: although Greenberg is clearly responding to Fry’s discussion of Le Compotier, he casts his discussion as about Cézanne in general. He does not specifically mention Le Compotier. Both because of the context of his discussion and because of the utility of the assumption for my purposes, I take him, like Fry, to be discussing Le Compotier.

5.  There is nothing about the gustatory realm that excludes it from aesthetic judgment. There are food critics, just as there are critics of painting, music, and dance. Food critics, like other art critics, give reasons for their judgments. They point to factors, like the slightly metallic aftertaste or a soupçon of lemon, which others may not have noticed, to back up their claims. In this they differ from folks like Jim and Kim who simply do not find the same food tasty.