Below is a version of a paper that Jacob Berger (CUNY Graduate Center) and I are working on. Any comments, criticisms, questions, etc. would be most appreciated.
Since at least Plato’s Republic, philosophers have debated whether art affords knowledge. Like Plato, many reject the notion that art has educational value. Others maintain that at least part of what makes some art valuable is that it imparts knowledge. Cognitivism as we construe it is twice existentially quantified: it’s the thesis that some of the value of some art is epistemic. Call that value, whatever exactly it is, the cognitive value of art (CVA). Cognitivism is therefore consistent with pluralism about aesthetic axiology; for example, a cognitivist may also hold that some art is valuable because of its purely formal properties.
One way to distinguish varieties of cognitivism is by the content of the CVA; cognitivists differ on what belongs in the CVA. One promising version of the view is that the content of CVA is moral. Call that content the CVA-M. What moral knowledge can the audience extract from art, and how can it be extracted? Here we can, someone crudely, distinguish four versions of moral cognitivism, depending on the normative content of the CVA-M. First, one could hold that the content of CVA-M is deontological: art teaches what’s permissible, obligatory, and forbidden, perhaps with an eye to the universalizability of motives. Second, one could hold that the content of CVA-M is consequentialist: art shows us which states of the world are good and bad, better, worse, and optimal. Thomas More’s Utopia, Soviet socialist realism, and much science fiction might serve as examples. Third, one might claim that the content of CVA-M illustrates the ethics of care. Zora Neale Hurston’s Their Eyes Were Watching God might be cited in this context. Perhaps the most attractive and most common version of moral cognitivism holds that the content of CVA-M is virtue ethical: art teaches its audience the subtleties of virtue and the ways in which virtues interact to produce characteristic patterns of behavior. Homer’s Iliad, for instance, is an investigation of the strengths and weakness of mêtis (artifice), as exemplified by Odysseus, and bíê (might), as exemplified by Achilles. Spenser’s Fairie Queene devotes each of its books to the narrative elaboration of a particular virtue: Redcross represents holiness, Sir Guyon temperance, Britomart chastity, Cambel and Telamond friendship, Artegall justice, and Calidore courtesy. It’s of course possible to subscribe to all four versions of moral cognitivism or even to hold an interactive view according to which art teaches us how, for instance, deontological and consequentialist normative concepts interact and sometimes clash. Sophie’s Choice might be an example.
Moral cognitivism has been attacked on a variety of grounds, but in this paper we will consider a critique of the view that has not been adequately explored. Our target is any version of moral cognitivism that includes virtue and vice in the CVA-M. Such views are open to an elaboration of what’s come to be known as the situationist challenge to virtue ethics. According to this challenge, social psychology suggests that most people are surprisingly susceptible to seemingly trivial and normatively irrelevant situational influences, which in turn means that talk of virtue (and vice) is psychologically unrealistic. If this challenge succeeds, virtue-based moral cognitivism is likewise threatened. After all, art cannot teach us about something that doesn’t exist.
The only paper replying to this challenge in the aesthetic domain is a forthcoming article by Noël Carroll, who argues that art need not teach us how actual virtuous or vicious people do act; instead, art may teach us how people with such character would act – and thus provides us with moral models. While we find Carroll’s proposal instructive, we argue that it faces several difficulties. We propose an alternative that emerges from struggling with the situationist challenge, arguing that the content of the CVA-M is best construed as interactionist. Art doesn’t (only) teach us how virtuous people do or would act. It teaches us how context and character conspire to produce action. We thus recommend a new account of the moral component of the CVA, which we illustrate with examples from literature and television.
2. Moral Cognitivism Introduced
The idea that art provides knowledge is rooted in Aristotle, who urged that, unlike the historian who depicts only how things did happen, the poet depicts much more than what a handful of people do over some circumscribed bit of time. Aristotle maintained that the purpose of narrative art is to reveal how people of certain sorts would act in relevantly similar circumstances:
[T]he poet’s function is to describe, not the thing that happened, but the kind of thing that might happen, i.e. what is possible as being probable or necessary. The distinction between historian and poet […] consists really in this, that the one describes the thing that has been, and the other a kind of thing that might be. Hence poetry is more philosophic and of graver import than history, since its statements are of the nature of universals, whereas those of history are singulars. By a universal statement I mean one as to what such or such a kind of man will probably or necessarily say or do.
For example, while the play Antigone specifically concerns Creon and Antigone, the audience learns things about human nature that go beyond facts about these two individuals. We learn a truth about human nature – namely, that strong-willed people with different aims are prone to clash catastrophically. As Carroll explains, “For Aristotle, it is the task of the poet to reveal these kinds of patterns in human life to audiences – to disclose how things are apt to go when certain personalities are put in play in the same place and time.”
In similar fashion, the moral cognitivist maintains that poets do not merely represent facts about how particular characters behave, but models for how certain character-types think, feel, decide, act, and flourish (or founder). And, as a result, art communicates knowledge about the nature of virtue and vice. For example, Antigone illustrates the virtues of fidelity and familial honor, whereas Creon exemplifies the vice of unbending commitment to state power.
Carroll urges that art functions as a forum to engage in literary analogues to philosophical thought experiments. Just as we may use thought experiments to explore the nature of, for example, semantic content, so too can we use literature to explore the nature of virtue and vice, and thereby gain useful knowledge about how we ought to behave. Carroll argues that artworks are particularly good at imparting moral knowledge in this way because many of them exhibit a specific sort of structure, which he calls the virtue wheel:
To promote and guide this reflection, the novel deploys a structure – which I believe is quite frequent in art, especially narrative art and literature – that we might call the virtue wheel or virtue tableau. A virtue wheel or virtue tableau comprises a studied array of characters who both correspond and contrast with each other along the dimension of a certain virtue or package of virtues – where some of the characters possess the virtue in question, or nearly so, or part of it, while others possess the virtue, but only defectively, or not at all, even to such an extent that their lack of the virtue in question amounts to the vice that corresponds to the virtue.
By exhibiting the structure of a virtue wheel, art presents opportunities for its audience to learn not only what virtuous or vicious people do, but also why they themselves should act virtuously and avoid from vice. Importantly, art performs this educative function precisely because the characters in narratives behave as actual people do and would act, and so provides moral models. As Carroll puts it, this kind of moral cognitivism “holds that fictional characters provide us with social information that we can apply to the world because fictional characters possess character-structures like those of actual people.”
And it would seem that art is particularly well-suited to perform this instructive function. Martha Nussbaum, for example, has famously argued that literature is an ideal medium of moral education because narratives employ “thick” vocabulary – vocabulary that is at once descriptive and normative – to explain and evaluate characters’ behavior. Because art is couched in these terms, it has the power to entice empathic readers to follow suit. Likewise, Daniel Hutto has recently proposed that we make sense of our own and one another’s conduct by constructing narratives in which people act for reasons, and that children only learn to make sense of human conduct by engaging in such reason-based narrative construction themselves. Art therefore bolsters this natural way that we relate to ourselves and others.
So far, we’ve distinguished varieties of moral cognitivism by the content of the CVA-M. Another, cross-cutting distinction, has to do with the cognitive task accomplished by art. According to Carroll’s so-called clarificationist version, art does not provide interestingly new information about morality, but it does help us to clarify and make vivid facts about morality that we already know. Carroll suggests that by providing concrete examples of people acting virtuously or viciously, art enables audiences to come to have a deeper understanding of what are otherwise relatively abstract moral notions. On the clarificationist picture, then, the content of the CVA-M is already known, but incompletely grasped; working through the artistic representation makes the audience’s grasp of the CVA-M firmer. For example, observing Sir Guyon over the course of a whole book of the Fairie Queene affords a nuanced picture of the virtue of temperance.
In contrast, Oliver Conolly and Bashar Haydar defend a version of moral cognitivism that they call propositionalism – the view that art can provide us with fresh knowledge of moral facts. On this view, the content of the CVA-M is unknown prior to engagement with art. Seeing Creon and Antigone crash together like matter and anti-matter teaches the audience a new truth about dominant personalities. Our arguments in this paper apply to both clarificationist and propositionalist varieties of cognitivism, since both assume that the content of the CVA-M is knowable and hence true.
3. The Situationist Critique of Virtue Ethics
To the extent that art is supposed to teach us about virtue and vice, moral cognitivism is open to a challenge based on a prominent criticism of virtue ethics. Although this paper chiefly concerns philosophy of art, it is helpful to say a word about the critique as it applies to virtue ethics. The emphasis of virtue ethics falls not on the rightness or wrongness of actions or the universalizability of motives, but on the sorts of people we ought to be. According to virtue ethics, a good person is someone who cultivates and exercises virtuous character, manifesting virtues such as wisdom, courage, generosity, honesty, and the like. Virtues are dispositions to notice, construe, prefer, deliberate, and act in characteristic ways. Courage, for example, is a disposition to notice threats, construe ambiguous cues astutely, prefer to overcome threats when possible, deliberate soundly about how to respond to threats, and act successfully on the basis of this deliberation.
Importantly, virtue ethicists are clear that these character traits must be deeply ingrained in specific ways. John Doris spells out this idea in terms of the consistency, stability, and evaluative integration of character traits. Consistency entails reliable manifestation in diverse conditions. The honest person doesn’t lie to his mother about money, but he also doesn’t lie to his friends about current events, nor does he cheat or steal. Stability entails reliable manifestation over repeated trials in similar circumstances. The courageous person is someone who responds and would respond successfully to the same type of threat today, tomorrow, and next year. Evaluative integration entails that possessing one virtue makes it more likely that one possesses other virtues. The open-minded person is more inclined to be humble than the closed-minded person.
One of the most innovative contemporary challenges to virtue ethics has been the attempt to marshal evidence from situationist social psychology against the empirical adequacy of the virtue ethical conception of character. As Doris construes it, situationism is the view that “behavioral variation across a population owes more to situational differences than dispositional differences among persons,” that “people will quite typically behave inconsistently with respect to the attributive standards associated with a trait,” and that “evaluatively inconsistent dispositions may ‘cohabitate’ in a single personality.” According to situationism, consistency and evaluative integration do not characterize the psychological lives of most people. Someone who seems physically courageous may quail in the face of social pressure. Being open-minded is as likely to accompany arrogance as humility.
Contemporary social psychology presents a plethora of evidence in favor of situationism. A classic example is John Darley and Daniel Batson’s so-called “Good Samaritan” study, conducted with participants from the Princeton Theological Seminary. Each participant answered a questionnaire to determine whether he viewed religion as a means, an end, or a quest. Participants were then asked to prepare a talk focused either upon job prospects for seminarians or upon the scriptural parable of the Good Samaritan, in which a robbed and beaten man is ignored by a Levite priest but helped by a lowly Samaritan. The moral of the story is naturally that one should emulate the compassionate Samaritan, not the sanctimonious clergy. Presumably the seminarians knew this moral lesson, and presumably each of them aimed to live by it.
To test whether the seminarians would themselves act on the lesson of the Good Samaritan, Darley and Batson arranged for each of them to encounter a distressed confederate slumped on the ground along the path to the location where they would give their talks. Some seminarians were informed that they had time to spare, others that they were running late. The experimenters covertly observed whether the participants stopped to help, as the Good Samaritan had done. Surprisingly, neither the participant’s view of religion nor the assigned subject of his speech correlated with willingness to help; the degree of hurry accounted for all of the observed differences in behavior. Despite the fact that they were reenacting the very parable about which they were about to lecture, a huge majority of the rushed seminarians (90%) failed to show compassion to the distressed confederate. By contrast, those who were not prompted to hurry offered help 63% of the time.
This example is proffered not as an argument for situationism, but as an illustration of the sort of evidence that situationists bring to bear against a robust notion of character. In light of such evidence, situationists admit that people’s personalities do differ in important ways. After all, some seminarians helped even when hurried or failed to help even when unhurried. But situationists claim that such differences are neither as robust as we tend to assume nor sufficiently robust to support the explanatory and predictive work that traits are meant to do. Personality variables seem to account for at most 10% of the variance in behavior, including morally evaluable behavior. In contrast, situational factors – from mood elevators and mood depressors to ambient sensory stimuli and the presence of bystanders – account for as much as 16% of the variance in behavior. Thus, while something like character may exist, it is a pale reflection of the robust sort of character that both ordinary people and virtue ethicists usually presuppose.
Equipped with this sort of psychological evidence, philosophers such as Doris, Gilbert Harman, and Mark Alfano have applied the lesson to virtue ethics. What the situationist evidence suggests is that most persons do not exhibit, and are seemingly unable to exhibit, the kinds of deeply ingrained personality traits that virtue ethics commends. Slight changes to situational contexts generate great disturbances in behavioral profiles. If character traits are unlike the virtual-ethical conception of them, then it appears that there are no such things as virtues, or at least that virtues are quite rare. And, consequently, being a good person cannot be dependent upon developing such virtues, lest (almost) no one be good.
Contemporary psychology has largely resolved the person-situation debate in favor of what is known as interactionism, which gives roughly equal weight to both character and situation as drivers of action. Interactionism holds not that character is totally impotent, but that it is both much less influential than people pre-theoretically suppose and much less influential than it would need to be in order to explain behavior on its own. It is beyond the scope of this paper to evaluate whether the situationist challenge succeeds. In what follows, the question we will explore is: If the situationist challenge succeeds, what are the implications for moral cognitivism?
4. Extending the Situationist Critique to Moral Cognitivism
The situationist critique has been extended to aspects of virtue epistemology, but until now there has been almost no discussion of its implications for philosophy of art. Just as virtue ethics may need to be revised to cope with situationist insights into human nature, so too may aesthetics need to be reoriented in light of these phenomena.
At the descriptive level, the extension of the situationist critique to moral cognitivism is straightforward: if the content of the CVA-M is meant to be virtue ethical (for example, couched in terms of virtue and vice predicates such as ‘courageous’, ‘persistent’, ‘greedy’, and ‘compassionate’), then the content of the CVA-M will be mostly false. Both virtue and vice are extremely rare, so any art that purports to show a wide variety of virtues and vices distributed across a population misrepresents psychological reality. Carroll’s virtue wheel, for instance, is meant to juxtapose several individuals who all nearly approximate the same virtue. Such clusters of people are hard if not impossible to find. Furthermore, art that fails to represent the interaction of personality with situational influences paints a misleading picture of actual human moral psychology. If the content of the CVA-M is false and misleading, it’s hard to see why it deserves to be called the cognitive value of art.
At the normative level, the extension of the critique is more complicated. First, if the claim of moral cognitivists is that by acquiring or clarifying the content of the CVA-M, people become morally better, then we have reason to doubt. Acquiring or clarifying the content of the CVA-M is acquiring or clarifying mostly false beliefs. It’s hard to see how such beliefs would be useful guides for moral behavior. Furthermore, if art motivates its audience to strive for virtue, it suggests that acquiring virtue is a psychologically realistic goal for most people, which is also quite dubious. Since audiences would presumably come away from art with the erroneous notion that they can and should cultivate consistently good character, they may waste time and effort attempting to strengthen personality traits when these traits may be impotent in the face of situational influences regardless of how assiduously they are cultivated. Second, if the claim of cognitivists is instead that acquiring or clarifying the content of the CVA-M enables people to interpret and evaluate others’ moral conduct and character, we again have reason to doubt. The models of virtue and vice contained in the CVA-M will rarely resemble actual people. What we need to interpret and evaluate others is an understanding of the interaction of personality and situation.
Thus, from both the descriptive and the normative perspective, it would appear that the CVA-M is threatened by the situationist challenge to virtue ethics.
5. Potential Replies
In a forthcoming paper, Carroll attempts to defend moral cognitivism from the kind of criticism developed in the previous section. He acknowledges that the situationist evidence suggests that art cannot teach us how virtuous or vicious people do act, but replies that art may still convey or clarify how people with certain character traits would act. Art thus provides us with a model for how we ought to strive to behave and furnishes us norms with which we can evaluate our own and others’ conduct:
Characters – such as those found in the great myths and later in the fictions (and especially the popular fictions) of secular culture – are a crucial, recurring mechanism for imparting social information about the way in which members of that culture, given their role or station in it, should comport themselves. In this fashion, fictional characters function to provide norms for evaluating ourselves and others.
Odysseus, for example, is not depicted as being susceptible to situational influences and is in that way unlike actual human beings. But Odysseus can nonetheless function as a model of cunning, courage, and leadership. So while (almost) no one can be as virtuous as Odysseus, we learn from reading the Iliad and the Odyssey what ideal cunning looks like. We learn that when one fails to act as Odysseus does, one falls short of the norms of cunning and courage. Though arguably no one can be immune to situational variables, being taught about an unrealistically virtuous person promotes better engagement with the idea of virtue. Art makes vivid certain traits so one can engage in the sorts of thought experiments about virtue that a wheel of virtue encourages.
This is an intriguing view and doubtless there is much to recommend for it, but as Carroll himself acknowledges, it may require some clarification or revision. In what follows, we’ll raise a few difficulties for the view as stated, which will dovetail into what we’ll argue is a better way of fleshing out the content of the CVA-M.
Perhaps the most obvious difficulty with Carroll’s view is that if characters in art provide models of good conduct, but most people are unable to emulate those models, then characters provide overly demanding norms. Carroll claims that we can use these norms to evaluate people’s behavior “whether they live up to those norms or fail.” But if personality variables account for at most 10% of behavioral variance, it is hard to see how most people or perhaps anyone could live up to such norms. The norms that art putatively teaches us, then, violate the famous Kantian dictum that ought implies can. Though one might question the validity of this maxim, it would be strange to think that art frequently models for us people whom we cannot emulate and provides norms that we can (almost) never live up to.
One might reply that aspiring towards unattainable ideals would at least point people in the right direction. Since character traits are not wholly impotent in the face of situational factors, there is some value to acquiring them. If at most 10% of the variance in behavior is due to personality factors, perhaps art can help people move from 3% to 8%. Or perhaps art can ensure that the input of personality aims at virtuous conduct rather than vicious conduct.
But it is unclear how useful such ideals might be, and this idea risks backfiring. For one thing, if the norms conveyed by art are too demanding, audiences might conclude that attaining a richly moral life is out of reach. If one knew that one could never act as courageously as Odysseus, one might conclude that one should never attempt to be courageous at all. On the flip side, if one incorrectly believes that one has cultivated some virtue, one may act in ways that call for that virtue though one lacks it. As Doris observes, if someone incorrectly believes that he has cultivated the virtue of fidelity, he may be more apt to place himself in compromising situations. Likewise, by obscuring the important role played by situation in determining behavior, fiction might lead the audience to ignore the vicissitudes of situation when attempting to act morally or interpreting and evaluating others’ behavior. In other words, the audience could end up in even worse shape both morally and epistemically than it would have been otherwise.
One might instead defend the artistic representation of unattainable norms by saying that, while art may represent things that don’t exist or may represent them in unrealistic ways, it nevertheless has value because it depicts exaggerated versions of character, hyperbolically bringing character traits into relief. On this view, when treating how characters behave across various situations, the artist is entitled to ignore the vagaries of any given situation in an effort to highlight the moral properties of interest. Such selective representation might be justified on the grounds that when one wants to showcase certain properties, one may emphasize them in ways that distort the facts on the ground. Just as the subway map of New York City puffs up Manhattan while deflating Staten Island because the main object of interest is the metropolitan train system, so does the tragedian puff up the headstrong personalities of Creon and Antigone while deflating the situational influences that partially govern behavior because the main object of interest is the clash of strong-willed individuals.
After all, representations of any kind invariably elide certain aspects or properties of the object represented. A topographical map represents elevation but not flora, fauna, political boundaries, wealth distributions, and so on. An image produced by an infrared camera represents electromagnetic radiation from one part of the spectrum but not others. If artists are interested in personality but not situation, then it seems that they may legitimately elide the influence of situation.
On a related note, one could claim that art tends to represent only the most extreme individuals. Aristotle himself attempted to justify this sort of exaggeration in tragedy because, claiming that art essentially concerns those who are unusual: “As tragedy is an imitation of personages better than the ordinary man, we should follow the example of good-portrait painters, who reproduce the distinctive features of a man, and at the same time, without losing the likeness, make him handsomer than he is.” It is plain that art frequently catalogues the exploits of the extraordinary rather than the ordinary. Our favorite novels and movies focus on the exploits of valiant heroes and base villains – those special few who break the everyday mold. Perhaps, then, art focuses on the unlikely few who manage to exhibit robust character, despite their situations.
We thus have two ways of defending moral cognitivism by saying that the content of the CVA-M is selective. Either it selects and emphasizes the personality traits of ordinary people, to the exclusion of situational influences; or it selects extraordinary people who are less susceptible to situational influence, to the exclusion of the great majority who are susceptible.
But even if some art does exaggerate character, it need not. Perhaps some artists, such as Sophocles, exaggerate character or focus on the few who exhibit robust character. Other artists are sensitive to the influence of situational factors. Some, such as Homer, catalogue the exploits of the heroic and unusual. Others depict the humdrum. In The Stranger, for example, Albert Camus describes how a surfeit of sunlight could lead to murder:
I was conscious only of the cymbals of the sun clashing on my skull, and, less distinctly, of the keen blade of light flashing up from the knife, scarring my eyelashes, and gouging into my eyeballs […] a great sheet of flame poured down through the rift. Every nerve in my body was a steel spring, and my grip closed on the revolver.
While Antigone may succeed in being educative because it exaggerates character or elides the situation, it would make no sense to say that The Stranger fails to teach us anything about morality because it emphasizes the power of situations or elides the importance of character. So it would seem that the content of the CVA-M is not exhausted by exaggerated representations of ordinary character or selective representations of extraordinary character. Both Carroll’s normative response and the exaggeration response just described leave us with an artificially impoverished understanding of the content or the CVA-M.
6. Interactionist Cognitivism about Aesthetic Value
In this section, we mount a defense of a new version of moral cognitivism that emerges from struggling with the situationist critique. One thing the critique makes vivid is that we are all embodied beings operating within particular situational contexts. And what the psychological evidence suggests is that these situations often have as much influence on our behavior as whatever characterological endowments we may possess.
Instead of ignoring or avoiding this fact, moral cognitivists may enshrine it in the CVA-M: art may teach us the myriad ways in which character and situation interact. Thus we propose what we call interactionist cognitivism, according to which art can and should educate us about the interplay of context and character in producing behavior. Some of this interplay will inevitably involve the influence on conduct of seemingly trivial and normatively irrelevant situational factors. On this view, the content of the CVA-M is not purely normative; art represents both how people do act and how they would act given various permutations of characters and situations. Furthermore, art need not represent individuals as being unrealistically virtuous or vicious, nor must it showcase unrealistic norms.
Interactionist cognitivism is resolutely pluralist. Whereas Carroll’s defense of cognitivism countenances only the normative ideals of moral heroism and sainthood, and the exaggeration defense countenances only facts about personalities, interactionist cognitivism countenances these as well as situational influences and the interaction of person and situation. We therefore think that our proposal should be seen as a friendly amendment to Carroll’s defense of cognitivism.
Interactionist cognitivism thus holds that art can educate along both moral and factual dimensions. On the factual dimension, art may encode knowledge of the frailty of character and the ways in which character and context interact. Audiences learn from art that situations play a much greater role in determining what we do than we might initially suppose, and that the power of our personalities can go only so far. Though audiences may not be capable of transcending the pressure of situation, art might actually disabuse us of the mistaken notion that we will always act in accordance with our character. We might learn that situations may force even the best of us to act wrongly, or the worst to act nobly. It may be hoped that this is the lesson drawn through watching, for example, The Experiment, the documentary-style dramatization of Milgram’s infamous investigations of obedience to authority.
On the moral dimension, art may educate us about realistic norms. Art could reveal, for instance, how people ought to behave given the limits of character. By seeing a generally courageous character succumb to situational influences, for example in the film Saving Private Ryan, one can learn what realistic courage looks like.
It is important to distinguish this view from what some read as, for example, Aristotle’s own position. We have so far portrayed Aristotle as holding that character is a matter of certain kinds of personality types, cashed out in terms of propensities to behave in certain ways. But some claim that Aristotle regarded characters as already involving broader situational concerns such as a character’s social role and interactions with other social roles. Carroll, for instance, claims that Aristotle views characters as “concrete representations of ideal types or ideas. In Antigone, for example, Creon stands for the claims of the state while his niece Antigone is a cipher for familial or tribal obligations.” According to this reading of Aristotle, if a character like Creon represents the claims of the state, then his personality already includes certain situational concerns. Regardless of Creon’s particular personality quirks, the audience learns that human beings in his position of power tend to act in certain ways. Unlike the initial reading of Aristotle discussed above in which people with characters akin to Creon’s invariably act in certain ways, this amended reading would suggest that people inhabiting Creon’s social role tend act in certain ways.
Whether or not this intriguing reading of Aristotle is an altogether apt one, one might think that it meets the situationist critique. Indeed, such a view has a parallel in the virtue-ethical literature. Robert Merrihew Adams has argued that affiliations and social roles can even be virtues or vices. For instance, friendship and participation in certain forms of kinship (for example, filial piety) may be virtues, and membership in odious organizations (for example, the Ku Klux Klan) may constitute a vice.
But our view is importantly different from these positions. Even if character is best seen as including one’s social station, the situationist critique still applies because many influential situational variables (for example, lateness, mood elevators, mood depressors, ambient smells, ambient sounds, etc.) have nothing to do with such affiliations or social roles. Our proposal is that art can educate about even these morally and socially irrelevant drivers of behavior.
Many works do explicitly explore this dialectic between character and context. Jane Austen’s writings, for example, present an effective demonstration. In chapter five of Persuasion, we read, “Sir Walter… [had] been flattered into his very best and most polished behavior by Mr. Shepherd’s assurances of his being known, by report, to the Admiral, as a model of good breeding.” Sir Walter is sometimes boorish, sometimes polished. He also cares a great deal how he is viewed by his social superiors, such as the Admiral, so when he is informed that the Admiral expects him to behave courteously, he lives up to those expectations. This is a phenomenon that Alfano calls factitious virtue, which emerges when someone behaves in accordance with a character trait not because he possesses the trait but because he thinks that others expect it of him.
Richard Wright’s novel Native Son powerfully illustrates the contrary phenomenon – what Alfano calls factitious vice, in which someone acts viciously not because he has bad character but because he thinks everyone expects him to behave so. The novel depicts Bigger Thomas, a young African American living in poverty in Chicago during the 1930’s. As Bigger encounters various forms of racism and humiliation, he is eventually driven to commit heinous crimes, including the murder of his employer and of his partner Bessie Mears. While the novel does not invite the reader to absolve Bigger of his crimes, literary critics have argued that we are to see Bigger’s actions as driven by society’s brutal expectations of black youth. Discussing Franz Fanon’s analysis of the novel, Harold Bloom says that “Bigger is meant to terrify us, and he does, but Wright has the skill to show us that all too frequently Bigger acts out of intense fear, a realistic terror of the world. […] Bigger kills only as a response to the world’s expectation that he must kill or die.” The novel movingly dramatizes how attributions of negative traits can lead to behavior which then seems to confirm the attribution.
A more recent example in a different genre, the fourth season of the television series The Wire, dramatizes the effects of negative expectations on child development. The series centers on the drug epidemic in Baltimore, and season four is devoted to its impact on the school system. Many students are “corner boys,” sucked into the drug trade at a young age. One adult character, Howard “Bunny” Colvin, expresses a genuine understanding of how expectations can function as self-fulfilling prophecies. In episode nine, “Know Your Place,” he says of these children, “They’re not fools. They know exactly what we expect them to be.” And in the next episode, “Misgivings,” he tells a school administrator,
You put a textbook in front of these kids, put a problem on the blackboard, or teach them every problem on some statewide test, it won’t matter. None of it. ’Cuz they’re not learning for our world. They’re learning for theirs. And they know exactly what it is they’re training for, and what it is everyone expects them to be.
As Colvin says, most of the corner boys end up on the street engaged in the drug business. The one genuine counterexample is Naymond Brice, the son of a soldier in the Barksdale drug empire. Colvin makes his positive expectations clear to Brice, which in turn make possible the boy’s eventual escape from his imprisoned father, his anti-social mother, and the milieu in which he was raised and which had communicated its sinister expectations to him.
Examples such as these show how useful interactionist cognitivism is in coping with the situationist challenge. Artists need not depict only moral examplars or exaggerated versions of character. They can show how situational influences partially govern behavior. Furthermore, they can depict how easy it is to fabricate evidence for character where no evidence or contrary evidence exists, a phenomenon known in social psychology as the Fundamental Attribution Error. Austen is also a master of this device, as the subplot involving George Wickham in Pride and Prejudice illustrates. With his charming manners, Wickham not only manages to impress the other characters with his rectitude – Lizzie Bennett included – but also to slander Fitzwilliam Darcy’s character. Through both subtle insinuation and overt gossip, he induces society to think him a martyr to Darcy’s pride. Manipulation of sympathy is one of the contributing factors to Wickham’s disastrous seduction of Lydia.
In this paper, we’ve attempted to show how, if the situationist challenge succeeds against virtue ethics, it also creates problems for moral cognitivism in philosophy of art. The key question is what the moral content of the cognitive value of art might be. That content, as most moral cognitivists have spelled it out, turns out to be false. We then considered several ways for cognitivists to retrench. They could follow Carroll in claiming that art teaches us not how virtuous and vicious people do act, but how they would act, if they existed. Alternatively, they could argue that art exaggerates weak character traits to emphasize their workings, or that art selects only extraordinary individuals to represent. Both of these responses are consistent with the more comprehensive response we developed in detail: interactionist cognitivism, according to which the moral component of the cognitive value of art is the interaction of personality and situation.
In closing, it’s worth noting that the situationist challenge may have further ramifications in philosophy of art. Recently, the field of virtue aesthetics – according to which artists ought to cultivate such virtues as being honest in what one seeks to produce and consumers of art ought to cultivate such virtues as good taste – has been developed. It’s not hard to imagine (as Peter Goldie already did) that if moral and epistemic virtues are susceptible to situationist critique, so might aesthetic virtues. This is an interesting topic for further investigation, which may be informed by the interactionist view sketched in this paper.
 See, for example, Peter Lamarque and Stein Haugom Olsen, Truth, Fiction, and Literature: A Philosophical Perspective, (Oxford: Clarendon, 1994).
 See, for example, Immanuel Kant, The Critique of Judgment, James Creed Meredith (trans.), (Oxford UP, 1952 ) and Monroe Beardsley, Aesthetics: Problems in the Philosophy of Criticism, (New York: Harcourt, Brace and World, 1958).
 Noël Carroll, ‘Art, Narrative, and Moral Understanding’, in Aesthetics and Ethics, Jerrold Levinson (ed.), (Cambridge UP, 1998), pp. 126-160; ‘The Wheel of Virtue: Art, Literature, and Moral Knowledge’, The Journal of Aesthetics and Art Criticism, 60: 1 (2002), pp. 3-26; ‘Character, Social Information, and the Challenge of Psychology’, (forthcoming); Matthew Kieran, ‘Art, Imagination and the Cultivation of Morals’, Journal of Aesthetics and Art Criticism, 54:4 (1996), pp. 337-51; Revealing Art, (London: Routledge, 2005; ‘Art, Morality and Ethics: On the (Im)Moral Character of Art Works and Inter-Relations to Artistic Value’, Philosophy Compass, 1:2 (2006), pp. 129-143; Oliver Conolly and Bashshar Haydar, ‘Narrative Art and Moral Knowledge’, British Journal of Aesthetics, 41:2 (2001), pp. 109-124.
 Gregory Nagy, The Best of the Acheans (Johns Hopkins Press, 1979), p. 45.
 For discussion and reply to several of these criticisms, see Carroll’s ‘The Wheel of Virtue: Art, Literature, and Moral Knowledge’.
 For discussions of this argument, see, for example, Mark Alfano, ‘Explaining Away Intuitions about Traits: Why Virtue Ethics Seems Plausible (Even if it isn’t)’, Review of Philosophy and Psychology, 2:1 (2011), pp. 121-136; Character as Moral Fiction (Cambridge UP, forthcoming); John Doris, ‘Persons, Situations, and Virtue Ethics’, Nous, 32:4 (1998), pp. 504-540; Lack of Character: Personality and Moral Behavior (Cambridge UP, 2002); Gilbert Harman, ‘Moral Philosophy Meets Social Psychology: Virtue Ethics and the Fundamental Attribution Error’, Proceedings of the Aristotelian Society, New Series 119 (1999), pp. 316-331; ‘The Nonexistence of Character Traits’, Proceedings of the Aristotelian Society, 100 (2000), pp. 223-226; Christian Miller, ‘Social Psychology and Virtue Ethics’, The Journal of Ethics, 7:4 (2003), pp. 365-392; ‘Empathy, Social Psychology, and Global Helping Traits’, Philosophical Studies, 142:2 (2009), pp. 247-275; Peter Vranas, ‘The Indeterminacy Paradox: Character Evaluations and Human Psychology’, Nous, 39 (2005), pp. 1-42.
 ‘Character, Social Information, and the Challenge of Psychology’.
 1451a37-1451b8. References to Aristotle are to the Bywater translation in Barnes (ed.) The Complete Works of Aristotle, (Princeton UP, 1984).
 Classics in the Philosophy of Art, (unpublished book manuscript).
 ‘The Wheel of Virtue: Art, Literature, and Moral Knowledge’, p. 9. For similar views about art as thought experiment, see, Edward Davenport, ‘Literature as Thought Experiment (On Aiding and Abetting the Muse)’, Philosophy of the Social Sciences, 13:3 (1983), pp. 279-306, and Peter Swirski, Of Literature and Knowledge, (New York: Routledge, 1983).
 ‘The Wheel of Virtue: Art, Literature, and Moral Knowledge’, p. 12.
 ‘Character, Social Information, and the Challenge of Psychology’, p. xxx.
 Poetic Justice: The Literary Imagination and Public Life (Boston: Beacon Press), p.10. The distinction between thick and thin vocabulary is due to Bernard Williams, who argues that thick concepts “express a union of fact and value. The way these notions are applied is determined by what the world is like (for instance, by how someone has behaved), and yet, at the same time, their application usually involves a certain valuation of the situation, of person or actions.” Ethics and the Limits of Philosophy, (Harvard UP, 1985), p. 129.
 Folk Psychological Narratives: The Sociocultural Basis of Understanding Reasons, (MIT Press, 2008).
 ‘Art, Narrative, and Moral Understanding’, p. 142.
 ‘Narrative Art and Moral Knowledge’, p. 109.
 Lack of Character, p. 22.
 Op. cit, pp. 24-25.
 ‘From Jerusalem to Jericho: A Study of Situational and Dispositional Variables in Helping Behavior’, Journal of Personality and Social Psychology, 27 (1973), pp. 100-108.
 Walter Mischel, Personality and Assessment, (New York: Wiley, 1968).
 David Funder and Daniel Ozer, ‘Behavior as a Function of the Situation’, Journal of Personality and Social Psychology, 44 (1983), pp. 107-112.
 See Mark Alfano, ‘Identifying and Defending the Hard Core of Virtue Ethics’, Journal of Philosophical Research, (forthcoming) for a summary and evaluation of the dialectic since 2002.
 A common virtue ethical reply to this argument is that virtue may be rare but real. See Christian Miller, ‘Social Psychology and Virtue Ethics’, The Journal of Ethics, 7:4, pp. 365-92, at p. 379, and Christine Swanton, Virtue Ethics: A Pluralist View, (Oxford UP: 2003), p. 30. Even if not all participants stopped to help the confederate, 10% of them did so – so perhaps those compassionate few are virtuous after all. It’s arguable, however, that even these few could be influenced in other circumstances by other morally irrelevant situational variables. Either way, virtue ethics would still be broadly psychologically unrealistic insofar as the majority of persons are influenced by such situational factors.
 Katherine Reynolds, John Turner, Nyla Branscombe, Kenneth Mavor, Boris Bizumic, and Emina Subašić, ‘Interactionism in Personality and Social Psychology: An Integrated Approach to Understanding the Mind and Behavior’, European Journal of Personality, 24:5 (2010), pp. 458-482.
 Mark Alfano, ‘Extending the Situationist Challenge to Responsibilist Virtue Epistemology’, Philosophical Quarterly, 62:247 (2012), pp. 223-249.
 ‘Character, Social Information, and the Challenge of Psychology’ (forthcoming).
 Op. cit. (forthcoming).
 Op. cit. (forthcoming).
 The Critique of Pure Reason, Paul Guyer & Allen Wood, (trans.), (Cambridge UP, 1998 ), A548/B576, p. 540.
 Op. cit. pp. 112 and 149-152.
 Op. cit. 1454b8-1454b11.
 Albert Camus, The Stranger, translated by Stuart Gilbert, (New York: Random House, 1942/1946), pp. 38-39.
 Stanley Milgram, Obedience to Authority, (New York: Harper Collins, 1974). For a virtue ethical perspective on this experiment, see Neera Badhwar, ‘The Milgram Experiments, Learned Helplessness, and Character Traits’, Journal of Ethics, 13:2-3, pp. 257-289.
 Classics in the Philosophy of Art (forthcoming).
 A Theory of Virtue: Excellence in Being for the Good, (Oxford UP: 2006), pp. 138-42.
 Jane Austen, Persuasion, (New York: Tribeca Books, 1818/2012), p. 21.
 Character as Moral Fiction, (Cambridge UP: forthcoming).
 Harold Bloom, Bloom’s Guides: Richard Wright’s Native Son, (New York: Infobase Publishing, 2007), p. 9.
 Lee Ross, ‘The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process’, in Berkowitz (ed.), Advances in Experimental Social Psychology, 10, (New York: Academic Press, 1977), pp. 173-220.
 See Peter Goldie, ‘Towards a Virtue Theory of Art’, British Journal of Aesthetics, 47: 4 (October 2007), pp. 372-387; ‘Virtues of Art and Human Well-Being, Proceedings of the Aristotelian Society Supplementary Volume, LXXXII (2008), pp. 179-195; ‘Virtues of Art’, Philosophy Compass, 5: 10 (2010), pp. 830–839; Matthew Kieran, ‘The Vice of Snobbery: Aesthetic Knowledge, Justfication and Virtue in Art Appreciation’, The Philosophical Quarterly, 60: 239 (2010), pp. 243-263; Dominic McIver Lopes, ‘Virtues of Art: Good Taste’, Proceedings of the Aristotelian Society Supplementary Volume, LXXXII (2008), pp. 197-211.
 Virtues of Art’, p. 835.
 With thanks to [redacted for blind review] for helpful comments on earlier drafts.