VICTIMS OF CIRCUMSTANCES? A DEFENSE OF VIRTUE ETHICS IN BUSINESS By Robert C. Solomon

Abstract: Should the responsibilities of business managers be understood independently of the social circumstances and "market forces" that surround them, or (in accord with empiricism and the social sciences) are agents and their choices shaped by their circumstances, free only insofar as they act in accordance with antecedently established dispositions, their "character"? Virtue ethics, of which I consider myself a proponent, shares with empiricism this emphasis on character as well as an affinity with the social sciences. But recent criticisms of both empiricist and virtue ethical accounts of character deny even this apparent compromise between agency and environment. Here is an account of character that emphasizes dynamic interaction both in the formation and in the interplay between personal agency and responsibility on the one hand and social pressures and the environment on the other.

Business ethics is a child of ethics, and business ethics, like its parents, is vulnerable to the same threats and challenges visited on its elders. For many years, one such threat (or rather, a family of threats) has challenged moral philosophy, and it is time it was brought out in the open in business ethics as well. It is a threat that is sometimes identified by way of the philosophical term, "determinism," and though its status in the philosophy of science and theory of knowledge is by no means settled, it has nevertheless wreaked havoc on ethics. If there is determinism, so the argument goes, there can be no agency, properly speaking, and thus no moral responsibility. But determinism admits of at least two interpretations in ethics. The first is determination by "external" circumstances, including pressure or coercion by other people. The second is determination within the person, in particular, by his or her character. In the former case, but arguably not in the latter, there is thought to be a problem ascribing moral responsibility.

The argument can be readily extended to business ethics. Versions of the argument have been put forward with regard to corporations, for instance, in the now perennial arguments whether corporations can be or cannot be held responsible. ^ One familiar line of argument holds that only individuals, not corporations, can be held responsible for their actions.

But then corporate executives like to excuse their actions by reference to "market forces'" that render them helpless, mere victims of economic circumstances, and everyone who works in the corporation similarly excuses their bad behavior by reference to those who set their agenda and policies. They are mere "victims of circumstances." They thus betray their utter lack of leadership. Moreover, it doesn't take a whole lot of research to show that people in corporations tend to behave in conformity with the people and expectations that surround them, even when what they are told to do violates their "personal morality." What (outside of the corporation) might count as "character" tends to be more of an obstacle than a boon to corporate success for many people. What seems to coutit as "character" in the corporation is a disposition to please others, obey superiors, follow others, and avoid personal responsibility.

In general philosophy, Kant tried desperately to separate determinism and moral responsibility, defending determinism in the domain of science and "Nature" but preserving agency and responsibility in the domain of ethics. "I have found it necessary to limit knowledge to make room for faith," as he put in one of his most concise but rather misleading bon mots. Other philosophers were not so bold. They were willing to accept determinism (even if conjoined with skeptical doubts) and somehow fit agency and responsibility into its domain. David Hume and John Stuart Mill, the two most illustrious empiricist promoters of this strategy, suggested that an act is free (and an agent responsible) if it "flows from the person's character,"^ where "character" stood for a reasonably stable set of established character traits that were both morally significant and served as the antecedent causal conditions demanded by determinism.

Adam Smith, Hume's best friend and the father of not only modern economics but of business ethics too, agreed with this thesis. It was a good solution. It saved the notions of agency and responsibility, it was very much in line with our ordinary intuitions about people's behavior, and it did not try to challenge the scientific establishment. So, too, a major movement in business ethics, of which I consider myself a card-carrying member, is "virtue ethics," which takes the concept of character (and with it the related notions of virtue and integrity) to be central to the idea of being a good person in business. Among the many virtues of virtue ethics in business, one might think, is that, as in Hume and Mill, it would seem to keep at bay the threat of situational ("external") determinism.

Such a solution seems particularly appropriate for business ethics because the concept of character fills the void between institutional behaviorism ("organizational behavior") and an overblown emphasis on free will and personal autonomy that remains oblivious to context, the reality of office work, and the force of peer and corporate pressures. It provides a locus for responsibility without sacrificing the findings of "management science." But 1 bave mixed feelings about the empiricist solution. On the one hand, it seems to me too weak. It does not account (or try to account) for actions "out of character," heroic or saintly or vicious and shockingly greedy behavior, which could not have been predicted of (or even by) the subject. And it does not (as Aristotle does) rigorously hold a person responsible for the formation of his or her character.

A DEFENSE OF VIRTUE ETHICS IN BUSINESS

Aristotle makes it quite clear that a wicked person is responsible for his or her character not because he or she could now alter it but because he or she could have and should have acted differently early on and established very different habits and states of character. The corporate bully, the greedy entrepreneur, and the office snitch all would seem to be responsible for not only what they do but who they are, according to Aristotle's tough criterion.

On the other hand, however, the empiricist solution overstates the case for character. (This is what some psychologists, and Gilbert Harman, refer to as the "attribution error.") The empiricists make it sound as if character is something both settled and "robust" (the target of much of the recent psychological literature). Character consists of such traits as honesty and trustworthiness that are more or less resistant to social or interpersonal pressures. But character is never fully formed and settled. It is always vulnerable to circumstances and trauma. People change, and they are malleable. They respond in interesting and sometimes immediate ways to their environment, their peers and pressures from above. Put into an unusual, pressured, or troubled environment, many people will act "out of character," sometimes in heroic but more often in disappointing and sometimes shocking ways. In the corporate setting, in particular, people joke about "leaving their integrity at the office door" and act with sometimes shocking obedience to orders and policies that they personally find unethical and even downright revolting.

These worries can be taken care of with an adequate retooling of the notion of character and its place in ethics, and this is what I will try to do here. But my real worry is that in the effort to correct the excesses of the empiricist emphasis on character, the baby is being thrown out with the bath toys. In recent work by Gilbert Harman and John Doris, in particular, the very notion of character is being thrown into question.** Indeed, Harman suggests that "there may be no such thing." Doris entitles his book, tellingly. Lack of Character. Both Harman and Doris argue at considerable length that a great deal of what we take as "character" is in fact (and demonstrably) due to specific social settings that reinforce virtuous conduct. To mention two often-used examples, clergy act like clergy not because of character but because they surround themselves with other clergy who expect them to act like clergy. So, too, criminals act like criminals not because of character but because they hang out with other criminals who expect them to act like criminals. Harman argues vehemently against what he calls the illusion of "a robust sense of character." Doris argues, at book length, a very detailed and remarkably nuanced account of virtue and responsibility without character. The conclusion of both authors is that virtue ethics, construed in terms of character, is at best a mistake, and at worst a vicious political maneuver.

It is worth saying a word about this "vicious political maneuver" that is the political target of Harman's and Doris's arguments. I share in their concern, and I, too, would want to argue against those who, on the basis of an absurd notion of character, expect people to "pick themselves up by their own bootstraps," blaming the poor, for instance, for their own impoverishment and thus ignoring social and political (not to mention medical and racial) disadvantages that are certainly not their fault. I, too, reject such a notion of character, but I am not willing to dispense with the very notions of character and the virtues in order to accomplish this.

So, too, in business ethics, there is a good reason to be suspicious of a notion of character that is supposed to stand up to overwhelming pressures without peer or institutional support. I would take Harman's and Doris's arguments as a good reason to insist on sound ethical policies and rigorous ethical enforcement in corporations and in the business community more generally, thus maximizing the likelihood that people will conform to the right kinds of corporate expectations. Nevertheless, something extremely important can get lost in the face of that otherwise quite reasonable and desirable demand. It is the idea that a person can, and should, resist those pressures, even at considerable cost to oneself, depending on the severity of the situation and circumstances. That is the very basis on which virtue ethics has proven to be so appealing to people in business. It is the hope that they can, and sometimes will, resist or even rise up against pressures and policies that they find to be unethical.

So whatever my worries, I find myself a staunch defender of character and the indispensability of talk about character in both ethics and business ethics.^ To quote my friend and colleague Ed Hartman, "the difference between Peter Hempel [one of the most wonderful human beings we ever met] and Richard Nixon is not just a matter of environment." In both everyday life and in business, there are people we trust, and there are people we do not, often on the basis of a substantial history of disappointments and betrayal. And we trust or distrust those people in much the same circumstances and under much the same conditions.

To be sure, character is vulnerable to environment but it is also a bulwark against environment. Character supplies that familiar and sometimes uncomfortable or even uncanny resistance to untoward pressures that violate our "principles" or morally disgust us or are damaging to our "integrity." It is character and not God or the Superego that produces that nagging inner voice called "conscience." (It has been suggested that conscience produces character rather than the other way around, but apart from religious predilections there seems to be little sound philosophical argument or empirical research to defend this.) One person refuses to obey a directive to short-change his customers while another refuses to cheat on her expense account despite the fact that everyone around her is doing so. It is character that makes the difference, though not, to be sure, all the difference.

Some of my concern with this issue is personal. Like most conscientious people, I worry about my integrity and character, what sorts of temptations and threats I could and would withstand. I feel ashamed (or worse) when I give into those temptations and humiliated when I succumb to (at least some of) those threats. I am occasionally even proud about those temptations and threats I have withstood. Philosophically ("existentially"), I worry about how we view ourselves when the balance of accounts is shifted over to causal and statistical explanations of behavior instead of a continuing emphasis on character, agency, and responsibility. Will that give almost everyone an excuse for almost everything?^ And professionally, I have made something of a reputation for myself as a "virtue ethicist" in business ethics, in the twisted tradition of Aristotle and Nietzsche, and virtue ethics requires a solid notion of character. But not a fixed and permanent notion of character. To be sure, many writers about the virtues, perhaps betraying their own insecurity, tend to describe good character and integrity in terms of rock and stone metaphors, suggesting that the truly virtuous person is capable of standing up against anything. (A handful of mostly legendary examples provide the paradigm.)

But I for one never said that virtue ethics requires a strong sense of autonomy, the ability to cut oneself off from all influences and pressures from other people and institutions and ignore one's personal "inclinations" and make a decision on the basis of one's "practical reason" alone. On the contrary, I have argued that one's inclinations (one's emotions, in particular) form the essential core ofthe virtues, and I have argued that so can Aristotle and (more obviously) Nietzsche. And one's emotions are largely reactive, responsive to other people and the social situations in which one finds oneself. Virtue ethics need not, and should not, deny any of this.

The "New Empiricism " Virtue Ethics and Empirical Science

Behind the attack on character and virtue lies another commendable motive. John Doris puts it well. Virtue ethics, he says, can be traced back to the momentous writings of Aristotle, 2500 years ago. Unfortunately, however, the social psychology on which virtue ethics rests is also 2500 years old. Even the work of Hume, Mill, and Adam Smith is pressing on 250 years, ancient times in the scope of modern psychology. As both Doris and Harman properly point out, there has been a great deal of research in the social sciences, much of it within the last 50 years, that ought to be taken into account. And this, they think, seriously undermines the claims of virtue ethics and its emphasis on character.

As so often in these discussions, there is an easy, but wholly misleading, analogy with physics. Paul Griffiths, for instance, compares our present "folk psychology" to Aristotle's obviously erroneous category of "superlunary objects," an arbitrary grouping wholly determined by ancient ignorance of astronomy.^ So, too, Harman juxtaposes our ordinary intuitions about morality with the findings from scientific research, arguing (by analogy) that just as we are wrong in our intuitions about classical mechanics so, too, we can be wrong in our moral intuitions. But I think there are very real questions about the extent to which modern empirical studies of human behavior have in any sense replaced rather than merely supplemented or possibly deepened our age-old "folk psychology" in anything like the way that modern astronomy and physics have introduced revolutions in the way we see the world. I do not doubt that many of our moral intuitions are erroneous or archaic, left over from earlier phases of human culture and no longer practical (what Nietzsche once called "the shadows of God"). Nor do I doubt that many of our moral judgments are based on hypocrisy, self-deception, and wishful thinking. But our moral intuitions are not like our intuitions in physics. There is no "matter of fact" independent of our intuitions and attitudes. (Against the moral realists of his day, Nietzsche insisted that "there are no moral facts," and contemporary authors such as Simon Blackburn have argued for a "quasi-realism" in which our personal intuitions and attitudes are ineliminable from moral concerns.^) The social sciences, our ordinary intuitions, and moral philosophy are all of a piece. There is no easy separation of "facts" about personal character and evaluations of moral merit.^ Character traits and virtueshonesty, trustworthiness, and a sense of fairnessare normative. They are not mere behavioral tendencies. All psychology, if it is psychology at all, is one or another version of "folk psychology" ("the only game in town," according to Jerry Fodor).

Harman and Doris attack virtue ethics in general and the concept of character in particular on the grounds that they do not survive experimental findings in the past few decades. Exhibit number one for both of them is the infamous Stanley Milgram experiments in which people with supposedly good character performed the most despicable acts when encouraged to do so by an authority (the experimenter). But though empirical research in social psychology can on occasion shock us, surprise us, annoy us, and sometimes burst our illusions, it all gets weighed and accounted for, whether well or badly, in terms of our ordinary folk psychology observations and the ordinary concepts of belief, desire, emotion, character, and interpersonal influences, interactions, and institutions. There are no Copernican revolutions and no Michelson-Morley experiments. The Milgram and other experiments such as those by Darley and Batson that play a central role in Doris's and Harman's arguments get rationalized and explained in all sorts of ways, but none of them in violation of the basic forms of psychological explanation that Aristotle would have found perfectly familiar.' Of course, there remains a debate about the relative influence of "external' (environmental) and "inner" factors (character), but the debate, which ever way it goes, remains within the framework of folk psychology and our ordinary psychological concepts.

We might be disturbed, for example, that so many subjects followed the instructions of an authority figure to the point of (what they thought was) the torturing of another human being, but the various explanations in terms of "obedience to authority" or the unusual circumstances of the experiment (how often are most of us told to punish anyone?) do nothing to challenge our ordinary moral intuitions. It just reminds us of something we'd rather not remember, that ordinary people sometimes act very badly in group and institutional situations. This should come as no surprise to those of us who do corporate and organizational ethics.

I would not want to rest my argument on a general and contentious claim about the social sciences, however. On the contrary, what is disturbing to me is Harman's and Doris's juxtaposition of virtue ethics against the social sciences. One of the virtues of virtue ethics, I have always thought, is its utter compatibility with the social sciences. It rests on (one or another) theory of human nature, and it is unashamedly a theory about how people are, not how they ideally ought to be. Kantian ethics is explicitly not so. Its main thrust lies in the domain of autonomy and it matters only marginally what people in fact want to do or normally do. In virtue ethics, by contrast, what people want to ensure and normally do makes a great deal of difference. And learning what people want to ensure and normally do is always relevant, even if only as a warning that our practices and institutions are offering up the wrong kind of role models and encouraging the wrong kinds of desires, ideas, and behavior. (Utilitarianism is also rigorously empirical and shares this virtue with virtue ethics, but it tends to emphasize the consequences of behavior and thus ignore the intentions and motivesand thus the characterof the agent.)

I have long been an advocate of cooperation between moral philosophy and the social sciences in business ethics. I think that the more we know about how people actually behave in corporations, the richer and more informed our moral judgments and, more important, our decisions will be. In particular, it is very instructive to learn how people will behave in extraordinary circumstances, those in which our ordinary moral intuitions do not give us a clue. All of us have asked, say, with regard to the Nazi disease in Germany in the Thirties, how we would have behaved; or how we would behave, think, and feel if we worked for a tobacco company. But even in an ordinary corporation (which is not the same as a university in which there is at least the illusion of individual autonomy and "academic freedom"), the question of "obedience to authority" comes front and center.

Thus, an experiment like the Milgram experiment is shocking precisely because it does not seem to presuppose any extraordinary context. Milgram's experiment, which would certainly be prohibited today, has to do with subjects inflicting potentially lethal shocks to victim-learners (in fact the experimenter's accomplices). Even when the victim-learners pleaded for them to stop, the majority of subjects continued to apply the shocks when ordered to do so by the authorities (the experimenters). One could easily imagine this "experiment" being confirmed in any corporation. ' But I find the use of such research to undermine the notion of character not at all convincing.'^ Harman, for example, argues that Empirical studies designed to test whether people behave differently in ways that might reflect their having different character traits have failed to find relevant differences. It is true that studies of this sort are very difficult to carry out and there have been few such studies. Nevertheless, the existing studies have had negative results. Since it is possible to explain our ordinary belief in character traits as deriving from certain illusions, we must conclude that there is no empirical basis for the existence of character traits.

But in addition to leaping from "very few studies" that are "difficult to carry out" to the conclusion that there is "no empirical basis for the existence of character traits," the whole weight of the argument comes to depend on iho. possibility of explaining our ordinary belief in character traits as "deriving from certain illusions." But what would such an explanation consist of? What illusions are we talking about? And what is our "ordinary belief in characier"? I will argue that it does not require the "robust" notion attacked by Harman. Doris is much more cautious and painstaking in his conclusions. He admits that empirical psychological studies are deeply flawed and limited especially in the fact that the studies he employs describe only particular behavior in particular (artificial) situations and not long-term patterns of behaviorwhich is what character is all about. He admits that "meaningful generalization outside of the laboratory" is "questionable."^"^ He even says, borrowing from Churchill on democracy, "I'll readily admit it: experimental psychology is perhaps the worst available method for understanding human life. Except, I hasten to add, for all of the other methods" (including the use of moral "intuitions" of armchair moral philosophy).'5 His main objection to those who champion virtue ethics, however, is that "they presuppose the existence of character structures that actual people do not very often possess" (12, 42, 68). But unless such structures are supposed to be indefensibly wooden and the "not very often" means "very rarely" this is a fairly weak claim that is perfectly compatible with what virtue ethicists require in terms of character.

What Is a Virtue and Whence Character?

Harman does a nice job of delimiting the ordinary notions of virtue and character, namely those that are most relevant to business ethics. He distinguishes character from various psychological disorders (schizophrenia, mania, depression). More dubiously, he distinguishes character from "innate aspects of temperament such as shyness or being a happy or sad person."^^ Kant, oddly enough, quite correctly insists that being happy (though an "inclination") can be a virtue, as it makes us more inclined to do our duty. But Harman is not just attacking the virtues. He is after character traits in general. Shyness, for example, is a non-moral example of a character trait. Harman considers this a prime example of "false attribution." But I think Jean-Paul Sartre has his eye on something very important when he refers to the citing of such a character trait as "bad faith," namely, where we point to a causal syndrome where we should be talking about decisions and the cultivation (in a very strong sense) of character.'^ There is a certain element of such Sartrianism (an insistence on existential choices rather than robust character) in Harman's argument (with which I quite agree), but this is a very different set of reasons for questioning or qualifying the concepts of character and the virtues.

Harman then considers such Aristotelian traits as courage, cowardice, honesty, dishonesty, benevolence, malevolence, friendliness, and unfriendliness. (Although it is not clear, contra Hume, that benevolence and malevolence are virtue and vice, respectively. Doing and not merely wishing, beneficence and maleficence, are the virtues in question.) Aristotle describes "the ordinary conception of such character traits" as relatively long-term dispositions to act in certain ways. (We might note again that Aristotle was describing the ordinary conception of his Mediterranean peers twenty-three hundred years ago.) Doris calls this long-term disposition to act in certain ways "globalism," which involves (a) consistency of character traits, (b) stability of character traits, and (c) the integration of various such traits, what in Aristotle is usually called "the unity of the virtues." It is what he ultimately claims to be "empirically inadequate."^^ Character traits involve activity, not just "possession," habits and operative desires and not just skills. Skills and knowledge may well be involved but are not sufficient for the attribution of the virtue (or vice). Furthermore, character traits must be broad based rather than narrow dispositions. (A particular fear does not signify cowardice.)

But the attribution of virtue (or vice) and the ascription of character traits are particularly tricky notions in Harman's and especially in Doris's discussions. To deny that a particular fear or phobia entails cowardice is not yet to leap to the "global" hypothesis that a virtue or character trait must be all-pervasive in one's personality. Doris discusses "local traits" (honesty in particular) and observes that people are sometimes honest in one sort of situation but not in another (108). This, of course, is no surprise. (Alfred Carr, among others, have often noted the inappropriateness of the virtues in at least some business settings.2) But in the defense of the virtues one need not insist on global virtues (or vices) any more than one should insist that each and every bit of behavior is the reflection of a virtue (or lack thereof). Doris's dramatic postmodern conclusion, "The Fragmentation of Character," the idea that there is no single "core of character" that alone explains our social behavior, is on the one hand (like most postmodern rhetoric) enormously overblown.

On the other hand, it is just a bland description of what we all recognize, when we are not being blindly moralistic or overly philosophical, that the virtues are contextual and only rarely "global" in nature. In the ordinary conceptions of character traits and virtues, Harman and Doris tell us, people differ in their possession of such traits and virtues. People are different, and these differences explain their differences in behavior. Harman: "We ordinarily suppose that a person's character traits help to explain at least some of the things the person does" (italics mine). But, he says, "the fact that two people regularly behave in different ways does not establish that they have different character traits. The difference may be due to their different situations rather than differences in their characters" (italics mine). But notice that there is no inconsistency whatever between insisting that a person's character traits help to explain their behavior and insisting that a difference in behavior may be due to the different situations in which two people find themselves. So, too, Doris's objection to globalism is that people (in experimental situations) fail to display the consistency and stability that explanations in terms of character require. But, again, the short-term experiments that he cites do not undermine our more ordinary long-term judgments about personal propensities and dispositions.

At best, they force us to face some hard truths about ourselves and consider other propensities and dispositions that may not be virtuous at all.

In our "ordinary conception" two people (one honest, one dishonest) in the same situation (discovering a lost wallet in the street, encountering a person in apparent desperate need, being ordered by an experimenter to "keep on punishing") will very probably act differently. But any philosopher worthy of his or her debating trophies will quickly point out that no two situations are sufficiently similar to make that case. It is only a very thin description of "the situation" (the experimental set-up) that makes it seem so. Subjects come from different backgrounds and different social classes. They are different genders. They may as a consequence have very different senses of the situation. I would not join Joel Feinberg in claiming that those students who do not stop for a stranger in need (in Darley and Batson's much-discussed "Good Samaritan" experiment) have a "character flaw," but neither would I conclude (with Doris) that their behavior is largely "situational."^i The student's way of seeing and being in the situation may be very different, and this, of course, is just what Aristotle says about character. It is, first of all, a kind of perception, based on good up-bringing. Thus I think Harman is being a bit disingenuous when he argues that "they must be disposed to act differently in the same circumstances (as they perceive those circumstances)." The question of character begins with how they perceive those circumstances.

In his subsequent discussion, Harman follows Nisbett and Ross (1991) in arguing that "people often choose the situations to which they are exposed."^2 "Thus clerics and criminals rarely face an identical or equivalent set of situational challenges. Rather they place themselves, or are placed by others, in situations that differ precisely in a way that induce clergy to look, feel, and act consistently like clergy and that induce criminals to look, feel, and act consistently like criminals" (Nisbett and Ross, 19).

Furthermore, in the presence of their peers, people "sometimes feel obliged even committed to act consistently" (ibid, italics in original). True enough (and Jean-Paul Sartre could not have put it better). Corporate managers and employees feel obliged and committed to act in conformity with corporate pressures and policies even when they are questionable or unethical, and they learn to rationalize accordingly. The question is, does any of this imply that we should give up or give in on character? Or should we say that character is both cultivated and maintained through the dynamic interaction of individuals and groups in their environment and they in turn develop those virtues (and vices) that in turn motivate them to remain in the situations in which their virtues are supported, reinforced, and not threatened? In Milgram's famous "shocking people" experiment in the early 1960s (just as America was getting more deeply involved in the morass of Vietnam), the experimental data were indeed shocking, even to Milgram and his colleagues who expected no such result. In the social context of the times, questions about obedience to authority (left over from the Nuremberg trails not so many years before) had a special poignancy, especially in the face of the soon to be challenged American "innocence" of the time.

It was very upsetting to find that good, solid, ordinary middle-class people could be ordered (but not coerced) to act so brutally (whether or not they had severe misgivings about their behavior at the timea matter of no small importance here). The facts of the experiment are beyond dispute. But what the experiment means remains highly controversial, and it does not deserve the central place in the attack on character that it is now receiving. Doris claims that "Milgram's experiments show how apparently non-coercive situational factors may induce destructive behavior despite the apparent presence of contrary evaluative and dispositional structures." Accordingly, he "gives us reason to question the robustness of dispositions implicated in compassion-relevant moral behavior.

Well, no. The disposition (virtue) that is most prominent and robust in this very contrived and unusual situation, the one that virtually all of the subjects had been brought up with and practiced everyday since childhood, was doing what they were told by those in authority. Compassion, by contrast, is a virtue more often praised than practiced, except on specially designated occasions (giving to the neediest at Christmas time) or stretching the term to include such common courtesies as restraining one's criticism of an unprepared student or letting the other car go first at a four-way intersection. (I would argue that such examples betray a lack of understanding of what compassion is.) Most often, people display compassion by "feeling sorry for" those much worse off than they, a very small expenditure of effort even when it is sincere. It seems to me that what the Milgram experiment showsand what subsequent events in Vietnam made all too painfully obviouswas that despite our high moral opinions of ourselves and our conformist chorus singing about what independent individuals we all are, Americans, like Germans before them, are capable of beastly behavior in circumstances where their practiced virtues are forced to confront an unusual situation in which unpracticed efforts are required.

In the Milgram experiment as in Vietnam, American subjects and soldiers were compelled by their own practiced dispositions to follow orders even in the face of consequences that were intolerable. Obedience may not always be a virtue. But that is not what is being challenged by Harman and Doris. They are denying (contrary to the empirical evidence) that people have robust dispositions. I would say, no. They are just looking at the wrong disposition.

In discussions of Vietnam, those who were not there (especially politicians) like to talk about the virtue of courage as the defining trait of the American forces. What they ignore, of course, is the very nature of the war. In several important memoirs by soldiers who served there. Bill Broyles and Tim O'Brien, it becomes clear that courage was just about the last thing on most of the soldiers' minds.24 They were terrified of losing legs and arms. They were moved by camaraderie and a sense of mutual obligation. (The virtue-name "loyalty" misses the mark.) The only discussion of courage in O'Brien's book has to do with a single heroic figure, a Captain Johansen whom he likens to Hector in Homer's Iliad. But this one character is exemplary in precisely the fact that he alone talked about and exemplified true courage.

But the absence of courage (which is not to imply anything like cowardice on the part of the American troops) had a great deal to do with the nature of this particular war. It lacked any sense of purpose or progress. It lacked any sense of meaning for most of the men. And so, in that moral vacuum, all that was left for most soldiers was the worry about their own physical integrity and their keen sense of responsibility for each other. The atrocities at My Lai and Thanh Phong followed as a matter of course. There was no context in which either character or courage could be exercised.

Which brings us back to the misgivings and feelings of discomfort experienced by some (not all) of the subjects and the "grunts" in Vietnam. Feelings of compassion (and other moral sentiments) may not be definitive in motivating behavior, especially if one has not faced anything like the awful situation in which the subjects and soldiers found themselves. But it does not follow that there is nothing more for virtue ethics to say about such cases. Experiments such as Milgram's are no longer allowed on college campuses, and for good reason. The feelings provoked in the subjects were too painful, and often with lasting damage.25 And this is nothing, of course, compared to the post-traumatic experiences of many of those who served in Vietnam.

The robustness of compassion must be measured not simply in terms of whether the subjects refused to continue with the experiment or not (most did) or whether the soldiers continued to ensure as they were ordered but by how powerful and upsetting the feelings they experienced both during and after the experiment. It is worth noting that there were a few sadists who actually enjoyed cruelty. There were others that were brutalized by the experiment and many who were brutalized by the war. That, it seems to me, should not be discounted. Bosses today are once again being forced to lay off thousands of their managers and employees. ("Market forces" is the inescapable explanation.) But there is all the difference in the world between those monsters like the infamous Al "Chainsaw" Dunlap who took such evident pride in cross the board cuts and virtual saints such as Aaron Feuerstein who felt so badly about having to lay off workers (after a fire gutted his factory) that he kept them on the payroll until the company got back on its feet.

The Milgram Experiment Revisited: A Model of Corporate Life?

Is corporate life nothing but the vectors of peer pressures, leaving very little or even no room for the personal virtues? Does social psychology show that this is not the case only for corporate grinds but for all of us? Empirically-minded philosophers love to find a single experiment, or perhaps two, that make this case for them, that is, which provide the basis for speculative excursions that go far beyond the (usually rather timid) findings of the social psychologists themselves. Harman's appeal to the two famous experiments by Milgram and by Darley and Batson are illustrative. Doris takes in a much wider swath of the social science literature, but even he is forced to admit, throughout his admirable book, that there are profound reasons for not generalizing from particular experiments to a good deal of "real life."

Regarding the Milgram experiment, Harman (following Ross and Nisbett) rejects as implausible any explanation in terms of a "character defect" and suggests instead the "step-wise character ofthe shift from relatively unobjectionable behavior to complicity in a pointless, cruel, and dangerous ordeal." I think that this is indeed part of the explanation. Milgram's subjects needed to have their callousness cultivated even as they dutifully obeyed the authorities (like the proverbial frog in slowly boiling water). The subjects could not have been expected to simply shock strangers on command. But where Harman adds that we are tempted to make the "fundamental attribution error" of blaming the subject's destructive obedience on a personal defect, I would say instead that what the Milgram experiment shows is how foolish and tragic the otherwise important virtues of conformity and obedience can be.

There is no "personal defect" on display here precisely because what the experiment shows is the consistency and stability of that virtue. And the fact that it is (like all virtues) not always a virtue is no argument against its status as part of the core of the explanation of the subjects' behavior. The rest of the explanation involves not just the incremental but also the disorienting nature of the situation.

But one third of the subjects in the Milgram experiment did quit. And those who did not were indeed confused. Is there no room for character in a complete explanation? Or do the differences between the subjects and their behavior and feelings demand such an explanation? Where are all of those studies on individual differences that would explain the differences (without necessarily taking anything away from the importance of the situation and the importance of the authority of the experimenters)? What about that voluminous literature not in social psychology but in the (artificially competing) field of personality theory, from Freud, Gordon Allport, David McClelland, and more recently, Costa and McRae? At the University of Minnesota, just a few blizzards away from Doris's previous base in Ann Arbor, the continuity and stability of character has become something of a minor industry.

If we want to play off moral philosophy and virtue ethics against the social sciences, let's make sure that all of the social sciences are represented and not just social psychology, which tends to define itself (artificially again, in competition with personality theory) as the study of the social dimensions of human behavior.

The other often-used case for "lack of character" is the case of the "good Samaritan," designed by Darley and Batson. Seminary students, on their way to give an assigned lecture (on "the good Samaritan") were forced to confront a person (an accomplice of the experimenter) on their way. Few of them stopped to help. It is no doubt true that the difference between subjects and their willingness to help the (supposed) victim can be partially explained on the basis of such transient variables as the fact that they were "in a hurry." And it is probably true as well (and not at all surprising to those of us who are not pushing "faithbased initiatives" these days) that people who were (or claimed to be) religious or who were about to talk on a religious topic of direct relevance to the experience did not act so differently as they would have supposed.

But does it follow that character played no role? I would say that all sorts of character traits, from one's ability to think about time and priorities to one's feelings of anxiety and competence when faced with a (seemingly) suffering human being all come into play. Plus, of course, the sense of responsibility and obligation to arrive at an appointment on time, which once again slips into the background of the interpretation ofthe experiment and so blinds us to the obvious.

As in the Milgram experiment, how much is the most plausible explanation ofthe case precisely one that the experimenters simply assume but ignore, namely the character trait or virtue of promptness, the desire to arrive at the designated place on time? It is not lack of character. It is a conflict of character traits, one practiced and well-cultivated, the other more often spoken of than put in practice. Theology students have no special claims on compassion. They just tend to talk about it a lot. And as students they have had little opportunity to test and practice their compassion in ways that are not routine.

In his discussion, Harman argues that people often choose the situations to which they are exposed. But on what basis do they make such choices? Surely part of the explanation is their wanting to act as they believe they ought to, with the knowledge that they are prone to temptation and peer pressure. A more obvious aspect of their choice is their judgment that they would feel more comfortable in one situation rather than another and that their comfort depends, in part, on their virtues. Thus clerics and criminals place themselves, or are placed by others, in situations that differ precisely because they induce clergy to look, feel, and act consistently like clergy and induce criminals to look, feel, and act consistently like criminals.

None of this eliminates situational factors as an explanation of behavior. On the contrary, it furthers them and explains why people "feel obliged and even committed to act consistently." None of this implies that we should give up or give in on character but rather tells us that circumstances and character cannot be pried apart and should not be used competitively as alternative explanations of virtuous or vicious behavior.

Harman notes that employers mistakenly think that they can gain useful information from interviewing potential employees. But such interviews, Harman argues, only add "noise" to the decision process. This may explain, by the way, Princeton's peculiar (and so far as I know unique) policy of hiring new professors without interviews, on the basis of the written journal alone. But I doubt that it has much justification in social science. First of all, it is a falsification of the interviewing process to think that what it provides is more information. Rather it provides an opportunity for employers (or their chosen representatives in "Human Resources") to "get the feel" of a candidate, see how much "in sync" they are, in order to anticipate how they will "get along." This explains why most interviewers describe themselves as typically having made a decision for all intensive purposes within the first minute or so, which would make little sense if the purpose of the interview was to gather "more information."

Second, of course, there are people who have the skill (not necessarily a virtue) to interview well and others who do not. This can indeed be misleading, and it is all the worse if the candidate is also skillful at deception, hiding his or her crasser motives and intentions in order to "make a good impression." But none of this undermines the importance or intelligibility of the interviewing process. It just means that interviewers should be on their toes and learn to ferret out insincerity and deception, skills on which most of them already pride themselves.

What is not debatable, it seems to me, is that people present themselves differently, whether or not their presentations accurately represent their virtues and vices (which longer exposure is sure to reveal). I have long argued that the subject of explanation is not just the behavior of an agent but the behavior of an agent-in-situation (or some such odd locution). In business ethics, in particular, the behavior in question is the behavior of an ''individual-within-the-organization," which is not for a moment to deny that this context may not be the only one of relevance in moral evaluation. Context is essential but it isn't everything. Virtues and vices are important for our explanations of human behavior, but they make sense only in the context of particular situations and cultural surroundings. There is no such thing as courage or generosity in abstraction, but it does not follow that there is no such thing as courage or generosity.

Conclusion: In Defense of Business Virtue Ethics

Virtue ethics has a long pedigree, going back to Plato and Aristotle, Confucius in China, and many other cultures as well as encompassing much of Medieval and modern ethicsincluding, especially, the ethics of Hume, Adam Smith, and the other "Moral Sentiment Theorists." But we would do well to remind ourselves just why virtue and character have become such large concerns in the world todayin business ethics and in politics in particular. The impetus comes from such disparate sources as the Nuremberg trials and American atrocities in Vietnam, teenage drug use and peer pressure, and the frequently heard rationalization in business and politics that "everyone is doing it." The renewed emphasis on character is an attempt to build a personal bulwark (call it "integrity") against such pressures and rationalizations and (though half-heartedly) to cultivate virtues other than those virtues of unquestioning obedience that proved to be so dominant in the Milgram experiments and in Vietnam atrocities such as My Lai.

Nevertheless, I share with Harman and Doris a concern that virtue ethics and talk about character is being overused and abused. Too often preachers of the virtues praise (in effect) their own sterling personalities without bothering to note how little there has been in their lives to challenge their high opinion of themselves. Too often, people are blamed for behaving in ways in which, given the situation and their personal backgrounds, it is hard to see how they could have acted or chosen to act otherwise. In contemporary politics, in particular, the renewed emphasis on character is prone to bullying and even cruelty, for example, as way of condemning the victims of poverty and racial oppression for their behavior and insisting that such people "boot-strap" their way to respectability.

Thus, I could not be more in agreement with Harman when he throws suspicion on the American conservative William Bennett. But Nietzsche beats him to it, a full century before The Book of Virtues: Then again there are those who consider it a virtue to say, "virtue is necessary"; but at bottom they believe only that the police are necessary.

I think that Harman's and Doris's ultimate aim is to take moral philosophy away from such vicious moralism and give it back to the good old empirical social engineer. Indeed, B. F. Skinner is never far from these new empiricist accounts, although neither Harman nor Doris would accept the absurdities of strict behaviorism. But once we have downplayed character and with it responsibility and put all of the emphasis back onto the environment, the "situation," all that is left is to design circumstances conducive to desirable behavior. To be sure, such design is important and essential and almost totally ignored by too many virtue ethicists today. If we are to combat intolerance, encourage mutual forgiveness, and facilitate human flourishing in contexts plagued by ethnic hatred, for instance, there is no denying the need for mediating institutions that will create the circumstances in which the virtues can be cultivated. Closer to home, the cultivation of the virtues in much-touted moral education also requires the serious redesign of our educational institutions. And much of the crime and commercial dishonesty in the United States and in the world today is due, no doubt, to the absence of such designs and character-building contexts. (The market, said the late great "Buddhist" economist E. F. Schumaker, "is the institutionalization of non-responsibility."^^) We need less moralizing and more beneficent social engineering.

I could not agree more with these aims. But the existentialist twist to which Harman alludes (that we choose our circumstances) and the postmodern turn encouraged by Doris (that we acknowledge that for the most part our circumstances make us) convince me not that we should eliminate talk of the virtues and character but fully acknowledge both the role of the social sciences {all of the social sciences) and stop preaching the virtues without due emphasis upon both personal responsibility and the force of circumstances. Like Doris, we should appreciate more such "out of character" heroic and saintly behavior (he mentions Oscar Schindler in particular) and the exigencies of context and circumstances. But we should insist, first and foremost, that peopleat any rate, people like us are responsible for what they do, and what they make of themselves.

Questions

2 things you understood well from this article it can be (a concept, idea, quote, point or argument)

1 thing that is challenging to understand from this article, what specifically? and why?;

1 thing that you really liked, disliked; agreed or disagree with from this article?