I. THE SITUATIONIST EXPERIMENTS

John Doris has argued that a group of experiments conducted by social psychologists have deep implications for ethics. The most significant of these experiments is Milgram's obedience experiment, in which subjects were instructed by an authority figure to administer painful electric shocks to a person in an adjoining room. The overwhelming majority complied, and roughly 65 per cent of subjects continued to administer shocks up to the maximum of 450 volts despite the apparent screams of pain from their victim.Footnote ¹ Milgram's original results have been widely replicated in different cultures and over time, which suggests that most of us would administer the shocks were we placed in the experimental scenario.

Several other experiments focused on helping behaviour have been grouped by Doris with the Milgram experiment. In the bystander experiment conducted by Latané and Rodin, subjects who were filling in a questionnaire were led to believe that a person in the next room had become trapped under a fallen object and was in pain. Seventy per cent of subjects who were alone when filling in the questionnaire offered to help the victim, but when the subjects were filling in the questionnaire in the presence of a nonplussed confederate who did not offer to help the victim, only 7 per cent offered to help.Footnote ² In the Princeton Seminary experiment conducted by Darley and Batson, seminarians were also given the opportunity to help a person who appeared to be in need. When the seminarians believed that they had no reason to hurry, many stopped to help. Yet when the seminarians believed that they were running late for a presentation the overwhelming majority of them offered no help.Footnote ³ The mood-effect experiments conducted by Isen and Levin showed that subjects who were in a good mood because they had found a small amount of money were much more likely to assist people who had dropped articles than were subjects who made no lucky find.Footnote ⁴ Similarly, the noise-effect experiments conducted by Matthews and Cannon show that subjects are much less willing to help others if there is a high level of background noise.Footnote ⁵

According to Doris, this collection of social psychology experiments support situationism, a position that consists of three core claims:

1. (A) Moral differences in behaviour across a population owe ‘more to situational differences than dispositional differences among persons’.

2. (B) People behave inconsistently, and any behavioural consistency ‘may be readily disrupted by situational variation’.

3. (C) Character is not ‘evaluatively integrated’ because a person can possess some good and some bad behavioural dispositions, and the one disposition might be good in some situations and bad in others.Footnote ⁶

Several philosophers have claimed that the situationist experiments have important implications for ethics, while others have downplayed their significance. Thus far, this dispute has been focused primarily on issues concerning the viability of traditional virtue ethics. Doris and Gilbert Harman argue that the experiments undermine traditional virtue ethics by showing either that we do not possess character traits at all,Footnote ⁷ or that we do not possess the kind of stable and integrated character traits that reliably produce only right actions.Footnote ⁸ Rachana Kamtekar, Christine Swanton, John Sabini and Maury Silver, and Jonathan Webber have claimed that Doris fails to show that we do not have the kind of character traits posited by Aristotle or by neo-Aristotelians.Footnote ⁹

However, the situationist experiments do not bear merely on issues concerning the status of traditional virtue ethics, but on more fundamental questions concerning moral practice. Ethicists are required to offer a psychologically informed recommendation as to how we should go about attaining our practical moral goals, a recommendation that should be both morally apt and useful. All of us must consider whether we should change the way we deliberate and act in light of situationist experiments. As we shall see, the consensus amongst philosophers is that situationism merely identifies a threat to be guarded against. I shall argue that these philosophers have overlooked the practical opportunities that come with situationism.

II. SITUATIONISM AS ALL BAD NEWS

The philosophers who have written about the practical implications of situationism tend to see situationism as all bad news, not in the sense that it would be better if we never found out about the situationist effects, but in the sense that there is nothing good about our susceptibility to situational forces. The disagreements between these philosophers are usually disagreements over exactly how bad the news turns out to be and over the appropriate methods to mitigate the situationist effects.

Traditional virtue ethicists such as Kamtekar suggest that the situationist effects might well produce wrong actions in some people some of the time, but that truly virtuous people are not susceptible to such effects. The appropriate moral of the story, Kamtekar would claim, is that we should be aware of the situationist effects and inculcate virtues that will allow us to resist them.Footnote ¹⁰ Similarly, psychologist Philip Zimbardo responds to the situationist experiments by setting out a ‘ten-step program for resisting the impact of undesirable social influences and . . . promoting personal resilience and civic virtue’.Footnote ¹¹ Webber suggests that the situationist effects might well disrupt many traditional virtues, but that we can still cultivate the virtue of placing ourselves only in those situations that are conducive to right action. From his perspective, the appropriate moral of the situationist story is that we be aware of the situationist effects and make it a habit to stay out of situations in which such effects move us to act wrongly.Footnote ¹²

Dana Nelkin sees the situationist experiments not as posing a ‘monolithic threat to either freedom or responsibility’, but as posing many specific threats to freedom and responsibility that must be assessed ‘on a case-by-case basis’.Footnote ¹³ She too, it seems, thinks of the situationist effects not as something positive or useful, but as a dangerous threat that often can be overcome. At one point, Nelkin does hint at the view that the situationist effects might sometimes be useful. She notes that the mood effect detected in the Isen and Levin coin experiments can increase the likelihood of our performing helpful actions, yet Nelkin is also worried that people who perform helpful acts because they are in a good mood are not acting for morally admirable reasons. Moreover, if such people are not aware of their real reasons for acting, Nelkin suggests that they might thereby fail to count as free and responsible agents.Footnote ¹⁴

Doris is comparatively even-handed in his comments on the implications of situationism for moral practice, yet even he tends to treat the situationist effects as a threat to be overcome. He claims that it is better that we know about the situationist effects, and that by avoiding the kind of global characterological evaluation typical of virtue ethics we will be able to predict behaviour more accurately.Footnote ¹⁵ Doris also maintains that ‘reflection on situationist moral psychology can help us to judge and act better in ethically trying circumstances’Footnote ¹⁶ and goes on to suggest that we do so by ‘changing or avoiding problematic situations’, just as a recovering alcoholic should avoid a relapse by staying out of bars.Footnote ¹⁷ Along with Nelkin, Doris thinks that the situational forces that shape our behaviour are disparate. The environmental conditions that led people to act wrongly in Milgram's obedience experiment are different from those that caused the Princeton Seminarians not to help the person in need, and different again from the conditions in the bystander experiment and mood-effect experiments. It seems that the practical conclusion to be drawn from situationism is that we ought to keep watch for a diverse range of environmental conditions that typically lead to moral failures.

Sabini and Silver argue that Doris has drawn the wrong practical conclusion from the situationist experiments. They claim that the apparently heterogeneous phenomena observed in situationist experiments are produced by a single psychological mechanism: fear of embarrassment.Footnote ¹⁸ The subjects in the Milgram experiment were unwilling to confront and disobey the authority figure due to their fear of embarrassment, and this led them to act in ways that conflicted with their own moral judgements and values. Rather than suffer the embarrassment of arriving late for their talk, the Princeton seminarians hurried by the person who appeared to be in need. Rather than suffer the embarrassment of speaking out in a room of nonplussed confederates, the subjects in the bystander experiment stood by quietly instead of responding to the cries for help. It is far less plausible to claim that fear of embarrassment was the motivation force in some of the other situationist experiments, yet Sabini and Silver dismiss those experiments as being of trivial significance for morality.Footnote ¹⁹ It is not clear that such a dismissal is warranted, but, for current purposes, let us grant that the single psychological mechanism of fear of embarrassment explains some of the most significant situationist phenomena.

After unifying the phenomena in this way Sabini and Silver conclude that Doris's suggestion that you ought to ‘be sensitive to situational features that may affect your behaviour in subtle ways’ is not only vague but misleading. In its place, Sabini and Silver offer a more focused recommendation:

Sabini and Silver admit that this advice is narrow, but claim that it is ‘quite practical’, and that the greater generality of Doris’ advice is ‘apparent, not real’. In their words, ‘when the situationist position is distilled, the remaining substance is exactly as narrow as our prescription’.Footnote ²¹ Sabini and Silver clearly express the view that situationism brings nothing but bad news, and the situational forces are purely a threat to be guarded against.

As we have seen, there is significant disagreement over the unity or disparity of the underlying causes of the situationist effects. Yet we have also seen that there appears to be a rough consensus amongst philosophers who have commented on the practical moral implications of situationism. The rough consensus is that the situationist effects can pull us off the rails, and that awareness of the situationist effects should prompt us to guard against their influence in our lives and the lives of other moral agents.

What is missing from the rough consensus, though, is acknowledgement of the fact that situational forces can also put us back on track, and can produce many important kinds of moral success. In the remainder of this article I will argue that the particular recommendations made by Sabini and Silver are deeply unsatisfactory, and that the more general recommendations made by the other philosophers also neglect the good news that comes with situationism. While they portray the situational forces as threats to morality, I will argue that such forces are much closer to being morally neutral. I will then attempt to explain why philosophers have focused only on the bad news that comes with situationism, and finally defend the claim that we should put situational forces to good use.

III. SLIPPERY SLOPES ARE MORALLY NEUTRAL

Sabini and Silver criticize Doris for offering advice that is so vague as to be useless. Unfortunately, part of the advice offered by Sabini and Silver themselves appears to be useless. According to Sabini and Silver, the incremental increases of voltage in the Milgram experiment constitute a

Thus, Sabini and Silver advise us to ‘stay away from slippery slopes’.Footnote ²³ Yet the existence of this kind of moral slippery slope requires only a series of incremental steps that stretch between a permissible action and an impermissible action. These slippery slopes yawn on all sides of us almost every time we act, simply because there are almost always impermissible alternatives to our right actions, and those alternatives could almost always be reached by incremental steps. For instance, it is permissible to reprimand a friend gently for failing to keep a promise, but there is a slippery slope between a gentle reprimand and an impermissible violent reprisal. Similarly, it is permissible to give many kinds of preferential treatment to our own children at the expense of other children, but there is a slippery slope between this permissible preferential treatment and impermissible disregard for the well-being of other children. In most cases the immediate proximity of these slippery slopes is no obstacle to our acting as we should. Yet, even if it were such an obstacle, it is not clear that we could stay away from such slippery slopes. In telling us to stay away from slippery slopes Sabini and Silver have not offered practically applicable advice.

Perhaps what Sabini and Silver mean is not that we should stay away from slippery slopes, but that we should avoid sliding down slippery slopes. Alternatively, they might mean that we should stay away from situations in which other people try to push us down slippery slopes. The problem with these suggestions is that most slippery slopes that stretch between right and wrong actions are slippery in both directions, and, of those that are not, many slide us exclusively towards right actions.

Many so-called slippery slopes that appear in moral arguments are slippery purely in that there are incremental steps stretching between a permissible action and an impermissible action. Since these steps are incremental in both directions, there is no reason to suppose that all people who find themselves on such so-called slopes will slide towards wrong rather than right actions. In effect, these so-called slopes are not inclined one way or the other. In contrast, some slippery slopes are steeply inclined in one direction. In the literal case, for instance, in which we step onto a muddy hill, gravity will slide us down but not up. Similarly, some political slippery slopes are steeply inclined. For instance, we can imagine scenarios in which granting a limited degree of power to a hitherto disenfranchised political group would enable them to use that power to gain more power, and this self-reinforcing structure would very likely lead to a sharing of power between political groups. As this example illustrates, some steeply inclined slippery slopes can slide us towards good states of affairs.

Given that slippery slopes can also slide individual people towards right actions, should we stay away from slippery slopes? Let us suppose that John is inclined to perform an action ø that is actually wrong, but that he mistakenly judges as being right. Further, let us suppose that there are a series of incremental steps between that action ø and an alternative action ψ that is actually right, but that John mistakenly judges to be wrong. Surely it would be good for John to slide along this slope, and end up performing the action that really is right, namely ψ. Arguably, it would be good for John to do so even if he continues mistakenly to regard action ψ as wrong, and hence considers himself when he does ψ to be acting immorally, or acting against his own better judgement. Sometimes, at least, it is good to slide down or be pushed down slippery slopes. A recommendation that we always should avoid doing so would be morally bad advice.

Perhaps what Sabini and Silver should have recommended was that we should avoid sliding down slippery slopes that end in the performance of wrong actions. This is sound advice, but part of it is superfluous. We should avoid performing wrong actions in general, whether or not we are led to perform them by sliding down slippery slopes. Since slippery slopes run towards right actions too, the distinguishing feature that we should be advised to avoid is not the slope, but the wrong act at one of its ends.

In this case Sabini and Silver have focused on particular examples in which a certain force or process causes us to fail morally, and have concluded that we should avoid succumbing to the force; in effect, that this force is all bad news. Instead, they should have noticed that this force can also be used to produce moral successes, and hence that, in some circumstances, it is a good thing that we are susceptible to this force.

IV. FEAR OF EMBARRASSMENT IS MORALLY NEUTRAL

According to Sabini and Silver the most important moral of situationism is not that we stay away from slippery slopes, but that we recognize the dangers of failing to act rightly due to fear of embarrassment. Situationism teaches us that all too easily we are swept along with the crowd or intimidated by authority figures. Hence, Sabini and Silver recommend that people ‘should be very wary when their moral perceptions seem to clash with others’, and that they should understand both that it is hard and that it is possible to confront other people who are doing wrong’.Footnote ²⁴ Sabini and Silver claim that this recommendation, and nothing more general, is all that is warranted by situationism. In fact, even if their unified interpretation of the situationist phenomena were correct, the conclusion that we ought to draw would differ in two respects from that drawn by Sabini and Silver themselves, and the second of these differences has important implications for ethics.

The first mistake in the explicit conclusion drawn by Sabini and Silver is comparatively superficial. They conclude that it is hard to confront people who are doing wrong, yet the experiments on which their case rests suggest that in some situations it will be hard to avoid confronting people who are doing wrong. For instance, let us imagine that we are in a room full of people, all of who are confronting a wrongdoer, and we are asked our opinion of the wrongdoing. In this case our fear of embarrassment would make it hard for us to refrain from confronting the wrongdoer. Perhaps what the situationist experiments actually suggest is that it is difficult to confront people who are doing wrong when those wrongdoers make up an overwhelming majority of those present, or when the wrongdoers are lone, uncontradicted authority figures who are perceived to be ‘institutional experts’.Footnote ²⁵ If we want to state Sabini and Silver's conclusion more carefully while sticking to the spirit of their argument, we should say that it is hard yet possible to confront overwhelming majorities or lone institutional authority figures who are doing wrong.

The second, deeper problem with the conclusion drawn by Sabini and Silver is that it is framed in terms of actions that are wrong. At first glance this framing seems appropriate, because the most important situationist experiments involve subjects performing morally wrong actions. Yet the psychologically and morally significant phenomenon observed in the situationist experiments is not that of people performing actions that are actually wrong, or even of people performing actions that they are justified in believing to be wrong. The significant phenomenon is that of people performing actions that they themselves ordinarily would judge to be wrong and that conflict with their own values.Footnote ²⁶ There is no reason to believe that the facts about whether their judgements are correct or justified are relevant to the psychological effects in question. For instance, it is plausible that people who have made unjustified and false moral judgements can be led to act in violation of those judgements by a fear of the embarrassment that would arise from contradicting overwhelming majorities or lone institutional authority figures.

As we have seen, the explicit conclusion that Sabini and Silver draw from situationism is ‘that it is hard and that it is possible to confront other people who are doing wrong’.Footnote ²⁷ This suggests that situationism brings only bad news: that situational forces are capable only of disrupting morality and leading to moral failures. We can easily be led astray, Sabini and Silver suggest, so ‘people should be very wary when their moral perceptions seem to clash with others’.Footnote ²⁸ The proper conclusion to be drawn from situationism, though, is that it is hard yet possible for a person to confront overwhelming majorities or lone institutional authority figures whom that person judges to be doing wrong. Because people are sometimes mistaken in their moral judgements and overwhelming majorities and institutional authority figures often judge correctly, the proper conclusion is not wholly negative, but is much closer to being morally neutral. The news that comes with situationism is that our fear of embarrassment, just like our susceptibility to slippery slopes, leads to many moral successes as well as failures.

The practical conclusion drawn by Sabini and Silver is more deeply flawed than the interpretations of situationism offered by Doris and the other philosophers we have considered. Nonetheless, the negativity of Sabini's and Silver's conclusion is emblematic of a broader failure to acknowledge and explore the positive implications of situationism. Kamtekar and Swanton are not particularly troubled by the situationist experiments, but that is because they think that the effects can be overcome by proper moral training, rather than because they think that situationism brings good news as well as bad. Webber and Nelkin suggest that it is better that we know about the situationist effects, but not that we have any reason to be glad that we are susceptible to them. Even Doris is surprisingly conservative, failing to set out a systematic account of when and how we might exploit those effects in order to pursue our moral ends. Why have philosophers tended to see situational forces merely as threats to be guarded against? In the following sections we shall see how either an inappropriately narrow conception of the situationist effects or an inappropriately narrow conception of our moral goals could give rise to the illusion that situationism is all bad news.

V. INAPPROPRIATELY NARROW CONCEPTIONS OF SITUATIONISM

If we focus too closely on the actual situationist experiments, it might seem that people can be led to act wrongly by situational forces but not that people can act rightly due to such forces. While the Isen and Levin mood-effect experiment suggests that finding a coin makes people more likely to offer help, the Milgram experiment, the bystander experiment and the Princeton Seminary experiment all show people failing to do the right thing due to situational forces. Yet there is no good conceptual or empirical objection to the claim that the situational forces that lead people to act wrongly in these experiments could elsewhere lead people to act rightly. Indeed, Bateson, Nettle and Roberts have demonstrated experimentally that people engage in far less dishonest freeloading behaviour in a university coffee room if a picture of a pair of watchful eyes is placed over the ‘honesty box’ into which payments are supposed to be made.Footnote ²⁹ It is plausible that imagined shame or the fear of embarrassment prevents these wrong actions, and that, as situationists would predict, this imagined shame or fear of embarrassment is triggered by variable environmental conditions rather than reliably produced in all conditions by the agents’ respective characters.

Even if we grant that fear of embarrassment can prevent moral failures, it could be argued that, by definition, situational forces are those that produce rather than prevent moral failures. There are a great variety of social psychology experiments and a variety of forces that shape behaviour in those experiments. If these forces do not fall neatly into a natural group, we could simply stipulate that situational forces are those that lead people to perform wrong actions that conflict with their own values or judgements. On this definition, fear of embarrassment would count as a situational force only in those cases in which it produced wrong action, and situationism would bring only bad news. Yet this ad hoc classification of situational forces runs counter to the psychological literature, which groups the Milgram experiment and the bystander experiment with other experiments involving non-moral judgement, such as Asch's group effect experiments concerning judgements as to the length of lines.Footnote ³⁰ According to the most natural classification of situational forces, those forces can lead to moral successes as well as failures.

VI. SITUATIONIST EFFECTS UNDERMINE VIRTUE AND AUTONOMY

Another reason for which people have failed to notice the good news that comes with situationism may be that they have an inappropriately narrow conception of our moral goals. For instance, that conception can be too narrow if it is focused exclusively on the achievement of virtue or autonomy, thereby ignoring other less demanding goals. Another way in which that conception can be too narrow is by being focused exclusively on the first-person synchronic deliberation that immediately precedes action, thereby ignoring the second-personal and diachronic considerations that should be part of moral deliberation.

First, let us consider the implications of focusing only on the most demanding of our moral goals. Suppose that our sole moral goal were to be fully virtuous and nothing less than fully virtuous, in the sense described by traditional virtue ethics. We can be fully virtuous in this Aristotelian or neo-Aristotelian sense only by possessing stable and broad good character traits that are integrated by practical wisdom so as regularly and reliably to produce right judgement, feeling and action rather than wrong judgement, feeling and action.Footnote ³¹ How would our susceptibility to situational forces affect our capacity to be fully virtuous? In some contexts, such as that occupied by the subjects in the Milgram experiment, situational forces reliably lead us to act wrongly. Our disposition towards such failures strongly suggests that we do not possess full Aristotelian virtue, and hence do not achieve our moral goal.Footnote ³² In other contexts in which overwhelming majorities or lone institutional authority figures approve of actions that are right, our fear of embarrassment might push us towards right actions that we were not otherwise inclined to perform. Yet, if we are moved to act rightly by fear of embarrassment rather than by courage, generosity, honesty, or other virtues, then it seems that we will not have achieved our goal of possessing virtues and acting virtuously. Thus, it seems that situational forces cannot lead us to be virtuous and can prevent us from being virtuous. If our only moral goal were to be virtuous, situationism would be all bad news.

Traditional virtue ethicists might reject this argument for a couple of reasons. They could point out that, on Aristotle's account, children do not act out of virtue, but acquire virtue by repeatedly being made to do what a virtuous person would do. Thus, traditional virtue theorists could argue that a child's fear of embarrassment can act as the motivating force behind that child's right actions during the process of her acquisition of virtue. If this were the case, fear of embarrassment might be a useful stepping stone to virtue, and even traditional virtue ethicists might have some reason to believe that the situationist results are not all bad news with respect to our goal of being virtuous. Unfortunately, though, situationism does not support the claim that fear of embarrassment is a stepping stone to virtue. Our fear of embarrassment appears not to be merely a juvenile motive that typically is replaced by more admirable motives in mature adults. Rather, it remains a deeply influential motive throughout adulthood, and hence a persistent obstacle to the possession and exercise of virtue.

Alternatively, traditional virtue ethicists might claim that fear of embarrassment itself is a virtue, and that action motivated by a fear of embarrassment can be virtuous action. In this case, susceptibility to at least some of the situational forces could be part of a virtuous character. This view is unconvincing. It is plausible that the fear of a particular kind of embarrassment – namely, embarrassment at one's own moral failures – is a virtue. A well-developed and finely tuned sense of shame or guilt is an admirable character trait. Yet the fear of embarrassment detected in the situationist experiments is not this morally sensitive and admirable sense of shame or guilt. The fear of embarrassment described by situationists often prompts people to perform acts that they themselves judge to be wrong, and that are in fact wrong.

Some non-traditional theories of virtue allow character traits to count as virtues simply because they are instrumentally good.Footnote ³³ These theories suggest that in environments in which fear of embarrassment reliably has good effects, such fear counts as a virtue. Nonetheless, it is implausible that traditional virtue ethicists would judge our general fear of embarrassment to be a virtue. Thus, fear of embarrassment can prevent us from achieving traditional virtue, and cannot lead us to achieve traditional virtue. If our only moral goal were to be fully virtuous in the Aristotelian or neo-Aristotelian sense, situationism would bring only bad news for morality. It is plausible that some of the philosophers who see situationism as all bad news, including Swanton, Kamtekar and Webber, do so because they focus closely on traditional virtue and thereby overlook other, less demanding moral goals.

Philosophers influenced by the Kantian tradition have claimed that our moral goal is not to be virtuous in the Aristotelian sense, but to act autonomously. There is much disagreement over exactly what is required for action to count as autonomous, but some suggestions include that the action be congruent with the agent's own second-order desires,Footnote ³⁴ that the agent not be alienated from her motives,Footnote ³⁵ that the motives be the product of rational deliberation or be capable of surviving rational scrutiny,Footnote ³⁶ that the causes of the action should also be the causes of the agent's judgement as to what should be done,Footnote ³⁷ and that the agent be aware of the relevant reasons counting for and against the action and be moved by those reasons.Footnote ³⁸ This diversity in conceptions of autonomy is mirrored by a diversity of opinions regarding the proper role of autonomy in moral theory. Perhaps autonomy is a prerequisite for moral agency and responsibility, and hence a prerequisite for any kind of moral success or failure. Alternatively, perhaps moral success requires autonomy, but non-autonomous actions can count as moral failures for which the agent should be held responsible. In any case, autonomy is widely considered to be a necessary component of moral success, and the richer, reason-responsive conceptions of autonomy might be seen as both necessary and sufficient for that success.

For current purposes we need not determine which is the most plausible account of autonomy. We simply should note that on a broad range of such accounts, the situationist explanation of our actions in the experimental scenarios threatens to undermine our autonomy. For instance, actions that are caused by fear of embarrassment or by mood effects might count as less than autonomous because the agent was moved by forces that conflict with her own second-order desires, or forces from which the agent was alienated, or forces that did not and would not survive rational scrutiny, or forces that were not the cause of the agent's judgement as to what she should do, or forces that would not provide moral reasons in favour of the action. People who are moved to act by situational forces are often internally conflicted and are not appropriately connected to their reasons for acting or to their own processes of reasoning.Footnote ³⁹ If our actions regularly are caused by situational forces, it seems that we are not autonomous agents. If we are not autonomous agents, then either we fail to qualify as morally responsible agents at all, or we are flawed moral agents who are incapable of real moral success. Even in those cases in which situational forces cause us to do what an autonomous and well-informed moral agent would do, if we have not acted autonomously, our achievement might appear to be negligible.

Doris himself is worried that situationism will undermine autonomy. His response is to defend a less demanding account of autonomy, according to which an action is autonomous if it the product of motives that can be integrated into the narrative of the agent's life. In some cases, Doris suggests, people who are moved by situationist forces are acting on motives that can be narratively integrated, and hence those people should be held responsible for their moral failures. In cases in which motives cannot be integrated, Doris concludes, we should at least have mixed feelings about holding agents responsible.Footnote ⁴⁰ Whether Doris’ argument is convincing or not, it is clear that people who give comparatively rich accounts of autonomy will judge that the situational forces can only block us from acting autonomously, and that they do so in many cases. If they also believe that our only moral goal necessarily includes our acting autonomously, then, along with traditional virtue ethicists, they will conclude that situationism is all bad news.

VII. MORAL GOALS BELOW VIRTUE AND AUTONOMY

The fact that several philosophers have seen situationism as being all bad news might well be explained by the fact that those philosophers assume that our only moral goal necessarily includes our being virtuous or our acting autonomously. However, these are not our only moral goals. We also have several goals that are less demanding, all of which regularly will be achieved by traditionally virtuous or autonomous agents, but all of which might also be achieved by people who are not virtuous or autonomous.

The first and second of these extra moral goals are that we be synchronically and diachronically resolute. We are synchronically resolute when we perform actions that we ourselves, at the time of acting, judge to be right. We are diachronically resolute when we act in accordance with our long-term moral plans and hence in accordance with our long-term moral values.Footnote ⁴¹ These two goals can come apart. For instance, a person can be synchronically but not diachronically resolute, as in circumstances in which she changes her moral judgement just prior to acting, acts in accordance with that new judgement, but then regrets having done so because that judgement conflicts with her long-term values.Footnote ⁴² In pursuit of these goals we guard against akrasia, coercion and wantonness, all of which can lead us to abandon our plans and to act in ways that we judge to be wrong. An action's being synchronically and diachronically resolute is necessary for its being traditionally virtuous and, on many accounts, for its being autonomous, but it is not sufficient for its being traditionally virtuous or for its being autonomous. Actions might be synchronically resolute but not virtuous because they are the product of a whimsical judgement rather than stable character, and actions might be diachronically resolute but not virtuous because they are not backed by the right kind of feeling or by correct reasoning. Similarly, actions might be synchronically and diachronically resolute but not autonomous because the agent's judgement was not connected to the action in the right kind of way.

The third and fourth of our additional moral goals are that we make correct moral judgements and that we make good moral plans. Clearly, it is much better to act in accordance with our correct moral judgements and plans than merely to be resolute, because judgements can be mistaken and plans ill-conceived. Judging and planning correctly are necessary conditions for traditional virtue and for rich, reason-responsive autonomy, but the possibility of akrasia implies that they are not sufficient for either.

Each of these four goals plays a role in our ordinary practical moral deliberations. We aim to make correct moral judgements and effective long-term plans, and to act in accordance with those judgements and with those plans. Arguably, though, there is something good about our achievement of any one of these goals, even if it is the only one of our moral goals that we achieve, and even if we do fall short of virtue or autonomy. It is good to judge correctly, even if we fail to act in accordance with that judgement, and there is something good about being resolute, even if we have judged incorrectly.Footnote ⁴³

There is another a kind of moral success that can be achieved independently of all of the goals that we have considered so far. In many cases, people judge incorrectly, plan poorly, and fail to act in accordance with those judgements and plans, but still perform actions that are actually right. Huckleberry Finn, as described by Jonathan Bennett, judges incorrectly that he ought to turn in the runaway slave Jim to the authorities and plans to do so, but he fails to act in accordance with his own moral judgement and plan, thereby performing the right action of not turning Jim in. There is something good about Huck's motives and his action, even though he has failed to be resolute and failed to make correct judgements and plans about what morally he ought to do.Footnote ⁴⁴ In other cases people do the right thing but act from motives that are not admirable at all, or even from motives that are deplorable, as when someone gives to charity purely in order to be admired. Obviously, it would be better if such people acted from admirable motives, but regardless of the quality of their motives, they achieve some kind of moral success when they do the right thing for the wrong reason.

There is a significant limitation on our ability to aim for the goal of actually right action. To judge that an action is right is to judge that it is actually right. From a first-person perspective, it is impossible for me to judge that ‘I judge that my possible action ø is right, but ø is actually wrong.’ This suggests that the only way in which we can aim to perform acts that are actually right is to judge correctly and to act in accordance with our judgements, in which case it seems misleading to say that we have an extra practical goal of performing actions that are actually right. Yet even if we were to admit that this is not a goal at which we can aim independently of the others, actually right action remains a target that we can succeed in hitting independently of the others. Moreover, we can judge retrospectively that we ourselves have hit this target while having missed all of the others. For instance, an older and wiser Huck might look back on his failure to turn Jim in to the authorities and judge that back then he was irresolute and judged incorrectly, but that he did what was right.

Importantly, though, when we plan which actions we ought to perform in the future, as opposed to which actions we should perform right now, the goal of right action can come apart from the goals of judging correctly and being synchronically resolute. By way of comparison, let us consider Odysseus. He wants to hear the Sirens’ song without dying, and he plans to perform the action that is practically right: namely, to keep sailing past the island. But Odysseus knows that he cannot achieve this goal via judging correctly and being resolute at the time of acting, because he knows that the Sirens’ song will render him incapable of judging correctly at the time of acting. Thus, Odysseus ties himself to the mast so that he cannot leap overboard and instructs his oarsmen to ignore his future requests. He plans to be synchronically irresolute, and thereby manages to be diachronically resolute and to perform the action that is practically right.

There are moral analogues to this case in which we are aware of our own tendencies to make false moral judgements when in particular situations, such as when we face certain strong temptations, or when our rational capacities are impaired by emotion or alcohol. This knowledge can lead us to arrange obstacles that will prevent us from acting in accordance with those future false judgements. For instance, we can tell our friends in advance that they should intervene if we get heated and potentially violent during an argument, or we can deliberately leave the car at home so that we are not able to put lives at risk by driving home drunk. Sometimes we plan to sacrifice the goal of being synchronically resolute in order to do what, in moments of calm reflection, we recognize to be morally right.

Thus, we have five moral goals that we would hope to achieve regularly, even if we fail to be virtuous or to act autonomously: we hope to be synchronically and diachronically resolute, to judge correctly, to plan well, and to perform acts that are actually right. As we have seen, the possible conflict between some of these goals adds complexity to our moral enterprise. That enterprise is further complicated by a parallel set of additional goals regarding the moral qualities of other people and their actions. We also should strive to make other people judge and plan correctly and act resolutely and rightly. It is even better if we can help those people become virtuous and autonomous.

Often we succeed in reaching some of these second-personal moral goals without reaching others. For instance, we use threats of punishment to make people act rightly without leading them to make correct moral judgements or without making them virtuous. As with the goals concerning our own behaviour, our goal of making other people resolute can conflict with our goal of making those people act rightly. These goals concerning the moral qualities of other people are connected in some way to the original set of goals we have considered. For instance, the fact that my action causes another person to act rightly counts in favour of the rightness of my own action. Nonetheless, it is possible to meet our goals concerning the moral qualities of others while ourselves failing to act rightly or judge correctly and failing to be resolute or virtuous. Conversely, we often meet our first-personal moral goals while failing to meet any of our second-personal goals.

With respect to most of our moral goals beyond our own virtuous and autonomous action, situationism does not bring only bad news. Situationism teaches us that our resolve to act in accordance with our own moral judgements easily can be destroyed by fear of embarrassment. Yet situationism also reminds us that, in some situations, we can exploit our own fear of embarrassment so as to make ourselves perform planned right actions that otherwise we would abandon when the time comes to act. For instance, we can make public commitments to perform acts of charity, knowing that fear of embarrassment at breaking these commitments will make us follow through. Similarly, we can use this fear to prevent ourselves from performing actions to which we are inclined, but which we judge to be wrong. We can make sure that we are accompanied by others in situations in which we know that, were we unaccompanied, we would be likely to perform wrong actions. Our own fear of embarrassment can function as the rope that ties us to the mast, so to speak, and this is good news with respect to our goal of being diachronically resolute and performing actions that are right. In so far as our fear of embarrassment can affect our moral judgements, and not just our actions, we might also have reason to be glad that some of those judgements are shaped by that fear.Footnote ⁴⁵ If fear of embarrassment sometimes leads us to make incorrect moral judgements, in other situations that same fear will pull our moral judgements into line with the moral facts.

It is with respect to our goal of making other people act rightly that situationism brings the best news. Of course, in many social environments the existing situational forces do not help people act rightly. Were we trying to make slave owners act rightly in the white society of antebellum Alabama, the existing situational forces would be likely to frustrate our aim. But when we are part of a society in which the overwhelming majority of people hold correct moral views on most issues we should be pleased to hear that, even if we cannot change the moral views of dissenters, we can cause many of those dissenters to act rightly by exploiting their fear of embarrassment. Moreover, we should design and construct social institutions which exploit that fear to a good end. Public officials often try to secure the effectiveness of important social institutions by enforcing costly sanctions against people who do not comply with those institutions. The situationist results support the view that institutions which foster compliance simply by exploiting our fear of embarrassment and our desire for the approval of our peers will be more effective and cheaper to run.Footnote ⁴⁶ This is important news, and good news, and is directly relevant to moral practice.

Of course, the moral of the story is not that situationism is all good news. There are plenty of examples from inside and outside of the laboratory in which situational forces produce disastrous moral failures. To some people it might seem that even in the so-called good cases in which we use situational forces to produce right action, such manipulation is a paternalistic, disrespectful and sinister form of mind control. It could be argued that we achieve no moral success whatsoever unless we respect the autonomy and rational dignity of human beings, and that situational manipulation actively violates that autonomy. Such objections stem from a genuine and well-justified concern that we value the autonomy of ourselves and others, but we do not and should not value such autonomy above all else and regardless of the consequences. When we face the prospect of serious wrongs and harms being inflicted on innocent people, concerns about respect for the autonomy of the would-be wrongdoer should be overridden by concern for the well-being of those who otherwise would be wronged.Footnote ⁴⁷ Manipulating people into acting rightly via situational forces is far from the ideal means of moral persuasion. If we were to adopt this method unthinkingly as a default or as a first port of call, there is a danger that we would come to neglect higher goals such as virtue and autonomy. Nonetheless, sometimes the best we can do in practical circumstances is to embarrass a would-be wrongdoer into performing a right action.