Hostname: page-component-745bb68f8f-d8cs5 Total loading time: 0 Render date: 2025-02-05T18:00:23.778Z Has data issue: false hasContentIssue false
  • English
  • Français

When Utilitarians Should Be Virtue Theorists

Published online by Cambridge University Press:  01 June 2007

DALE JAMIESON
DALE JAMIESON*
Affiliation:
New York UniversityDale.Jamieson@NYU.edu
Rights & Permissions [Opens in a new window]

Abstract

The contrast typically made between utilitarianism and virtue theory is overdrawn. Utilitarianism is a universal emulator: it implies that we should lie, cheat, steal, even appropriate Aristotle, when that is what brings about the best outcomes. In some cases and in some worlds it is best for us to focus as precisely as possible on individual acts. In other cases and worlds it is best for us to be concerned with character traits. Global environmental change leads to concerns about character because the best results will be produced by generally uncoupling my behavior from that of others. Thus, in this case and in this world, utilitarians should be virtue theorists.

Type
Research Article
Information
Utilitas , Volume 19 , Issue 2 , June 2007 , pp. 160 - 183
Copyright
Copyright © Cambridge University Press 2007

1. I begin with an assumption which few would deny, but about which many are in denial: human beings are transforming Earth in ways that are devastating for other forms of life, future human beings, and many of our human contemporaries. The epidemic of extinction now underway is an expression of this. So is the changing climate. Ozone depletion, which continues at a very high rate, is potentially the most lethal expression of these transformations, for without an ozone layer, no life on Earth could exist. Call anthropogenic mass extinctions, climate change and ozone depletion ‘the problem of global environmental change’ (or ‘the problem’ for short).Footnote 1

2. Philosophers in their professional roles have by and large remained silent about the problem. There are many reasons for this. I believe that one reason is because it is hard to know what to say from the perspective of the reigning moral theories: Kantianism, contractarianism and common-sense pluralism.Footnote 2 While I cannot fully justify this claim here, some background remarks may help to motivate my interest in exploring utilitarian approaches to the problem.

3. Consider first Kantianism. Christine Korsgaard writes that it is ‘nonaccidental’ that utilitarians are ‘obsessed’ with ‘population control’ and ‘the preservation of the environment’.Footnote 3 For ‘a basic feature of the consequentialist outlook still pervades and distorts our thinking: the view that the business of morality is to bring something about’ [sic].Footnote 4 Korsgaard leaves the impression that a properly conceived moral theory would have little to say about the environment, for such a theory would reject this false picture of the ‘business of morality’. This impression is reinforced by the fact that her remark about the environmental obsessions of utilitarians is the only mention of the environment in a book of more than four hundred pages.Footnote 5

It is not surprising that a view that renounces as ‘the business of morality’ the question of what we should bring about would be disabled when it comes to thinking about how to respond to global environmental change. The silence of Kantianism on this issue is related to two deep features of the theory: its individualism, and its emphasis on the interior. Some Kantian philosophers have tried to overcome the theory's individualism but this is difficult since these two features are closely related.Footnote 6 Kant was not so much interested in actions simpliciter as the sources from which they spring. But if our primary concern is how we should act in the face of global environmental change, then we need a theory that is seriously concerned with what people bring about, rather than a theory that is (as we might say) ‘obsessed’ with the purity of the will.Footnote 7

4. Contractarianism has difficulties in addressing environmental problems in general and global environmental change in particular for at least three reasons. First, it generally has a hard time coping with large-scale cooperation problems and the difficulties with assurance to which they give rise. Second, contractarianism has a difficult time with negative ‘externalities’ – the consequences for me (for example) when you and another consenting adult agree to produce and consume some substance that pollutes the air. It may be possible to overcome these problems, at least in principle, through various revisions of the core theory. But the deeper problem with contractarianism is that it excludes from primary moral consideration all those who are not parties to the relevant agreements.Footnote 8 Yet much of our environmental concern is centered on those who are so excluded – future generations, distant peoples, infants, animals, and so on.

5. Common-sense pluralism is hampered by its intrinsic conservatism.Footnote 9 Although common-sense pluralists morally condemn obvious forms of bad behavior, they are ultimately committed to the view that most of what we do is perfectly acceptable. The role of moral philosophy is primarily to explain and justify our everyday moral beliefs and attitudes rather than seriously to challenge them. From this stance they criticize utilitarianism for being too revisionist and utilitarians for being no fun.Footnote 10 But what produces global environmental change is everyday behavior that is innocent from the perspective of common sense: building a nice new house in the country, driving to school to pick up the kids and, indeed, having kids in the first place, to mention just a few examples.Footnote 11 By the standards of common sense, a moral theory that would prescribe behavior that would prevent or seriously mitigate global environmental change would be shockingly revisionist.

6. Some may say that the reigning moral theories have little to say about our problem because it is not a moral problem. No doubt climate change (for example) presents all sorts of interesting and important scientific and practical challenges, but this does not make it a moral problem.Footnote 12

The question of what is (and is not) in the scope of morality is itself an interesting and important question worthy of extensive treatment, but here I will confine myself to only a few remarks. Deontologists might not consider global environmental change a moral problem because, on their view, moral problems center on what we intend to bring about, and no one intends to bring about global environmental change.Footnote 13 Similarly, Kantians who reject the idea that ‘the business of morality is to bring something about’ might also have reason to exclude our problem from the domain of morality. But whatever one's official view about the scope of morality, the question of how we should regulate our behavior in the face of climate change, ozone depletion and mass extinctions is important for anyone who cares about nature or human welfare – and these concerns have traditionally been thought to be near the center of moral reflection.

7. For present purposes I assume that our problem is a moral problem. I investigate utilitarian approaches to our problem because utilitarianism, with its unapologetic focus on what we bring about, is relatively well positioned to have something interesting to say about our problem. Moreover, since utilitarianism is committed to the idea that morality requires us to bring about the best possible world, and global environmental change confronts us with extreme, deleterious consequences, there is no escaping the fact that, for utilitarians, global environmental change presents us with a moral problem of great scope, urgency and complexity.

However, I would hope that some of those who are not card-carrying utilitarians would also have interest in this project. Consequences matter, according to any plausible moral theory. Utilitarianism takes the concern for consequences to the limit, and it is generally of interest to see where pure versions of various doctrines wind up leading us. Moreover, I believe that the great traditions in moral philosophy should be viewed as more like research programs than as finished theories that underwrite or imply particular catechisms. For this reason it is interesting to see how successfully a moral tradition can cope with problems that were not envisioned by its progenitors.Footnote 14

8. While Korsgaard castigates utilitarianism for its environmental obsessions, many environmental philosophers see utilitarianism as a doctrine that celebrates consumption rather than preservation. Specifically, it has been accused of preferring redwood decks to redwood trees and boxes of toothpicks to old growth forests. Other environmental philosophers argue that utilitarianism cannot account for the value of biodiversity, ecosystems or endangered species, and go on to condemn the theory for ‘sentientism’ and ‘moral extensionism’. According to these critics, rather than presenting us with a new environmental ethic, utilitarianism is the theory that has brought us to the edge of destruction.Footnote 15

But utilitarianism has an important strength that is often ignored by its critics: it requires us to do what is best. This is why any objection that reduces to the claim that utilitarianism requires us to do what is not best, or even good, cannot be successful. Any act or policy that produces less than optimal consequences fails to satisfy the principle of utility. Any theory that commands us to perform such acts cannot be utilitarian.Footnote 16

As I understand utilitarianism, it is the theory that we are morally required to act in such a way as to produce the best outcomes. It is not wedded to any particular account of what makes outcomes good, of what makes something an outcome, or even what makes something an action.Footnote 17 Moreover, having good theoretical answers to these questions does not mean that we will always know what is right when it comes to practical decision-making. And even when we think we know what is right we may change our minds in the light of reflection, analysis or experience. If utilitarianism is true, embracing the theory may be the first step towards doing what is right, but it is certainly not the last.Footnote 18

9. Utilitarianism is a highly context-sensitive moral theory. Since my concern here is with how a utilitarian should respond to an actual moral problem, I need to make some simplifying assumptions in order to produce responses that are more definitive than ‘it depends’. So in what follows, I will assume that the utilitarian in question holds fairly generic and reasonably traditional views about the matters mentioned in the previous paragraph (e.g. that well-being is at least one of the things that are good, that my causing something to occur or obtain is part of what makes something an outcome of my action, etc.). I will also assume that taken together these views imply that, all things considered, global environmental change is bad (or at least not best). Furthermore, I will assume that the utilitarian in question is a person whose psychology is more or less like mine, and that we have roughly the same beliefs about how the world is put together. I do not mean anything fancy by this – only that, for example, our decision-making is not decisively affected by our belief that this world is just a training ground for the next, that most of the world's leaders are agents of an alien conspiracy, or that I am as likely to be a brain in a vat as a guy with a job. Given this background, in the face of global environmental change, a utilitarian agent faces the following question: how should I live so as to produce the best outcomes?

10. Part of what should be taken into account in answering this question is that global environmental change presents us with the world's biggest collective action problem. Together we produce bad outcomes that no individual acting alone has the power to produce or prevent. Moreover, global environmental change often manifests itself in ways that are quite indirect. The effects of climate change (for example) include sea level rises, and increased frequencies of droughts, storms, and extreme temperatures. These effects in turn may lead to food shortages, water crises, disease outbreaks, and transformations of economic, political and social structures.Footnote 19 Ultimately, millions may die as a result, but climate change will never be listed as the cause of death on a death certificate. Because our individual actions are not decisive with respect to outcomes, and we are buffered both geographically and temporally from their effects, many people do not believe that their behavior has any effect in producing these consequences.Footnote 20 Even when people do see themselves as implicated in producing these outcomes, they are often confused about how to respond, and uncertain about how much can reasonably be demanded of them.

For a utilitarian, this much seems clear: agents should minimize their own contributions to global environmental change and act in such a way as to cause others to minimize their contributions as well. However, in principle, these injunctions could come apart. It is possible that the best strategy for a utilitarian agent would be hypocrisy: increasing my own contributions to the problem could be necessary to maximally reducing contributions overall (perhaps because my flying all over the world advocating the green cause is essential to its success). Or asceticism could be the best strategy: paying no attention to anyone's contributions but my own might be the most effective way for me to reduce overall contributions to the problem.Footnote 21 There may be particular utilitarian agents for whom one of these strategies is superior to a ‘mixed’ strategy. However, it is plausible to suppose that for most utilitarian agents under most conditions, the most effective strategy for addressing the problem would involve both actions primarily directed towards minimizing their own contributions, and actions primarily directed towards causing others to minimize their contributions.Footnote 22 This would seem to follow naturally (but not logically) from the fact that we are social animals who strongly influence others and are strongly influenced by them.

11. In light of these considerations, how should a utilitarian agent live in order to address the problem? I believe that one feature of a successful response would be non-contingency. Non-contingency requires agents to act in ways that minimize their contributions to global environmental change, and specifies that acting in this way should generally not be contingent on an agent's beliefs about the behavior of others.

The case for non-contingency flows from the failure of contingency with respect to this problem. Contingency, if it is to be successful from a utilitarian point of view, is likely to require sophisticated calculation. But when it comes to large-scale collective action problems, calculation invites madness or cynicism – madness, because the sums are impossible to do, or cynicism because it appears that both morality and self-interest demand that ‘I get mine’, since whatever others do, it appears that both I and the world are better off if I fail to cooperate. Indeed, it is even possible that in some circumstances the best outcome would be one in which I cause you to cooperate and me to defect.Footnote 23 Joy-riding in my ’57 Chevy will not in itself change the climate, nor will my refraining from driving stabilize the climate, though it might make me late for Sierra Club meetings. These are the sorts of considerations that lead people to drive their ’57 Chevies to Sierra Club meetings, feeling good about the quality of their own lives, but bad about the prospects for the world. Nations reason in similar ways. No single nation has the power either to cause or to prevent climate change. Thus nations talk about how important it is to act while waiting for others to take the bait. Since everyone, both individuals and nations, can reason in this way, it appears that calculation leads to a downward spiral of non-cooperation.Footnote 24

This should lead us to give up on calculation, and giving up on calculation should lead us to give up on contingency. Instead of looking to moral mathematics for practical solutions to large-scale collective action problems, we should focus instead on non-calculative generators of behavior: character traits, dispositions, emotions and what I shall call ‘virtues’. When faced with global environmental change, our general policy should be to try to reduce our contribution regardless of the behavior of others, and we are more likely to succeed in doing this by developing and inculcating the right virtues than by improving our calculative abilities.Footnote 25

12. This may sound like a familiar argument against act-utilitarianism. Act-utilitarianism is the theory that directs agents to perform that act which brings about the best outcome, relative to other acts that the agent could perform. Some philosophers have argued on conceptual grounds that agents who are guided by act-utilitarianism would not produce the best outcomes. This is because certain goods (e.g. cooperation, valuable motives, loving relationships) are inaccessible to, or unrealized by, agents who always perform the best act.Footnote 26 Thus, rather than being ‘direct utilitarians’ who focus only on acts, we should be ‘indirect utilitarians’ who focus on motives, maxims, policies, rules or traits.

The first point to notice is that it does not follow that act-utilitarians do not bring about the best world from the fact (if it is one) that certain goods are inaccessible to, or unrealized by, act-utilitarians. The world may be constructed in such a way that the best state of affairs is not one in which these values obtain, however important they may be taken individually. For example, the pleasure of drinking fine wine is inaccessible to, or unrealized by, a teetotaler, but it does not follow from this that the teetotaler's life is not the best life for him to lead, all things considered (i.e. the one that produces the most utility). By declining the pleasures of wine, the teetotaler may mobilize resources (both financial and energetic) that allow him to realize more utility than he otherwise would if he did not abstain from alcohol.Footnote 27

However, what I have said thus far is consistent with the rejection of act-utilitarianism, but my main concern here is not with the architecture of various versions of utilitarianism. My focus is on the moral psychology of a utilitarian agent faced with the problem, rather than on the conceptual structure of value. I agree that such a utilitarian agent should not adopt act-utilitarianism as a decision-procedure and try to transform herself into a moment-by-moment, act-utilitarian calculating device. One reason is because it is not possible for the attempt to succeed. We are cognitively and motivationally weak creatures, with a shortage of time, facts and benevolence. Our very nature as biological and psychological creatures is at war with the injunction, ‘transform yourself into a moment-to-moment, act-utilitarian calculating device and act on this basis’. There is no reason to think that attempting to live an impossible dream will produce more good than any other course of action.

This seems so obvious that I sometimes (darkly) wonder who invented act-utilitarianism, when, where, and for what purpose. As a theoretical construct it has its uses, but the idea that a utilitarian moralist must embrace a psychologically impossible doctrine on pain of inconsistency is to misunderstand the very project of moral theorizing.Footnote 28

Clearly Bentham and Mill were strangers to this doctrine.Footnote 29 They were promiscuous in their application of the principle of utility to acts, motives, rules, principles, policies, laws, and more besides.Footnote 30 Rather than beginning with the principle of utility and then demanding that people become gods or angels in order to conform to it, they start from a picture of human psychology which they then bring to the principle. While conforming to the principle of utility is supposed to make us and the world better, embedding the principle in human psychology is what makes the principle practical. Bentham and Mill were aware of the fact that the world comes to people in chunks of different sizes: sometimes we must decide between acts, at other times between rules or policies. Indeed, acts can express rules and policies, and rules and policies are instantiated in acts. One of the most difficult problems we face as moral agents is trying to figure out exactly what we are choosing between in particular cases.Footnote 31 Yes, textbook act-utilitarianism is a non-starter as an answer to our question, but who would have thought otherwise?Footnote 32

Ultimately, the most important problem with act-utilitarianism is also a problem with indirect views that focus on motives, rules, or whatever. All of these accounts are ‘local’, in that they privilege some particular ‘level’ at which we should evaluate the consequences of actions that are open to us. Rather than adopting any such local view, we should be ‘global’ utilitarians and focus on whatever level of evaluation in a particular situation is conducive to bringing about the best state of affairs.Footnote 33 Derek Parfit saw this point clearly when he wrote: ‘Consequentialism covers, not just acts and outcomes, but also desires, dispositions, beliefs, emotions, the color of our eyes, the climate and everything else. More exactly, C covers anything that could make outcomes better or worse.’Footnote 34

13. Some may sympathize with my rejection of utilitarian calculation, but think that in appealing to the virtues I have thrown myself into the arms of something worse. There are other, safer, havens for refugees from utilitarian calculation, it might be thought.

Some may say that what is needed to address our problem is coercive state power, not virtuous citizens. I do not see these as mutually exclusive alternatives. Legitimate states can only arise and be sustained among people who act, reason and respond in particular ways. The mere existence of a collective action problem does not immediately give rise to an institution for managing it, independent of the values and motivations of actors. Indeed, if it were otherwise, we would not be confronted by our problem. While it is true that our problem cannot fully be addressed without the use of state power, this observation does not answer or make moot the questions that I am asking.

Others may say that the solution to our problem consists in developing collective or shared intentions of the right sort. One version of this view holds that individual agents need to form intentions ‘to play one's part in a joint act’ or to ‘see themselves as working together [sic] to promote human well-being’.Footnote 35 It may be that such intentions would have an important role to play in successfully addressing our problem, but questions remain about what exactly such intentions consist in, how they arise, what sort of people would have them, and exactly why and in what circumstances they would be adopted.Footnote 36 My investigation is meant to address these further questions. In this respect my account can be seen as complementary to, or even perhaps as part of, the project of investigating shared or collective intentions as solutions to collective action problems.

14. It is now time for me to say something more constructive about my conception of a virtue. Julia Driver's account is helpful as a first approximation: a moral virtue is ‘a character trait that systematically produces or gives rise to the good’.Footnote 37 Clearly this account should be supplemented to reflect the fact that the emotions are closely associated with the virtues.Footnote 38 Emotions play an important role in sustaining patterns of behavior that express such putative virtues as loyalty, courage, persistence, and so on. Without emotions to sustain them, it is difficult to imagine how parenting, friendship and domestic partnership could exist among creatures like us.Footnote 39

Even if Driver's account were supplemented in this way, it would still remain quite generic, since there are different understandings of such expressions as ‘character trait’, ‘systematically’, ‘produces’ and ‘gives rise to’. Moreover, this account would leave many important questions unanswered, including those about the relations between the virtues and human flourishing, and about the relations between the virtues themselves. However, answering these questions is not required for my purposes. What matters to me is the contrast between calculative and non-calculative generators of action, and I use ‘the virtues’ as the name for a large class of the latter.Footnote 40

Some virtue theorists will not be very welcoming of this project. They would deny that an account of the sort I want to give constitutes a version of ‘virtue ethics’. For they hold that ‘What is definitive of virtue ethics. . . is that it makes virtues not just important to, but also in some sense basic in, the moral structure.’Footnote 41 Perhaps in deference to this view, what I should be understood as exploring is when an account of utility-maximizing requires a theory of virtue.Footnote 42

15. Here is a reminder of what I am claiming. Given our nature and the nature of our problem, non-contingency is more likely to be utility-maximizing than contingency. This is because contingency is likely to require calculation, and calculation is not likely to generate utility-maximizing behavior. Thus, in the face of our problem, utilitarians should take virtues seriously. Focusing on the virtues helps to regulate and coordinate behavior, express and contribute to the constitution of community through space and time, and helps to create empathy, sympathy and solidarity among moral agents.

16. The most serious problem with the idea that non-contingency should be an important part of a utilitarian theory of how to respond to our problem is that it is in tension with an underappreciated, but extremely important, general feature of utilitarianism: non-complacency. Non-complacency refers to the fact that ways of life and patterns of action should be dynamically responsive to changing circumstances, taking advantage of unique opportunities to produce goodness, and always striving to do better.

Consider first how non-complacency counts against some versions of indirect utilitarianism, especially those motivated by the desire to produce moral judgments that are more closely aligned with common-sense morality than the judgments that act-utilitarianism would seem to deliver.Footnote 43 Views motivated by this desire can lead to a kind of moral complacency that is at odds with any theory that is directed towards producing the best outcomes. Consider two examples.

Suppose that I am a motive-utilitarian who acts on the set of motives that produces more utility overall than any other set of motives that I could have. Imagine that in a one-off situation it is clear that I could produce the most good by acting in a way that is horrific from the point of view of common-sense morality, and that this action is not consistent with my set of standing motivations. A conscientious utilitarian should struggle to perform this one-off act. If she fails in her struggle, she should regret her failure – not because a utilitarian should value regret for its own sake, but because feelings of regret are a characteristic response to the failure to do one's duty. Such feelings of regret may also have a role to play in steeling the agent so that in the future she can perform such one-off acts, however repugnant they may seem to her. Someone who complacently comforted herself with the knowledge that her motives are the best ones to have overall ought to be suspect from a utilitarian point of view, for she acts in a way that she knows is wrong and does not even try to do better.

A similar story can be told about someone who knows he ought to save a stranger rather than his brother in some moment of stress. Such a person, insofar as he is a utilitarian, cannot really be satisfied by telling himself that on the whole he does better acting on the intuitive level rather than ascending to the critical level. He would be like a pilot who on the whole does better flying at 30,000 feet rather than ascending to 40,000 feet, comforting himself about the importance of acting on the basis of good rules of thumb while he is headed directly towards a fully-loaded 747. He may not be able to bring himself to do the right thing, but more than shoulder-shrugging is called for.

Non-complacency should lead a utilitarian to moral improvement in two ways. First, she should be sensitive to the fact that circumstances change. What is the best motivational set in an analog world may not be best in a digital one. Moving from Minnesota to California may bring with it not only a change of wardrobe, but also a different optimal motivational set. Second, a utilitarian should constantly strive to shape his motivational set in such a way that his behavior is ever more responsive to particular situations. Broad motives and rules of thumb are starting points for a utilitarian agent, but not where he should aspire to end his struggle for moral improvement.

The problem is that non-complacency, which seems to me to be important and underappreciated by indirect utilitarians, appears to be in tension with non-contingency, which is required in order to address large-scale collective action problems. Virtues give utilitarians a way of making human behavior inflexible enough to deal with collective action problems, but outside the context of collective action problems it is flexible patterns of behavior that generally are needed for utility-maximizing.

17. One approach would be to relax the demand of non-complacency by giving up utilitarianism in favor of progressive consequentialism. Progressive consequentialism requires us (only?!) to produce a progressively better world rather than the best world. Abandoning the maximizing requirement of utilitarianism in favor of a diachronic duty to improve the world would help relieve, but not entirely resolve, the tension between non-contingency and non-complacency. For as long as non-contingency is in the picture there are going to be conflicts between the character traits that it evokes, and the demand of non-complacency that on at least some occasions we act in ways that are contrary to what these traits would manifest. Relaxing the demands of duty will make these conflicts rarer but will not eliminate them entirely.Footnote 44

18. Another, complementary, approach is to develop a highly domain-specific account of the virtues. When it comes to global environmental change, utilitarians should generally be inflexible, virtuous greens, but in most other domains they should be flexible calculators.

The problem with this is that life is not very good at keeping its domains distinct. Suppose that my friend Peter asks me to give him a lift to an Oxfam meeting and that this is the only way that he will be able to attend.Footnote 45 However, I am an inflexible, virtuous green when it comes to global environmental change. My green dispositions cause my hand to tremble at the very thought of driving, and I cannot bring myself to give Peter a lift to the meeting. If I were a globally flexible calculator instead, I would not care in what domain utilities are located. If driving Peter to the meeting would produce better consequences than my refusing, then I would give Peter a lift. Thus it would seem that non-contingency in the domain of global environmental change may not contribute to realizing what is best overall.

One response would be to say that in this case I should calculate about whether to calculate. In one way this response is correct and in another way it is wrong. As theorists we should try to identify those cases in which calculation is likely to lead to optimal outcomes and those in which it will not, and this requires calculating the utility of calculating in various domains (as indeed we did informally in the previous paragraph). But as utilitarian agents we should not calculate about whether to calculate, for this would defeat the very possibility of inculcating the character traits that make us virtuous greens. And anyway, such higher-order calculation threatens an infinite regress of calculations as well as generally straining psychological credulity.Footnote 46

So what should I say to Peter? First, the problems of global environmental change are so severe and the green virtues so generally benign that the domain over which they should dominate is very large. Second, the green virtues would never take hold if their particular expressions were systematically exposed to the test of utility; so if we think that having green virtues is utility-maximizing overall then we ought not to so expose their expressions (except in extreme cases, of which, I have been assuming, this is not one). So too bad for Peter and his Oxfam meeting.

But the problem of calculation reappears with the words, ‘except in extreme cases’. For a utilitarian, the commitment to non-contingency must include such an ‘escape clause’. If this were an extreme case (suppose that the lives and well-being of the entire population of a medium-sized African country turned on Peter attending the Oxfam meeting) and I could not bring myself to give Peter a lift, then I would be no better than one of those compulsive rule-worshipers whom utilitarians love to bash. But without calculation, how can I know whether or not this is an extreme case?

Part of the answer is that we are simply able to recognize some extreme cases as such: we just do it. When the house is on fire, a child is screaming, atrocities are being committed and civilizations threatened, moral mathematics are not needed in order to see that the patterns of behavior that are generally best may not be up to it in the present case. Of course there may also be cases in which calculation would be needed in order to see that it would be best to break patterns of behavior given to us by the green virtues. But on these occasions the virtuous green will just have to forgo the best, trusting in the overall utility-maximizing power of the green virtues.

19. There is a further challenge to which I have already briefly alluded (in section 11). If others are having a good time changing climate, destroying ozone and driving species to extinction, and the green cause is hopeless, then it appears that I am morally obliged to join in the fun. A utilitarian should not, at great cost to herself, plow through the snow on her bike while everyone else is blowing past her in their gas-guzzling ‘suburban utility vehicles’ (SUVs). If the world is to be lost anyway, then the morally responsible utilitarian will try to have a good time going down with the planet. If the best outcome (preventing global environmental change) is beyond my control and the worst outcome would be for me to live a life of misery and self-denial in a futile attempt to bring about the inaccessible best outcome, then the best outcome that I can produce may involve my living a high-consumption lifestyle. But everyone can reason in this way and so we may arrive at the conclusion, not just that it is permissible to live like a normal American, but that utilitarians are morally obliged to do so. This seems truly shocking.

There are really two arguments here. The first argument concerns the decision process of a single agent; the second claims that the first argument generalizes to all similarly situated agents.

Consider the second argument first. This argument trades on equivocating as to whether or not the best outcome is in fact accessible to an agent. Imagine a world of only two agents, Kelly and Sean. From Kelly's point of view, if it is clear that Sean will fail to behave in an environmentally friendly way, then it may be best for Kelly to fail to do so as well. But if Sean is in the same position with respect to her decision as Kelly, then it cannot be taken as given that Sean will not engage in the environmentally friendly behavior, for that is just what she is reasoning about. If there is any point to her reasoning about this, then the environmentally friendly behavior must be accessible to her, contrary to what we assumed when we considered Kelly's decision process. The apparent generalization of the first argument introduces an equivocation that is not implicit in the first argument itself.Footnote 47

The first argument should not be confused with what might be called the Nero objection. This objection states that, just as Nero fiddled while Rome burned, so a utilitarian agent should fiddle (or its functional equivalent) while global environmental change ravages the planet. Since Nero's fiddling was morally horrendous, the functionally equivalent utilitarian fiddling must be morally horrendous as well. However, Nero's fiddling and that of the utilitarian are not equivalent in relevant respects. What is horrendous about the image of Nero fiddling while Rome burns is that he probably set the fires, or could have had them put out. Rather than making the best of a bad situation, he was making a bad situation.Footnote 48 This is clearly forbidden by utilitarianism.

Here is a better account of the first argument. In the domain of global environmental change-relevant behavior, what we want is inflexible green behavior, but even here it should not be too inflexible. Suppose that there is some threshold of cooperation that must be surpassed if global environmental change is to be mitigated. If this threshold will not be surpassed regardless of what I do, then it might be best for me to act in some other way than to exemplify green virtues. But calculating about whether the threshold has been met seems to defeat the advantage of inflexibility that green virtues are supposed to deliver. Moreover, if the calculation delivers the result that I ought to behave in a way that is environmentally destructive, then this seems to contradict the result that we know morality must deliver. It is for reasons such as these that some people think that moving from a focus on actions to a focus on character does not solve collective action problems.

Whether or not the shift of focus from actions to character succeeds in solving the problem depends on exactly what the problem is. If utilitarianism really implied that I should throw tequila bottles out of the window while commuting to work in my SUV, this result would not on the face of it be any more shocking than some other possibilities that utilitarianism can countenance in various hypothetical situations: for example, that in some cases I might be morally obliged to hang innocent people, torture prisoners or carpet-bomb cities. The reason that these objections do not sway anyone with utilitarian sympathies is because, by hypothesis, all of these cases presuppose that my acting in these horrific ways would produce the best possible world.Footnote 49 If the world is in such a deplorable state that hanging innocent people would actually constitute an improvement, that is surely not the fault of utilitarian theory. On the other hand, if the assumption that the contemplated act is optimal is not in play, then the critic is making the ubiquitous error (discussed earlier) of purporting to show that utilitarianism directs agents to act in ways that make the world worse or less good than it could be. As we have seen, utilitarianism can have no such implication.

If the best outcome is truly inaccessible to me, then it is not obviously implausible to suppose that I have a duty to make the best of a bad situation.Footnote 50 When I was a kid, growing up in a neighborhood that would certainly have been a ‘first-strike’ target had there been a nuclear war between the Americans and the Russians, we often seriously discussed the following question. Suppose that you know that They have launched their missiles and that We have retaliated (or vice versa), and that in twenty minutes the planet will be incinerated. What should you do?Footnote 51 The idea that we should enjoy the life that remains to us may not be the only plausible response to this question, but it is surely not an implausible one.

What many people find grating about this answer, I think, is the idea that we have a duty to enjoy life in such a situation. Some might agree that it would be prudentially good to do so, but find it outrageous that morality would be so intrusive, right up to the end of the world. When it comes to the case in which the green cause is hopeless, it might be thought that matters are even worse. It is one thing to say that it is permissible or excusable to abandon our green commitments in such circumstances; it is another thing entirely to say that we have an affirmative duty to join the ranks of the enemy, and to enjoy the very activities that destroy the features of nature that we cherish.Footnote 52

This objection has proceeded under the assumption that we might find ourselves in circumstances in which we know that living according to our green values would be entirely ineffectual, and that we would enjoy helping ourselves to the pleasures of consumerism. On these implausible assumptions, the objector is correct in claiming that utilitarianism would require us to join the side of the environmental despoilers. However, there is nothing really new in principle about this kind of case. It is another example of either the demandingness of utilitarianism, or of how utilitarianism holds our ‘ground projects’ (and therefore our integrity) hostage to circumstances beyond our control.Footnote 53

It is not my task here to defend utilitarianism as anything more than a plausible research program. However, it is surely old news that utilitarianism can require us to break familiar patterns of behavior that are dear to our hearts when doing so would realize what is best. Of course this would be difficult to do, and most of us, most of the time, would not succeed in doing what is right. (No one said that it was easy to be a utilitarian.) But our failures to do what is right would not count against doing what is best as a moral ideal, anymore than the human proclivity for violence should lead us to give up on peace as a cherished moral value. Or so it seems at first glance.

However, the most important point is this. My present concern is not with alternative realities or possible worlds; it is facts about this world that are relevant for present purposes. I am concerned with how a utilitarian agent should respond to the problem of global environmental change that we actually face here and now. Global environmental change is not like the case of an impending interplanetary collision that is entirely beyond our control. Nor is it an ‘all or nothing’ phenomenon. Collectively, we can prevent or mitigate various aspects of global environmental change, and an individual agent can affect collective behavior in several ways. One's behavior in producing and consuming is important for its immediate environmental impacts, and also for the example-setting and role-modeling dimensions of the behavior.Footnote 54 It is a fact of life that one may never know how one's long-term projects will fare, or even how successful one has been in motivating and enlisting other people to pursue them, but this is as much grounds for optimism as pessimism. Nor does an environmentally friendly lifestyle have to be a miserable one.Footnote 55 Even if in the end one's values do not prevail, there is comfort and satisfaction in living in accordance with one's ideals.Footnote 56 All of this taken together suggests that real utilitarian agents here and now should try to prevent or mitigate global environmental change rather than celebrate its arrival.

However, presently there is no algorithm for designing the optimal utilitarian agent.Footnote 57 Nor is there an algorithm for constructing the perfect constitution, which constrains majority rule when it should, but does not prevent its expression when it should not.Footnote 58 Nevertheless, we have better and worse people and constitutions, and sometimes we know them when we see them. It might be nice to have a calculus that we could apply to constitutions and character, but absent this, we can still go forward living our lives and organizing our societies. These responses may not satisfy those who are concerned with the logic of collective action or who believe that every question must admit of a precise answer. But they should go some way towards satisfying those who like me are concerned with the moral psychology of collective action, and are willing to accept Aristotle's view that deliberation can never be completely divorced from practical wisdom.

20. What I have argued thus far is that despite various conundrums and complexities, in the face of global environmental change, utilitarians should be virtue theorists. While it is not my task here to provide a full account of what virtues utilitarians should try to develop and inculcate, I will conclude with a brief, tentative sketch of what might be called the ‘green virtues’.Footnote 59 My goal is not to construct a complete account of the ideal utilitarian moral agent, but only to provide a sample of how we might think about the green virtues that such an agent might exemplify.Footnote 60 There is a modest literature on this subject, and a fair amount of experience with, and reflection on, green lifestyles, on which we can build.Footnote 61

Abstractly we can say that the green virtues are those that utilitarians should try to exemplify in themselves and elicit in others, given the reality of global environmental change. Practically, it seems clear that green virtues should moralize such behavior as reproduction and consumption. As Alan Durning writes,

When most people see a large automobile and think first of the air pollution it causes rather than the social status it conveys, environmental ethics will have arrived. Likewise, when most people see excess packaging, throwaway products, or a new shopping mall and grow angry because they consider them to be crimes against their grandchildren, consumerism will be on the retreat.Footnote 62

21. Green virtues fall into three categories: those that reflect existing values; those that draw on existing values but have additional or some what different content; and those that reflect new values. I call these three strategies of virtue-identification preservation, rehabilitation and creation. I will discuss each in turn, offering tentative examples of green virtues that might fall into these various categories.

Thomas Hill Jr offers an example of preservation.Footnote 63 He argues that the widely shared ideal of humility should lead people to a love of nature. Indifference to nature ‘is likely to reflect either ignorance, self-importance, or a lack of self-acceptance which we must overcome to have proper humility’.Footnote 64 A person who has proper humility would not destroy redwood forests (for example) even if it appears that utility supports this behavior. If what Hill says is correct, humility is a virtue that ought to be preserved by greens.

Temperance may be a good target for the strategy of rehabilitation. Long regarded as one of the four cardinal virtues, temperance is typically associated with the problem of akrasia and the incontinent agent. But temperance also relates more generally to self-restraint and moderation. Temperance could be rehabilitated as a green virtue that emphasizes the importance of reducing consumption.

A candidate for the strategy of creation is a virtue we might call mindfulness. Much of our environmentally destructive behavior is unthinking, even mechanical. In order to improve our behavior we need to appreciate the consequences of our actions that are remote in time and space. A virtuous green would see herself as taking on the moral weight of production and disposal when she purchases an article of clothing (for example). She makes herself responsible for the cultivation of the cotton, the impacts of the dyeing process, the energy costs of the transport, and so on. Making decisions in this way would be encouraged by the recognition of a morally admirable trait that is rarely exemplified and hardly ever noticed in our society.Footnote 65

Although I have been speaking of individual agents and their virtues, it is easy to see that institutions play important roles in enabling virtue. Many of these roles (e.g. inculcation, encouragement) have been widely discussed in the literature on virtue theory. However, it is also important to recognize that how societies and economies are organized can disable as well as enable the development of various virtues. For example, in a globalized economy without informational transparency, it is extremely difficult for an agent to determine the remote effects of her actions, much less take responsibility for them.Footnote 66 Thus, in such a society, it is difficult to develop the virtue of mindfulness.

22. I close by gathering some conclusions. If what I have said is correct, the contrast typically drawn between utilitarianism and virtue theory is overdrawn. Utilitarianism is a universal emulator: it implies that we should lie, cheat, steal, even appropriate Aristotle, when that is what brings about the best outcomes. In some cases and in some worlds it is best for us to focus as precisely as possible on individual acts. In other cases and worlds it is best for us to be concerned with character traits. Global environmental change leads to concerns about character because the best results will be produced by generally uncoupling my behavior from that of others. Thus, in this case and in this world, utilitarians should be virtue theorists.Footnote 67

The central morals of this article are these. Philosophically, we should ask when, not whether, utilitarians should be virtue theorists. Practically, we need to develop a catalog of the green virtues and identify methods for how best to inculcate them. Some may consider this an ‘obsession’ produced by allegiance to a particular moral theory, but to my mind this is not too much to ask of those who are philosophizing while human beings are bringing about the most profound transformation of Earth to occur in fifty million years.Footnote 68

References

1 While ‘global environmental change’ may seem a clumsy or misleading expression, it has come to be the standard way of referring to this cluster of problems in the scientific and policy literatures; see e.g. the website for The Encyclopedia of Global Environmental Change (http://www.wiley.co.uk/wileychi/egec/). For an overview of these problems see The World Resources Institute, The United Nations Environment Programme and The World Bank, World Resources 2000–2001 (New York, 2000), also available on the web at http://wristore.com/worres20.html.

2 Some would modify this list of the reigning moral theories by adding or substituting contractualism or virtue ethics.

3 Creating the Kingdom of Ends (New York, 1996), p. 300.

4 Korsgaard, Creating, p. 275. Cf. Annette Baier who thinks that contemporary moral philosophers have not yet escaped the clutches of Kant (Postures of the Mind (Minneapolis, 1985), p. 235).

5 However Korsgaard does briefly discuss the moral status of plants and animals in The Sources of Normativity (New York, 1996), ch. 4, and she extensively discusses Kantian views of animals in her University of Michigan Tanner Lecture, ‘Fellow Creatures: Kantian Ethics and Our Duties to Animals’, Tanner Lectures on Human Values, vol. 25, ed. Grethe B. Peterson (Salt Lake City: University of Utah Press, 2005), pp. 77--110.

6 See for example the work of Onora O'Neill collected in her Constructions of Reason: Explorations of Kant's Practical Philosophy (New York, 1989). Korsgaard tries to overcome the interiority of the theory by focusing on ‘how we should relate to one another’ as the subject matter of morality (Creating, p. 275).

7 There are interpretations of Kant, perhaps most notably that of R. M. Hare (see e.g. Freedom and Reason (Oxford, 1965)), which emphasize the idea of universalizability and de-emphasize the notion of the good will. This is not the reading of Kant with which I am concerned here, in part because it has become less influential in recent years, but also because (at least in this respect) it blurs the distinction between Kantianism and utilitarianism.

8 This is quite clear in the work of David Gauthier and Jan Narveson, for example. For an early discussion of these problems see my ‘Rational Egoism and Animal Rights’, Environmental Ethics 3 (1981).

9 Although there are many differences and disagreements among them, and some would reject the charge of conservatism, I associate this view with British philosophers such as Jonathan Dancy and Stuart Hampshire, and American philosophers such as Susan Wolf.

10 Anti-revisionists come in different stripes, but for one version see the introduction to Judith Jarvis Thomson, The Realm of Rights (Cambridge, 1990); on the second point, see Susan, Wolf, ‘Moral Saints’, Journal of Philosophy 79 (1982), esp. p. 422Google Scholar. For a utilitarian response to such claims, see Peter Singer, How Are We to Live? Ethics in an Age of Self-Interest (Buffalo, 1995).

11 On the environmental consequences of American reproductive behavior, see Charles, A. S. Hall, Gil Pontius, R. Jr, Lisa, Coleman and Jae-Young, Ko, ‘The Environmental Consequences of Having a Baby in the United States’, Wild Earth 5 (1995)Google Scholar.

12 There is room for drawing various subtle distinctions here. Jürgen Habermas claims that ‘[h]uman responsibility for plants and for preservation of whole species cannot be derived from duties of interaction, and thus cannot be morally [sic] justified’, but goes on to say that ‘there are good ethical reasons [sic] that speak in favor of the protection of plants and species’. See his Justification and Application: Remarks on Discourse Ethics, trans. Ciaran Cronin (Cambridge, 1993), p. 111.

13 For further discussion of deontology and the role of intentions in shaping moral constraints, see Nancy (Ann) Davis, ‘Contemporary Deontology’, Companion to Ethics, ed. Peter Singer (Oxford, 1991), and the references cited therein.

14 I hope it is clear that my intention thus far has been only to show that, on a first approximation, in comparison with its rivals, utilitarianism appears well positioned to address the problem, and in this regard is worthy of detailed investigation. I do not mean to suggest that alternative approaches, however resourceful, are totally incapable of providing interesting responses to our problem.

15 For such criticisms see Callicott, J. Baird, ‘Animal Liberation: A Triangular Affair’, Environmental Ethics 2 (1980)Google Scholar; Holmes Rolston III, ‘Respect for Life: Counting what Singer Finds of no Account’, Singer and his Critics, ed. Dale Jamieson (Oxford, 1999); Eric Katz, Nature as Subject (Lanham, 1997); John, Rodman, ‘The Liberation of Nature’, Inquiry 20 (1977)Google Scholar; and Mark, Sagoff, ‘Animal Liberation and Environmental Ethics: Bad Marriage, Quick Divorce’, Osgood Hall Law Journal 22 (1984)Google Scholar.

16 Cf. Korsgaard, who insightfully writes that ‘[u]sually the “standard objections” that one school of thought raises against another are question-begging in deep and disguised ways’ (Creating, p. xiii).

17 In characterizing utilitarianism in this way, I chime with Liam Murphy (Moral Demands in Nonideal Theory (New York, 2000), p. 6) rather than with Shelly Kagan who uses the term ‘consequentialism’ for what I call utilitarianism; see his discussion in Normative Ethics (Boulder, 1998). For further discussion of these terms, see my ‘Consequentialism’, in ‘Ethics and Values’, Encyclopedia of Life Support Systems (EOLSS), ed. R. Elliot, developed under the auspices of the UNESCO (Oxford, 2002), available on the web at http://www.eolss.net. See also my Ethics and the Environment: An introduction (Cambridge, in press), ch. 4.

18 Indeed it may not even be the first step. Utilitarianism may imply that utilitarianism should be an ‘esoteric morality’. Whether or not it has this implication depends on facts about particular people and societies. For discussion of esoteric morality see Henry Sidgwick, The Methods of Ethics, 7th edn. (London, 1907), p. 490; and Derek Parfit, Reasons and Persons (New York, 1984), pt. 1 (esp. ch. 1).

19 For the most recent, authoritative and systematic account of the consequences of climate change, see Climate Change 2001: Impacts, Adaptation & Vulnerability, ed. James J. McCarthy, Osvaldo F. Canziani, Neil A. Leary, David J. Dokken and Kasey S. White (New York, 2001), and the updates found on the web at http://www.ipcc.ch/. See also my ‘The Epistemology of Climate Change: Some Morals for Managers’, Society and Natural Resources 4 (1991).

20 On this general issue see Jonathan Glover, ‘“It Makes No Difference Whether or Not I Do It”’, Applied Ethics, ed. Peter Singer (New York, 1986); and Parfit, Reasons, ch. 3.

21 It should be obvious that I am using ‘hypocrisy’ and ‘asceticism’ as technical terms; a full-blooded analysis of these concepts would reveal richer and more subtle conditions for application than what is suggested by the text.

22 Since such a strategy may well involve the construction and inculcation of norms, I believe that nothing I say here is inconsistent with Philip Pettit's discussion of norms as responses to collective action problems in part III of his Rules, Reasons, and Norms (Oxford, 2002). One way of relating our accounts would be to say that the account that I develop is a (relatively) thick description of what utilitarian agents would have to be like in order for relevant norms to emerge and to reduce their own contributions to the problem. Although my focus is primarily on individual agents, the argument generalizes to all similarly situated utilitarian agents. Moreover, I believe that the importance of individual agents in addressing collective action problems is not fully appreciated by many theorists (see sect. 19 for further discussion).

23 I discuss this objection further in sect. 19.

24 For further argument to this conclusion see Donald Regan, Utilitarianism and Cooperation (New York, 1980).

25 While the virtues, as I understand them here, are non-calculative generators of behavior, their exercise does not exclude deliberation. I am indebted to Steve Gardiner and Jerrold Katz for helpful discussion of these points.

26 For some important discussions of these points see Regan, Utilitarianism; Allan, F. Gibbard, ‘Rule Utilitarianism: Merely an Illusory Alternative?’, Australasian Journal of Philosophy 43 (1965)Google Scholar; Robert, M. Adams, ‘Motive Utilitarianism’, Journal of Philosophy 73 (1976)Google Scholar; and Peter, Railton, ‘Alienation, Consequentialism, and the Demands of Morality’, Philosophy and Public Affairs 13 (1984)Google Scholar.

27 Some may feel the pull of this example, but find it out of the question that a life without friends could be utility-maximizing. But if we assume that utility-maximizing behavior is frequently associated with acting on agent-neutral reasons, then it is not difficult to see why strong personal relationships might lead us to act in less than optimific ways. Of course, even if this is true there is no question that many of us here and now would do worse by abandoning our friends and setting ourselves up as rootless cosmopolitan utility-maximizers. For a recent discussion of some of these issues, see Elizabeth, Ashford, ‘Utilitarianism, Integrity, and Partiality’, The Journal of Philosophy 97 (2000)Google Scholar.

28 My quarrel here is not with those who have distinguished act- from rule-utilitarianism as part of an investigation of the varieties of utilitarianism, but rather with the way in which this distinction has subsequently been canonized and then read back into the tradition. For an excellent study in the former spirit see David Lyons, Forms and Limits of Utilitarianism (Oxford, 1965).

29 For a contrary view see Henry R. West, An Introduction to Mill's Utilitarian Ethics (New York, 2004). But see also Fred Berger, Happiness, Justice, and Freedom: The Moral and Political Philosophy of John Stuart Mill (Berkeley, 1984).

30 Cf. Michael Slote's discussion of Bentham in ‘Utilitarian Virtue’, Midwest Studies in Philosophy Volume XIII Ethical Theory: Character and Virtue, ed. P. French, T. Uehling Jr and H. Wettstein (Notre Dame, 1988).

31 Onora O'Neill has written insightfully about this in the context of Kantian ethics (Constructions, ch. 9). See also Stanley Cavell, The Claim of Reason (New York, 1979), pp. 263–7.

32 In unpublished work I have tried to develop a perspective on the purposes of moral theorizing that I believe are implicit in the tradition of consequentialist moral philosophy. I discuss these ideas under the rubric ‘naturalized moral theory’. For the beginnings of such an account see my ‘Method and Moral Theory’, Companion, ed. P. Singer.

33 This distinction between global and local utilitarianism derives from the felicitous distinction between global and local consequentialism drawn by Philip Pettit and Michael Smith, who argue persuasively for the superiority of the global view in their ‘Global Consequentialism’, in Morality, Rules, and Consequences: A Critical Reader, ed. B. Hooker, E. Mason and D. Miller (Edinburgh, 2000). See also Shelly Kagan's ‘Evaluative Focal Points’ in the same volume.

34 Parfit, Reasons, p. 25.

35 For the first view see Christopher Kutz, Complicity: Ethics and Law for a Collective Age (New York, 2000), p. 11; for the second, Murphy, Moral Demands, p. 96 (note, however, that Murphy's remark is in the context of a larger investigation of an individual's moral duty of beneficence under conditions of partial compliance). Other approaches to collective or shared intentions advocate revising our conceptions of agents or of intending, rather than focusing on the content of intentions. For example, John Searle holds that jointly intentional action can only be explained by postulating an irreducible form of intending that he calls ‘we-intending’ (in his Intentionality (Cambridge, 1983), ch. 3); for discussion see Kutz, Complicity, ch. 3.

36 Christopher McMahon (in his Collective Rationality and Collective Reasoning (New York, 2001)) tells us that the solution to prisoners’ dilemmas (a class of problems closely related to our problem) is to treat them as pure coordination problems. However, in prisoners’ dilemmas each agent is better off defecting whatever other agents do while this is not the case in pure coordination problems. Since prisoners’ dilemmas have a different structure than pure coordination problems, clear, convincing motivation is needed for why we should view them in the way that McMahon suggests, and some account needs to be provided of what agents would have to be like in order to act in the preferred way. In the absence of such accounts, this gambit seems merely to change the subject. For further discussion, see Gerald Gaus, ‘Once More Unto the Breach, My Dear Friends, Once More’, Philosophical Studies 116 (2003); and Michael Weber, ‘The Reason to Contribute to Cooperative Schemes’, in the same issue. My brief remarks in this paragraph are not meant to minimize the contributions of McMahon, Kutz and others, but only to suggest that more detailed work needs to be done.

37 Julia, Driver, Uneasy Virtue (New York, 2001), p. 108Google Scholar.

38 Here I agree with Rosalind, Hursthouse, On Virtue Ethics (Oxford, 1999)Google Scholar, pt. 2. Driver also discusses the relations between the virtues and the emotions, but I am not clear what her considered view is on this matter.

39 Cf. Robert Frank who has argued that emotions promote self-interest by solving commitment problems, in his Passions within Reason (New York, 1988).

40 However, not all non-calculative generators of action count as virtues. Some are too trivial, others are vices, and still others would be too far from the traditional notion of a virtue even for me to call virtues.

41 James Griffin, Value Judgement: Improving Our Ethical Beliefs (New York, 1996), p. 113; see also Michael Slote, From Morality to Virtue (New York, 1992). For a more relaxed view about what counts as virtue ethics see Julia Annas, The Morality of Happiness (New York, 1993).

42 An objection to virtue theory that is beginning to gain currency draws on results from social psychology that show that contextual factors are stronger predictors of behavior than facts about individual character. For such objections, see Gilbert Harman, ‘Moral Philosophy Meets Social Psychology: Virtue Ethics and the Fundamental Attribution Error’, reprinted in his Explaining Value and Other Essays in Moral Philosophy (Oxford, 2000); and John, Doris, ‘Persons, Situations, and Virtue Ethics’, NOUS 32 (1998)Google Scholar, and his Lack of Character: Personality and Moral Behavior (New York, 2002). Because I am not committed to any particular account of the virtues, much less to one that makes them radically internal to agents rather than relative to contexts, I do not believe that this objection threatens the claims that I advance here.

43 Bernard Williams fastens onto a somewhat similar point in his critique of Hare's ‘two-level’ theory (see his ‘The Structure of Hare's Theory’, Hare and Critics, ed. D. Seanor and N. Fotion (Oxford, 1988)). But while Williams emphasizes the psychological untenability of living simultaneously at both the ‘intuitive’ and ‘critical’ levels, my criticism is specifically aimed at someone who rests content with rules of thumb when she is committed to the view that morality requires her to do what is best.

44 There is quite a lot more to be said about progressive consequentialism. I say a little more in ‘Consequentialism’, and Robert Elliot discusses this view under the rubric ‘improving Consequentialism’ in his Faking Nature (New York, 1997).

45 Let us assume that in this case the benefits and harms do not cross domains: the benefits of Peter attending the meeting attach only to famine relief and the harms of my driving are confined to their contribution to global environmental change.

46 Such problems are much discussed in the economics literature under the rubric of ‘optimal stopping rules’. See, for example, G. J. Stigler's classic, ‘The Economics of Information’, Journal of Political Economy 69 (1961).

47 There are ways of trying to revive the second argument by casting it in probabilistic terms, but I cannot consider that possibility here. My understanding of a range of such cases has benefited greatly from discussions with Scott (Drew) Schroeder.

48 The locus classicus for this image of Nero is Gibbon, but recent scholarship suggests that Nero has been maligned: that he neither set the fires, nor was indifferent to the destruction they caused. See Miriam T. Griffin, Nero: The End of a Dynasty (London, 1984).

49 R. M. Hare makes a similar argument with respect to slavery; see his ‘What is Wrong with Slavery’, Philosophy and Public Affairs 8 (1979).

50 Here I break with Christopher Kutz (Complicity, pp. 124–32) who rejects what he calls ‘consequentialism’ for failing to explain why it is wrong to participate in a bad practice whose occurrence is overdetermined. For an alternative view to Kutz's, see Frank Jackson, ‘Group Morality’, Metaphysics and Morality: Essays in Honour of J. J. C. Smart, ed. Philip Pettit, Richard Sylvan and Jean Norman (Oxford, 1987). Intuitions about overdetermination cases seem to run in different ways, depending on particular cases and how they are described; a full treatment of this problem is beyond the aspirations of this article. I have benefited here from reading unpublished work by Dan Moller.

51 This question is similar to one many of us may face in our future (or, arguably, face now): what should you do knowing that, in some specified amount of time, you will surely die? And, of course, we should not be too confident that the question from my youth may not yet again become relevant.

52 This objection echoes a remark of C. S. Lewis to the effect that if one is about to be swept over a waterfall one does not have to sing praises to the river gods.

53 This latter objection to utilitarianism was a constant theme in the work of Bernard Williams and has stimulated an enormous literature. To begin at the beginning with the famous case of Jim and the Indians, see his ‘A Critique of Utilitarianism’, in J. J. C. Smart and Bernard Williams, Utilitarianism: For and Against (Cambridge, 1973). For an unusually insightful discussion of the ‘demandingness’ objection see Murphy, Moral Demands, chs. 2–3.

54 See Ziva Kunda, Social Cognition: Making Sense of People (Cambridge, 1999), pp. 501–6.

55 Contrary to what one might think reading the newspapers, relationships between subjective reports of well-being and economic measures (such as per capita GDP) are equivocal and complex. An easy way into these issues is through the home page of Ed Diener, one of the leading researchers in the study of subjective well-being (http://www.psych.uiuc.edu/~ediener/).

56 One way of developing this thought in a decision-theoretic context would be to follow Alexander Schuessler (in his A Logic of Expressive Choice (Princeton, 2000)) in distinguishing the ‘expressive’ from the ‘outcome’ value of a choice. This distinction may also help explain our intuitions in cases of overdetermined harms (mentioned in n. 50). The deepest general philosophical discussion of these issues that I know is Thomas, Hill Jr, ‘Symbolic Protest and Calculated Silence’, Philosophy and Public Affairs 9 (1979)Google Scholar. However, Hill focuses mostly on obviously malevolent acts and practices rather than the apparently innocent ones implicated in global environmental change.

57 David Lyons discusses a similar point when he talks about the ‘moral opacity’ and ‘moral ambiguity’ of utilitarianism (in ‘The Moral Opacity of Utilitarianism’, Morality, ed. Hooker et al.), though I'm not certain exactly what conclusion he wants to draw from his discussion.

58 Jon Elster has extensively discussed the analogy between individual and collective pre-commitment and restraint, most recently in his Ulysses Unbound (New York, 2000).

59 James Griffin points out (Value Judgement, p. 106), that the problem of calculation returns here to haunt us, since in order to identify virtues it appears that we need to be able to determine exactly which character traits are utility-promoting. To some extent this is a problem that will have to be faced by any theory that takes both character and consequences seriously.

60 A full account of the ideal utilitarian agent facing our problem would have to find a place for vices as well, as I was reminded by Corliss Swain. Indeed, it is plausible to suppose that vices such as greed would be as important in explaining and motivating behavior as the virtues that I mention here.

61 For a start on the literature of green virtue theory, see Ronald L. Sandler, Character and the Environment: A Virtue-Oriented Approach to Environmental Ethics (New York, 2007); Ronald Sandler and Philip Cafaro (eds.), Environmental Virtue Ethics (New York, 2005); and Louke van Wensveen, Dirty Virtues: The Emergence of Ecological Virtue Ethics (Amherst, 1999).

62 Alan Durning, How Much Is Enough? The Consumer Society and the Future of the Earth (New York, 1992), p. 138.

63 In his ‘Ideals of Human Excellence and Preserving the Natural Environment’, Reflecting on Nature: Readings in Environmental Philosophy, ed. Lori Gruen and Dale Jamieson (New York, 1994).

64 Hill, ‘Ideals’, p. 108.

65 Cooperativeness would be another important characteristic of agents who could successfully address our problem (as well as collective action problems generally). Surprisingly, this characteristic appears to be neglected by both ancient and modern writers on the virtues (Hume may be an exception). Perhaps a virtue of cooperativeness is a candidate for creation, or perhaps, though not itself a virtue, cooperativeness would be expressed by those who have a particular constellation of virtues. For discussion of the importance of cooperativeness to morality, see Robert A. Hinde, Why Good is Good: The Sources of Morality (London, 2002).

66 There is a growing literature on this topic. See, for example, David C. Korten, When Corporations Rule the World (West Hartford, 1995).

67 Roger Crisp reaches a similar conclusion in ‘Utilitarianism and the Life of Virtue’, Philosophical Quarterly 42 (1992).

68 Earlier versions of this article were presented at the Utilitarianism Reconsidered conference in New Orleans LA; the Department of Philosophy at Edinburgh University; the Sub-faculty of Philosophy at the University of Oxford; the Center for Values and Social Policy at the University of Colorado; the Australasian Association of Philosophy meeting in Sydney; the International Conference on Applied Ethics at the Chinese University of Hong Kong; the Department of Philosophy at the University of Wisconsin, Madison; the Minnesota Monthly Moral Philosophy Meeting; the Philosophy Program at the Graduate Center of the City University of New York; and the Department of Philosophy at Yale University. I am deeply grateful for all of the interesting discussion provided by these audiences. I thank especially David Copp, Roger Crisp and James Griffin for helpful comments. The origin of this article goes back many years to a conversation with Barbara Herman about the scope and domain of morality; while nothing I say here will settle the differences between us that were expressed that afternoon, I want to thank her for causing me to think so long and hard about this problem.