Introduction

Many significant environmental problems are longitudinal collective action environmental problems. They arise from the cumulative unintended (and often unforeseen or unforeseeable) effects of a vast amount of seemingly insignificant decisions and actions by individuals who are unknown to each other and distant from each other (spatially, temporally, and socially).Footnote 1 Global climate change is perhaps the paradigmatic longitudinal collective action environmental problem. However, many other environmental issues, such as pollution, fisheries depletion, human population growth, deforestation, and ozone layer degradation, exhibit these features as well. It takes quite a lot of people, almost all of them acting without environmental malice (though often with environmental ignorance or indifference), to precipitate the environmental problem.Footnote 2

Longitudinal collective action environmental problems are likely to be effectively addressed only by an enormous number of individuals each making a nearly insignificant contribution to resolving them. However, when a person’s making such a contribution appears to require social, personal, or economic costs the problem of inconsequentialism arises: given that a person’s contribution, although needed (albeit not necessary), is nearly inconsequential to addressing the problem and may require some cost from the standpoint of the person’s own life, why should the person make the effort, particularly when it is uncertain (or even unlikely) whether others will do so?

A common response to the problem of inconsequentialism is to emphasize the need for political solutions to longitudinal collective action problems and, therefore, the need for political activism on the part of individual agents, i.e., acting in support of developing adequate political arrangements and sufficient compliance incentives to address the problems. However, even if it is true that longitudinal collective action environmental problems are likely to be resolved only through political arrangements and that acting in support of developing these is justified, this does not adequately address the problem of inconsequentialism. Consider, for example, global climate change. The problem of inconsequentialism arises regarding activism in support of political arrangements and institutions to address global climate change, since the activities of the vast majority of us are inconsequential to bringing about the national and international agreements and activities necessary to address the problem. Moreover, that one ought to actively support aggressive national greenhouse gas emission reduction policies (or other systematic changes related to fossil fuel consumption) does not tell a person what she ought to do with respect to reducing her own emissions, particularly in the absence of such policies. In addition, even if there is a political arrangement in place, the problem of inconsequentialism still arises if the costs of non-compliance are lower than the costs of compliance (or the benefits are greater than those of compliance). Thus, the problem of inconsequentialism arises when there is no political arrangement in place to address a longitudinal collective action environmental problem, there is such an arrangement but it is inadequate to address the problem, or there is an adequate arrangement but the costs to the agent of contributing are greater than those of not complying (even given the arrangement’s mandates, sanctions, and incentives to encourage participation); and appeals to the need for political activism (or promoting systematic change) do not address what, if anything, one ought to do with respect to the problem itself.

The problem of inconsequentialism is old news in environmental ethics, and several types of responses to it have been offered. Appeal to agent benefit responses attempt to show that making an effort to address longitudinal collective action environmental problems are not as costly to the agent as might initially appear, e.g., that making the effort to address the problem is itself beneficial to the person or that, while there are some initial costs, making the effort to address the problem is beneficial to the person in the long run.

Appeal to duty responses attempt to shift the focus of assessment from the costs and benefits involved in acting to the acting itself, e.g., that a person is required to do their duty with respect to the problem, regardless of whether others do theirs; that a person is required to act in ways that would resolve the problem if others were to act likewise; or that a person is required to do their fair share to address the problem, no matter how small the share.

Appeal to amplifying effects responses attempt to show that an individual’s contributions to addressing longitudinal collective action problems are not as inconsequential as they might initially appear, e.g., that in making the effort, a person sets an example that could lead to more widespread participation to address the problem; that the sum of a person’s actions over her lifetime is not inconsequential to the problem; that if systematic change to address the problem is to occur, it must be preceded, precipitated, and lead by committed individuals.

Appeal to character responses attempt to shift the focus of assessment from the costs and benefits involved in acting to the agent herself, e.g., that making the effort to address the problem is a matter of personal integrity; that to fail to make the effort to address the problem is to be complicit with those who do not; or that to fail to make the effort to address the problem is to be indifferent or insensitive to the problem.

In this article I argue that this array of justifications for making the effort to respond to longitudinal collective action environmental problems is, on the whole, particularly well supported by virtue-oriented normative theories, on which character traits are evaluated as virtues and vices consequentially or teleologically and actions are evaluated in terms of virtues and vices. The broader theoretical context or meta-normative underpinnings of such a normative theory could be virtue ethical (e.g., neo-Aristotelian) or utilitarian (e.g., virtue utilitarian). If ethical theories are to be assessed on their theoretical and practical adequacy, and if providing a compelling response to the problem of inconsequentialism is an instance of such adequacy, then this is a reason for preferring virtue-oriented ethical theory over non-virtue-oriented Kantian and utilitarian theories (e.g., act utilitarianism and global utilitarianism).

Here, then, is a summary of this article’s central argument:

  1. (1)

    Many significant environmental problems are longitudinal collective action problems that give rise to the problem of inconsequentialism.

  2. (2)

    An adequate environmental ethic will justify acting to address significant environmental problems.

  3. (3)

    Therefore, an adequate environmental ethic will provide a cogent response to the problem of inconsequentialism—i.e., it will explain why a person ought to make the effort (or take on the costs) to make a (even insignificant) contribution to addressing longitudinal collective action environmental problems.

  4. (4)

    Virtue-oriented ethical theory supports well several types of responses to the problem of inconsequentialism, whereas non-virtue-oriented forms of utilitarianism and Kantianism do not.

  5. (5)

    Therefore, virtue-oriented ethical theory is preferable to those theories (and environmental ethicists should be virtue-oriented ethicists), all other things being equal.

Thus far I have attempted to motivate 1–3. The remainder of the article will focus on premise 4 and the inference to 5. I begin by identifying the shortcomings of act utilitarian and Kantian ethics with respect to longitudinal collective action problems, and use these to help articulate what the structure of a normative ethic well resourced for addressing the problem of inconsequentialism would be like. I then argue that virtue-oriented ethical theory exhibits that structure. Finally, I discuss two utilitarian alternatives to virtue-oriented ethical theory, emulator utilitarianism, and global utilitarianism.

Act Utilitarianism

According to act utilitarianism, an action is right to the extent that it brings about the best (or good enough) consequences of all the courses of action available to the agent.Footnote 3 Act utilitarian reasoning provides the clearest formulation of the problem of inconsequentialism. Almost any action performed by almost any agent will have a vanishingly small effect on longitudinal collective action environmental problems.Footnote 4 Many of these same actions will have burdens for the agent and those close to her (i.e., family and friends), in terms of, for example, time, economic costs, social costs, and professional costs. In such cases the local utility of actions that contribute to longitudinal collective action problems (or that fail to address the problems) will outweigh the inconsequential global utility of those actions. Therefore, act utilitarianism cannot explain why we ought to act or live in ways responsive to longitudinal collective action problems when doing so has costs or sacrifices associated with it. This suggests two features that a normative theory well resourced for addressing the problem of inconsequentialism will exhibit:

  1. (1)

    That discrete actions are not evaluated entirely on the basis of the outcomes (or likely outcomes) of the action.

  2. (2)

    That it evaluate on the basis of patterns of behavior or activities throughout a person’s life or patterns among people or communities.

Advocates of act utilitarianism might respond to the claim that act utilitarian reasoning gives rise to the problem of inconsequentialism by arguing that it trades on a mistaken version of the theory. It is not the consequences forecasted by agents that are relevant to the rightness of actions, but the action’s actual consequences (or, perhaps, what they are in fact likely to be). Therefore, it is open to act utilitarians to argue that while many agents might believe that their actions to address longitudinal collective action environmental problems have high local costs (and low local and global benefits) often this is not the case. That is, they might offer versions of appeal to agent benefit or amplifying effects responses. However, both these types of responses have significant limitations in the context of act utilitarian theory. For appeal to agent benefits responses, the key limit is that the local (dis)utilities associated with the effort made by an individual to address some environmental problem are likely to be related to her particular perspectives. A person who does not care much about the environmental problem is not likely to feel good (a local utility) about having made an effort to address the problem. A materialistic person who lives a high consumption lifestyle is more likely to feel constrained (a local disutility) by lower levels of consumption. Act utilitarianism has difficulty overcoming this limitation because it does not evaluate the appropriateness of an agent’s perspectives or have such evaluations inform evaluations of the rightness of action. This suggests that a normative theory resourced for supporting appeal to agent benefit responses to the problem of inconsequentialism will have the following features:

  1. (3)

    That its evaluative focal points (i.e., what it directly evaluates) include attitudes or perspectives of people.

  2. (4)

    That evaluation of attitudes and perspectives bear on evaluation of actions.

Appeal to amplifying effects responses, e.g., that by moving others to act or setting a habit/trajectory in one’s own life a person’s actions are not really inconsequential to longitudinal collective action environmental problems—are also limited within act utilitarianism. With respect to effects on others, it often will not be the case that what a single person does in a single act will impact the actions of others or be crucial to bringing about systematic reform, particularly in situations where others are not already acting responsively or motivated to do so. With respect to the cumulative effects over a person’s lifetime, even a person’s entire lifetime impacts on any longitudinal collective action problem is likely to still be inconsequential. Moreover, refocusing evaluation from individual acts to lifetimes requires moving to a non-act version of utilitarianism.

In addition, and more important for present purposes, act utilitarian appeals to amplifying effects responses fail to explain why one should make the effort to address the problem even if others are not (or are not likely) to do so. Act utilitarianism, since it evaluates discreet actions indexed to a particular time, place, and context, is not able to decouple the rightness of action from what others are doing (or are likely to do). If act evaluation is not indexed to the actual world or possible worlds relevantly similar to the actual world, act utilitarianism sacrifices its intuitive plausibility, which rests on the idea that the right action is the one that would bring about the best consequences of those available to the agent at this time and place. It is for this reason, i.e., that rightness of action is tied to what others do or are likely to do, that act utilitarianism does not support well several of the duty-oriented responses to the problem of inconsequentialism, e.g., that one should do one’s fair share even if others do not do theirs. On act utilitarianism, what one agent ought to do often heavily depends on what other agents are doing (or are likely to do). This limitation suggests an additional feature that a normative theory well resourced for addressing the problem of inconsequentialism will have:

  1. (5)

    That evaluation of an agent’s actions is not overly contingent on the actions of others.

There are three primary ways that an advocate of act utilitarianism might respond to the set of criticisms presented above. First, she might argue that it is not a criterion of adequacy for an environmental ethic that it provides normative resources to respond effectively to the problem of inconsequentialism.Footnote 5 Second, she might argue that some or all of the features identified in this section as conducive to an ethical theory’s capacity to respond to the problem of inconsequentialism are not needed. Third, she might argue that some version of act utilitarianism can exhibit some or all of the features. I have already addressed the first and second of these responses by providing reasons for believing it is an adequacy criterion and that the features identified are crucial to meeting it. With respect to the third response, it may be that a sufficiently sophisticated version of act utilitarianism can exhibit the features. However, such an act utilitarianism would be (in the terminology of this article) a virtue-oriented form of act utilitarianism. Therefore, it would not belie the central argument of this article, which aims to show that virtue-oriented ethical theory, regardless of the basic theoretical construct out of which it is developed, is particularly well resourced for responding to the problem of inconsequentialism. The target of the arguments in this section is non-virtue-oriented forms of act utilitarianism.

Kantian Ethics

Kantian ethics appears less susceptible to the problem of inconsequentialism, since it evaluates actions on the basis of their maxims (i.e., principles from which they are performed or that they express), rather than on the basis of what they bring about. The fact that an agent’s actions might be inconsequential in resolving (or creating) an environmental problem is, therefore, not material to the evaluation of the action. It is one’s duty to act on appropriate maxims, those that conform to the moral law, i.e., that can be consistently willed to be universal law and that do not treat something with inherent worth as a means only and not an end in itself, without regard to the consequences (or lack thereof) of an act (Kant 1997). In this way, Kantian ethics appears to reject the sort of reasoning that gives rise to the problem of inconsequentialism.

However, Kantian ethics struggles with a different dimension of longitudinal collective action environmental problems: that the problems often result from unintended byproducts of activities, where those byproducts are not a necessary means to the ends sought. Consider, for example, global warming. The continuing build up of greenhouse gases in the atmosphere is not sought by those whose actions contribute to it, nor is it a means to any end that is sought. Carbon dioxide emissions are not a means to producing energy; it is a byproduct of the fossil fuel energy production process. If the process did not produce carbon dioxide, it would make no difference to the point of the process or to whether its aims are achieved. Moreover, when a person contributes, through her consumption, to the amount of greenhouse gases in the atmosphere, those who are affected by the resultant climactic and environmental changes do not stand as a means to the consumer’s ends, e.g., health, convenience, comfort, recreation, or status. If the emissions did not occur, did not contribute to global climate change, or did not result in ecological and social problems, the consumptive actions and practices would still have the same intended ends and means of accomplishing them.

Because longitudinal collective action environmental problems are a by-product of activities that have other intended ends and do not stand as means to those ends, Kantian ethics has difficulty explaining what is wrong with the actions or practices that give rise to them. If one buys a terribly inefficient vehicle to take the kids to soccer practice, to raise one’s status, or just because one likes driving it there is no person (or animal or plant, for that matter) treated as a mere means in one’s doing so. It does not violate or offend the value of other human beings (or animals or plants). There is no contradiction in willing the maxim as one appropriate for anyone (or everyone) else. The same is true of having children and population growth. The ecological challenges associated with global population increases are not a sought end or a means to any sought end of those who choose to have children.Footnote 6 Therefore, those ecological challenges do not inform the maxim (or principle) that is evaluated. This suggests a crucial feature that a normative theory well resourced to respond to the problem of inconsequentialism will exhibit:

  1. (6)

    That evaluation of actions is sensitive to the significance of consequences, even when those consequences are unintended byproducts of actions.

One way that a Kantian might attempt to incorporate this feature is by focusing on the possibility of moral agency itself. That is, a Kantian might emphasize that longitudinal collective action problems can undermine necessary conditions of moral agency, e.g., material goods, such as potable water and food, needed for adequate cognitive functioning (Westra 2005). Therefore, acting on a maxim that if universally adopted gives rise to them violates the moral law, since it renders acting on maxims itself impossible or involves disregarding the value of rational agency. However, while the environmental problems associated with global warming (and other longitudinal collective action environmental problems) are likely to be severe, they will not make moral agency (particularly for those who are highly consumptive, and so well resourced to adapt) impossible. Nor will they make the behaviors that gave rise to the problem, e.g., driving fuel inefficient vehicles, impossible. So, with respect to most (if not all) actions that contribute to longitudinal collective action environmental problems, universalizing their maxims would not result in undermining moral agency or otherwise disregarding the value of rational agency.

An advocate of Kantian ethics might concede that it is difficult to establish a perfect duty to refrain from contributing to collective action environmental problems, but argue that Kantian ethics can generate at least imperfect or meritorious duties not to contribute to (and to address) them. One way this might be done is by focusing on the ways in which they rely on a limited resource base. However, if the finitude of resources needed for performing an action are grounds for not being universalizable, e.g., that a maxim to travel in whatever vehicle one pleases, including those that are inefficient and run only on fossil fuels, cannot be universalized because there is only a finite amount of fossil fuels—then Kantian ethics will be reduced to absurdity. It would mean that farming violates the moral law due to the finitude of farm land and being a practicing doctor violates the moral law due to the finitude of patients. So this reply succeeds in establishing a duty to respond to longitudinal collective action environmental problems at the cost of rendering Kantian ethics non-viable.

Another way a Kantian might seek to establish an imperfect duty to respond to longitudinal collective action environmental problems is by appeal to the duty to be benevolent. On Barbara Herman’s interpretation of Kantian benevolence, “What one may not do is fail to give aid when true needs are threatened for reasons of mere self-interest” (Herman 1993, p. 156). If not giving aid for reasons of self interest is impermissible, contributing to a threat to others for reasons of self interest would seem a fortiori impermissible. Thus, failing to reduce one’s greenhouse gas emissions, and thereby contributing to the problems associated with global warming, for reasons of self-interest would appear to violate the Kantian duty of benevolence. However, this response is limited for several reasons. First, even on Herman’s view, a person does not fail to act benevolently if she is not capable of providing the necessary aid. As Herman puts it, “Still one may fully acknowledge the claim of need but be in no position to help because one does not possess what is needed” (Herman 1993, p. 156). This is the situation with respect to longitudinal collective action environmental problems. For the vast majority of us, our actions are inconsequential to the problem. We do not “possess what is needed.” Second, if we must only sometimes act responsively, as is the case with imperfect duties, it is not likely that longitudinal collective action environmental problems will be effectively addressed. Third, because benevolence is an imperfect duty, it could be satisfied in ways independent of longitudinal collective action environmental problems, e.g., by volunteering at soup kitchens or tutoring in need children. Therefore, an imperfect duty of benevolence (or even an imperfect duty to address longitudinal collective action environmental problems) will not require responsiveness adequate to the problem.

The type of response to the problem of inconsequentialism most at home within Kantian ethics is appeal from duty. But this response only works if one’s duty (or the domain of justice or fairness) involves acting to address longitudinal collective action environmental problems. It turns out, for the reasons discussed above, that it does not.Footnote 7

As with act utilitarianism, there are three primary ways that an advocate of Kantianism might respond to the criticisms presented above. First, she might argue that it is not a criterion of adequacy for an environmental ethic that it provides normative resources to respond effectively to the problem of inconsequentialism. Second, she might argue that having unintended byproducts inform evaluation of actions is not needed to provide responsiveness to the problem of inconsequentialism. Third, she might argue that some version of Kantian ethics can exhibit this feature. I have addressed the first two of these above. However, with respect to the third, if a version of Kantian ethics can incorporate this and the other features conducive to addressing the problem of inconsequentialism, then it would be a virtue-oriented form of Kantianism, and as such would not belie this article’s central argument. The point of this and the preceding section was not to show that no normative theory developed out of act utilitarian or Kantian ethical frameworks can address the problem of inconsequentialism. It was to demonstrate why standard formulations of those types of theories fail to adequately do so, and use this as a basis for generating an account of features that a normative theory well resourced for responding to the problem would have.

Virtue-Oriented Ethics

The limitations of familiar versions of act utilitarian and Kantian ethics with respect to longitudinal collective action environmental problems suggested that a normative theory well resourced for responding to the problem of inconsequentialism would have the following features:

  1. (1)

    That discrete actions are not evaluated entirely on the basis of the outcomes (or likely outcomes) of the action.

  2. (2)

    That it evaluate on the basis of patterns of behavior or activities throughout a person’s life, as well as patterns among people or communities.

  3. (3)

    That its evaluative focal points (i.e., what it directly evaluates) include attitudes or perspectives of people.

  4. (4)

    That evaluation of attitudes and perspectives bear on evaluation of actions.

  5. (5)

    That evaluation of an agent’s actions is not overly contingent on the actions of others.

  6. (6)

    That evaluation of actions is sensitive to the significance of consequences, even when those consequences are unintended byproducts of actions.

These features characterize a two-tiered virtue-oriented normative theory, where tier one is teleological or consequential evaluation of character traits and tier two is evaluation of actions, practices, and policies in light of those traits. Here is a generic formulation of such a theory:

  • Theory of virtue: A character trait is a virtue to the extent that its possession is generally conducive to promoting the good; and a character trait is a vice to the extent that it is generally detrimental to promoting the good.

  • Principle of right action: An action is right to the extent that it is virtuous.

This theory of virtue and principle of right action are severely underspecified. Among the features in need of explication are: what constitutes a character trait; the level of generality at which traits are evaluated (e.g., all moral agents, all human beings, particular cultures, or particular cultures in particular environmental circumstances); the class of worlds over which traits are evaluated (e.g., the actual world, possible worlds very similar to the actual world, or worlds in which all relevant agents possess the trait); the nature and source of the good; what is constitutive of the good; what it is to promote the good; and what it is for an act to be virtuous (e.g., what a virtuous person would do, what a virtuous person would advise one to do, or that it hit the target of the operative virtues).

Nevertheless, this generic theory of virtue and principle of right action exhibit the features enumerated above. The theory of virtue evaluates attitudes or perspectives of people directly (3), and it does so on the basis of patterns of behavior throughout a person’s life and among people generally (2). Actions are evaluated in terms of the virtues (4). Because the theory of virtue evaluates character traits on the basis of outcomes, promoting the good, even when those outcomes are unintended or byproducts and actions are evaluated in terms of the virtues, act evaluation is informed by unintended consequences (6), without acts being evaluated only on the basis of their consequences (1). Finally, evaluations of actions are not highly contingent on the actions of others, since an act is evaluated on the extent to which it is virtuous, which is independent of whether or not others act virtuously (5).

Because it exhibits these features, such a virtue-oriented ethical theory supports well several of the reasons proffered for being responsive to longitudinal collective action environmental problems. It clearly does with respect to appeal to character responses, since whether an action is virtuous bears directly on its rightness. If an act exhibits indifference to the welfare of others (possibly including non-humans), and indifference of this sort is a vice, then it is at least prima facie wrong to act in that way. The same holds for other character traits that advocates of appeal to character responses might invoke, such as complicity, fairness, or personal integrity. If these character traits are virtues and failing to alter one’s behavior in response to longitudinal collective action environmental problems is contrary to them, then according to a virtue-oriented approach, a person acts wrongly if she does not act responsively.

Appeal to duty responses are also well supported by virtue-oriented ethical theory. Within a virtue-oriented theory, an action is right to the extent that it is virtuous and not vicious, wrong to the extent that it is vicious and not virtuous. Categories of “obligatory,” “permissible,” and “supererogatory” can be generated by setting thresholds along the continuum of virtuousness and viciousness.Footnote 8 There is a threshold below which actions within certain domains (e.g., those with implications for the welfare of other people) are sufficiently vicious or lacking in virtue that they are not permissible, and this generates duties or obligations not to act in those ways (Swanton 2003). Thus, virtue-oriented theory has the capacity to ground duties or obligations. Moreover, whether something is obligatory is a matter of the extent to which it is virtuous or vicious, which, as discussed above, is not contingent on whether other people are acting virtuously or viciously. So even if others are not doing their fair share (or acting in ways that would, if everyone so acted, address the problem) virtue-oriented ethical theory can support the claims that a person has an obligation or duty to do hers.

Appeal to agent benefit responses are also well supported within a virtue-oriented approach. Many virtue-oriented theories are eudaimonistic or, at least, include eudaimonia (or agent welfare or well-being) among a plurality of goods that virtue is conducive to promoting (Sandler 2007). Moreover, because virtue-oriented ethics evaluates perspectives and outlooks (i.e., character traits) directly, it can evaluate agent benefit counterfactually with respect to her perspectives. For example, there is a considerable body of empirical evidence that suggests that it is detrimental to a person to be highly materialistic, i.e., to value highly the possession of material goods in one’s life, be desirous to possess them, be emotionally invested in their presence and absence, and orient one’s life to accumulate them (Kasser 2002; Kasser and Kanner 2003). If a person is highly materialistic, it might be false that reducing consumption would be better for her, given her dispositional background; but it is probably true (given the research) that it would be better for her not to be highly materialistic. Furthermore, a person who is not materialistic, and prioritizes relationships and self-realization in her life, will not see reducing consumption as a sacrifice. This is an instance of the integrating effect of virtue. For a person who possesses the virtues, acting virtuously, even when it appears from other perspectives to involve limitation and sacrifice, is in general beneficial and regarded as such (McDowell 1980; Foot 2001).

The final type of response to the problem of inconsequentialism is the amplifying effects response. There are limitations to this type of response in genera, i.e., independent of their being situated within a particular ethical theory. For example, even the cumulative effects of a person’s actions over her lifetime are usually inconsequential in their impact on many longitudinal collective action problems; and it is not often the case that what a single person does impacts the behavior of many others or is crucial to bringing about systematic reform. However, to the extent that certain dispositions are conducive to bringing about such amplifying effects, they will inform character trait evaluation within virtue-oriented ethical theory. Moreover, virtue-oriented ethical theory has the resources to justify modifying one’s behavior in response to longitudinal collective action problems even if other are not likely to do so, which is crucial if one is to set an example or lead others to change. Therefore, the amplifying effects responses are reasonably well supported within a virtue-oriented framework, even as they are, in general, not as compelling or robust as the other types of responses.

Thus, overall, virtue-oriented normative theory is significantly better resourced for meeting the challenge of inconsequentialism than are standard versions of Kantian ethics and act utilitarianism. Therefore, environmental ethicists should be virtue-oriented ethicists, all other things being equal. This is not to claim that environmental ethicists should be virtue ethicists (as the term is often used). It is possible develop a virtue-oriented normative theory from neo-Aristotelian and other virtue ethics approaches to ethics. But there are also forms of utilitarianism that are virtue-oriented ethical theories. It is possible, for example, to construct a form of virtue utilitarianism on which a character trait is a virtue to the extent that it is conducive to promoting utility and an action is right if it is virtuous (e.g., done from or expresses virtue). Similarly, it is possible to construct a version of rule-utilitarianism that is a virtue-oriented normative theory. Brad Hooker, for example, has defended a version of rule internalization utilitarianism (he calls it rule consequentialism) on which an action is right if it is consistent with a code of rules the general internalization of which promotes consequences as good or better than any alternative code (Hooker 2000). If internalization is understood dispositionally (i.e., in terms of character traits), this is a virtue-oriented normative theory. As discussed above, it may even be possible to formulate virtue-oriented forms of Kantian ethics and act utilitarianism.

Emulator Utilitarianism, Global Utilitarianism, and Virtue-Oriented Ethical Theory

Dale Jamieson has recently advocated a version of utilitarianism that might be called emulator utilitarianism. Emulator utilitarianism is, like all utilitarianisms, committed to bringing about what is best [or, at least, what is good enough or “more good than not” (Driver 2001, p. 91)]. However, it is not committed to any particular theoretical framework or level of evaluative focus. On Jamieson’s view, “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” (Jamieson 2009, p. 1). Jamieson argues, as I have, that Kantian ethics is not conducive to justifying appropriate responsiveness to global environmental challenges. He further argues, as I have, that the nature of global environmental challenges favors an evaluative focus on “character traits, dispositions, emotions, and what I shall call ‘virtues’” (Jamieson 2009, pp. 9–10). This is because, as I also argued, these can accommodate non-contingency, i.e., that an agent’s actions are not “contingent on [her] beliefs about the behavior of others” (Jamieson 2009, p. 9) and are uncoupled from the actions of others (Jamieson 2009, p. 28).Footnote 9 However, Jamieson does not conclude from this that environmental ethicists should be virtue-oriented ethicists. Instead, because he endorses emulator utilitarianism, he concludes that “in this case and in this world, utilitarians should be virtue theorists” (Jamieson 2009, p. 28). In this way, Jamieson concurs with the conclusions argued for above, i.e., that the problem of inconsequentialism is a significant problem in environmental ethics and that virtue-oriented ethics is best resourced to address it, but disagrees that this implies virtue-oriented ethical theory is superior to utilitarian ethical theory, even with respect to such problems.

Because emulator utilitarianism and virtue-oriented ethical theory generate nearly identical normative theories, both structurally (i.e., regarding the structure of the normative theory) and substantively (i.e., regarding the content of the norms), which is preferable cannot be adjudicated on that basis,Footnote 10 but only on broader theoretical considerations. A fundamental point on which they come apart is their flexibility with respect to evaluative focal points. In virtue-oriented ethical theory, character traits are always the primary level of evaluation and actions are always evaluated in terms of virtues and vices. In emulator utilitarianism, the levels of evaluative foci and their relationships to one another can shift depending upon context. The proper level is the one that brings about the best consequences. Jamieson believes that this is an advantage of emulator utilitarianism over “indirect views that focus on motives, rules, or whatever” (Jamieson 2009, p. 12), which includes virtue-oriented ethical theory:

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 that is conducive to bringing about the best state of affairs (Jamieson 2009, p. 12).Footnote 11

However, there is a reason to want to privilege some level of evaluation over others. Doing so prevents pervasive irresolvable contradictory evaluations, which arise if no level of evaluation is privileged over any other. As Jamieson recognizes, due to its non-contingency, a virtue-oriented normative theory will often reach a different evaluative conclusion regarding an action than does act utilitarianism. This is why he advocates being a virtue-oriented ethicist and not an act utilitarian with respect to longitudinal collective action environmental problems. But on emulator utilitarianism, it is utilitarianism that is operating to justify virtue-oriented normative theory. There is thus a conflict, with respect to what action is right, between emulator utilitarianism operating on the act level and emulator utilitarianism operating on the theory level. Which level takes precedent? Within emulator utilitarianism there is no theoretical basis for privileging either level. Nor is there any practical basis for privileging either level. Emulator utilitarianism is committed to bringing about the best consequences. But, given act level evaluation it would bring about the best consequences to perform one action (the one with greater local utility), while given normative theory level evaluation it would bring about the best consequences to follow a norm that favors a contrary action (the one that contributes to addressing the problem).

Jamieson might respond that the reason for privileging a theory level evaluative focus is that this level is effective in addressing longitudinal collective action environmental problems, whereas an act level evaluative focus is not. However, this only shows that one should not adopt act utilitarianism as a normative theory with respect to such problems. It does not show that, when deciding how to act in a particular instance, it would not be better to act in accordance with act utilitarianism. In fact, it would be better in terms of bringing about the best state of affairs in that instance. The reason for this is that any time one acts in a way relevant to a longitudinal collective action environmental problem, one is also just plain acting. This is the source of the tension within emulator utilitarianism. Simultaneous with supporting virtue-oriented evaluation of actions in order to address the problem of inconsequentialism, it supports equally well act utilitarian evaluation of actions that gives rise to the problem. When one asks, “from the perspective of emulator utilitarianism, should I evaluate this act as a virtue-oriented ethicists or an act utilitarian”‘the answer is “both” or “either.” They are equally theoretically and practically justified; they both bring about the best consequences (one from the normative ethics perspective, the other from the act perspective). There is thus a conflict between two levels of utilitarian evaluation, neither of which is more basic. This is not merely a methodological or implementation problem, it is a significant theoretical problem. Because of it, environmental ethicists should not be emulator utilitarianisms that emulate virtue-oriented ethical theory. They should be virtue-oriented ethicists.

Global utilitarianism differs from emulator utilitarianism in that global utilitarianism has several levels of evaluation—actions, policies, character traits, etc.—that are distinct from each other. An action is right to the extent that it brings about the best consequences. A character trait is a virtue to the extent that it brings about the best consequences. A policy is right to the extent that it brings about the best consequences. And so on. Unlike with emulator utilitarianism, in global utilitarianism the structure and foci of normative evaluations never change. All levels are always evaluated, and they are always evaluated directly in terms of consequences. As Derek Parfit has put it:

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. (Parfit 1984, p. 25).

Therefore, on global utilitarianism it is never the case that actions are evaluated in terms of the virtues, even when the virtues are those character traits the possession of which bring about the best consequences. As Julia Driver has put it, “Character traits are simply another thing that, like action, can be evaluated along consequentialist lines” (Driver 2001, p. 72).

Global utilitarianism does not give rise to the contradictory evaluation problem generated by emulator utilitarianism. The problem arises for emulator utilitarianism because, with respect to longitudinal collective action environmental problems, evaluating acts according to act utilitarianism and evaluating acts according to virtue-oriented ethics are equally justified. In contrast, global utilitarianism always only evaluates actions according to act utilitarianism. However, due to the independence between character evaluation and act evaluation within global utilitarianism, it is subject to the problem of inconsequentialism in the same way as act utilitarianism, despite the fact that it involves consequentialist evaluation of character traits. (That is, it fails to exhibit features (1), (4), and (5) of a theory well resourced for responding to the problem on inconsequentialism, even as it exhibits (2) and (3).) It often is the case that the local utilities associated with an action that is complicit with a longitudinal collective action problem outweigh the (inconsequential) non-local utilities.

Global utilitarianism is perhaps a bit better off than act utilitarianism. Since it evaluates character traits in addition to actions, it can support counter-factual claims of this sort: if the agent were non-materialistic (i.e., virtuous) then it would not be right for her to consume so much. Nevertheless, because character and act evaluations are independent, on global utilitarianism it will still be right for the materialist to act consumptively if, due to her materialism, there are high local disutilities associated with acting non-consumptively. The rightness of her action is informed by her materialism. In contrast, virtue-oriented normative theory evaluates actions according to what a person’s dispositions ought to be (i.e., in terms of what is the virtuous thing to do). Thus, global utilitarianism fails to address several core aspects of the problem of inconsequentialism.

Conclusion

Longitudinal collective action environmental problems are not the only environmental problems, but they are among the most significant. Virtue-oriented ethical theories are better resources for justifying individual responsiveness to such problems than are standard version of Kantian and non-virtue-oriented utilitarian theories. If providing a compelling response to the problem of inconsequentialism is an instance of theoretical or practical adequacy for environmental ethics, then this is a reason for preferring virtue-oriented ethical theory over those other ethical theories. This is not an all-things-considered argument for virtue-oriented ethical theories. There may be reasons, independent of the considerations discussed here, for preferring other approaches to normative ethics. But if the arguments above are correct, then environmental ethicists should be virtue-oriented ethicists, rather than non-virtue-oriented utilitarians or Kantian ethicists, all other things being equal.