In this essay, I revisit the familiar contrast between anthropocentrism and nonanthropocentrism in the context of environmental virtue ethics (EVE). My interest in EVE concerns its potential to motivate and justify concern for the more-than-human world. Environmental ethicists have long described the transition toward nonanthropocentrism as a dramatic shift in worldview (Taylor 1986; Rolston 1988; Norton 1991). The apparent failure of deontological and consequentialist theories to justify this shift—they arguably achieve much success once the shift has been made—has moved some to look to virtue ethics as offering a possible third way. On this view, the normative force of nonanthropocentric ends is to be explicated in terms of the virtues, and in terms of the various strengths of virtue ethics in general, including, but not limited to its ends of human flourishing and the strengthening of community bonds (Hursthouse 1999; Swanton 2003). Other features of virtue ethics also explain its attractiveness to environmental ethics. Its adaptability to ethical and scientific naturalism, its resistance to codifiability, and its elevation of the significance of emotional intelligence imply that virtue ethics is impressively adequate to contemporary insights in moral psychology and social behavior.Footnote 1 Consequently, virtue ethics seems primed to situate human morality within a much broader (more-than-human) natural context. Yet as much as I find this promise energizing—it meshes with my own metaphysical sensibilities—I hesitate when I think of my students, colleagues, and neighbors who just cannot seem to get the force of nonanthropocentrism, that is, who just cannot embrace the view that the more-than-human-world ought to matter for them morally. When I try to understand why I am bothered by this, turning to philosophy for consolation, I find guidance and even a glimmer of hope, if not comfort, in the moral philosophy of David Hume.

In what follows, I bring several features of Hume’s moral thought to bear on the practical (and some theoretical) implications for nonanthropocentrism in EVE. Hume is relevant to EVE for a variety of reasons, not least of which is the significant extent to which his own moral philosophy is arguably virtue ethical.Footnote 2 For this essay, where the difference between worldviews is under scrutiny, I locate Hume’s critique of the so-called “sensible knave” at the center. With the knave, Hume confronts a legitimate threat to his sympathy-based morality in the form of a prudent, calculating, and deceptively amoral agent who exploits the moral virtues of his sympathetic associates. After examining Hume’s response to the knave, I argue that his discussion elicits a useful analogy for diagnosing the practical challenge EVE faces when it includes nonanthropocentric ends in its justifications of the virtues. By identifying the practical limitations in making the case for morality to knaves, Hume reveals that the most convincing case for morality lies in realizing the effects of the full range of virtues. It is these that the knave is unfit to enjoy. Nonanthropocentric accounts of EVE find themselves in a similar situation, one that ultimately directs inquiry to the effects of specific virtues in order to underwrite confidence in the nonanthropocentric worldview.

A Dichotomy of Worldviews

There are many senses of anthropocentrism, both normative and descriptive. The sense that I employ in this paper refers to the worldview according to which environmental objectives are approved when they benefit human individuals and communities. I take this view to be consistent with the idea that in many cases protections ought to be extended to nonhuman animals, species, and ecosystems to the extent that such protections are beneficial to human consumptive, educational, or recreational interests or are otherwise of negligible cost. Thus, according to this view, the more-than-human world is either of instrumental value to humans or its value is undetermined, but anthropocentrists so-described will likely include ecological principles in their long-range plans for sustaining human communities. In this sense, anthropocentrism is a kind of environmental prudentialism. Nonanthropocentrism, by contrast, refers to the worldview that extends moral considerability to features of the more-than-human world. By this view, nonhuman animals, species, or ecosystems may have value for their own sakes, and because of that possibility, the choice between protecting them or satisfying human interests is an intelligible one.

In consequentialist and deontological environmental ethics, these worldviews typically inform the ends of actions, with the implication that nonanthropocentrism includes consideration of the non-instrumental value of the more-than-human world as a meaningful end of certain actions; anthropocentrism does not. Most virtue ethical approaches, however, distinguish between justifying virtues in terms of their ends and evaluating actions. Actions reflect an agent’s degree of success in possessing a virtue. From the agent’s perspective, the choice to perform one action rather than another will reflect her sense of which virtues are relevant to the situation (along with other salient facts about her—e.g., her moral maturity, her emotional profile—and her understanding of the context of action). Her action will be the correct one, then, to the extent that it adequately hits the target of the relevant virtues, and even then, depends on her unique capabilities as an agent.Footnote 3 Her action will not be correct because it is anthropocentric or nonanthropocentric, or even because it respects the intrinsic value of things (or not). Rather, her action will be correct because it is compassionate, prudent, sincere, friendly, etc. This analysis suggests that EVE may be able to avoid having to reveal a preference for a particular worldview, since actions (and agents) are to be evaluated relative to the virtues rather than value-theoretical ends. But this suggestion fails to consider the limits of virtues for specifying the scope of their application. For example, helpfulness that responds exclusively to the plights of humans is different from a helpfulness that also recommends responding to the plights of nonhumans, even though in both cases helpfulness is substantively identical (i.e., giving assistance). In practice, of course, helpfulness may be fully discharged if the context only requires consideration of human interests, so my point is not that nonanthropocentrism must be apparent in all specific acts of virtue. Rather, when it should be (from the standpoint of the worldview), a nonanthropocentric end will not be explicable specifically from an account of helpfulness.Footnote 4 I take it as relatively uncontroversial that EVE can be adapted to anthropocentrism. The goal of a sustainable human future is consistent with virtue theoretical ends of human flourishing and strengthening community bonds. I also find unproblematic the idea of nonanthropocentric EVEs, in the theoretical sense that virtues can be defined relative to and justified by nonanthropocentric ends. But if I find myself inclined toward nonanthropocentric applications of virtue and my anthropocentric (but helpful) neighbor does not, EVE appears no better prepared than other theoretical approaches to compel a shift in worldview. And as I discuss later, this issue emerges within pluralistic EVEs, too. Of course, demanding such a result from EVE is perhaps to hold it to a standard that other theories cannot meet; that there are no independent justifications for a worldview is an ineliminable feature of moral life after all. It is in light of this situation that Hume’s knave becomes relevant.

Hume’s tale of the knave is important for EVE for two primary reasons. First, Hume is able to diagnose a significant source of the difficulty (if not failure) of virtue education in the form of preoccupation with a single virtue. In the case of the knave, the virtue is prudence, in the sense of careful self-regard. On Hume’s account prudence capitalizes the knave’s reason and precludes his ability to appreciate other sorts of virtuous ends. Second, Hume’s confirmation of the benefits of possessing a wider set of virtues than the knave suggests that EVE has much to gain by attending to the typical effects of certain virtues. In other words, if Hume’s moral agents gain confidence for their moral outlook (which the knave lacks) by their realization of the benefits of virtue, then perhaps the confidence needed to underwrite nonanthropocentrism in EVE can also be located in the typical effects of certain virtues, and importantly, in virtues that do not already rig the game on the side of the nonanthropocentric worldview.

Hume’s Sensible Knave and the Hegemony of Prudence

Hume’s account of the knave occurs in his Enquiry Concerning the Principle of Morals. For Hume, the knave provokes practical questions concerning the aims and intended audience of moral education. Not everyone, it turns out, will be receptive to the virtues. Knaves in particular are constitutionally unfit for moral education owing to their sympathy deficit. Although this strict naturalistic sense of knavery is important for Hume’s sympathy-based morality, it is not applicable to the analogy developed in this paper. Trivially, it is true that some persons are simply psychologically or cognitively precluded from moral agency, but this is not what makes the knave interesting. Rather, it is Hume’s depiction of a character who is preoccupied with a narrow but rational self-interest, someone who might even pass for prudent, that makes the knave significant. For the knave so-described unfits himself for morality by attending only to what is prudent. He cannot see the sense in what the (other) virtues provide for their possessors even if he understands (and sees opportunity in) the fact that they value them.

Internal to Hume’s project, the knave represents an important challenge to his sentiment-based theory of virtue because the knave, in the bulk of his actions, appears to possess a trait others in his moral community find praiseworthy, namely his tendency to act prudently in his rational self-interest. The sad news for the knave’s moral associates is he is so thoroughly self-interested and rational that he free-rides on their benevolence, helping himself to all of the benefits of their kindness without ever feeling motivated to reciprocate in kind. The main problem the knave poses for Hume concerns his apparent unresponsiveness to sympathy, which for Hume is a necessary condition for morality. Because the knave lacks sympathy, no degree of argument or other means of persuasion can convince him to be moral. Significantly, the knave exposes Hume’s moral theory as a whole to a significant vulnerability: that there could be those among us who lack sympathy, and who, protected by our morality, deceive us into promoting their own ends. As Hume explains, the knave “in particular incidents, may think that an act of iniquity or infidelity will make a considerable addition to his fortune, without causing any considerable breach in the social union and confederacy. That honesty is the best policy, may be a good general rule, but is liable to many exceptions; and he, it may perhaps be thought, conducts himself with most wisdom, who observes the general rule, and takes advantage of all the exceptions” (Hume 1975).

The knave, being sensible, effectively employs his instrumental reason to optimize his pursuit of his self-interested ends. Consequently, it is no argument against the knave to protest that he is acting irrationally; he already has reason on his side doing his selfish work for him. Hume is thereby resigned to “confess that, if a man think that this reasoning [that the knave cannot be persuaded to morality by reason] much requires an answer, it will be a little difficult to find any that will to him appear satisfactory and convincing. If his heart rebel not against such pernicious maxims, if he feel no reluctance to the thoughts of villainy or baseness, he has indeed lost a considerable motive to virtue” (Hume 1975).

One obvious practical question, then, is what can be done about the knave, and how can the rest of us, that is, those of us who are sympathetically motivated to be moral, protect ourselves against such a person? Hume recommends two responses to sensible knavery. The first is that, in a community dominated by knaves, it is best to act like a knave. Otherwise, one risks exploitation by one’s knavish peers; moral integrity is a weakness in a society in which generosity and trustworthiness are consistently used against their possessors. Fortunately, Hume believes this first response is probably unnecessary, and he entertains the empirical hypothesis that most human communities are more characteristically moral than they are knavish. That is, most human communities reflect the widespread activity of sympathy in their moral norms, and for Hume, these norms allow a much greater degree of evaluative richness than prudent motives alone could support. Hume’s second response, therefore, is to encourage non-knaves to retrench their commitment to the full spectrum of human values with its normative richness, which the knave (because of his egoistic constitution) is unfit to enjoy.

On Hume’s view, then, the knave is seriously missing out on the benefits of morality, and Hume admonishes the knave for denying himself the superior happiness that stems from self-reflection and morality:

[the sensible knave], if he has any tincture of philosophy, or even common observation and reflection, will discover that they themselves are, in the end, the greatest dupes, and have sacrificed the invaluable enjoyment of character, with themselves at least, for the acquisition of worthless toys and gewgaws (Hume 1975).Footnote 5

To be sure, the knave has much in common with the rest of us, particularly an agreeable penchant for pleasure-seeking. But Hume also asks,

what comparison between the unbought satisfaction of conversation, society, study, even health, and the common beauties of nature, but above all the peaceful reflection on one’s own conduct; what comparison, I say between these and the feverish, empty amusements of luxury and expense? These natural pleasures, indeed, are really without price; both because they are below all price in their attainment, and above it in their enjoyment (Hume 1975).

Of course, the knave lacks the ability to appreciate the goods that Hume describes here. So, other than the unlikely possibility that the knave might possess enough “common observation and reflection” to divert himself from his moral shortsightedness, it is probably best to conclude that Hume’s rebuke is not really meant for knaves. Rather it is an injunction to the rest of us to be thoughtful, to reflect on the sources of our moral and other pleasures, and to inquire as to the character one needs to achieve such satisfaction. What such reflection will show is that knavish self-interest—and by affiliation, a disproportionate weighting of prudence—are unlikely to engage us with the wealth of other values that tend to show up in relation to our creaturely constitutions. In short, Hume is appealing to his readers to reflect on their standards for virtue and to take notice of how those standards do not all reduce to the pleasure attained by doing well for oneself.

An additional point needs to be made here. In light of Hume’s call for reflection, it is important not to embellish Hume’s critique with his otherwise optimistic assessment of human nature as generally benevolent. Hume consistently regards prudence—acting in one’s best self-interest—as a virtue.Footnote 6 Indeed, prudential considerations are instrumental for Hume in explaining how we are able to act upon the perceived advantages of cooperation and adhering to custom (Hume 1978). Nonetheless, we risk knavery, and its corresponding deprivation of the full range of virtuous pleasures, if we reduce all evaluations to standards of prudence.

Anthropocentrism and Environmental Virtue: On the Limits of EVE

By this point it should be apparent that Hume’s discussion of the knave bears resemblance to the relation between anthropocentrism and nonanthropocentrism in EVE. Reiterating my earlier qualification, true sensible knavery (sympathy deficient persons) is not the issue here. Rather, Hume’s discussion also draws attention to the possibility of a more attenuated knavery that involves preoccupation with a single virtue, specifically prudence. Whereas the true knave is unfit for morality owing to his nature, the prudence-dominated agent unfits herself for appreciating the benefits of the wider range of non-prudential virtues by allowing her attachment to self-interested (instrumental) rationality to occlude her appreciation of social and noneudaimonistic goods made possible through possession of other virtues. This kind of situation, it should be noted, is especially plausible in Humeanism, since practical rationality is constrained by more dominant sentiments. But even if one does not endorse Humeanism, the point still holds that an agent could act in ways that are prudently approvable (for oneself and others) yet fail to appreciate the benefits that Hume’s otherwise moral agents are able to enjoy as part of their richer morality. Moreover, such a person will present a particular challenge for moral education because of the totalizing effect of her prudence-informed practical rationality. Such a person may even prove to be hopelessly ineducable relative to a fuller set of virtues and thereby fail to gain full entry into the moral worldview. I take this situation to be quite comparable to the one that arises with anthropocentrism/nonanthropocentrism. If the analogy I am offering is strong, anthropocentrism and its accompanying practical rationality occludes the estimable nonanthropocentric effects of virtue. Those who reason well enough about their ends or are dispositionally in near enough proximity may be gradually educated into the full (nonanthropocentric) implications of many virtues, but the bulk of agents who have unfit themselves to appreciate the more-than-human world will have to be persuaded through other means than by appeal to nonanthropocentric values and virtues.

Hume’s diagnosis of what constrains the morally virtuous in pressing their case to knaves and the less virtuous is also useful to consider here. One practical limitation the virtuous may confront occurs if their morality is too closely identified with a worldview that few are fit to share. For Hume, this sort of practical constraint appears most starkly in his skepticism about the reach of religious ethics that are too otherworldly and hence unrealistic or too demanding (Hume 1975). (Considering the great extent to which so many people do find these sorts of ends convincing might imply that Hume is just wrong on this one. However, it would explain why religious justifications for morality are equally unconvincing to the religiously uninitiated or disaffected). A second limitation occurs with respect to the skill of the moral educator in communicating her moral norms. In the Treatise, Hume develops his account of the “general point of view,” a maneuver in moral evaluation necessary for contravening the potential narrowness of first-person sentimental projections of value by fixing moral evaluation to norms held by a wider moral community (Hume 1978). There he also develops his so-called “narrow circle requirement.” This evaluative construct limits the expansiveness of the general point of view by designating effective evaluators as those who familiarize themselves with and apply the norms relative to an agent’s particular moral community, where community is defined broadly as those with whom one has regular “commerce” (Hume 1978). In stipulating these standards, Hume is not espousing moral relativism so much as he is articulating methods of achieving consensus across varying scales of moral community. As a result, Hume should be read as seeking to accommodate significant flexibility in how norms are applied. On the one hand, Hume’s standards caution against crude judgments of highly localized interpretations of virtue: e.g., politeness requires removing one’s shoes in their house, even if we do not expect that in ours. On the other hand, when the effects of conduct typically exceed narrow circles through trade and travel (and it’s fair to include the effects of military engagement, international policy, global communication, and the economic and ecological impacts of agriculture and industry), moral discourse becomes more dynamic, bringing in more players and viewpoints, and increasing the demand for skill in moral discernment, and for philosophers perhaps, skill in virtue explication. When taken descriptively, Hume’s position on evaluation may give us pause to consider that the moral challenges of our age reflect the collapse of so many narrow circles at so many levels of human and environmental interaction: as an ever-increasing global community we are being challenged to find a center amidst varying scales of “commerce” unprecedented in history. Taken normatively, which I think we can (Haught 2006; Sayre-McCord 1996), Hume is enjoining moral evaluators to avoid the risk of too tightly clinging to the narrowness of their own understanding of virtue. In the present context, these standards expose the practical difficulty that a nonanthropocentrist will have in conveying the normative force of her virtues to her anthropocentrist peers. At the same time, however, they reveal the strongest hand the nonanthropocentrist has to play, which is to call out the anthropocentrist’s conduct for its impacts that demonstrably reach beyond any one “narrow circle,” and by Hume’s account, criticism of that conduct is fair game.

Not everyone agrees. In articulating her own Humean approach to EVE, Jennifer Welchman interprets the reach of Humean ethics along exclusively anthropocentric lines. Advocating a stewardship ethic, Welchman urges that moral agents are called to “develop a greater sensitivity to the noneconomic needs, values, and interests that natural entities and environments serve if we are to develop enduring commitments to preserve them” (Welchman 1999, emphasis added). Developing this Humean account of enlightened anthropocentrism, Welchman adds:

If species or entities about which virtually no one cares, has cared, or are ever likely to care can be saved from extinction only by demanding enormous sacrifices of things for which many people have cared, do care, and will continue to care, then we can and perhaps must permit their destruction. As a rule, uncertainty about our understanding of the natural systems involved favors preservation. However, in those instances where both our certainty and the costs are very high, I cannot see why we must accept the costs (Welchman 1999).

Here, Welchman’s precautionary rule tempers the sting that nonanthropocentrists might feel in her defense of anthropocentrism; it appears that only in extreme circumstances will the Humean anthropocentrist allow concern for her own species to override concern for the well-being of others. Yet nonanthropocentrists get stung again later—this time more painfully—at the conclusion of her discussion, when she writes, “To those who believe both that we need to embrace nonanthropocentric principles of value if we are ever to respond adequately to human threats to the environment and that theoretical justification of such principles can move us to act upon them, what I have to offer is certainly not the loaf they desire” (Welchman 1999). In light of Welchman’s self-imposed confinement of Humean morality to anthropocentrism, it might be unfair to criticize her argument for failing to accommodate the nonanthropocentric viewpoint. Notably, it is Welchman’s skepticism that such a position could be motivating (for moral creatures such as ourselves) that underscores the extent to which she argues that Humean style constraints on virtue ascription ought to be recognized.Footnote 7

Where Welchman and I agree is that anthropocentrism impairs the ability of agents to be motivated by nonanthropocentric norms, and it is not a failure of practical rationality. The force of Hume’s critique of the knave is that knaves are difficult to dislodge from their dispositions and appeals to what should matter to them can often fail. As a result, Welchman and I also agree that there are limits on what philosophers can expect of EVE. Advocacy of nonanthropocentric virtues through systematic arguments of their beneficial effects for the more-than-human world is unlikely to reach the hearts of anthropocentrists and the even more environmentally vicious. Where I disagree with Welchman is that Hume’s moral philosophy implies that anthropocentrism is the only rational option consistent with environmental ethics. Ask any nonanthropocentrist, and they would be puzzled by the suggestion that their virtues or their ends are irrational. In part, this is because the rationality at stake here is not merely a practical one, but a theoretical one concerning the relation between virtues, their typical effects, and their justifications. But there are important limits to EVE on this account as well.

To see what they are, it is useful to consider Ronald Sandler’s responses to both the practical limitations of EVE and the responsibility of EVE to provide justification for valuing nature nonanthropocentrically. One of the strengths of Sandler’s approach is that it explicitly shows how EVE can be inclusive of nonanthropocentrism while acknowledging the diversity of senses in which agent or community ends are also constitutive of virtues. Sandler’s pluralistic teleological account of the virtues carves out a range of ends that both determine and are supported by the virtues. Not surprisingly, among these diverse ends are those consistent with anthropocentrism. For Sandler, these agent and community-directed ends fulfill the eudaimonistic function of virtue ethics. Noneudaimonistic ends also appear in Sandler’s argument, and he appropriates and refines Paul Taylor’s theory of inherent worth to inform and justify the relation between virtues and environmental entities that possess a good of their own. On Sandler’s view, this means that environmental collectives such as species and ecosystems lack intrinsic value, but they are nonetheless frequently the proper targets of environmental virtue because of their instrumental relation to individual organisms (and some collectives, such as ant colonies) that also possibly possess goods of their own.

Sandler understands that the pluralistic teleological theory runs into practical limitations. He mentions these as part of his assessment of three adequacy conditions for environmental ethics in general. The first two assert that an environmental ethic is adequate to the extent that it can justifiably criticize environmentally destructive actions, and that it can guide action and policy, and on both counts Sandler is comfortable that his theory passes the test. The third adequacy condition, however, produces a potential obstacle for Sandler’s theory with its demand for practical efficacy. As he explains, an environmental ethic must provide “arguments, reasons, or justifications that are efficacious in moving people to perform the actions or adopt the policies that are recommended” (Sandler 2007). Sandler proceeds to describe the contemporary scene involving American environmentalist attitudes, noting the many instances of gaps between belief (e.g., “environmental protection is good”) and motivation (e.g., “I’m not ready to change my consumptive and recreational practices”).Footnote 8 Despite these practical impediments, even if many Americans fail to appreciate environmental values or lack the relevant virtues, Sandler argues that advocacy for removing obstacles to environmental virtue (e.g., facilitating opportunities for wilderness experiences or promoting environmental awareness education) would nonetheless be warranted. Some of the warrant is diminished since it is unknown to what extent such actions would be effective. However, the motivation for these actions is virtuous, aiming at increasing the prevalence of environmentally virtuous behavior, even if the outcomes of specific educational practices is uncertain.

Is EVE Necessarily Nonanthropocentric?

In explaining his methodology, Sandler distinguishes his pluralistic teleological approach from preceding efforts in EVE. One of these, which he calls the “extensionist” approach, is typified by explicating virtues for their appropriateness for environmental ethical ends. The extensionist approach is weakened, however, by its tendency to ignore justifications for the ends of virtues: “it is not possible to employ the extensionist strategy without a background account of what makes a character trait a virtue” (Sandler 2007). Sandler’s approach attempts to remedy this weakness by including nonanthropocentric justifications of the virtues as part of a wider set of justified ends. Specifically, he calls upon Taylor’s theory of inherent worth to underwrite a select set of noneudaimonistic virtues, what Sandler refers to as virtues of respect for nature, and what makes a virtue of respect for nature a virtue is the extent to which it fits an agent well for respecting the inherent worth of living organisms and some environmental collectives.

For the sake of space, I will not reiterate Sandler’s argument for inherent value. Its present importance stems from its resemblance to Hume’s injunction to the knave. Citing Taylor, he writes

By [the attitude of respect for nature, Taylor] means that it represents entry into a particular moral outlook, not a position justified by some already legitimated outlook. It is justified instead by a ‘belief-system that underlies and supports the attitude’ and shows ‘that it is acceptable to all who are rational, factually informed, and have a developed capacity of reality awareness’ (Sandler 2007).

If this sounds familiar, it is because Sandler (via Taylor) is echoing Hume’s response to knavery, and it likely to have similar results. For nonanthropocentrists, the argument grounds their confidence that the targets of their virtues are indeed rational and responsive to truly worthwhile things. For anthropocentrists, the effect of the argument is less clear, contingent on the extent to which their rationality is depleted already by their anthropocentric commitments. Nonetheless, the argument is very likely the best one available to nonanthropocentrists in support of their virtues, just as the argument for the richness of virtue was for Hume against the knave. Yet as an important caveat, doing virtue theory in this way does not provide an argument for taking up the nonanthropocentric perspective by appealing to practical or theoretical rationality; rather, it assumes the perspective will be convincing to reason once one dwells within it. To the uninitiated, much of the success of the argument depends on effective proselytism rather than garnering rational assent. This circumstance arises because nonanthropocentric virtues imply a capacity to grasp outcomes of modes of discernment that cannot be demonstrated by rational argument. To put it another way, to the anthropocentrist, the claim that the attitude of respect for nature will be acceptable to all who have a developed capacity of reality awareness can easily be turned around: concern for nonhumans over human interests is just as likely to imply reality blindness. The good news for EVE is that virtue ethics does not have to depend exclusively on rational assent. Learning, practicing, and possessing the virtues can all be instrumental in preparing one to dwell within a particular worldview. Once the worldview is taken up, the rational justifications for the virtues fall into place, and it’s here that Sandler’s defense of the attitude of respect for nature can be deployed.

Interestingly, Sandler’s pluralism may also undercut the force of the nonanthropocentric justification for the noneudaimonistic ends. It leaves open the possibility that an agent can switch-off between eudaimonistic and noneudaimonistic ends, since “Virtues of respect for nature do not have special status within the virtue-oriented approach to environmental ethics” (Sandler 2007). Moreover, Sandler explicitly wants to avoid the implication that the attitude of respect for nature is all-encompassing, that it “is the moral attitude” (Sandler 2007). However, I wonder that he is not overstating his need to resist such implications. First, the inherent worth that justifies the virtues of respect for nature is conjunctively included with other determinations of virtue. As Sandler explains, “the inherent worth of living things provides only one end against which character traits are evaluated, the dispositions constitutive of the virtues of respect for nature will be informed also by other ends” (Sandler 2007, emphasis added). Second, Sandler’s argument for differential compassion—intended to show how an environmentally ethical compassion avoids the practical pitfalls of biocentrism—implies that whatever switching-off is allowed by virtue, it does not permit a choice of targets of compassion: the well-being of individuals, whether human or nonhuman, is always primary (Sandler 2007).Footnote 9 Therefore, the pluralism of Sandler’s theory does not imply that an environmentally virtuous person must (sometimes) choose between a belief-system that accommodates the inherent worth of things and one that does not. If this is right, then Sandler’s view is both consistent with and arguably a gain for nonanthropocentrism. But it also means that Sandler’s EVE elevates the importance of nonanthropocentric discernment by fixing the noneudaimonistic ends of virtues to an arguably nonanthropocentric belief-system. EVE allows that virtuous persons will sometimes act in ways that put the interests of persons and communities ahead of nonanthropocentric ends. However, a virtuous person will not be one who shares the anthropocentrist’s perspective that environmental responsiveness is limited to the benefits of that conduct for persons and their communities. Moreover, the reason that the anthropocentrist holds the perspective that he does is that, from the perspective of EVE, he lacks the necessary worldview to make the noneudaimonistic ends of the nonanthropocentrist’s virtues intelligible.

Sandler’s argument is much more comprehensive than I can do justice to here, but in much of Character and Environment he struggles to resist the nonanthropocentric implications of his overall account. For example, in explaining environmental decision making, Sandler offers the following anthropocentric rationale for avoiding environmental degradation: “any environmentally virtuous person will be opposed to unnecessary environmental degradation that compromises some peoples’ access to environmental goods” (Sandler 2007). However, it does not appear that Sandler intends to restrict environmental virtue only to ends that accommodate human interests, and that the virtuous person is one who also recognizes that “there is nearly always an alternative to some environmentally insensitive behavior or policy that does not compromise the demands of other virtues” (Sandler 2007). Therefore, with Sandler there is an ambivalence—perhaps a necessary one given the demands of cogency for his pluralistic account—over how limited we should understand the reach of the Sandler/Taylor belief-system across the range of virtues to be. Understandably, there are all sorts of circumstances, where environmental considerations are irrelevant to exercising virtue where one would expect the relevant eudaimonistic ones to function. One does not have to be a nonanthropocentrist to grade a test impartially or be attentive to a child’s needs. But when environmental considerations are warranted, I read Sandler as arguing that the virtuous person will be inclined to give the more-than-human world due consideration. The rationale for doing so is nonanthropocentric even if “there is no presumption that what an environmental virtue recommends will be subordinate” (Sandler 2007).

If I am correct to tease out this ambivalence in Sandler’s account, then it supports my view that EVE duplicates the situation of morality with respect to the knave when EVE aspires to include responsiveness to nonanthropocentric value. As a practical matter, the anthropocentrist is not likely be persuaded by appeals to the rationality of nonanthropocentric ends, but nonanthropocentrists may find such arguments reassuring, underwriting confidence in their virtues as they are able to integrate them with a richer moral horizon than their anthropocentric associates are fit to appreciate. There is something troubling about this image, however, in that it reinforces what is perhaps an artificial sense that there is an option between anthropocentrism and nonanthropocentrism. However, with Hume and the knave, knavery is not an option if one wants to be moral. Comparably, anthropocentrism should not be an option if one wishes to be environmentally virtuous. I think Sandler would agree.

EVE and Discernment

Arguably, one of the strengths of EVE is that there is much to be learned about morality from consideration of the virtues themselves (an area where both Welchman and Sandler excel). As an ethical theory, EVE must meet demands of adequacy, applicability, coherence, and consistency, but EVE invites a separate level of reflection on the profiles of specific virtues across a wide sweep of cases. Sandler consistently reminds his reader that the exercise and even excellence of virtue is contextual and agent-relative. As a result, not all that is ethically important about virtue is exhausted at the level of theory, even as pluralistic a theory as Sandler’s. The knave’s unfitness for virtue also generates the importance of looking to the specific effects of virtues in Hume’s theory. By understanding the profile of each virtue, a sense of what the knave is missing may start to become clear. The knave fails to understand generosity, for instance, because he cannot possess the virtue. Non-knaves may be similarly unclear about the virtue, but for them education is still an option. By acting in ways consistent with generosity, they put themselves in position to enjoy its effects. In EVE, explication of relevant virtues would likewise aim to shed light on how they give rise to the nonanthropocentric worldview. This exercise will be less suspicious to anthropocentrists if it does not already assume the validity of a nonanthropocentric worldview. As a theoretical exercise in virtue explication, it will be important to begin with virtues that do not necessarily imply nonanthropocentrism. Thus, the goal is to identify virtues that typically elicit effects that could engage their possessors to shift perspective to a wider, more inclusive moral outlook.

To sketch how this approach might work, consider humility and wonder. Thomas Hill notably describes humility as supporting a family of other virtues relevant to environmental attitudes, such as cherishing (that which is important to us) and gratitude (Hill 1983). Adding to Hill’s considerations, humility’s profile highlights the importance of relation, typically a relation of dependence on the part of the humble person to something else. Humility gives leeway to something other than oneself, yielding to its presence, giving it a voice, acknowledging the significance of its otherness. As a result, humility is consistent with nonanthropocentrism by fitting the agent to discern her particular modes of relating to otherness. To be humble is often to put oneself into a proper relation with that which is other and sometimes greater than oneself. To be virtuous, humility cannot refer to self-ingratiation or obsequiousness, but it clearly points in the direction of the value of something more-than-oneself, bringing it close to nonanthropocentrism.

For its part, wonder has similar effects for expanding an agent’s horizon of concern. In Book II of the Treatise, Hume’s comprehension of wonder is closely tied to a critical feature of human psychology, namely the love of truth (Hume 1978). Wonder, as curiosity, helps Hume explain why we take pleasure in mathematics and games, the latter of which, though not intrinsically valuable, are pleasing because generally “life is so tiresome a scene” (Hume 1978). More seriously, Hume notes that lacking wonder might make us indifferent to our own reality. For Hume, this implicates wonder strongly in the level of interest one takes in “morals, politics, natural philosophy, and other studies, where we consider not the abstract relations of ideas, but their real connexions and existence” (Hume 1978).

In environmental ethics, Holmes Rolston considers that we may have a duty to “Keep life wonderful!,” to sustain scientific, aesthetic, philosophical, religious, and recreational enjoyment of nature as well as to reveal nature’s intrinsic value (Rolston 1988). Conversely, wonder might also be attractive within a narrower anthropocentrism, suggesting that wondrous human civilization could substitute for a wondrous more-than-human world. The suspicion lingers, however, that parents and educators have failed if wonder is to be cultivated within limits that confine it to the purely human. If it is not blameworthy to wonder only at what is near and familiar, it is nonetheless consistent to appreciate it for the apparent boundlessness of its effects. Wonder may be responsible as well both for liberating the objects of our desires from our claims on them as ends and for informing us of potential ends for our desires. In making the case for wonder’s importance for love, Luce Irigaray claims that the object of wonder remains “impossible to delimit, im-pose, identify (which is not to say lacking identity or borders): the atmosphere, the sky, the sea, the sun” (Irigaray 1984). The value in inculcating a sense of wonder in children is, for Martha Nussbaum, relevant for informing them of potential objects of love and compassion: “Wonder, as non-eudaimonistic as an emotion can be, helps move distant objects within the circle of a person’s scheme of ends” (Nussbaum 2001). For his own part, Rolston has been a champion of the view that scientific and aesthetic engagement with the natural environment is not only a consequence of wonder but an activity that enhances wonder, bringing with it increased possibilities for value. In short, if wonder has these effects, not only is wonder thereby important for liberating us from life’s “tiresome” scene, but wonder is a virtue especially critical for fitting its possessor to discern the kinds of value that underwrite the nonanthropocentric stance in EVE.

“What are the benefits of morality if I am perfectly content with my own self-interested outlook?” Were Hume’s knave so curious, Hume might have a chance to convince him of those benefits. But only by inviting the knave to learn the virtues, getting him to practice them, and ultimately to put him in position to enjoy their effects is Hume likely to have success. Even then, the knave may not warm up to the moral outlook. Such is the predicament of EVE with respect to the nonanthropocentric worldview. However, like Hume, EVE also possesses a resource in explicating specific virtues to account for the values that agents discern in the world. As such, the knave and Hume’s response may provide more than an interesting analogy to EVE and its response to anthropocentrism. It may also reinforce a sense that moral philosophy is itself motivated by virtue, which, if properly discerned, fits one to live a life that is richer in value than the alternative.