Care ethicists have disagreed for decades over whether and how to incorporate virtue theory into their ethical framework. At the heart of these disagreements lie pivotal questions about the nature of care and the place it ought to occupy in moral philosophy. Yet in spite of the vigor of these debates and their profound implications for the ethics of care, there is lingering unclarity about a number of issues in this area. One of these is a sort of conceptual unclarity about the options that are available to those who think of care as a virtue, which of these options care ethicists have explored, and which options are most promising. Another is comparative; it remains unclear where the most prominent care ethicists stand in relation to one another on this topic, but also in relation to philosophers of other stripes who acknowledge care. A third sort of unclarity persists over the relationship between care ethics and virtue ethics.
This article aims to offer a little more clarity on all of these fronts by way of a meta‐analysis of the ways in which care is and can be treated as a virtue. I begin in the first section with a study of what I call the “virtue thesis” about care. The virtue thesis can be cashed out in a number of ways, and I argue that a weaker variety is widely endorsed by care ethicists. In parts II–IV, I develop a taxonomy of the approaches available to proponents of any version of the virtue thesis. The three approaches I distinguish are: analogical approaches, which treat care as analogous to some pre‐existing virtue; supplementalist approaches, which translate care into a novel virtue; and bundling approaches, a subset of which cardinalize care or treat it as a constellation of virtues. Parts V and VI engage in an evaluative project, defending the virtue thesis and analogical approach, arguably the least popular approach among care ethicists, from some likely objections.
I. The Virtue Thesis
As I understand it here, the virtue thesis claims just that care is a virtue. There are several things to note about this formulation; the first is that the virtue thesis excludes the interesting and not implausible notion that care is virtue. That claim warrants its own analysis—here, the claim is simply that care is one virtue. As it stands, the virtue thesis is also noncommittal about definitions of its two relata and ambiguous about the nature of their relationship beyond a general identity claim. This identity claim can be spelled out in a number of ways, and we ought, in particular, to distinguish between stronger and weaker versions of the virtue thesis. According to a strong virtue thesis, care is just a virtue. A strong virtue thesis exhausts the set of conceptual categories into which care falls. Care is a virtue and nothing more. A weak virtue thesis, by contrast, claims that care is at least in one sense a virtue. A weak virtue thesis is open‐ended about the set of conceptual categories into which care can fall; care is a virtue, but it can also denote an emotion, a profession, an activity, and so on.
There is no reason at the outset to suppose that care ethicists are hostile to the virtue thesis per se. The common argument that care and virtue ethics are not equivalent does not entail any particular stance on the relation between care and the virtues qua moral concepts. The opponent of the virtue thesis who takes this route must argue that care and virtue ethics are distinct precisely because care is not a virtue in any way in order to rule out all of its variants, and this particular argument has yet to gain traction in the literature. Though they tend to oppose the strong virtue thesis, care ethicists seem to be generally open to something like the weak virtue thesis. Nel Noddings, for instance, deliberately leaves room in her account for a weaker variant of the virtue thesis. Like most care ethicists, she accepts that:
The virtue sense of caring is still significant. We do, after all, say such things as, “He is a caring person,” “They are a caring family,” “Nurses are more caring than doctors.” When we understand the relational sense, however, the virtue sense takes on new meaning. (Reference NoddingsNoddings 2002, 19)
On this point there is some consensus. Virginia Held too has done much to distinguish care from virtue ethics, but she also accepts that “to be caring is no doubt a virtue, [though] the ethics of care is not simply a kind of virtue ethics” (Reference HeldHeld 2006, 19). Here Noddings and Held represent a number of prominent care ethicists (see also, for example, Reference TrontoTronto 2013, 36). Many of care ethics’ most renowned proponents see no inconsistency in distinguishing care ethics from virtue ethics while simultaneously endorsing the weak virtue thesis.
This consensus alone is enough to warrant deeper excursions into the virtue thesis, but we ought perhaps to briefly note that the virtue thesis is one for which several arguments are already in circulation. In the passage above, Noddings argues that the virtue thesis simply enshrines moral common sense. To be caring is intuitively to possess a particular kind of character; to respond but also to be responsive. Other reasons for adopting the virtue thesis have also been explored in greater depth than I can offer here (see, for example, Reference Dillon and SnowDillon 2017). So the virtue thesis is also philosophically interesting for its apparent defensibility. But a number of questions now arise. How should we cash out the claim that care is a virtue? What relation, if any, does this establish between care and the other virtues? Does care add anything of value to virtue theory, and if so, what? Sections II–IV go some way toward answering these questions by way of an analysis of three different approaches to the virtue thesis.
II. Analogical Approaches
One way of framing care as a virtue is to equate it with one that is already established. Care's precinct overlaps with those of several other virtues to such an extent that they represent potential analogues. Love of some sort will be among the strongest candidates for such a tack, since care ethicists have modeled much of their moral advice on maternal or parental love (see especially Reference RuddickRuddick 1989 and Reference BaierBaier 1994, which speaks of “the natural virtue of parental love”). Admittedly, love is regarded most often not as a character trait but as a type or a trait of relationships, and is normally associated not with virtue but rather with emotions, desires, attitudes, and other mental events or states. Yet there are other sorts of love for the analogist to appeal to. Of these friendship and charity seem most favorable. Our contemporary moral lexicon restricts these virtues to smaller spheres, but among virtue theorists they still govern vast swaths of moral life—enough, perhaps, to be synonymous with care.
These analogues have already been met with some resistance, however. Held anticipates an analogy with charity and explicitly rejects it in The Ethics of Care: Personal, Political, and Global:
Some people suggest that caring is close to the Christian virtue of caritas, but caritas is equivalent to charity. Care, however, is not the same as charity—when we take care of our children we are not being charitable—and being caring is not the same as being charitable. Valuing care is entirely independent of any religious foundation, and is the stronger for this, since those not sharing a given religious tradition have few reasons to attend to arguments that appeal to that tradition. Understanding the value of care can be based on a universal experience of having been cared for and being able to engage in caring. (Reference HeldHeld 2006, 44)Footnote 1
But Held's argument against charity is not sufficient to rule out charity in toto. To begin with, theism is not necessary for charity; virtue theorists can and have defended nontheistic versions thereof. And though Held's other quarrel with charity in the passage above does not stand or fall with theism, it does not undermine a premise common to all accounts of charity. Held is right to suggest that even if we disentangle charity from its theological context and define it as a dispositional sort of universal goodwill/benevolence, the resulting picture of charity is incapable of encapsulating care in its entirety. We cannot characterize the relationship between a loving parent and a child as a generalizable love, a universal goodwill, or a manifestation of some broader regard for humanity.Footnote 2 But virtue theorists seem to have something rather different in mind when they address charity. Philippa Foot, for instance, defines charity as a virtue that “attaches us to the good of others” (Reference FootFoot 2002, 44), and this is perfectly consistent with pluralism about the kinds of attachment one can have, and about their relative weights. So Held's attack on charity successfully topples only one sort; the others remain potentially viable analogues for care.
What's more, even if we grant that charity and care are not analogous, it does not follow that care has not appeared at all in normative ethics in the form of a virtue. Indeed, a comprehensive study of Thomistic caritas will find it rooted in Aristotle's philia, possibly a more promising candidate for an analogue of care. Aristotle's tripartite theory of friendship is nuanced enough to encompass both relations with our near and dear and relations with our compatriots, and is also, at points, reminiscent of analyses of care. Identifying certain marks of friendship, Aristotle writes:
some people define a friend as someone who wishes and does what is good, or what appears to be good, for the sake of his friend; or someone who wishes his friend to be and to live for his own sake—this is the attitude of mothers towards [sic] their children, or of friends who have come into conflict. Others define a friend as someone who spends time with another and chooses the same things as he does; or someone who shares in the sorrows and joys of his friend—and this quality too is found in mothers in particular. It is by one or other of these characteristics that friendship is defined as well. (Reference AristotleAristotle 2004, 1106a2–10)
Here and elsewhere, Aristotle picks up on several major themes in care ethics.Footnote 3 He explicitly accounts for relations between mothers and children, and between friends (in the colloquial sense), two relations that take center stage in the ethics of care. He also conceives of a friend as “another self” (allos autos), thereby rejecting hard self–other distinctions on which atomistic accounts of human nature rely. Aristotle also emphasizes the import of interaction and/or cohabitation with philoi, a feature many care ethicists will find attractive, and Book VIII of the Nicomachean Ethics explicitly allows for philia between members of a political community, so Aristotelian philia is (debatably)Footnote 4 broad enough in scope to cover occupational and institutional cases of care.
This bodes well for philosophers who wish to construe care as an analogue of some established virtue, and indeed commentators have already noted some of these resemblances and used them to mount analogies between care and philia. Howard Curzer, for instance, has argued that Aristotle's philia amounts to an account of care. In “Aristotle: Founder of the Ethics of Care,” he writes:
Aristotle cannot be an ethics of care advocate without a concept of care, but Aristotle does have a concept of care. Although the Greek terms philēsis and its infinitive version to philein are typically translated as “love,” or “friendly feeling,” or “friendly affection” by Aristotle's translators, Aristotle uses philēsis and to philein to mean approximately what advocates of the ethics of care mean by “caring” and “care.” (Reference CurzerCurzer 2007, 221)
According to Curzer, Aristotelian philia falls well within care ethical orthodoxy. Like care, philia involves attending to needs, and it covers all of the paradigmatic cases of care identified by care ethicists. So there is already some momentum behind analogical approaches to the virtue thesis.
III. Supplementalist Approaches
Adopting an alternative strategy, some have argued that care is a novel virtue with its own distinctive set of excellences, skills, capacities, and so forth. Margaret McLaren addresses the relationship between care and virtue ethics in her “Feminist Ethics: Care as a Virtue” (Reference McLaren, DesAutels and WaughMcLaren 2001), concluding that care ought to be “fostered as a feminist virtue” (111) alongside virtues like compassion and empathy. Care is, on her view, incapable of standing as an independently viable normative theory, because its association with women perpetuates stereotypes about women and femininity, because of its apparent inability to provide meaningful advice beyond the private realm (due to its particularism), and because it is so often binarized and juxtaposed with justice—unhelpful commonplaces, in McLaren's eyes (109–10). McLaren's preferred approach is Aristotelian, and in defending the integration of care and virtue ethics, she recounts what she takes to be the features of Aristotelian virtue ethics that render it hospitable to care as a moral virtue. Care ethics and virtue ethics, she tells us, share a relational ontology, a sensitivity to particulars, and appreciation of the moral salience of social and political context, and she takes these commonalities to indicate that virtue theory can accommodate care. As McLaren sees it, care ethics ought to be treated as a subspecies of virtue ethics rather than an independent and comprehensive moral theory in its own right because virtue ethics is capable of accounting for care and avoids the pitfalls of standalone ethics of care. The most natural way to merge the two is to treat care as a virtue.
Raja Halwani, defending a similar take, defines care as an Aristotelian virtue, and though he relies on an indistinct account of care, it is clear that he believes care is a virtue with no existing analogue. On Halwani's view care is, like the other virtues, a character trait regulated by phronesis and necessary for human flourishing. Marshaling his evidence, he argues:
it is no violation of our understanding of caring to think of it as a deep trait of a person; or as enduring (a caring person is one who is liable to stay a caring one, everything else being equal); or as acquired: while we might have the capacity to care, caring properly is a trait acquired by training and good upbringing. Caring is also an excellence because, simply, it is a good trait to have. (Reference HalwaniHalwani 2003, 182, original emphasis)
Halwani justifies this last claim along eudaimonist lines, writing that it is “an obvious point to make that without proper care human beings cannot generally grow up to lead mentally and emotionally healthy lives” (183), and that “if intimate relationships are essentially characterized by caring, and flourishing is constituted by intimate relationships… then the necessity of caring to a flourishing life stares us obviously in the face” (183). On Halwani's account, then, care satisfies virtue's areteic standard because of its necessity for proper human development, and because it is partly constitutive of intimate relationships, which are themselves necessary for human flourishing.
Daniel Putman, too, argues that care is a virtue, citing three reasons for thinking so. The first is that it “increases the possibility of forming and maintaining a community” (Reference PutmanPutman 1991, 232). On Putman's view, care is a virtue that promotes cohesion in society, and serves to draw out the best in each person qua a member of various social units (232–33). Care is thus a virtue for all, but especially for teachers and those whose job it is to mold others into productive and caring members of society or to return them to that state. The second reason that care is a virtue for Putman is that it motivates caring agents to use their intellectual resources to act well (233). Care thus plays a role similar to the virtues of character in Aristotle's and Halwani's systems, in that it establishes certain ends as good or desirable and prompts the intellectual/calculative virtues to deliberate about particular sorts of action in particular contexts.Footnote 5 In doing so, caring also “opens up avenues for new experiences and development by the agent herself” (233). A third reason offered by Putman to treat caring as a virtue is that it allows us to achieve MacIntyrean internal goods (233–35). Appealing to Noddings's discussions of engrossment, Putman argues that the virtue of caring enables one to be “completely present to the practice, whether it be a conversation or an athletic event” (234). This “engrossment” is, on his view, essential if we are to appreciate the goods internal to caring relations and practices.
Putman is unclear about what exactly it means to be “present to” the object/relation of care, but Alasdair MacIntyre sorts internal goods into two categories, both of which are plausible here. MacIntyre separates the excellence of a practice's products/performance from “the good of a certain kind of life” (Reference MacIntyreMacIntyre 2011, 22). Care may be a virtue in that it enables one to excel at meeting needs, either by bringing about some good in the care‐recipient or manifesting some good in the caregiver. The life of a carer may also be a good life. Certainly there are elements that are dirty, unglamorous, and unpleasant, but philosophers like Eva Kittay and Sara Ruddick have shown just how much value there is to be found in a life of care. So there are several ways in which care may generate MacIntyrean internal goods.
IV. Bundling Approaches
Both of the approaches above treat care as a unitary virtue, albeit one that serves a diverse array of functions ranging from the detachment of care from a normative theory ill‐suited to feminist aims (McLaren), to the explanation of how political communities are maintained (Putman). But we can also distinguish approaches that treat care as a bundle of virtues. There are variants of this approach that do not assume a virtue thesis about care—it is possible to conceive of care as something other than a virtue while still treating it as a bundle thereof. And this is, in fact, an approach that care ethicists themselves have shown some sympathy for. Daniel Engster, for instance, identifies attentiveness, responsiveness, and respect as central “virtues of caring.” The virtues of caring are, in his words:
those qualities necessary for meeting the aims of caring. They are constitutive of caring in the sense that one cannot successfully achieve the aims of caring without them, or at least do so with any regularity. (Reference EngsterEngster 2005, 19; see also Reference EngsterEngster 2015, 19)
Yet Engster's definition of care is functionalist; he defines care as an activity or practice (Reference EngsterEngster 2005, 53–54), but one that cannot be performed successfully in the absence of certain traits. The aims of care, in turn, are “to help individuals to meet their basic needs, develop or maintain their basic capabilities, and live as much as possible free from suffering” (54) and these supply the criteria for successful care and are therefore the aims that necessitate virtue in caring agents. The first virtue, attentiveness, is necessary because successful caring requires us to notice when our charges are in need, and to respond appropriately in a way that satisfies the need. Responsiveness also serves a major corrective function, in that carers may be otherwise prone to ignoring crucial information and feedback supplied by the cared‐for, and respect relates to the recognition of the cared‐for as an agent “worthy of attention and responsiveness” (55) and “capable of understanding and expressing their needs” (55). This bundle of virtues is thus a necessary condition for, and is partly constitutive of, good care, but care is not thereby a virtue. Engster is not alone in taking this approach. Ruddick spends much of Maternal Thinking exploring the virtues of motherhood, including humility, cheerfulness, and forgiveness (she addresses all three together in Reference RuddickRuddick 1989, 72–75). Like Engster, Ruddick seems reluctant to conceive of care as a virtue, but she is open to the inclusion of virtues underneath care.
Not all bundling approaches reject the virtue thesis, however—some do treat care as a virtue. This particular kind of bundling approach I call the “cardinalizing approach.” The next section of part IV expounds upon this alternative to the analogical and supplementalist approaches, and the final section of this part briefly explains how cardinalizing approaches to care can be used to hybridize the two.
Cardinalizing Approaches
A cardinalizing approach to the virtue thesis assumes that the virtues can be nested within or subordinated to one another in some way. This assumption opens up conceptual space for notions of care as a cardinal virtue under which other virtues may be ordered, precedent for which appears in several strands of virtue theory. One of these is Platonic. Plato is well‐known for his conception of justice (dikaiosunē) as a cardinal virtue consisting in the proper functioning of the disparate parts of the soulFootnote 6 temperance, courage, and wisdom are, on the Platonic account, components of the virtue of justice because justice consists partly in the oversight of these parts by reason. A theory of cardinality also appears in Thomistic virtue ethics. Like Plato's, Aquinas's set of cardinal virtues includes prudence, justice, courage, and temperance. His bases for classifying them as such are also similar to Plato's: they represent excellences over particular spheres of human life, and encompass smaller virtues that preside over more specific areas (Reference AquinasAquinas 1912, I–II.61). There are thus a number of virtues spanning several strands of virtue theory to which care might bear structural similarity. We need not attempt to shoehorn care into the place of one of these. The point here is simply that virtues can be said, very generally, to contain or govern one another, and we can imagine a care ethicist borrowing from a number of figures in the tradition in order to generate an account of care like this, which retains its status as a virtue in its own right but also conceives of care as a bundle of other virtues.
Cardinalizing approaches to care are not unpopular among care ethicists either. Ruth Groenhout, for instance, seems to favor this view. Care ethics, she tells us, “centers on what sorts of people we must be in order to care properly for one another” (Reference GroenhoutGroenhout 1998, 176, emphasis added), which is just to say that character traits are its primary moral currency, and when reflecting upon care ethics’ roots in maternal experience, she writes that “a consideration of mothering practices [can] offer guidance in thinking about what sort of person I ought to be” (175, emphasis added). For this reason, Groenhout prefers “to say that Care resembles virtue, rather than is a variant of virtue theory” (172). She tells us unambiguously that there are “central moral character virtues that make up Caring” (175), and she lists altruism, empathy, and sociality as three of these. These are not, to be sure, the only character traits relevant to an assessment of caring agents in her view, but Groenhout sets most others aside because care ethics shares them with other moral theories (178–79). Thus, the primary difference between care ethics and virtue ethics seems, on Groenhout's view, to be that care ethics makes use of a different set of virtues, rather than different conceptual cornerstones. Groenhout views these three virtues as unique to care ethics, and premises her distinction between virtue ethics and care ethics not on differing foundational concepts, but on differing sets of virtues. By these lights, Groenhout appears to be committed to cardinalizing care.Footnote 7
Groenhout is arguably not an outlier in taking this position. Although Joan Tronto explicitly rejects the strong virtue thesis (Reference TrontoTronto 2013, 35–36), she has no quarrel with weak ones (36), and her discussions of attentiveness, responsibility, competence, and responsiveness in Moral Boundaries can be plausibly read as accounts of different virtues. She refers to these as “the four ethical elements of care,” to be distinguished from “the four elements of care,” which are:
caring about, noticing the need to care in the first place; taking care of, assuming responsibility for care; care‐giving, the actual work of care that needs to be done; and care‐receiving, the response of that which is cared for to the care. (Reference TrontoTronto 1993, 127)
It quickly becomes apparent in the passages that follow that the four ethical elements of care are dispositional excellences with respect to the four elements of care.Footnote 8 Tronto even goes so far as to juxtapose the ethical elements of care with what seem to be vices—attentiveness, for example, is opposed to ignorance. Groenhout and Tronto are not the only major care ethicists to take a cardinalizing approach to the virtue thesis. Noddings also treats certain virtues as integral components of (good) care, arguing that a set of virtues is both necessary and sufficient for what she calls “caring‐as‐virtue” (Reference NoddingsNoddings 2002, 124). So there is some support among those who endorse a virtue thesis for the thought that care amounts to a virtue that oversees the operation of other, more localized virtues.
Hybrids
Cardinalizing care is not necessarily a departure from the other two approaches we have discussed. If one believes that some existing cardinal virtue is analogous to care, then one could argue that there is nothing new to be added to the list of virtues, since all of care's normative content already exists in virtue theory. Similarly, one could argue for care as a cardinal virtue using a supplementalist approach alone. Such an account would argue that new virtues ought to be added to the list, and that those new virtues are governed by the virtue of care, itself a novel cardinal virtue. The cardinalizing approach to care could thus instantiate rather than challenge the other two approaches surveyed here.
Cardinalizing approaches can also be used to hybridize the analogical and supplemental approaches. A hybrid of these two approaches will claim that some care‐related virtues have already appeared in virtue theory, but that new virtues must still be recognized. This can take one of two forms, depending on whether one wishes to cardinalize care or to order care underneath a cardinal virtue. Either we maintain that some existing cardinal virtue equates to care, but that this virtue contains at least one hitherto unrecognized virtue, or we maintain that care is contained by an existing cardinal virtue but is itself a new virtue. In the latter case, no analogue for care exists, but at least some of care's normative content has already been accounted for under the precinct of an existing virtue. Given her acknowledgment of significant overlaps between care and virtue ethics and her claim that altruism, empathy, and sociality distinguish care ethics from virtue ethics, it seems fair to suppose that Groenhout endorses such a hybrid. And indeed this is an unsurprising position for care ethicists to take, given their apparent preference for bundling and the varying degrees of traffic seen by the virtues they typically list. Attentiveness, responsiveness, empathy, sociality, and other virtues in that vein are repeatedly cited by care ethicists as virtues partially constitutive of care,Footnote 9 but are not usually referred to as virtues in (Aristotelian) virtue theory. Other virtues, such as compassion, humility, courage, and honesty, are also commonly acknowledged elements of care, but are fixtures in the literature. So hybrids of the analogical and supplemental approaches look to be leading contenders among the various kinds of virtue thesis.
V. In Defense of the Virtue Thesis
In this part I raise and respond to three foreseeable objections to the virtue thesis. I begin with the thought that virtue as a conceptual category is insufficiently sensitive to gender, and is therefore ill‐suited to care's feminist aims. I then raise two apparent inconsistencies between the care ethical claim that care is natural and the standard Aristotelian picture of virtue. On the one hand, care's naturality appears inconsistent with the claim that virtue is habituated, and on the other it appears inconsistent with concepts of virtue as a corrective. I argue that none of these problems pan out, and this provides a solid foundation for an evaluation of the competing approaches laid out in parts II, III, and IV.
One reason to reject the virtue thesis is Maureen Sander‐Staudt's. Attacking Halwani's and McLaren's Aristotelianism, she maintains that interpreting care as a virtue is antithetical to certain feminist ends because “gender neutral definitions of virtue… prevent them from highlighting the gender sensitive distinction between care as a motive and care as an end” (Reference Sander‐StaudtSander‐Staudt 2006, 23). According to Sander‐Staudt, male or masculine virtue is typically associated with caring motivations (“caring about”) whereas female or feminine virtue is more often associated with caring practices (“caring for”). In defining virtue as a matter of both motivation and practice, she argues, the virtue thesis blurs a moral distinction that tracks gender and, for that reason, presents an account antithetical to feminist aims. Sander‐Staudt assumes a monolithic virtue of care that, in its generality, cannot be tailored to gender‐sensitive defects. Her “caring about” and “caring for” are supposed to delimit areas where we can, and qua members of a gender are more inclined to, fail to care as we ought. Yet Sander‐Staudt is not prescribing one sort of care for men and one for women. Sander‐Staudt (rightly) wants both men and women to care about and for others, so a morally ideal agent on her view would be caring in both senses. And if that is the case, then it becomes unclear why Sander‐Staudt's picture of care differs substantively from an appropriately compartmentalized virtue of care. Sander‐Staudt's morally ideal agents are still caring in a broad, amorphous sense, but they also exhibit excellences in particular areas of care—a portrait of care that, it seems, is just as tenable if we treat care as a unitary but subdivided virtue. The problem here applies only to particular theories of virtue.
Additional issues arise once we begin talking of care as a natural phenomenon. Care ethicists univocally assert that care is in several senses natural or innate. We are enmeshed in caring relations from birth and so we typically remain for the rest of our lives, and more than this, we are naturally disposed to nurture and maintain these relations. Consider, for example, Halwani's remarks above, and Noddings's comments on our inborn tendency to care:
[Care is] that condition toward which we long and strive, and it is our longing for caring—to be in that special relation—that provides the motivation for us to be moral… as human beings we want to care and be cared for. (Reference NoddingsNoddings 2013, 5–7)Footnote 10
In contrast, there are at least two ways in which virtue seems not to come naturally. First, virtues are supposed to be habituated—“virtues arise in us neither by nature nor contrary to nature, but nature gives us the capacity to acquire them, and completion comes through habituation” (Reference AristotleAristotle 2004, 1103a24–26). According to the dominant view, virtues are a product of a lengthy process whereby particular patterns of thought, affect, and action become engrained. Thus Aristotle concludes at the end of Book X of the Nicomachean Ethics that “the person who is to be good must be nobly brought up and habituated” (1180a). Second, virtues are supposed to be correctives. The inculcation of virtue is, on the dominant view, a process of setting our dispositions aright in areas where our actions, thoughts, and feelings tend to miss the mark.Footnote 11 If we were born perfectly just or kind, we would have no need of correction, and no reason to afford justice or kindness the special moral status of virtues any more than we afford that status to self‐love.Footnote 12 So one might also think that care's naturality rules it out as a virtue because our motivations and desires in the sphere of care typically do not need to be corrected, and therefore care cannot represent an excellence of character.
It is open to the care ethicist to reject the standard picture of virtue, but rather than jettisoning these claims about virtue, I think it can be shown that these inconsistencies are merely apparent. To begin with, natural inclinations to care are unproblematic as far as habituation and moral education are concerned. Virtue theorists do not deny the existence of natural inclinations or tendencies; indeed they have often deliberately left room for them by treating such inclinations and tendencies as integral to the formation of character and virtue. Julia Annas exemplifies this in Intelligent Virtue, writing:
we teach children to be fair and honest not by teaching them what they should do and then trying to interest them in having a new motivation to do this. Rather, we try to educate and form motivations that are present already. (Reference AnnasAnnas 2011, 10)
So the mere fact that we tend to be motivated to form friendships, to tell the truth, to share, and so on, does not entail that friendship and honesty are not habits we pick up and perfect. Rather, these inclinations will (in the best of cases) provide the loam from which virtues grow. Likewise, an inclination to care is not evidence that care is not habituated and cannot qualify as a virtue. Just the opposite—it unites care with virtues like friendship, benevolence, and justice. Like the other virtues, any natural inclination to care will invariably need to be refined during the process of enculturation. We do not know by default that some people enjoy surprises and others do not, or that one cares for sick people by taking them to the doctor.Footnote 13
A similar point can be made in response to the claim that care does not need correction. The claim that care is natural cannot be taken to entail the claim that good care is natural without an overly romantic picture of care. Caring relations, as many care ethicists have stressed, are difficult. It is all too easy to care too much or too little, to care for the wrong person, to care for the wrong reasons, and so on. Human beings generally seem prone to overestimating the urgency or severity of our own needs, and conversely, to downplaying or being insensitive toward the needs of others. We are all familiar, after all, with tropes such as the self‐absorbed miser who is blissfully unaware of the dire situations of those nearby, and we are familiar also with more mundane cases of failure to properly respond to the needs of the underprivileged, the suffering, the elderly—even, sometimes, when they are right under our noses.Footnote 14
Excessive responsiveness is also familiar. One does not have to look far to find well‐meaning caregivers who smother their charges and thereby do them harm, and indeed care ethicists have often cautioned us against just that. But to avoid these shortcomings, one must have a particular sort of character. One must be disposed to notice the need for care and to respond rightly in action, affect, and thought. One must be properly motivated by the existence of needs, and one's values must be so calibrated as to reliably meet them well. So without denying that we are naturally embedded in caring relations, it seems reasonable to suppose that caring well requires a corrective disposition. And this is particularly clear when we begin to move away from the usual paradigms of caring relations. Our inclination to romanticize care seems to wither entirely when we consider the ways in which we fail to care for the environment, for victims of domestic violence, for the impoverished, and for nonhuman animals; it is glaringly obvious that we fail to care as we should for those needs. Care may well be natural, but good care does not come naturally.
VI. In Defence of Analogism
Suppose some version of the virtue thesis holds water. An additional question now arises as to which of the three approaches above one ought to adopt. In this part, I defend the analogical approach as the most promising, despite the initial appeal of the other two. The first section contends that the cardinalizing approach, perhaps the most attractive at the outset, rests on mischaracterizations of the virtues subordinate to care. The second section argues that the supplementalist approach fails, not because the approach is philosophically inadequate, but because it assumes the failure of the analogical approach—an assumption that is, at least at this time, untenable. The third section argues that the analogical approach is not as unattractive as care ethicists have sometimes thought, and that it can be made to share one of the most attractive features of cardinalizing approaches.
Against Cardinalism
As noted above, the cardinalizing approach, and the bundling approach more generally, have garnered significant support from care ethicists. The primary reason for this seems to be that it neatly explains central elements of care; good care involves a set of interlocking excellences and is, therefore, a constellation of virtues rather than a single, undifferentiated virtue.Footnote 15 Here, however, I want to challenge the notion that care is best thought of as a cardinal virtue governing a set of subordinate virtues. I shall mount two arguments, the conjunction of which seems to me to suggest at best that we ought to suspend judgment on cardinalism, and at worst that cardinalism fails. The first is that many of the virtues identified by care ethicists may not in fact qualify as virtues. The second is that those subordinate traits that do qualify as virtues are not governed entirely by care, and thus do not map onto conventional understandings of cardinality.
One point to make against cardinalizing care is that some of its subordinate virtues may not qualify as excellences of character at all, because they can be possessed also by the wicked. Take, for instance, attentiveness. Attentiveness is more or less a disposition to notice particularities (especially minutiae) that need to be factored into practical deliberations if one is to care well (see, for example, Reference TrontoTronto 1993; Reference EngsterEngster 2005). People who are attentive notice the symptoms of need—a pained expression, an empty pantry, sunken eyes, and so on. That moral exemplars are attentive is without doubt; one can hardly care well if one is not paying attention. But we also seem to find this trait in the wicked. Those who are patently spiteful or hateful or manipulative are just as capable of recognizing the needs of those around them as good carers are. Indeed, what is often most harrowing about wicked individuals like Hannibal Lecter and Charles Manson is how very attentive they prove to be. But if one believes, as many philosophers and nonphilosophers do (see, for example, Reference BadhwarBadhwar 1996; Reference HursthouseHursthouse 1999), that the wicked cannot by definition possess moral virtues (either because the virtues necessitate one another, or because they necessitate moral decency), then attentiveness cannot be a virtue. Those committed to the claim that attentiveness is a sort of excellence are better off classifying attentiveness not as a moral excellence, but perhaps as a morally neutral perceptual capacity that can be put to good or evil uses. And one would likely have to say the same of virtues like competence and cheerfulness. These may well be good traits to possess in some sense, but they do not seem to be moral excellences insofar as the wicked can and do exhibit them. Again, it is entirely open to the care theorist to reject this picture of virtue, but at the very least it is not obviously wrong, and we require a reason to prefer an alternative picture.
This argument does not empty the set of virtues that could be governed by care, and care ethicists will likely complain that I have mischaracterized attentiveness. Attentiveness, it might be said, ought to be further specified; the virtue of attentiveness lies within the context of caring relations (thus Ruddick, for example, refers to attentive love). Sociopaths may be attentive, but because their relationships are not caring, their attentiveness falls short of virtue. Furthermore, even if we concede that attentiveness can be put to evil uses, the same appears to be true of courage, but most of us will balk at the idea that courage is not a virtue. The mere fact that particular traits admit of vicious iterations does not show that they cannot admit of virtuous iterations.
Even if we grant for the sake of argument that attentiveness is like courage in this respect and that virtues can sometimes serve wicked ends,Footnote 16 there is another point to be made here against care's cardinality. Those virtues that remain seem not to slot neatly into conventional understandings of cardinality. As it is usually understood, cardinality is a function of generality. Charity is a cardinal virtue because its sphere is abstract or broad enough to encompass smaller, more localized virtues. But this does not describe care's relation to its subordinate virtues. Care does not encompass responsiveness, empathy, or responsibility, because all of these virtues appear to fall also within the precinct of justice, kindness, and friendship. Kindness surely involves a sort of responsiveness to others, and one certainly cannot possess the virtue of friendship if one is not empathetic. Virtues like attentiveness and competence also seem to fall well within the ambit of practical wisdom. There is a long history in virtue theory of using perceptual analogies to illustrate practical or moral wisdom,Footnote 17 and what is attentiveness if not a sort of perceptual skill or disposition? So it seems that many, possibly even all, of the virtues typically ordered underneath care in fact belong to several virtues. Good care might involve the virtues, but the virtue of care does not seem to contain them. And this, of course, is inconsistent with theories of cardinality like that of Plato and Aquinas, which rest on a process of abstraction.
Against Supplementalism
I argue in this section that the supplemental approach too ought to be abandoned, at least in the interim. My reason for this is simply that the supplementalist approach assumes the failure of the analogical approach, and thus a burden of proof falls upon its proponent to prove as much—a burden that, hitherto, has gone unmet. The addition of care to the list of virtues in virtue theory assumes that care has not appeared already. To add care to a list of virtues, one of which already contains all of care's normative content, is simply to introduce a synonym for that pre‐existing virtue. So the supplementalist approach makes a meaningful contribution to the virtue theoretic tradition only if care has not been acknowledged in some form already (if, in other words, the analogical approach has failed).
In itself this is not enough to show that the supplementalist approach is redundant. But it is not at this point clear that the assumption of care's novelty is tenable. Care ethicists have sometimes argued, as we saw in part II, that particular virtues are poor analogues for care. And this will be part and parcel of any case against analogical approaches. But there are two methodological problems with the usual strategy for dismissing analogical approaches. The first is that they are rarely defended in much depth. Extant virtues tend to be given short shrift, acknowledged for their prima facie resemblance to care but quickly shelved in order to pursue other philosophical aims. Without dismissing those aims, this is a tack that invites rebuttals such as the one I offered in part II against disanalogies with charity. Philosophically robust disanalogies forthcoming, however, we have at best a relatively weak reason to think that any particular analogue fails. Greater attention must be paid to the virtues neighboring care if we are to have confidence in the need for an additional virtue of care. Perhaps more concerning, however, is the usual strategy itself. Arguments against analogies with charity do not rule out analogies with friendship or other kinds of love. Care ethicists must show at minimum that there is good reason to suspect that all initially plausible analogues fail. In‐depth studies of particular virtues will be integral to this, but the study of a single virtue, no matter how thorough, cannot justify the rejection of the analogical approach tout court. Pending such arguments, we ought to withhold judgment on the supplemental approach for its reliance on an undefended and nonobvious assumption. Until we are given reasons to doubt the analogical approach itself, rather than reasons to doubt particular instances thereof, there is no good reason to suppose that we need to add care to the list of virtues.
In Defence of Analogism
Many care ethicists will no doubt find these conclusions, that both the cardinalist and supplementalist approaches ought to be rejected in the interim, unpalatable, for they imply that analogism is in fact to be preferred. In this section I wish to make two points in defense of the analogical approach. One is that the analogical approach is just as capable of capturing care's complexity as cardinalism, and the other is that the care‐ethical preference for weak virtue theses mitigates certain drawbacks that dissuade care ethicists from the analogical approach.
One of the most attractive features of the cardinalizing approach is that it enables care ethicists to draw on a wide range of specific, localized virtues. Attentiveness, empathy, and other subordinate virtues are traits any good caregiver will possess because they are indispensable elements of good care. In rejecting a cardinalizing approach in favor of an analogical one, it might be thought that we have sacrificed this advantage. But this is not the only means by which care ethicists can draw on other normative concepts in care's neighborhood. Another takes the form of the “unity” or “reciprocity theses.” The unity and reciprocity theses denote sets of claims about the way virtues necessitate or imply one another. Strictly speaking, unity theses are the stronger—according to one popular understanding, they posit that the distinctions between different virtues are artificial (see, for example, Reference AnnasAnnas 1993, 73–74 and Reference TonerToner 2014). Virtue is really one single thing, artificially subdivided by our moral language. Reciprocity theses, by contrast, do not erase the boundaries between particular virtues, but they blur them by claiming that the possession of one virtue necessitates possession of them all. Usually, this is because acting well is at least sometimes impossible without a collection of virtues because each virtue's province overlaps with others’. Aristotle's is perhaps the quintessential statement of the reciprocity thesis. For Aristotle, what distinguishes “natural” virtues from “full” virtue is phronesis (“practical wisdom”) (Reference AristotleAristotle 2004, VI). Phronesis is a calculative or deliberative virtue that enables its possessor to act well. It does not deliberate about ends; that is the task of the virtues of character like courage, generosity, and temperance. Rather, phronesis takes those ends and deliberates about how best to further them. Phronesis does not, moreover, call on one virtue at a time. Instead, it produces what Foot calls “all‐things‐considered reasons” (Reference FootFoot 2001, ch. 4)—it takes input from all of the relevant virtues in order to guide us in any given scenario. In the mind of the phronimos, mercy can curb justice, kindness can encourage honesty, and courage can promote generosity. The virtues thus collectively render their possessor appropriately responsive to moral considerations in a way that no virtue could do alone.
Care can be plugged into a notion of reciprocity like this one to generate judgments like what the cardinalist hopes to offer. Take, for instance, courage. It seems reasonable to suppose that care and courage overlap—meeting needs sometimes requires us to face our fears and put ourselves at risk. Consider Huckleberry Finn's refusal to return Jim to Miss Watson. Toward the end of The Adventures of Huckleberry Finn, Huck finds himself faced with an apparent moral dilemma. He can either turn his friend, Jim, in and see Jim returned to Miss Watson, his owner, or he can continue to help Jim on his way to the free states. Huck believes that siding with Jim amounts to a kind of theft that is severe enough to condemn him to hell, but after a great deal of internal conflict he musters the courage to accept this (“all right, then, I'll go to hell,” he says defiantly [Reference TwainTwain 1885, 217]). This is an unusual case, to be sure, but the point generalizes—courage will often be relevant where care is concerned. And if the virtues are reciprocal in the way outlined above, then we can explain such judgments. Both virtues provide essential inputs for phronesis. Huck could not have cared well for Jim if he did not possess a measure of courage.
Reciprocity theses also offer us a set of mechanisms with which to dissolve aspects of the care–justice debate. If care and justice are virtues, then we can expect that the virtue of care will often curb the virtue of justice, so that punishments are meted out, for instance, with an eye to their impact on an individual's or community's well‐being, and that justice will sometimes curb the virtue of care, such that those holding public office, for example, are not nepotistic (in such cases, justice might incline us to say that it is not really caring to show partiality in this particular context) and so forth. The reciprocity theses may thus help to resolve debates at the normative level over whether justice or care take precedence by providing a means of showing that conflicts between them are merely apparent. Temperance too seems to be intertwined with care as a matter of common sense, especially since self‐care has seen significant uptake among care ethicists as a subcategory of care (see, for example, Reference EngsterEngster 2005, 66). So there is no shortage of virtues with which we want care to reciprocate. It seems, then, that analogical approaches can share in at least one feature that care ethicists have found attractive about cardinalism.
Opponents of analogism may concede this point, but attack the approach on other fronts. Another likely complaint is that the analogical approach threatens to subsume care ethics under the banner of virtue ethics, since care is construed as one virtue among many. Since the virtue ethicist's repertoire includes virtues other than care, the thought goes, its ethical theory is more inclusive than the care ethicist's and care ethics thus represents the study of one area within virtue ethics. But if this is a problem, it is one shared by the other two approaches. The supplementalist approach, of course, merely adds care to the list of virtues. It does nothing to expand care's ambit, and it is difficult to see how it could resist the implication that ethics is broader than mere care. The cardinalizing approach may seem to fare better here, since it appears to expand care's scope to include a set of other virtues. But merely cardinalizing care does not expand its scope—cardinalism, at least as we have discussed it here, does not stretch care to cover more moral terrain; it merely reconceives the elements of care as virtues. Care's scope is already set.
The problem may not be as severe as it seems, however. Most care ethicists have a complex and multifaceted concept of care. Care is a virtue, but it is also other things besides, including a foundation or precondition for virtue (or maybe a “meta‐virtue”). This, after all, is why care ethicists have typically not opted for stronger virtue theses. The standard position among care ethicists is that care is not to be taken as a virtue alone, because there are other understandings of care in play in their normative theory. And it is entirely possible, once we accept this, for care ethicists to view the virtue of care as one among many, but to add that all of the virtues are encompassed or accounted for by some other concept of care—possibly as a kind of relation. Indeed, care ethicists have sometimes claimed that care is “prior” to virtue in some sense (see, for example, Reference EngsterEngster 2007). Thus, some have argued that care can explain justice—two normative concepts that may be isolable when looked at as virtues, but overlap when looked at through other lenses. And this suggests that there are more fundamental concepts than virtue at work. Care ethicists can therefore explain virtues other than care by suggesting that another sort of care grounds them. In endorsing a weak version of the virtue thesis, care ethicists seem to avoid the worry that analogical approaches collapse care ethics into virtue ethics.
An Underexplored Avenue
Care's relation to virtue is a complex matter, and a full account thereof requires far deeper and far broader analysis than I have offered here. I hope at the very least, however, to have gone some way toward sorting out the various positions taken in this field, and to have offered reasons to prefer some over others. I have argued that care is commonly regarded as a virtue, and that we have good reason to treat this as a viable option. I have also suggested that those who wish to treat care as a virtue have at least three distinct options for doing so, and contended that the analogical approach is the most defensible of these at the moment. More work needs to be done here in order to fully explore and evaluate the options available to care ethicists, however. Care ethics has matured into a viable alternative to other normative theories, but much remains to be done in the way of conceptual clarification and detailing. Hopefully care ethicists will do more to flesh out their theory of virtue in the future.
Acknowledgements
Thanks to Cheshire Calhoun, Joan Tronto, Virginia Held, Richard Hamilton, Elizabeth Fraser, Sandra Lynch, and Blanca Ybarra for helpful comments on this paper. Thanks also to Hypatia's anonymous referees.
