Empathy with vicious perspectives? A puzzle about the moral limits of empathetic imagination

Abstract

Are there limits to what it is morally okay to imagine? More particularly, is imaginatively inhabiting morally suspect perspectives something that is off-limits for truly virtuous people? In this paper, I investigate the surprisingly fraught relation between virtue and a familiar form of imaginative perspective taking I call empathy. I draw out a puzzle about the relation between empathy and virtuousness. First, I present an argument to the effect that empathy with vicious attitudes is not, in fact, something that the fully virtuous person can indulge in. At least one prominent way of thinking about the psychology of the virtuous person excludes the possibility that the virtuous could emotionally apprehend the world in a less than virtuous way, and empathizing with vicious outlooks does seem to run afoul of that restriction. Then, I develop an argument that runs in the contrary direction: virtue in fact requires empathy with vicious outlooks, at least in some situations. There is reason to think that a crucial part of being virtuous is ministering effectively to others’ needs, and there is also reason to think that other people may need to be empathized with, even if their emotional outlooks are at least minorly vicious. Finally, I outline two different solutions to this puzzle. Both solutions hold some promise, but they also bring new challenges in their train.

1 Introduction

One question we can ask about the limits of imagination is a moral one: are there limits to what it is morally okay to imagine, and if so, what are they? This subject has attracted interest from a couple of quarters. There are treatments that deal with the ethics of fantasizing, especially sexual fantasizing, when the content of the fantasy itself seems morally problematic or, in John Corvino’s terms, “naughty” (Corvino, 2002).¹ The ethics of imaginatively engaging with works that offer morally bad perspectives is also a live topic in the philosophy of art.² The puzzle I will develop here is concerned with moral limits on a form of imaginative experience that has not been squarely at the center of either of these discussions. I will be dealing specifically with a form of imaginative experience I will call empathy. I use “empathy” to pick out a familiar kind of imaginative perspective taking that has two distinctive features. First, it is interpersonal: it involves imaginatively adopting the perspective of another person. While we can coherently speak of empathy with fictional characters, with the dead, and with possible future people, my examination will focus on empathy with living people with whom one actually interacts. Second, empathy is emotionally charged. As I will explain in further detail below, empathy involves experiencing an emotional response to an imagined scenario, an emotional response that matches or at least harmonizes with the emotion of the person with whom one empathizes.³

Empathy is commonly regarded as an important instrument of moral goodness.⁴ It has been argued that it is critically important for securing and sustaining moral motivation, that it plays a crucial role in shaping our understanding of others’ needs, and that it is a central means by which we form moral judgments.⁵ It seems natural enough to say, then, that a capacity for this distinctive form of imaginative experience will be required for moral virtue.⁶ And it might also seem like a relatively obvious and unobjectionable extension of that claim to say that the broader and deeper a person’s ability and willingness to empathize with others, all things being equal, the more virtuous that person is.⁷

This latter thought starts to look considerably less platitudinous, however, once we attend to the fact that the perspectives which present themselves as potential targets for our empathy are not always themselves morally innocent. As we go through life, we are presented with opportunities to empathize not just with straightforwardly decent attitudes such as joy at spending time with friends, or grief over the loss of a loved one, but also with those everyday attitudes that seem at least venially bad, such as creeping jealousy at a friend’s success, a temptation to selfishly keep the best goods for oneself, or a twinge of schadenfreude in the face of a rival’s failure. So we can ask: Is empathy with these less noble attitudes something that the virtuous person will characteristically engage in? Or is such empathy in fact incompatible with virtue? Is it perhaps the case that the morally virtuous person is actually more limited in what she can empathetically imagine than more mediocre folks are?

I will focus on drawing out a puzzle about the relation between empathy and moral virtuousness. First, I will present an argument to the effect that the fully virtuous person cannot empathize with vicious attitudes. At least one prominent way of thinking about the psychology of the virtuous person excludes the possibility that the virtuous could emotionally apprehend the world in a less than virtuous way. If that is right, the same sort of restriction should carry over into the imaginative domain. Empathetic emotions are concerned with scenarios we are imaginatively representing, rather than believing in, but that difference will not mean that the virtuous person is free to indulge in less than virtuous ways of emotionally picturing those scenarios. Then, I will develop an argument that runs in the contrary direction: virtue in fact requires empathy with vicious outlooks, at least in some situations. There is reason to think that a crucial part of being virtuous is ministering effectively to others’ needs, and there is also reason to think that other people may need to be empathized with, even if their emotional outlooks are at least mildly vicious. According to this second argument, then, it is a mistake to conceive of the virtuous person as specially imaginatively limited.

My principal goal is to bring this puzzle into view. However, in the final section of this paper, I will also outline two possible solutions to the puzzle. They both have promise, but they do introduce significant new challenges in their train.

2 Virtuousness precludes empathy with vicious attitudes: the argument

In this section, I will develop and examine an argument for two interconnected claims, namely that (1) a fully virtuous person will, because of their virtue, not be able to empathize with vicious perspectives, and (2) the more virtuous a person is (other things being equal), the more constricted their ability to empathize with vicious perspectives will be. By a “vicious perspective,” I mean: an outlook on the world or some portion of it that is morally bad because and insofar as it involves seeing as viable or attractive morally bad possible courses of action, because it involves not seeing as viable or attractive relevant morally good possible courses of action, and/or because it more generally involves a misrepresentation of the morally salient qualities of the object(s) of the perspective.⁸

I am using the label “vicious” quite broadly, here, in two different respects. First, I do not restrict application of the term “vicious perspective” to only the perspectives of people who are not at all attracted to good courses of action, nor to only the perspectives of people who go on to actually do bad things. A person could have a vicious perspective, in my sense, and yet find the strength to act well. For instance, if the possibility of shoplifting for the simple thrill of it looks mighty attractive to you, if it tantalizes you, and yet you force yourself to pay for your items nevertheless, your perspective counts as somewhat vicious by my lights despite your laudable ultimate choice. It might be more precise, not to mention more respectful of the Aristotelian tradition, to use the label “vicious, akratic, and/or merely continent perspectives” for the ways of apprehending the world that I want to pick out with the label “vicious”—I’m avoiding doing so only because it is rather clunky. Second, while the word “vicious” commonly connotes a significant degree of moral seriousness, I don’t restrict the class of vicious perspectives in this way. A vicious perspective might be gravely bad (for example: seeing one’s fellow human beings as nothing but disposable playthings fit for abuse), or it might be merely mildly bad (for example: seeing as attractive an opportunity to take more than one’s fair share of some trivial good). Between these two extremes there is a wide and well-populated range. In this paper, I will primarily focus on empathy with what I regard as mildly vicious perspectives. A diet of more extreme examples would, I think, risk steamrolling over some of the intuitions I will be concerned to draw out in the second half of the paper.

The argument for the twin conclusions of this section hinges upon two characterizations, one of virtuous perception, and one of empathy. Let me briefly present both, and then elaborate upon them.

• 1. Virtuous perception: A virtuous person relating to the world in a characteristically virtuous way will apprehend it in a particular emotional light. Specifically, she will not be blind to, indifferent to, or repulsed by considerations that recommend acting in a virtuous way. Nor will she be tempted by considerations that might, to others, appear to recommend acting in a way contrary to virtue. Those latter considerations will not even figure in her practical appraisal of the situation.

• 2. Empathy as emotional imaginative perspective taking: Empathy involves emotionally picturing the world or some segment thereof as if from the point of view of the target of one’s empathy. We succeed in empathizing just insofar as we manage to imaginatively recreate, first-personally, the other person’s emotional perspective on their situation.

2.1 Virtuous perception

The idea that virtue consists at least partly in a characteristic form of evaluative apprehension is one that has roots in multiple virtue ethical traditions.⁹ In some form or another, it continues to be widely endorsed today. One of the people who has done the most to develop and defend this idea in our time is John McDowell, and I will be focusing on McDowell’s particularly influential articulation of what distinctively virtuous perspectives involve.¹⁰

McDowell offers his neo-Aristotelian account of distinctively virtuous vision as an alternative to a popular two-factor model of practical thought. On this model, an intention is the product of two elements: first, a set of beliefs about the facts that obtain in a situation, a set of beliefs which is both motivationally inert and equally available to folks all across the spectrum of virtuousness, and second, a desire which is distinct from that aforementioned set of beliefs, and which may vary independently from it.¹¹ That two-factor picture is misleading, the thought goes, because there is no such thing as a neutral apprehension of the facts that obtain. Depending upon where you are on the path to virtue, different features of a given situation will show up as relevant, and different action possibilities will call out to you as viable. McDowell particularly emphasizes that the perception of the virtuous is different in kind to the perception characteristic of less than virtuous people. To bring out the particularity of virtuous perception, McDowell introduces the concept of silencing. He tells us of a virtuous person who recognizes a particular requirement to act: “[T]he virtuous person conceives the relevant sorts of situation in such a way that considerations which would otherwise be reasons for acting differently are silenced by the recognized requirement” (1998, p. 92).

Silencing is not an entirely transparent notion. Here are a few possible ways of specifying what silencing involves:

• (1) The virtuous person does not see any considerations which would counsel acting in a way contrary to virtue as reasons.

• (2) Considerations which would counsel acting in a way contrary to virtue are accorded no weight at all in the virtuous person’s deliberations about the thing to do.

• (3) Considerations which would counsel acting in a way contrary to virtue are, in the case of the virtuous person, excluded from having any motivational pull.¹²

These characterizations focus on what might at first look like different conditions: the condition of appearing as a reason, the condition of figuring in deliberation, and the condition of engaging motivationally. For McDowell, though, these conditions actually cannot be teased apart from each other. The overall picture that emerges is this: considerations which would counsel acting in a way contrary to virtue do not get weighed in the agent’s deliberation because the agent does not see them as providing any reason for action at all. And saying that the agent does not see them as providing any reason for action just is another way of saying that the agent’s inclinations are not engaged by those considerations. Of a given consideration which would counsel acting in a way contrary to virtue, namely a prospective enjoyment that cannot be pursued consistent with a given requirement of virtue, McDowell writes: “[H]is [a virtuous person’s] clear perception of the requirement [of virtue] insulates the prospective enjoyment…from engaging his inclinations at all. Here and now, it does not count for him as any reason for acting in that way” (1998, p. 91).

To illustrate: Suppose that I, a morally mediocre person, go to the market and find a pile of delicious pears marked at a high price but unguarded by their inattentive seller. The idea of simply pocketing a pear might occur to me as an attractive possibility. I might resolve not to do it, or try to pay the option no mind—but still, there it is, exerting some pull. But that act of pocketing simply won’t show up for my virtuous counterpart as an option, because the virtuous person will not construe this unguarded pear stand as an opportunity for underhanded gain, even for a moment. Instead, she will maintain a clear, unshakable view of the situation as an opportunity to acquire something good through fair compensation. As Phillipa Foot writes in support of this conception: “The fact that a man is tempted to steal is something about him that shows a certain lack of honesty: of the thoroughly honest man we say that it ‘never entered his head,’ meaning that it was never a real possibility for him” (2002, p. 11).¹³

Importantly, McDowell and his fellow travelers do not hold that the virtuous person will be simply insensible to the pleasure of getting something without having to pay for it. The virtuous person can, compatibly with her virtue, recognize and even attend to this fact: if the circumstances were different, the ends that are now not at all motivationally or rationally “live” for her might have been worth pursuing. She can recognize as attractive the possibility of securing a free pear, were that possibility to arise under relevantly different conditions (a pear giveaway!). But that does not mean that she can, consistent with her virtue, see nabbing the pear as at all attractive under the conditions that actually obtain.

Virtuous apprehension of the world is sometimes glossed in terms of emotional outlook. I don’t want to insist that all evaluative apprehension, virtuous or otherwise, must be understood as emotional. Nor do I want to insist that all emotions must centrally involve evaluative apprehension; the category of emotion may be too diverse for that. But it nevertheless seems attractive and appropriate to refer to emotion when describing the special outlook of the virtuous person, because we can say of at least of a broad and important class of emotions that they centrally involve evaluatively charged ways of seeing. Take, for instance, the emotion of grief. Suppose I’ve just lost a loved one. What will my grief involve? Well, I feel leaden and exhausted. There is a dull ache in my chest, and tears spring to my eyes unbidden. My thoughts return again and again to all the plans we had made together. The everyday pleasures I normally would enjoy now feel like so much dust in my mouth, and the only thing that looks attractive is crawling back into bed. This emotion is composed of complex motivational, somatic, and attentional features. But there is something that unifies these components, such that they all count as elements of grief. And I think it is plausible that the unifying factor is the perception of the death as an awful loss, one that calls out for tears, solemn attention, and retreat from the world.

If it is true of virtuous people that they have particularly clear, undistorted evaluative perceptions, ones that get things right, and if it is true that emotions are plausibly understood as syndromes that typically centrally feature evaluative perceptions, then it stands to reason that virtuous people will characteristically have emotions that reflect, in an unusually accurate and unclouded way, the actual evaluative facts on the ground, and they will not have emotions that involve clouded or inaccurate evaluative perceptions.¹⁴ So we can say, for instance, that the really virtuous person will not feel jealousy, if it is true (as it seems plausible to say) that jealousy involves seeing the possibility of hatefully undermining a rival as attractive, and if it is also true that hatefully undermining a rival is not in fact a morally good course of action.

Of course, not all theorists fully embrace the view of virtuous vision I’ve sketched out here. Some worry that, in refusing to count individuals who experience temptation as virtuous, McDowell and those who agree with him make virtue too rare a thing.¹⁵ Others object that the view does not sit well with their sense that, at least in tragic cases where the right action will also bring grave harm, the virtuous should feel emotionally torn.¹⁶ The view is not without its attractions, though. First, it does feel like there is something right about the thought that the virtuous person is distinguished by clarity of moral vision, and that part of that clarity is a kind of singularity: morally bad possibilities simply do not get a look in. The proposal that virtuous vision is singular and clear yields a highly intuitive consequence in cases like the unguarded pears, even if it feels less obviously well equipped to handle tragic or quasi-dilemmatic cases.¹⁷

What is more: one hallmark of Aristotelian ethics is the distinction between the truly virtuous and the merely continent person. Both are said to act rightly, but the virtuous person’s action is set apart by a particular form of ease or grace. McDowell’s view of virtuous vision allows for a sharp and straightforward characterization of the difference between the merely continent and the truly virtuous. Because considerations which would counsel acting in a way contrary to virtue do not show up to tug at the will of the virtuous person, other possibilities do not cloud, crowd, or distort her apprehension of the thing to do. Conversely, we can say with McDowell, the merely continent are characteristically torn between different ways of seeing their situation, attracted to different courses of action that are simply unthinkable to the truly virtuous. Because she is emotionally buffeted by these different attractions, the continent person’s choice of right action is hard won, whereas the virtuous person’s choice is made without struggle and exhibits (in McDowell’s words) a sort of “sublimity” (1998, p. 91). The ability to differentiate between these two sorts of character is widely recognized as a valuable element of nuance in our thinking about what it takes to be good and who we ought to emulate.¹⁸ Virtue theorists who reject the claim that virtue entails singularity of moral vision themselves admit it thereby becomes difficult to retain the distinction between continence and virtue.¹⁹

2.2 Empathy as emotional imaginative perspective taking

We’ve reviewed an influential and attractive characterization of the virtuous person as apprehending the world in a particular clear emotional light, one that is distinct from the way that other folks can see it. Now let me turn to my description of what empathy involves. Empathy (as I understand it here) necessarily involves a form of imaginative “transportation.” In empathizing, one imaginatively occupies the other’s position and considers their situation from that vantage point. And, when efforts to empathize succeed, one secures an acquaintance with another’s emotional outlook on the world (or at least some part of it) from the inside.

There are earlier hints of it in Hume, in Montaigne, and perhaps even, if one squints, in Plato, but Adam Smith is the one who first properly develops the notion of empathy as imaginative emotional perspective taking. His detailed description of how empathy operates has had a long philosophical afterlife, I think in no small part because it feels true to life. Smith’s characterization of empathy seems to pick out a kind of process that we recognize as familiar. To get a grip on what empathy in my sense might involve, then, let us follow along with Smith’s treatment of one of his own examples, a case in which an observer comes across a mother grieving the recent death of her child. The empathetic process begins with the immediate recognition that the mother is deeply sad. Her facial expression, gestures, and posture all serve to make that clear to the observer. Having noted this emotion, the observer seeks information about the mother’s situation, and begin to “enter into” that situation (1982 [1759], p. 12). This “entering” is a matter of imaginatively re-centering one’s perspective, an activity one can facilitate by telling oneself a particular sort of story. Were I the observer, I might set the scene for myself by imagining that I have lost a beloved child, that I had fully expected her to outlive me and had counted on our having many years together, that the death has occurred quite recently, and so on.

Once we have re-centered our perspective through this sort of imaginative effort, Smith says, “[a] passion arises in our breast from the imagination” (1982 [1759], p. 12). Importantly, this is not the same as saying that we merely think of the emotion we might have in the given situation. We may think of questions like “how would you feel?” as invitations to empathize, and sometimes they may be exactly that, but not insofar as they are invitations to engage in a predictive exercise that can be done in a dispassionate, third-personal way. Empathy is not a cool predictive endeavor. When I empathize, I redeploy my own emotional sensibilities, directing them at the situation that I am imaginatively inhabiting, and that I believe mirrors the actual situation of the other.²⁰ So, I allow my thoughts to be directed in the ways characteristic of grief. The imagined loss absorbs my attention, while other matters fall away from view. My emotional processing of the imagined situation may also involve a somatic response. I might experience a sinking feeling in my stomach, or my eyes might well with tears. The apparent isomorphism between the original grief of the mother and my empathetic response gives us strong reason to conclude that when we succeed in empathizing, we do not merely guess at what emotion we might have in such a case. Nor do we imagine that we are feeling some emotion. Rather, when we empathize, we actually have an emotional experience, one that to some extent corresponds to the other’s original emotion.

Crucially, the empathizer retains an awareness that her empathetic emotion is oriented not toward a circumstance that actually obtains in her case, but rather toward a thought, one that corresponds to what the other believes her situation to be.²¹ That awareness has two consequences. First, it provides a check on the kind of emotional expressions or behaviors we will exhibit. Thought-directed emotions don’t prompt action in the same way that their belief-directed counterparts do. Empathetic grief at the thought of losing a child will not move us to wail and rend our clothes, because we retain a firm grip on the fact that our child is not actually dead. And second, that awareness naturally invites us to complete the process of empathizing by forming a belief about how our thought-oriented emotion compares to what we believe to be the original emotion of the other. Often, we simply assume that our thought-oriented emotion mirrors the other’s original emotion (“I feel awful grief at the thought of the loss; she must feel awful grief, too.”). But we might also notice that our emotion fails to match the other’s apparent emotion in valence, intensity, or orientation. Smith suggests that we should judge ourselves to have succeeded in empathizing just insofar as we are able to actually match the other person’s outlook along these various dimensions (1982 [1759], p. 16). So, for instance, if you feel completely anguished by your loss, but I can only work my way into feeling mildly annoyed at the thought of losing a child, we can say that my empathetic efforts have not been very successful.

Now, to draw the two parts of our analysis together. Successfully empathizing with another requires experiencing an emotion that is the thought-directed analogue of the other’s original emotion. Emotions typically centrally involve the presentation of their objects in a particular evaluative light—as lovely and deserving of delight, or as awful and deserving of dread, and so on.²² So, empathizing with someone typically entails emotionally apprehending the object of their concern (or, rather, the imaginative analogue of that object) in the same evaluative light that they do. We have also posited, with McDowell, that the virtuous have a distinctive way of seeing the world, a special sensibility, that is not shared with the merely continent, the akratic, or the classically vicious. Their evaluative apprehensions, including their emotional apprehensions, will be systematically different from other people’s, because they will be uniquely accurate, clear, steady, and singular. It would seem to follow, then, that the virtuous and the non-virtuous will struggle to empathize with each other across the perceptual divide.²³

One might at this point object that the fact that the empathizer is operating in an imaginative context has not yet been sufficiently taken into account. Perhaps if we do properly factor it in, we can forestall the conclusion that the virtuous and the non-virtuous will not be able to empathize with each other’s emotions. One could propose that while contra-moral considerations cannot show up for the virtuous person when she is responding to a situation she believes is actual for her, her virtue will not at all be compromised if when she merely (knowingly) imagines a scenario, she is attracted to non-virtuous courses of action available within that scenario. The problem, though, is that we would then need a non-ad hoc justification for thinking that the assessment of virtue should be very differently affected depending upon whether the person believes the given scenario is one that is actual for them or not.

Now, there is one prima facie reason to think this that stands out to me: we do not act on our emotional evaluative apprehensions when they are a response to what we take to be imagined scenarios. A foolish, angry perception of ruinous revenge as an attractive option is neatly insulated from generating bad behavior when we experience it as a response to a scenario that we are self-consciously imagining, but if we actually believed that, say, our friend had betrayed us, we might act on that angry perception.

If we thought that our virtue depended upon whether we in fact act badly, then it would make sense to say that the virtuous are free to experience all sorts of contra-moral evaluative apprehensions within an imaginative context, since those apprehensions won’t affect their behavior. But it is worth underlining that this response will not be adequate by the lights of McDowell and his fellow travelers’ conception of virtue. As I’ve mentioned, one of the major benefits of that conception was supposed to be that it is able to count the truly virtuous and the merely continent as unalike in virtue even though they both consistently act in accordance with virtue.

Interestingly, McDowell and those sympathetic to his account have actually suggested, in their characterization of virtuous vision, that the perspectives of the truly virtuous are in some sense inaccessible to the non-virtuous, not available for insider access even through strenuous imaginative effort. McDowell suggests that the choices of the virtuous will be comprehensible to the non-virtuous insofar as those choices are driven by “independently intelligible” desires; the virtuous and the non-virtuous share some basic desires in common, and for the non-virtuous those shared desires “constitute possible points of entry for an outsider trying to work his way into an appreciation of a moral outlook” (1998, pp. 83–84). But, importantly, McDowell holds that those coincidences will “take an outsider only some of the distance toward a full understanding” of the virtuous and their choices, because so much of the virtuous person’s outlook cannot be captured in terms of desires the non-virtuous can be expected to share (1998, p. 85).²⁴ What virtue ethicists have not highlighted, in their efforts to call attention to the gulf that separates the virtuous and the non-virtuous, is the plausibility of the parallel claim that the virtuous will be similarly hampered when it comes to imaginatively inhabiting the perspectives of those who are considerably less morally excellent.²⁵

It is important at this point to stress that empathy is not all-or-nothing. It comes in degrees. We can think of a quite deep empathy, one which captures more of the details of the situation and secures a more precise match with the original emotion of the target of empathy (I empathize with your stubborn, itching hankering for those particular pears, a longing specifically sparked by and centered upon their forbidden status), or a coarser sort of empathy (I empathize with strongly desiring some good even though that good it not permitted to me), or, at the extreme end, a really very vague sort of empathy (I empathize with hankering after something). I am not claiming that a person who is halfway decent will, in virtue of that fact, be unable to at all empathize with other people’s vicious attitudes. Rather, the thought is that increasing virtue translates into a diminishing ability to empathize deeply or precisely with vicious emotions.²⁶ Presumably, even a merely modestly virtuous person will not be able to summon up a feeling of yearning when she tries to imagine doing evil for evil’s sake. Someone who is still further along the path to virtue will find herself at a loss when it comes to seeing as attractive the act of taking something that she is not permitted. And as for someone who is so far down the path to virtue that she doesn’t feel the force of any sort of temptation to do other than what is right? Well, it looks like she will have reached a point where even quite coarse empathy with large classes of non-virtuous but very ordinary human emotions will be out of reach.

The thought that our empathetic abilities can be quite dramatically constrained by our virtue is perhaps surprising in light of the longstanding tendency to treat a broad and deep capacity for empathy, understood as emotional imaginative perspective taking, as a mark of great virtue. But the fact that a thought is surprising is not, in the end, a decisive strike against it. Perhaps, one might think, we should hold both that the highly virtuous are very constrained in their ability to empathize with the rest of us, and that this is in no way regrettable or concerning from a moral point of view.

I do not want to dismiss that possible conclusion out of hand, but I do want to raise some doubts about it. Let me turn to a first apparent reason to worry about the thought that the virtuous will be limited in their ability to empathize with the non-virtuous. I don’t think that this worry is as concerning as it first appears, but spending a moment with it will help us to get clearer on what exactly is and is not available to the virtuous person. We might think that the argument I’ve given so far implies that the virtuous person will suffer from the following sort of morally significant handicap: ordinary people’s choices will be less predictable to the virtuous person than they will be to her less excellent counterparts.²⁷ I have in mind one of the great clichés of detective fiction: the investigator exploits his own inner demons, his darker impulses, in order to effectively imaginatively channel others’ bad emotional perspectives, and thereby anticipate the villain’s next move. But virtuous people will not share those dark sensibilities, on the sort of view we have been considering, and so they won’t be able to engage such sensibilities empathetically in order to make an accurate prediction. This could be a problem because it seems like an ability to accurately predict what other people will think and do can be critical to identifying the morally best action. And folks who struggle to identify the morally best action are, by hypothesis, hampered in their virtue.

It is important, however, not to overstate or mischaracterize the gaps in understanding that virtue might entail. Virtuous people need not be naïfs who are simply taken aback, perpetually surprised, by others’ vicious attitudes and choices. There is nothing to bar a virtuous person from acquiring extensive third-personal knowledge about typical patterns of vicious thought, perception, and behavior, such that she can become quite adept at guessing what stupid, petty, or downright terrible things the rest of us are likely to do. If McDowell and others are right about virtue, and if I am right about what empathy involves, the virtuous person simply won’t be able to effect that prediction via an imaginative emotional “inside track,” in the way that the hard-bitten detective with a dark streak does. Now, it may be that this “outsider’s” approach to predicting vicious behavior is more laborious and unwieldy than our detective’s is. A simulative approach to predicting behavior and action can be remarkably mentally economical, and managing without this sort of inside track may well take up considerable attentional resources.²⁸ Still, if the aim of accurate prediction is important enough to acting well, we can expect the virtuous person to apply herself to it with alacrity, and to develop relatively sophisticated predictive powers as the result of her efforts.²⁹

The ability to predict non-virtuous people’s choices accurately does not seem to be beyond reach of the very virtuous in principle, although it may be a hard thing for virtuous people to secure in practice. But this thought about accurate prediction points in the direction of a distinct concern that has me in its grip. The worry is also about a gap in the virtuous person’s understanding of others, but a gap of a different sort. I will now turn to this other argument that without a capacity for empathy with vicious outlooks, a person will be missing out on something of positive moral importance.

3 Virtuousness (sometimes) requires empathy with vicious attitudes: the argument

I’m going to consider an argument for the claim that the virtuous person will characteristically empathize with vicious outlooks, at least in some situations. It has two premises.

• 1. Effective ministry: A virtuous person characteristically effectively ministers to others’ needs.

• 2. Need for empathy: Because being empathized with offers relief from a particular form of suffering, we sometimes need to be empathized with, even in cases where our own emotional outlook is at least venially vicious.

I think that both of these premises are intuitively attractive. Let me start with Effective ministry.

3.1 Effective ministry

There is widespread agreement among both ancient and modern ethicists that the virtuous person is distinguished from the rest of us more mediocre types not only in virtue of her particular dedication to morally good general aims, but also in virtue of her unusual ability to non-accidentally succeed in achieving those aims.³⁰ The virtuous person is no bungler. She is not just very concerned to be kind (to take just one example of a virtuous general aim). She is also alert to opportunities for kindness and keenly sensitive to the many factors that will affect what counts as kindness in the given circumstances. She will register those features and respond to them adroitly, nimbly avoiding the kinds of inappropriate intrusiveness or condescension that less skilled folk with ambitions to be kind may lapse into.

The virtuous person is exquisitely attuned to moral reasons of all sorts, and facts about others’ needs constitute one important source of moral reasons. By “others’ needs,” I mean those goods which are required for their flourishing.³¹ Flourishing is a notoriously difficult concept to characterize completely, but we do not need a fully general theory of flourishing to be able to say some compelling things about what is normally required for human flourishing. Strong candidates for necessary components of or conditions for (most) humans’ flourishing include: taking joy in the exercise of our capacities for creativity, learning, friendship, and self-determination; engaging in meaningful and rewarding activities; and being recognized for both our dignity and our lovability. So, in addition to the goods we require for simple survival, it seems reasonable to say that our needs generally include (at least) companionship, love, recognition, and adequate opportunities to learn, to invent, and to choose. We can also pick out a set of “negative” needs, given considerations about the things that get in the way of flourishing. It is widely thought that the enemies of flourishing include a mixture of material and psychological ills. Ill health, poverty, and lack of liberty plausibly constitute obstacles to flourishing, but so do boredom, insecurity, depression and loneliness.³² Insofar as these adverse conditions prevent us from flourishing, we need to be free of them.³³

A virtuous person will recognize others’ genuine needs, see them as providing reasons for her, and respond to those needs with alacrity and finesse. In ways direct and indirect, the virtuous person will strive to remove barriers to others’ flourishing.

With that link between virtue and responsiveness to need established, it is time to look at the second premise. I will lay out the argument that one genuine need a person can have is the need to be empathetically understood, even when her emotional outlooks are less than virtuous.

3.2 The need for empathy

My defense of this latter claim begins with a methodological point. Susan Wolf writes of philosophical analyses of friendship that “[i]t is hard to give sense to debates about the value of friendship in the absence of considerations about what features of life give people joy and comfort and keep depression and despair at bay” (2007, p. 167). To generalize the thought: debates about the value of human social institutions, relations, and practices should take seriously observations about what seems to make us happy and what seems to make us suffer. I will adopt this guideline: if we seem to suffer from the absence of a particular sort of treatment, then there is at least prima facie reason to think that that sort of treatment is in fact valuable and contributes to our flourishing.

Adam Smith writes about our valuation of others’ empathy with us: “Nothing pleases us more than to observe in other men a fellow-feeling with all the emotions of our own breast; nor are we ever so much shocked as by the appearance of the contrary” (1982 [1759], p. 15). Of the more particular case in which someone is swept up in a negative emotion, Smith goes so far as to say of that person that “[h]e longs for that relief which nothing can afford him but the entire concord of the affectations of the spectators [that is, empathizers] with his own. To see the emotions of their hearts, in every respect, beat time to his own, in the violent and disagreeable passions, constitutes his sole consolation” (1982 [1759], p. 22). Smith is being more than a little dramatic, here, but I still think he has hit upon something right. But what, exactly, is the nature of the consolation that others’ empathy might be said to offer us?

Part of our interest in being empathized with may sometimes have to do with a desire to have others agree with us. We want others to imaginatively emotionally apprehend things in the same light that we do in part because we want them to agree that this is actually the right light in which to see things, and because getting them to emotionally apprehend things in the same light may be a first step in securing their endorsement of our outlook. But there is more to our concern to be empathized with than that. After all, we do sometimes crave others’ empathy even in cases where we disapprove of our own emotional outlook as ultimately incorrect. So, for instance, we may believe of our own twinge of envy at a competitor’s success that it is not, in fact, the right way to feel. We are in this case not looking for others to endorse our feeling. And yet: that does not stop us from wishing, at least sometimes, that someone would empathize with our envy!

If a desire for endorsement is not (or not always) behind our interest in being empathized with, then what is? To answer this question, it is helpful to return once more to basic questions of what is involved in having an emotional experience. I’m going to suggest that the first-personal experience of an emotion affords a direct appreciation of our emotion’s intelligibility, at least in cases where two further conditions to be specified below are satisfied. This is true even for emotions that we do not endorse. And, much of the time, what we are really seeking in other people’s empathy is just a direct appreciation of the intelligibility of our own emotions.

I’ve already said that emotions can typically profitably be understood as centrally involving a presentation of the world or some part of it in a particular evaluative light. Let’s take a look at an emotion that is a good candidate for being at least mildly vicious in order to get a grip on the kind of appreciation I have in mind. Suppose I have just settled in for a much-anticipated TV and popcorn session when I get a call on the phone: it slipped my mind, but I had promised that I would help a friend move her things to her new apartment. I feel frustrated! The prospect of lounging on the couch still looks eminently attractive to me, while the thought of carrying through on my freely given promise leaves me utterly cold. I feel a kind of itchy irritation at the thought of having to haul myself out of the house in order to help. My thoughts turn to all the many things I have already done for my friend (conveniently skimming over all the things she has done for me)—what sacrifices I have made! And here is just one more to add to the pile! My friend’s request now looks unduly onerous rather than reasonable. It feels like a little dragging of feet and muttering under the breath are called for. In sum, my frustration presents a decidedly filtered view of the situation as a whole.

Because I am not entirely swept up in the emotion, I may conclude that my frustration is not all-things-considered correct—I know that this isn’t how I ought to see things. But even if I believe that I shouldn’t be seeing things this way, my frustration nevertheless makes sense to me, in that it reflects the situation’s apparent evaluative features. After all, the couch looks tempting, and the good deed looks like an unreasonable burden. When we recognize that our frustration reflects the situation’s apparent evaluative features, the seeming presence of those features makes our emotion itself look correct. The situation looks vexatious, and so seeing it as vexatious seems right, even if we know that this is not, in fact, a correct evaluative apprehension. Let us say that our emotion is intelligible to us insofar as we register it as reflecting the apparent evaluative properties of the given situation.

It is not the case that we grasp all of our emotions as intelligible, all of the time. First, emotions that do not centrally involve conscious evaluative perceptions will not be intelligible to us in this sense.³⁴ And second, we appreciate our emotions’ intelligibility only when we reflectively attend to them.³⁵ But when we reflectively attend to an emotion we are feeling, one that centrally involves a conscious evaluative perception, our emotion will be intelligible to us even if we do not endorse it.³⁶

Earlier, I suggested that empathy involves experiencing an emotion that is the thought-directed analogue of the other’s original emotion. We can combine that with the thought I’ve been developing here about the connection between experiencing an emotion and appreciating its intelligibility to yield the following suggestion: when I empathize with another person’s emotion, I also secure a first-hand appreciation of the apparent appropriateness of that emotion.³⁷ That suggestion is, I think, a plausible way of spelling out what we mean to pick out with fuzzy but evocative phrases like “seeing where you are coming from” or simply “getting it.”

Importantly, a first-hand appreciation of the apparent propriety of an emotion is not available to me if I find myself unable to see things in the relevant emotional light, despite my best empathetic efforts. Suppose you see the suffering of your adversary as amusing. It delights you and looks worthy of celebration. If, when I imagine confronting the suffering of an adversary, I feel saddened and can find nothing in that suffering that even looks like it calls out for anything but solemn regret, then I will not be able to regard your delighted emotion as responsive to the situation’s apparent features. I can know that schadenfreude is a normal emotion to feel in such circumstances. And I need not be at all surprised by your schadenfreude: I could recognize that it is consistent with your preferences and commitments.³⁸ But still, I will not see for myself how the suffering of a rival seems to invite schadenfreude rather than sadness. Let’s call what I would be missing, that first-hand appreciation of the intelligibility of another’s emotion, humane understanding.³⁹

Is humane understanding something that we need? Does it contribute to our flourishing?

Well, we can observe that being humanely understood does very much matter to us, at least some of the time and in some contexts. Interestingly, we seem to value humane understanding non-instrumentally. We can bring its value for us into view by thinking about cases where others fail to humanely understand us. Consider the case where I am frustrated at having to leave my couch to do a good deed. I may choose to keep my feeling to myself, but I might also be inclined to share the story of my frustration with others, my friends, for instance. And in doing so, my motivation may be to seek their humane understanding. My frustration is eminently intelligible to me, even if I don’t ultimately judge that it is morally correct. And I may just feel a need for someone else to appreciate that intelligibility firsthand.

Now imagine how disappointed I might feel if my effort to secure my friend’s humane understanding were met with this response: she cannot work her way into seeing the comfort of the couch as at all attractive in my circumstances, because, after all, there was compelling moral reason to get up and go. And, relatedly, she cannot work her way into perceiving the invitation to carry through on the promise as frustrating; for her, the thought of an opportunity to do good cannot show up as anything other than steadily and thoroughly attractive. What she can offer, based on her sophisticated third-personal knowledge of how less virtuous people tend to think and behave, is no more than the assurance that my frustration is not surprising. It’s what she would expect from folks at roughly the same stage in their progress toward virtue as I am. My emotion is not, to her, humanely understandable. It is merely predictable.

I think, in a circumstance like this, I would not be content with having my predictability registered. What I crave is for my friend to imaginatively get down in the trenches with me, appreciating firsthand how my situation seems to call out for the emotional response I am having. It can be painful to feel that our outlook has not been appreciated in this way. We will not, unless we are unduly churlish, blame someone who cannot humanely understand our vicious emotion because they cannot experience contra-moral considerations as attractive. But we may be disappointed nevertheless: there is a sort of intimacy that that person cannot afford us, an intimacy whose absence is experienced as a kind of hurt.

I have suggested that we are sometimes pained or discomfited by others inability to humanely understand our emotional outlooks, even in cases where we acknowledge our own outlooks to be at least mildly vicious. It will be recalled that I earlier adopted this general rule: if we seem to suffer from the absence of a particular sort of treatment, then there is at least prima facie reason to think that that sort of treatment is in fact valuable and contributes to our flourishing. I am nudging us toward the conclusion that not receiving the sort of humane understanding which empathy affords is a source of suffering.

Now, a one-off disappointment that our friends cannot humanely understand our frustration at having to leave the couch might seem pretty far off from the kind of suffering that actually impedes flourishing. But the threat to flourishing that not being humanely understood imposes starts to look more serious once we consider cases in which our non-virtuous perceptions are concerned with more serious matters. So, suppose I confront a situation in which I must choose whether to expose a sloppy manufacturing process at my plant, one which may be putting many people’s lives at risk. But if I do what I in fact know to be the right thing, I will risk losing a number of comforts which I have enjoyed up to this point, including my nice salary and my warm relations with my colleagues. Imagine, further, that while I recognize that this is an opportunity to save others, I am still powerfully attracted to the possibility of staying shtum, and thereby protecting my very nice life. Torn between the need to save lives and the possibility of holding on to the life goods I have in hand, I am deeply anguished. Now, my anguished outlook in these circumstances qualifies as a vicious perspective, according to the broad characterization of vicious perspectives with which we began, because it involves seeing as viable or attractive a morally bad course of action (staying silent). A virtuous person would not experience the choice as agonizing, because for her there really would be no choice, but I myself experience deep conflict.

I think my already bad situation would be made significantly worse if, when I sought a confidante’s empathy with my agony, I were met with this sort of regretful response: “Well, since you do want to know the truth, I’m afraid I just can’t get into experiencing the possibility of saving my own skin as at all attractive in these circumstances. When I imaginatively put myself in your position, the only thing that looks at all appealing is the thought of exposing the truth.” I suppose I can only really speak for myself, but in these circumstances I think I would experience the discovery that my outlook is beyond my confidante’s empathy as seriously distressing, not just mildly disappointing. If I were already struggling with a feeling of isolation, her admission that she cannot empathize, no matter how sincere or gently framed, would redouble my sense of painful loneliness.⁴⁰

I am not trying to persuade you that the suffering sometimes involved in not being humanely understood always entails that we have an actual need to be humanely understood, one whose satisfaction will contribute to our flourishing. Sometimes, it may in fact be better for us if we are effectively frozen out of others’ humane understanding altogether. So, for instance, it might be best for us if our truly monstrous emotions are treated as beyond the pale, not at all candidates for empathetic engagement. If we are the right sort of person, after all, that treatment might just be the thing that pushes us to reform our emotional outlook. But I still think we should be hesitant to say that humane understanding is not something we ever need, or something we need just insofar as our emotions are actually virtuous, and no more. The idea that it will be best for us to be shut out of humane understanding unless and until our outlook is virtuous may treat too lightly the pain of not being understood.⁴¹

In this section, I’ve made the case that virtuous people characteristically minister effectively to others’ needs. I’ve also tried to offer some reason to think that, qua potential targets of empathy, we can suffer in virtue of not receiving empathy’s characteristic form of understanding, humane understanding, from others. And furthermore, I’ve suggested that this suffering can be something we need to have relieved. Taken together, these thoughts suggest that virtuous people ought, in keeping with their character, to be able to offer empathy to people even when those people’s perspectives are vicious. Not always, of course, for empathy will not in every case be needed or even wanted.⁴² But some of the time, still.

4 Paths forward

I’ve now made the case for the second of the two opposing arguments concerning the relation between being virtuous and empathizing with vicious perspectives. Both arguments have their appeal, but taken together they yield an incoherent picture of what virtue involves. To resolve this puzzle, which claims should we be willing to give up or to modify? I will briefly consider two answers to this question. The solutions I consider are only two possibilities among many. Still, they strike me as particularly appealing because they do not require us to surrender either the thought that we sometimes need for our vicious emotions to be humanely understood, or the thought that the virtuous have a distinctive way of emotionally apprehending the world. Here are those solutions, in summary:

• (1) Virtuous multiplicity: The virtuous person does characteristically apprehend the world in a particularly clear emotional light, such that she does not see contra-moral possibilities as attractive. This mode of apprehension (call it her “home sensibility”) is the one she uses to guide her own action and planning. But other less virtuous sensibilities are also first-personally available to her, and she can slip into them in for the purposes of empathizing without compromising her virtue.

• (2) Division of moral labor: Virtuous people are limited in the emotional outlooks they can humanely understand. However, it does not necessarily impugn a person’s virtue if they cannot respond to everyone’s real needs. That inability just means that there is an important moral role for non-virtuous people who can (in virtue of their own moral imperfection) meet others’ empathetic needs.

Consider first Virtuous multiplicity. Some virtue theorists, discouraged by the demandingness of McDowell’s conception of virtue, have suggested that the virtuous person need not always be wholehearted. In difficult situations where real goods morally must be foregone, she can be torn in her emotional apprehension of her own circumstances.⁴³Virtuous multiplicity is not the same as the proposal that the virtuous person may sometimes be emotionally torn. That latter suggestion could not resolve the puzzle I’ve presented, because it would not provide us with the resources to explain how the virtuous could empathize with, for example, a bit of eminently human schadenfreude. The person who feels schadenfreude, even if they are sheepish about it, cannot readily be cast as emotionally torn between various real and important goods.

Rather, Virtuous multiplicity suggests that the virtuous have one sensibility they use in navigating the world, and that sensibility exhibits the kind of clarity and sublimity McDowell associates with distinctively virtuous vision.⁴⁴ But, in addition, the virtuous have alternate sensibilities available to them, ones they can also inhabit first-personally. Perhaps these alternate available sensibilities are just the ones that served as the virtuous person’s home sensibility at previous stages in their moral development. Virtue theorists generally agree that we are not born apprehending the world in perfectly virtuous ways. One might therefore wonder: even if the current home sensibility of the virtuous exhibits the kind of clarity and singularity we have been discussing, can’t the virtuous harness their memory of their past experience in order to imaginatively slip into seeing things in a less morally refined way?⁴⁵ We could think of this move as akin to reverting to a language that was once our mother tongue, but that has since been supplanted by another in our ordinary thought and conversation. Perhaps deliberately re-adopting an old jealous outlook for the purposes of empathizing is no more problematic than is resurrecting one’s childhood French. More ambitiously, we might even allow that the virtuous can shift into sensibilities that never functioned as their home sensibility, if they are sufficiently mentally flexible.

What are the prospects for the proposal that we can emotionally apprehend the world in less than virtuous ways without impugning our virtue, so long as we are not deploying our home sensibility?

This solution to the puzzle must walk a tightrope. The alternate perspectives must be ours enough that we can picture the world through them, whilst also being not ours enough that deploying them does not compromise our virtue. A first question to confront is whether we actually are capable of genuinely inhabiting sensibilities other than our home sensibility, such that we actually emotionally apprehend things in their characteristic light. Some theorists do treat it as obvious that when it comes to empathetic imagining, we can borrow emotional outlooks that are not our own.⁴⁶ That assumption has been met with powerful challenges, however. Peter Goldie, for one, forcefully argues that there is exactly one sensibility we can actually bring to bear in both imaginative and non-imaginative contexts, and that is our actual, singular, current sensibility. We can reason about other sensibilities in a third-personal way, but we cannot deploy them first-personally (Goldie, 2011).⁴⁷ It is at least true that our previous home sensibilities often seem broadly inaccessible now. We are bemused by our own past passions, and while we can give some account of our past attitudes (“Of course I adored pop punk, I thought I was a total rebel”), we often still find ourselves totally unable to see the object of our attitudes in the same light we once did (“Now it just sounds silly, I can’t find it at all cool any more”). We would need to investigate whether this observation generalizes to the point that it renders Virtuous multiplicity psychologically implausible.⁴⁸

Even if it is psychologically realistic to maintain that the virtuous have first-personal access to multiple distinct sensibilities, Virtuous multiplicity also faces this further question: just how plausible is the claim that so long as one is deploying one’s home sensibility, one’s virtue is not at all impugned by empathy with vicious perspectives? That claim seems to permit too much: it implies, for instance, that virtuous people can feel delighted at the thought of murdering a child, so long as that delight is part of an empathetic effort and does not at all figure in their home sensibility. This consequence is difficult to accept, but to avoid it, one would need to explain why some empathetic efforts are not compatible with virtue even if they do not implicate one’s home sensibility.⁴⁹

Virtuous multiplicity merits further exploration, but it also faces real challenges. Some of these are neatly avoided by Division of moral labor. This latter resolution of the puzzle does not require us to modify the picture of the virtuous person’s distinctive psychology that we developed in Sect. 2. Rather, it involves rethinking what we expect of the virtuous. In Sect. 3, I argued that the virtuous are experts at ministering to people’s needs. We might tweak that widely held commitment, and say instead that virtuous people are in general expert at ministering to others’ needs, but that there is at least one real need that they cannot meet, thanks to their virtue. That doesn’t mean they are not moral paragons, it just means they cannot do it all, morally speaking.

If we think that from a moral point of view, it is good for people’s needs to be met, then it will turn out that it is also a good thing that the less-than-virtuous are not solely surrounded by the perfectly virtuous. We can accept that it is good for there to be people who have the clarity of moral vision unique to the virtuous, too. But those of us who are less far along the path to virtue can offer something morally important that our more virtuous counterparts cannot, according to Division of moral labor, because the scope of what we can empathize with is different.

Division of moral labor may draw some support from the fact that in non-ethical domains, divisions of labor are frequently beneficial. And there is something tempting about the thought that it might be morally best for (some of) us to be a bit bad.⁵⁰ But accepting Division of moral labor would mean giving up some popular and well-entrenched virtue ethical commitments. For one thing, we would have to surrender the widespread view that the really virtuous only fail to meet needs when factors external to their own character intervene.⁵¹ The proposal also threatens the common and not unappealing tenet that the best response to a given need will always be just that response that the maximally virtuous person would give.⁵² Given its destabilizing potential, then, Division of moral labor is also not a solution we should adopt incautiously.

I have now provided some sense of the further questions raised by my two proposed solutions to this paper’s central puzzle. Whether or not these solutions prove tenable, I hope to have at least shown that the ethics of empathetic imagination are more complex, and more worthy of philosophical attention, than we might have thought. An adequate account of empathetic imagination’s moral status must acknowledge how empathy matters to those on its receiving end, while also grappling with worries about how our imaginative activities bear on our moral character. To arrive at a coherent view of the moral limits on empathy for vicious perspectives, we may have to rethink some quite basic assumptions and intuitions about virtue, imagination, and their relation to each other.⁵³