When Love Means Letting Go: Individual Autonomy and the Right to Die

Photo by Christopher, Creative Commons

By Dr. J. David Vellman
Professor of Philosophy and Bioethics
New York University

Kant argued that suicide is immoral when committed for the purpose of escaping from unhappiness. I have tried on one or two occasions to reconstruct Kant’s argument, by offering a particular interpretation of the Formula of Humanity, which says that a person has a value that makes him an end in himself.[2] The statement that a person is an end, I interpret as expressing the fact that we ought to care about some things for the person’s sake, by caring about them out of concern for him. A person is an end in the sense that he is that for the sake of which — out of concern for which — some things are worth caring about. This conception of how a person can be an end yields an argument against escapist suicide, I argued, when it is combined with a particular conception of a person’s good, which is what the escapist attempts to serve by cutting his life short.

I borrowed the latter conception from Stephen Darwall, who contends that a person’s good consists in what it would be rational to care about for the person’s sake in the sense of caring about it out of concern for the person.[3] Since this structure of concerns is the one that explains how a person can be an end, in my view, I suggested that a person’s good stands to the value of a person in a relation analogous to (though of course distinct from) that of means to an ordinary end, the former member of either pair being worth caring about for the sake of, or out of concern for, the latter. I then argued that escapist suicide entails a practical irrationality analogous to that of sacrificing an end for the sake of the means to it.

Such an irrationality is committed, for example, by people who grub for money. Money has value only as a means to happiness (let’s assume), but money-grubbers make themselves unhappy in the pursuit of money, thereby sacrificing happiness for the sake of something that is valuable only for its sake. A person makes a similar mistake, I argued, if he sacrifices himself for the sake of something that is valuable only for his sake by committing suicide to promote his own good. In either case, one thing (money, an end to unhappiness) is preferred to another (happiness, the person himself) even though it is worth caring about only out of concern for that to which it is preferred.

This analogy is open to two objections. Let me start with the one that I know how to answer.

Creative Comnmons

The answerable objection goes like this. The reason why money is valuable for the sake of happiness is that it is instrumental to producing happiness, but the reason why relieving someone’s unhappiness is valuable for his sake is not that it’s instrumental to producing him. So whereas destroying happiness defeats the purpose of money, destroying oneself doesn’t clearly defeat the purpose of ending one’s unhappiness. Ending one’s unhappiness is good for oneself in a sense that doesn’t entail its having a purpose at all: its ultimate end is not a purpose but a person. What, then, is there to be defeated?[4]

The answer to this objection is that it misinterprets the analogy between a person and an ordinary end. Of course there is no instrumental relation between ending a person’s unhappiness and the person himself, but the instrumental relation between money and happiness is not what makes grubbing for money irrational anyway. What makes it irrational to seek money at the cost of happiness is that money is worth wanting only out of a desire for happiness. Seeking money at the cost of happiness thwarts the desire out of which money is worth caring about to begin with. The instrumental relation of money to happiness explains why the one is worth caring about out of concern for the other, but the irrationality in the case is generated by the resulting relation between these concerns, one of which depends on the other but is served in such a way as to frustrate it: the desire for money is served in such a way as to frustrate the desire for happiness, on which it depends. A similar relation obtains between ending one’s unhappiness through suicide and the value of the person himself: the one is worth caring about only out of concern for the other. A similar irrationality will be generated, then, if killing oneself thwarts the concern out of which ending one’s unhappiness is worth caring about, to begin with.

But now comes the second objection, which is not so easily answered: How does killing oneself thwart the concern out of which ending one’s unhappiness is worth caring about? An end to one’s unhappiness is worth caring about, I claim, out of concern for one’s value as a person. But how is that latter concern frustrated when one pursues an end to unhappiness by means of suicide? This question becomes especially pressing in light of Kant’s view about a person’s value and its proper mode of appreciation. According to Kant, a person is a self-existent end, which is to be valued as it is, given that it exists, rather than as a thing to be brought into existence. And the mode of appreciation proper to such an end is respect, which is an attitude of self-restraint, inhibiting us from violating the person’s autonomy.

Well, killing a person does seem to violate his autonomy, to say the least. But when the victim is also the killer, his killing may be an exercise of his autonomy, too — so how can it qualify as a violation? Maybe escapist suicide, at worst, is a case of autonomy violating itself, a case toward which the attitude of respect must be, at worst, ambivalent. In that case, suicide would not exactly thwart the concern that underlies its own motivation.[5]

Surely, though, the argument was already in trouble once respect for autonomy was introduced. Respect was introduced in the role of that appreciation for a person’s value out of which it makes sense to care about the person’s good. But respect is simply the wrong attitude for that role. Rather than sort out whether suicide is more expressive than destructive of autonomy, let us consider alternative attitudes to which the motive for escapist suicide might more plausibly be subordinate.

Wikimedia Commons

Darwall proposes that the response out of which things are worth valuing when they are valuable for a person’s sake, is an attitude that he calls “sympathetic concern” for that person. This proposal will be vacuous, of course, if sympathetic concern must be defined as a concern for the person’s interests: to define a person’s interests as comprising whatever is worth valuing out of concern for his interests would be tautologous. But Darwall argues that sympathetic concern can be identified without being defined, since it is a natural kind of affective response, which can be singled out by paradigm instances.

Yet I suspect that sympathy, like respect, is ill-suited to serve as the concern in relation to which a person’s interests should be defined. Although sympathy is not the same as empathy, it is an empathic response — a variant of empathy — and so it focuses on the feelings of its object. What sympathetic concern for a person disposes us to care about will therefore tend to be the state of his feelings, and the resulting conception of a person’s good must consequently have a bias toward hedonism. While I don’t think that the very concept of a person’s good should rule out hedonism, I don’t think that it should rule in its favor, either.

Moreover, what would make sense to care about out of sympathetic concern for a person is not necessarily what we would judge to be in that person’s interest. For example, parents who think that the welfare of their child requires them to administer punishment or harsh medicine may be inhibited from doing so precisely by sympathetic concern. In order to do what’s best for the child, they may have to overcome their sympathetic impulses — which suggests that sympathy is not always a reliable guide to the well-being of its object.[6]

Suppose that the parent punishes the child, saying, “I’m doing this for your own good.” The parent would be unlikely to elaborate by saying, “So, you see, I’m doing it out of sympathy for you.” They both know that insofar as he feels sympathy, the parent is acting in spite of it. If he offers any elaboration on the claim to be acting for the child’s good, he is more likely to say, “So, you see, I’m doing this because I love you.” Thus, if the child’s good is that which is worth caring about out of some attitude toward the child, then the relevant attitude is not sympathy but love.[7]

Maybe this emendation to Darwall’s conception of well-being can yield an emendation to the argument that I rested upon it — my argument for a rational obstacle to escapist suicide. Fashioning such an emendation will be my aim here.

If a person’s good is to be analyzed as that which is worth caring about out of love for the person, then the relevant form of love must be carefully distinguished from the romantic or sexual emotion that goes by the same name — the love of falling or being “in” love — since romantic or sexual love is largely possessive, even self-seeking. The same goes for various other attitudes that tend to accompany love, such as attachment. We can feel attached to people without loving them, and we can love people to whom we are not especially attached.

Creative Commons

There is a sense of the verb ‘to love’ that denotes a kind of solicitous behavior that characterizes the familial relationships with which the emotion is conventionally associated. The “loving mother” or “loving husband” of standard obituaries was loving in this sense — actively attentive, affectionate, caring. Many philosophers assume that the emotion of love must be the attitude that naturally motivates such behavior, hence a benevolent form of affection. There certainly is such an attitude, and there is nothing wrong with calling it “love”.

Yet benevolent affection is an emotion that we can recognize animals as manifesting toward their young; it is also felt by children for their special toys, by gardeners for their flowers, and by philatelists for their stamps. This emotion is unlikely to reveal anything that we don’t already know about the interests of these beloved objects. The philia of philately is not, I think, an emotion whose nature will help us to understand what is best for stamps. The philatelist does indeed take loving care of his stamps, but he does so under the guidance of some antecedent conception of what it is for stamps to be in good condition. He cannot simply love his stamps and let his heart be his guide.

Yet there is an emotion, also called love, which is indeed a guide to the interests of the beloved. This emotion is present in many loving relationships, where it can be difficult to disentangle from benevolent affection, but it can be absent from such relationships as well. People can take loving care of companions or wards with genuine feeling that nevertheless amounts to no more than fondness, because it falls short of that fiercer emotion that can only be called love. Conversely, they can feel the latter emotion for someone without being thereby moved to treat him lovingly, as becomes especially clear in relationships carried on at arm’s length. Students can love a teacher — or patients, a doctor — without having any inclination to cuddle or coddle him; there can be love between colleagues who would never presume to take care of one another; and even loving friendships can be characterized by a formality that rules out intrusions into one another’s lives. Finally, what I have described as the fiercer and more compelling form of love can coexist with, can indeed give rise to, the very opposite of benevolent affection, in the form of hostility or even hate. You can want to hurt someone you love, and both your love for him and your desire to hurt him can still be wholehearted.

This last possibility is not incompatible with my suggestion that love is a guide to the interests of the beloved. The suggestion is not that love necessarily involves a desire for the good of the beloved or — to correct the order of analysis — that a person’s good is that which loving him would necessarily involve a desire for. Rather, the suggestion modifies Darwall’s account by saying that a person’s good is that which is worth caring about, or which makes sense to care about, out of love for that person. And the acknowledgment that love needn’t involve a desire for the beloved’s good is perfectly compatible with the claim that it provides a natural motive or reason for such a desire. Or — to correct the order of analysis once again — the acknowledgment that what is wanted by a lover need not be good for the beloved is compatible with the claim that what is worth wanting, or makes sense to want, out of love for the person is indeed what is good for him.

The question is what love makes it appropriate or rational to care about. More specifically, the question is what can constitute a person’s good by virtue of being that which loving him makes it appropriate or rational to care about — a question to which the answer must be more substantive than “the person’s good”. Here I want to adopt a suggestion from Connie Rosati, who has criticized Darwall’s sympathy-based account of well-being as follows:[8] “When we appreciate the value, as it seems to us, of a work of art, we endeavor to preserve it in its valuable condition. Likewise, when we appreciate, as it seems to us, the value of a person, we seek to preserve the person in her condition as the valuable being she is. Just what attitude might capture this idea without itself involving concern for a person’s welfare, I will not venture to guess. But care or sympathetic concern seems not to be it.”

Clearly, loving someone is a way of appreciating his value. Rosati’s analogy therefore suggests that what it makes sense to care about out of love for a person is the preservation of the value or the valuable condition to which love is an appreciative response

The next question, of course, is what sort of value or valuable condition is appreciated by love. Some philosophers would understand this question as equivalent to the question what we love people for; and they would answer it by citing the various qualities for which we love our parents, spouses, children, and friends — their fine characters, their fetching looks, their sense of humor, or their shared history with us. But I have difficulty believing that the value to which we respond in loving people is conferred on them by the miscellaneous qualities that we profess to love them for. As many philosophers have pointed out, we often profess to love people even for their flaws; but surely loving them doesn’t give us reason for wanting those flaws to be perpetuated.

My view is that loving a person “for” some quality is not a matter of responding to a value conferred on him by that quality.[9] Rather, I claim, the qualities for which we love someone are qualities that show us or remind us or symbolize for us that value to which we respond by loving him. They are signs of his value, not its substance. To find someone’s crooked smile endearing is not to find him more valuable in virtue of smiling crookedly; it is rather to find the smile emblematic of what is valuable about him, which would still be valuable even if his smile were straight.

The value that makes people proper objects of love is a topic that seems to be surrounded by paradox. In loving someone, we treasure him as special and irreplaceable; yet we love more than one person, and we believe that almost everyone is worthy of being loved by someone. If everyone is worthy of being loved, however, then everyone is worthy of being treasured as special, and so everyone must be special — in which case, there must be nothing special, or at least nothing especially special, about anyone. There’s a paradox for you. Here is another. We love our own children above all other children, and yet we don’t honestly believe that our children are more valuable than others. How can we be so selective in appreciating a value that we acknowledge to be virtually universal?

Niko Kolodny has argued that our reasons for loving a person lie in our relationship with the person — his being our parent, spouse, child, or friend. What makes someone special, according to this view, is that he is special to us by virtue of sharing a particular relationship with us. Everyone can be special, then, because everyone can be special to someone, with whom he shares a similar relationship.

Although I find much to admire in Kolodny’s discussion of love, I find his thesis unpersuasive. We probably cannot love people with whom we aren’t acquainted, but I think that we can indeed love acquaintances with whom we have no significant relationship — love them at first sight or from afar. We can also love people whose relationship with us we do not value at all, as when divorcing couples still love one another despite looking back on their marriage as a disaster from day one. People who are estranged from their parents or siblings generally say that refusing to have any further dealings with these people does not entail loving them any less.[10]

At a more fundamental level, I find love as Kolodny conceives it to be self-centered, since it responds to a value that the beloved has, not because of what he is in himself, but because of what he is to us. Love so conceived is a response to a fundamentally egocentric value, a value that others have in virtue of the part they play in our lives. I doubt whether love is really so egocentric.

In responding to a related objection, Kolodny says:[11]

Nevertheless, it may be replied, even if we are not valued only instrumentally on the relationship theory, we are still valued extrinsically and nonfinally. Our relatives value us, it might be said, in the way one might value a now useless pen that once belonged to Winston Churchill: as an [intrinsically] worthless object that merits a certain response only because it is associated with something of final worth. This analogy, however, is misleading in at least two respects. First, our relatives do not deny that we are finally valuable. However else they view us, they view us as persons, and hence as beings with final value. Second, it is not the case that our relatives’ valuing us is optional, given that they value their relationships to us. According to the relationship theory, their valuing us is constitutive of their valuing their relationships to us, in the sense that they cannot respond appropriately to the value of their relationships to us without also valuing us. Admiring Churchill, by contrast, does not require fetishizing his possessions.

Here Kolodny concedes that when people love us, they regard us as valuable in ourselves, because they regard us as persons. Yet according to Kolodny, this appreciation of our value as persons is distinct from people’s love for us, which is based instead on our relationship with them.

Iris Murdoch, novelst / Wikimedia Commons

In my view, appreciation for someone’s value as a person is not distinct from loving him: it is the evaluative core of love. I do not mean that love is a value judgment to the effect that the beloved has final value as an end in himself. Love is rather an appreciative response to the perception of that value. And I mean “perception” literally: the people we love are the ones whom we succeed in perceiving as persons within some of the human organisms milling about us. Only sometimes in this throng do we vividly see a face or hear a voice or feel a touch as animated by the inner presence of a self-aware, autonomous other — a person who is self to himself, like us. Iris Murdoch says that “love is the extremely difficult realisation that something other than oneself is real”[12] — that someone other than oneself is real, I would say, in the case of love for a person. A sense of wonder at the vividly perceived reality of another person is, in my view, the essence of love.

It is because the reality of other persons is not directly perceptible to us that we love people for their faces and voices — and even their flaws — which somehow alert us to the presence of another inner life alongside our own. Not every smile strikes us quite forcibly with the presence of the person behind it — not one in a million — which is why we love but one in a million, perhaps for his smile. We do not see most people for what they are, even if we know what they are. And what are they, that they are worthy of being loved, except other persons like ourselves?

This explanation for the selectivity of love still doesn’t solve the puzzle of everyone’s being worthy of love and hence worthy of being valued as special. Even if each of us can value only a few people as special, the thought that everyone deserves to be so valued seems to imply that everyone actually is special — in which case, no one is special, after all.

The solution to this puzzle, I think, is to realize that what makes a person special is not a value that sets him apart from others; it’s a value that calls for appreciating him by setting him apart, a mode of appreciation that considers him alone. The key to this solution is that values are normative, in the first instance, not for actions or choices but rather for appreciative attitudes.[13] To be valuable is to be worthy of being valued in some way — that is, worthy of being the object of some appreciative response. This conception allows us to understand a kind of value that is not merely incommensurable but constitutively incomparable, because it is properly appreciated by a response that essentially involves a refusal to make comparisons, an insistence on cherishing its object in isolation from others.

Love is just such an attitude. We treasure the object of our love as special, not by comparing him favorably with alternative love objects, but rather by focusing appreciative attention solely on him, shunning any thought of alternatives. Each person is special in the sense that he deserves to be valued singularly in this manner, as he is in himself. In this sense, each person can be literally beyond compare.

I believe that deserving to be valued singularly, without comparison, is what Kant had in mind when he spoke of a person as a self-existent end. The appearance of paradox in this doctrine is due to a confusion between being valued singularly, without comparison, and being evaluated as singular in comparison with others. The latter cannot be deserved by everyone, obviously, but the former can.

Thus far I have said that love is a noncomparative response to another’s personhood as vividly perceived through the medium of those characteristics which we are said to love the person for. But what can be said about the response itself? How do we respond to another person in loving him?

For an answer to this question, I draw on a comparison that many philosophers find counterintuitive — namely, a comparison between love and Kantian respect. The comparison shouldn’t be counterintuitive, I claim, because love is a moral emotion: in particular, it is the emotion by which moral sensibilities are first implanted in children and by which the moral sensibilities of adults are enlivened or, if necessary, revived. If loving someone were not somehow akin to respecting him, love could not be the moral education that it is.[14]

Now, Kant characterizes the response to a person that he calls respect by saying that it “checks” or “arrests” our self-seeking motives, which might otherwise move us to use the person merely as a means to our ends.[15] I think that love resembles respect in being an arresting awareness of another’s value — a description that I intend to be understood both phenomenologically and functionally. Like wonder, awe, and amazement, these emotions give us the feeling of being pulled up short, brought to attention, riveted, transfixed. And in each of these emotions some other, distracting motivational tendency is actually being arrested, though it is not the same tendency in every case. Whereas respect arrests our self-interested designs on a person, love arrests our emotional defenses against him, leaving us emotionally vulnerable to him.[16] In colloquial terms, loving someone lays our heart open to him, leaving us emotionally disarmed and susceptible to all manner of other emotions toward him.

This difference between respect and love is reflected in their motivational potentials. Because respect for a person checks our self-interested motives toward him, its motivational force tends toward restraint, abstinence, and noninterference. Because love for a person checks our emotional defenses against him, its motivational force favors involvement and engagement. Respecting someone, we take care not to do various things to him or to let various things happen to him; loving someone, we are open to caring about him in all sorts of ways.

The foregoing descriptions of love are too abstract to convey the feeling, of course; indeed, their abstractness serves only to make them seem phenomenologically false. I have tried to describe the feeling of love by saying that it is an arresting awareness of value, similar to other arresting responses such as wonder and awe, and that it arrests our emotional defenses, so that it results in an opening of the heart. I hope that these admittedly vague and metaphorical descriptions find at least some resonance in the reader’s experience of love. I have tried to think of a familiar experience that will similarly resonate with my claim that this arresting awareness of value is, more specifically, an awareness of personhood. Oddly enough, the best example I can find is one that doesn’t involve actual personhood at all: it’s the experience of loving a dog.

It’s not so odd, really. Precisely because a dog isn’t a person, we can more readily notice when we start to see him as one. If he’s the right dog and we have the right rapport with him, we come to see him looking back at us with what seems like intelligent self-awareness, which makes his habitual obedience seem more like respect for us, and his instinctual affection more like love. Looking into his eyes, we seem to see someone there, someone who can reciprocate these interpersonal emotions. And having seen someone there, we are susceptible to feeling that form of love which I have described as fiercer than mere benevolent affection. We are also susceptible to feeling not just irritated by the dog, if he misbehaves, but betrayed, because we have trusted him, when we should have known that all there is to trust is his training.

I am quite sure that my feelings for my late poodle were a response to the experience of seeing someone there in his eyes. In clearheaded moments, I don’t believe that there really was someone there, but I am still under the illusion after his death, remembering him as I would a deceased person — not a lost toy for which I felt a fond attachment but a beloved personal presence, even though he was only a dog.

Murdoch’s description of love as the realization that something is real other than oneself may seem to imply that loving oneself is either unavoidable or impossible.[17] If things other than oneself are the ones whose reality is difficult to realize, then perhaps one’s own reality is obvious and the realization constitutive of love is unavoidable in reference to oneself. Or perhaps one can never have the requisite realization in reference to oneself, because it must be the realization of reality in something else. In fact, however, I think that Murdoch’s description explains self-love more convincingly than the alternatives. Specifically, it better explains why loving oneself is possible but by no means easy, a moderately difficult accomplishment.

Of course, most people think well of themselves, and most also favor their own interests. If love were just a form of flattery or favoritism, then self-love would be virtually universal. But our test for whether people genuinely love themselves comes when we love them, thereby responding to their value in a way that models what self-love on their part would be. And when we love people, we frequently find that their self-flattery and self-favoritism fall somewhat short of love.

Genuine self-love is elusive because it requires a vivid awareness of one’s personhood, consisting in one’s rational autonomy. One can rarely avoid being vividly present to oneself as the conscious subject of feelings and behaviors, but one can easily be blinded to one’s own autonomy or to the moral valence of that capacity. One can consequently raise emotional defenses against oneself, defenses that take the familiar forms of repression and dissociation. One feels threatened by one’s unruly impulses, because one is blinded to one’s own capacity to tame them with the force of respect and love, and so one is moved to deny having such impulses or being responsible for the behavior that might express them. Self-love enables one to accept the presence of unruly impulses, to accept oneself as subject to them, because it involves the vivid awareness of something in oneself that can be trusted to manage them. And self-love thereby facilitates the lifting of repression and the healing of dissociation.[18]

When I say that loving someone is a response to the value of rational autonomy, I am not saying that we love him for being rationally autonomous. My view, as I have said, is that the qualities “for” which we love someone are the qualities that serve as signs or symbols of his rational autonomy in our eyes. To my knowledge, this view is the only way to explain why we can love someone for his flaws.

Sometimes the recognition of a brushstroke as a flaw is what alerts us to its surroundings as a work of art, as having a value without which that stroke would be just another daub of paint. Similarly, an actor’s portrayal of a character’s weaknesses can be what makes the character seem real — really a person, that is, having those powers of rational autonomy against which human traits can stand out as weaknesses. A trait that would be merely a nuisance in another animal can be, in a person, the foil that casts his personhood into relief, the exception that proves the rule. We can love someone for his flaws, then, because our seeing them as flaws can be what alerts us to the fact that we are seeing a person, with the capacities against which they stand out as flaws.

My conception of love, when combined with the views of Darwall and Rosati, favors an Aristotelian conception of a person’s interests. What it makes sense to care about out of love for a person is the unimpeded realization of his personhood, which might be described as his flourishing, in that sense of the term which is used to translate Aristotle’s ‘eudaimonia’. Caring about the self-realization of the beloved is not intrinsic to the emotion of love itself; it is one of the further responses to which love makes us susceptible by disarming our emotional defenses. But it is the further response that most naturally ensues when our defenses have been disarmed in response to the value of the beloved in himself, since it is a desire to see that value brought to its fullest realization.

We can observe this aspect of love in the feelings of parents for their young adult children, who are just coming into the full realization of their personhood. What I found natural to care about out of love for my adolescent children was, to begin with, that they find direction — goals in pursuit of which to exercise their powers. Nothing makes parents of young adults fret more than seeing their children adrift.[19] And once my children adopted some directions — and there were many different directions over the years — I found myself caring about their progress in those directions, no matter how little intrinsic value I might have been inclined to see there in advance. In a quick succession of years, I became deeply interested in lacrosse and Morris dancing, poetry slams and photography, and specifically in the accomplishments of a particular midfielder, Morris dancer, poet, or photographer, because these were the directions that my children had set for themselves. Of course, I eventually learned to appreciate some of these accomplishments intrinsically: I would realize with amazement that I was cheering as my son walloped a schoolmate with a metal stick, or that I was applauding choreography that previously would have struck me as no more than quaint. But I learned to appreciate these accomplishments, to begin with, because they were the ones that my children had chosen to cultivate. In other words, I learned to appreciate them out of love for my children.

These examples introduce three distinct but related values. First is the value of my children as persons. Next is the value of their good, which consists in whatever it makes sense to care about out of an appreciation for their value. I have suggested that the relevant mode of appreciation is love, and that what it makes sense to care about out of love for them is the realization of their autonomy — their exercise of the capacity to which my love is an appreciative response. In loving my sons, I respond to the powers constitutive of their personhood, and it then makes sense for me to care about their exercise of those powers, bringing their personhood to fruition. And their exercise of those powers, because it is that which it makes sense to me to care about out of appreciation of their value, is what constitutes their good.

Finally, there is the value of the ends in whose pursuit my children realized their autonomy. Caring about their ends is not quite the same as caring about their interests, though the two are deeply entangled. Playing lacrosse wasn’t essential to my son’s good: he could have exercised his powers in many other pursuits instead. But of course his autonomy could be properly exercised only in pursuits of his own choosing, and lacrosse was what he chose. I might have thought that lacrosse was a bad choice — indeed, a choice inimical to his interests, if I had thought that it would expose him to serious injury that would damage his prospects for autonomous pursuits in the future. In that case, I would have seen a conflict of value between his interests and his ends. This conflict would have been of the maddening sort that arises between values that are at least partly traceable to a common source.

The value of my son’s end, playing lacrosse, was distinct from the value of his good, which was the realization of his autonomy. But these two values were at least partly connected, because his end derived some of its value from its being that in pursuit of which he chose to realize his autonomy — a direction in which he decided that his flourishing would unfold. I will have more to say about this connection in a moment.

Because respect and love respond to the value of a person, they are responses out of which we do or want things for the person’s sake, thereby taking the person as our end. But what we do or want for the person’s sake out of love is rather different from what we do or want out of respect. The difference is perhaps clearest in our stance toward the person’s ends.

Respect for a person restrains us from interfering with his pursuit of his ends, and it can also restrain us from taking a stance of indifference to whether he has the wherewithal to pursue them. But insofar as we merely respect someone, his success in attaining his ends doesn’t matter to us, so long as we leave him free to do his best. I have now suggested that love engenders a different attitude, leading us to care about the full realization of personhood in the beloved through his autonomous endeavors. I have illustrated this suggestion by describing our tendency to care about the endeavors of someone we love even if we see no intrinsic value in them.

Creative Commons via Flickr

I think that this manifestation of love occurs in the reflexive case as well — that is, in love for oneself. Of course, one is already motivated toward one’s ends simply by virtue of having adopted them, to begin with: they are things that one wants, or at least aims to attain. And yet there are plenty of things that one wants or aims to attain without feeling that they really matter or that attaining them is of any importance. One doesn’t take them seriously or care about them deeply. But if one loves oneself, then one will care about one’s ends, not merely out of having adopted them as ends, in the first place, but also out of concern for the realization of one’s autonomy through the pursuit and attainment of whatever ends one has adopted. One will have, in other words, a second-order concern for one’s ends, out of love for the self who has chosen to invest his autonomy in pursuing them. And this second-order concern will transform one’s ends from objects that one merely desires or aims at into objects about which one genuinely cares.

(The idea of such a transformation by second-order attitudes will be familiar from the work of Harry Frankfurt. And I think that the point I have just made about self-love, in particular, is present in some of Frankfurt’s writings, though not in the writings primarily devoted to the topic of love. In the Appendix, I examine the relation between Frankfurt’s views on the subject and my own.)

I now return, at last, to the argument against escapist suicide, which occasioned the foregoing reflections on love and personal good. Do these reflections provide materials for repairing the argument?

On the conception of personal good that I have developed, a person’s good does not include his happiness essentially: it includes his happiness only because happiness is one of his ends. His good consists in the full realization of his rational autonomy, which is what would make sense to care about out of appreciation for his value as a person. His happiness makes sense to care about out of love for the person only because it is one of the ends in pursuit of which he must fully realize his autonomy. Indeed, it makes sense for him to care about, rather than merely desire, only out love for himself, and only as one of many ends whose pursuit would bring his autonomy to full realization.

Here escapist suicide impinges on the relevant concerns in the irrational manner that I described at the outset. In caring about our own autonomy out of self-love, we care about its full realization, which cannot come in the pursuit of a single end such as happiness, much less in the exercise of a single, one-time choice such as suicide. Out of respect for a person we can restrain ourselves from interfering with a single choice on his part, but in loving the person we want to see his autonomy brought to fruition more broadly. And because our own happiness is worth caring about only out of self-love, it is worth caring about only out of an appreciative response that extends to more than any one exercise of autonomy. Insofar as a person still has a variety of ends that he is capable of pursuing autonomously, they make sense for him to care about out of self-love. But of course suicide brings all of his pursuits to an end. Suicide therefore thwarts the concern out of which happiness is worth wanting in the way that entrenches it in the person’s good. In sum, self-interested suicide is irrational.

My own view is that hastening death becomes morally appropriate only in the context of deterioration or suffering that compromises autonomy to an extent that can make talk of suicide inappropriate. But my conclusion about self-interested suicide can be stated more generally so as to encompass end-of-life decisions that are not solely on the shoulders of the prospective decedent. The conclusion is that we should not favor ending someone’s life out of sympathy for him or concern for his happiness; we should favor ending his life only when we can do so out of love.

I think that sympathy or benevolence toward a loved one can tempt us to indulge his expressed wish for assistance in dying even when our love for him rebels at the thought. Such are the cases in which we should hesitate, in my view. We must of course distinguish carefully between loving perception of the person as he really is and attachment to him as he formerly was. Reluctance to let go of what is already gone should not determine our response. But neither should sympathy or benevolence, when they are not seconded by genuine, clear-sighted love.

I have elsewhere endorsed the slogan ‘death with dignity’, which I interpret as meaning “death while dignity is still mostly intact, before it suffers further, irrevocable deterioration”.[20] Unfortunately, however, the word ‘dignity’ is not generally understood in the morally relevant, Kantian sense; it is often used to denote grounds of self-esteem, such as youth, good looks, and independence. Maybe, then, I should advocate retiring the slogan ‘death with dignity’ in favor of ‘death with love’, meaning “death only as love would allow”.


Dr. Harry Frankfurt

On the topic of “caring”, Frankfurt says: [21]

We often devote our time and effort and other resources to the pursuit of goals that we desire to attain because we are convinced of their intrinsic value but that we do not really consider to be of any importance to us. … Suppose someone is planning to attend a concert that is to be devoted to music he particularly enjoys. There are easily imaginable circumstances in which he might emphatically and sincerely declare that, although he certainly does want to go to the concert, it is not something that he regards as being at all important to him. Consider the following scenario. The prospective concertgoer is asked by a close friend for an important favor. Doing the favor will make it impossible for him to get to the concert. He agrees gladly to do the favor, but incidentally mentions to his friend that doing it will require him to change his plans for the evening. Upon hearing this, his friend becomes confused and apologetic, expresses a reluctance to impose upon his good-natured readiness to forgo the concert, and begins to withdraw the request for the favor. At this point, the music lover interrupts him, saying: “Don’t worry about the possibility that you may be taking too much advantage of my friendship for you. The fact is that going to this concert is not at all important to me. I really don’t care about missing it.”

In order for this person to care about going to the concert, according to Frankfurt, his desire to go would have to be such as to persist even if he decides not to, and “the persistence of his desire must be due to the fact that he is unwilling to give it up” — that is, “the fact that he is disposed to support and sustain his desire”.[22] Caring thus has the hierarchical structure that is familiar from so much of Frankfurt’s work in moral psychology.

Frankfurt says that caring about things matters to us independently of whether they are intrinsically worth caring about. It matters because it is the means by which we give our lives coherence and unity, by supporting and sustaining some of our desires over time:[23]

Suppose we cared about nothing. In that case, we would be creatures with no active interest in establishing or sustaining any thematic continuity in our volitional lives. We would not be disposed to make any effort to maintain any of the interests, aims, and ambitions by which we are from time to time moved. … From our point of view as agents … whatever coherence or unity might happen to come about … would be merely fortuitous and inadvertent. It would not be the result of any deliberate or guiding intent on our part. Desires and volitions of various hierarchical orders would come and go; and sometimes they might last for a while. But in the design and contrivance of their succession we ourselves would be playing no concerned or defining role.

Because caring about things is our way of giving coherence and unity to our lives, Frankfurt believes, it has an importance beyond that of the particular things that we care about: “The value to us of the fact that we care about various things does not derive simply from the value or the suitability of the objects about which we care. Caring is important to us for its own sake, insofar as it is the indispensably foundational activity through which we provide continuity and coherence to our volitional lives. Regardless of whether its objects are appropriate, our caring about things possesses for us an inherent value by virtue of its essential role in making us the distinctive kind of creatures that we are.”[24]

The choice of objects to care about can thus be governed by the value of caring itself rather than the value of the objects:

What makes it more suitable, then, for a person to make one object rather than another important to himself? It seems that it must be the fact that it is possible for him to care about the one and not about the other, or to care about the one in a way which is more important to him than the way in which it is possible for him to care about the other. … The person does not care about the object because its worthiness commands that he do so. On the other hand, the worthiness of the activity of caring commands that he choose an object which he will be able to care about.[25]

Now, Frankfurt doesn’t say exactly why it matters or should matter to us that our lives have continuity, coherence, and unity. But he does say that it will matter to us insofar as we love ourselves:[26]

Parents express their love … by doing what they can to ensure that their children actually have genuine interests and are therefore not condemned to lives that are chaotically fragmented and empty of meaning. Thus, their concern may extend also to helping their children to become capable of loving, to encouraging and assisting them to find love. This suggests that a person who loves nothing may nonetheless be able to show that he loves himself by attempting to alter whatever personal characteristics may impair his capacity to love and by making suitable efforts to find things to love.

Frankfurt conceives of love as a selfless identification with, and concern for, the interests of the beloved. In the essay from which this last quotation is drawn, he says that “the true interests of anyone … are defined and determined by what he loves.”[27] Yet he goes on to say, in the quoted passage, that love for someone gives rise to a concern for something beyond what he loves — a concern, that is, for his having something to love, so that his life will have coherence and unity. And the value for the beloved of having something to love in the first place cannot be explained by his loving anything antecedently. Hence, Frankfurt seems to presuppose a “true interest” on the part of the beloved that is prior to that which is defined and determined by what he loves.

I would say that the interest in question is the interest that every person has in the fullest realization of his rational autonomy, which is in his interest because it is what would be most natural to want for his sake, out of love for him. A person most fully realizes himself as a person by having sustained and coherent pursuits, and so his caring about some pursuits in this fashion is what would make sense for anyone to want out of an appreciative response to his personhood. Like Frankfurt, I believe that it is what we want for our children, out of love for them; and I agree with Frankfurt that it is also what we want for ourselves, out of self-love.


1    This is a substantially revised version of a paper that appeared in Ethics 118 (2008): 191-212, http://dx.doi.org/10.1086/523746. The first version of the paper was presented to a workshop on value at Columbia University. I am grateful to the participants for helpful discussion: Ruth Chang, Jonathan Dancy, Jim Griffin, Ulrike Heuer, Tom Hurka, Shelly Kagan, Frances Kamm, Maggie Little, Véronique Munoz-Dardé, Peter Railton, Joseph Raz, Jacob Ross, Michael Smith, and Larry Temkin. A subsequent version was presented to the philosophy department at the University of Miami and to the Legal Theory Workshop at Yale Law School. Thanks to Shelly Kagan and Ruth Marcus for additional comments on the latter occasion. Finally, it was presented to a conference organized by Jeanette Kennett at the Centre for Applied Philosophy and Public Ethics of the Australian National University.

2    “A Right of Self-Termination?”, chapter 3 of this volume; “A Brief Introduction to Kantian Ethics”, in Self to Self: Selected Essays (New York: Cambridge University Press, 2005), pp. 16-44; and “Reading Kant’s Groundwork”, in Ethics: Essential Readings in Moral Theory, ed. George Sher (New York: Routledge, 2012), pp. 343-359.

3    Stephen Darwall, Welfare and Rational Care (Princeton, NJ: Princeton University Press, 2004).

4    Objections to my argument were first raised by Frances Kamm in “Physician-Assisted Suicide, the Doctrine of Double Effect, and the Ground of Value”, Ethics 109 (1999): 586-605. I am now dissatisfied with the responses I made to Kamm in my appendix to “A Right of Self-Termination?”, which I have omitted from the version reprinted as chapter 3 of this volume. I was prompted to revisit this debate by a discussion of the latter article at the Colloquium on Legal, Political, and Social Philosophy at New York University (led by Thomas Nagel and Ronald Dworkin), especially by Nagel’s comments on that occasion.

5    I am tempted to reply that escapist suicide is not autonomous, precisely because it is irrational. But I am trying to demonstrate its irrationality on the grounds that it thwarts the concern underlying its own motivation; if that concern is respect for autonomy, then I am at risk of arguing in a circle. For if escapist suicide is irrational because it thwarts respect for autonomy, and it thwarts that concern because it isn’t autonomous, then the reason why it isn’t autonomous cannot be that it is irrational.

6    Of course, one might argue that punishing the child would express a higher or better-informed sympathy than sparing the child. But such an argument would seem to rely on a prior conception of the child’s interests, as a basis for privileging one form of sympathy as higher or better-informed than another; so it seems to reverse the order of analysis in Darwall’s account, by identifying the relevant valuing attitude in terms of well-being rather than vice versa.

7    The following discussion of love is an attempt to summarize and expand upon my discussion in “Love as a Moral Emotion”, Ethics 109 (1999): 338-374, reprinted in Self to Self, pp. 70-109.

8    Connie S. Rosati, “Darwall on Welfare and Rational Care”, Philosophical Studies 30 (2006): 619-635, p. 626.

9    Of course, we can love someone without there being anything for which we love him. See D. W. Hamlyn, “The Phenomena of Love and Hate”, Philosophy 53 (1978): 5-20, p. 8: “To be loved full-stop is simply to be loved without there being anything that the love is for. In such a situation there is likely to be some explanation why the love came into being, and it is possible with some objects of love for one to love them for the fact that and because of the circumstances in which the love came into being; but there seems to me no necessity that it should be like that — the circumstances may explain the continuance of the love but they may not be what the love is for. I suggest that love is possible where there is nothing that the love is for.”

10    See Hamlyn, “The Phenomena of Love and Hate”, p. 9: “It might be [said] that it must at least be true that the lover desires the beloved, wants to be with him/her/it, or something of that kind. I am not sure that even this has to be true. Suppose that someone has got to the point of recognizing the absolutely disastrous character of a relationship. It is possible for them to renounce it and any desire for its continuance while still loving the person concerned.”

11    Niko Kolodny, “Love as Valuing a Relationship”, The Philosophical Review 112 (2003): 135-189, p. 156. I have substituted the word ‘intrinsically’ for ‘extrinsically’ in the original passage, on the assumption that the latter is a misprint.

12    Iris Murdoch, “The Sublime and the Good”, in Existentialists and Mystics: Writings on Philosophy and Literature, ed. Peter Conradi (New York: Penguin, 1997), 205-220, p. 215.

13    I borrow this conception of value from Elizabeth Anderson, Value in Ethics and Economics (Cambridge, MA: Harvard University Press, 1993).

14    Harry Frankfurt says, “The function of love … is not to make people good.” I disagree. (Or I would insist that making people good is, if not the function of love, at least one of its effects.) Frankfurt’s statement appears in “The Dear Self”, Philosophers’ Imprint 1, no. 1 (2001), http://www.philosophersimprint.org/001000/

15    Actually, Kant says that what respect “checks” is our self-love, but Kant isn’t thinking of love at all, in my view. ‘Self-love’ for Kant means “self-interest”.

16    I suspect that aesthetic appreciation is an arresting awareness in the same sense, responding to the beauty of an artwork, e.g., in a way that leaves us emotionally vulnerable to its content. I believe that Kant describes respect as an arresting awareness. What is arrested in respect, according to Kant, is self-love — though I believe that ‘love’ is a misnomer, since what Kant has in mind is self-interest.

17    On the subject of self-love, see Frankfurt, “The Dear Self”. I will have more to say about this essay in the Appendix, below.

18    It also facilitates self-forgiveness.

19    Frankfurt makes a similar point in “The Dear Self”, p. 10. (I quote the relevant passage at n. 26, below.) See also Frankfurt’s essay “On the Usefulness of Final Ends”, in his Necessity, Volition, and Love (Cambridge: Cambridge University Press, 1999), pp. 82-94.

20    See “A Right of Self-Termination?”, chapter 3 of this volume.

21    Harry G. Frankfurt, “On Caring”, in Necessity, Volition, and Love, 155-180, p. 159.

22    Ibid., p. 160.

23    Ibid., p. 162.

24    Ibid., pp. 162-163.

25    Harry G. Frankfurt, “The Importance of What We Care About”, in his The Importance of What We Care About (Cambridge: Cambridge University Press, 1988), p. 94.

26    Frankfurt, “The Dear Self”, p. 10.

27    Ibid., p. 8.

From Beyond Price: Essays on Birth and Death, by J. David Vellman (2015)