Daniel Callcut, University of North Florida
Bernard Williams has explored how various personal ideals–ideals of self-fulfillment or of romantic love–can come into conflict with and threaten to undermine morality’s authority over thought and conduct. One of his central aims was to show that the paramount normative standing traditionally accorded to morality is more precarious than moral philosophers, especially Kantian moral philosophers, have tended to admit. Here I focus on the potential conflict between love and morality. I think that a pervasive tendency of Kantian responses to Williams is to make the potential conflict between love and morality seem less troubling only by offering insufficiently robust conceptions of love. In this paper I begin to make this case. What I aim to do here is motivate and then criticize J. David Velleman’s Kantian conception of love in light of Williams’ concerns.
- The Potential Tension between Love and Morality
In “Persons, character and morality,” Bernard Williams discusses a scenario imagined by Charles Fried, in which a man is confronted with two drowning people, at equal risk, and where one of those drowning is the man’s wife (Fried 227). Fried, it seems, tries to offer a justification for the man saving his wife over the other person that shows the man’s actions to be morally fair or, at least, not unfair: the man’s actions are not unfair to the stranger because in such situations, where one person only can be saved and there is no more impartial reason to save one person over the other, then it is permissible to act on one’s partial reasons and to save one’s loved one. More specifically, Fried suggests that, once the impartial perspective does not point one way or another, the chance fact that one of the drowning people is the man’s wife is allowed to come into play because there is a sense in which this is ultimately no more unfair than using a lottery to decide whom to save: after all, it could just have easily been the stranger’s spouse on deck, and thus the other person’s good fortune to be saved.
Williams objects to the suggestion that “moral principle can legitimate [the man’s] preference, yielding the conclusion that in situations of this kind it is at least all right (morally permissible) to save one’s wife” (Williams, 18). This suggestion, in his well-known formulation, “provides the agent with one thought too many” (18). What Williams is primarily concerned about here is a normative tension between love and morality: a tension between the ideals of morality and the ideals of romantic or other love. He is also concerned with what might be called a psychological tension between love and morality: if the man really loves his wife, is it realistic to expect him to be capable of giving due regard to the stranger, considering the permissibility of saving his wife, and so on? But the main concern he is expressing is that, even if the man is capable of considering the permissibility of his action, is this not too cold and detached a thought to be compatible with what an ideal of love would suggest? Hence not only is love, seen realistically, in potential tension with morality, but the very ideals of love and morality are shown to be, at least potentially, in conflict.
- The Counterfactual Condition
Some of the responses to Williams suggest that one can accept his claim about what the man’s fully spelled out thought should be and yet reject the suggestion that this in any way raises difficulties for impartial morality. For just because it is morally permitted that the man save his wife, why must this be his motivating thought? Perhaps all that impartial morality demands of the man is that he satisfy a ‘counterfactual condition’: were it impermissible that he save his wife, then he would not save her. 1 In response, Williams might challenge the suggestion that what justifies the man’s action in this circumstance need not appear in his thoughts while he is acting. Shouldn’t the impartial outlook of morality suggest an ideal of character that is a little more conscientious? Shouldn’t the man check that “it is all right” that he save his wife? What status does the stranger earn in his thoughts? Is the stranger’s life, in this situation, ‘nothing’ (at the level of phenomenology) to him? If the stranger is nothing to him in this context, is that morally acceptable? How far can morality let motives and justifications come apart? These are important questions. But what I want to stress here is that even if Williams is wrong to think that impartialist conceptions of morality must give people “one thought too many,” his larger point, about the potential tension between love and impartial morality, survives. First we must note that though the counterfactual condition may assuage some of Williams’ concerns, by itself it surely cannot make these concerns go away.
To help bring this out, let us go back to Fried’s scenario, but let us alter it a little. Imagine that the man is the captain of the ship (an alternative scenario that Fried brings up) and that the man can either save his wife, whom he loves deeply, or (to alter the original case further) a group of strangers who are stranded nearby. Now, one might think, it is no longer permissible for the man to save his wife; indeed, I take it to be (at least prima facie) morally impermissible. After all, it is the captain’s duty to treat each of his passengers fairly and equally, and this means that a large group of easily rescuable people must take priority over a single individual. How could he morally justify saving his wife ahead of all the others? So now the situation is that it is impermissible for the man to save his wife and, if he is to satisfy the demands of morality, he will not save her. Williams’ point is that if the man does truly and deeply love his wife it is by no means obvious that he can or should abandon her to save the others. In other words, it is not realistic, or attractive, to think that love has built into it the condition that one’s devotion runs only so far as morality permits. 2 But insofar as one agrees that this is so, it seems one must accept the potential conflict between love and morality. To paraphrase Williams, one runs that risk if one is to love at all.
There are two main ways of trying to avoid this conclusion. One way is to offer a more ‘moralized’ account of love such that love and morality, for instance, are not even in potential conflict. The other way is not to shape love around morality, but rather to alter the content of morality so that it allows more room for partiality. For the remainder of this paper I shall examine and reject a Kantian response that takes the first route. 3
- Impartial Love?
- David Velleman, in “Love as a Moral Emotion,” is responding to those such as Williams who find a tension between the value we place upon our loved ones and the equal moral value of everyone. Those who find such a tension disagree as to how great the consequences are for morality or our conceptions of morality; some think the tension points to a need to abandon what one may call impartialist conceptions (be they Kantian or consequentialist) of morality and the moral life; others find this an extreme and overblown response to a manageable tension. Velleman thinks that both sides of this dispute are mistaken in their common assumption that love and morality are “even potentially at odds to this extent” (341). If they were, then “love would have to be, if not an immoral emotion, then at least non-moral” (341). But love, according to Velleman, is a “moral emotion” (341). Velleman’s aim is thus to dissolve the question of “how two divergent perspectives [those of love and morality] can be accommodated” (341) by replacing the faulty conception of love which gives rise to it.
Velleman wants to defend a conception of love inspired by Kant. In truly loving someone, Velleman argues, what our love is centrally responding to is that person’s rational nature. Now, as Velleman notes, one issue that such a view must face is “how being valued on so generic a basis is compatible with being valued as special” (366). Part of the answer, he argues, can be found in the Kantian distinction between dignity and price, since the notion of dignity explains the way that we can all share the value of being special: our status as a person entitles each of us to be seen as irreplaceable. But there is a further puzzle to which his view gives rise, namely, how some people rather than others turn out to be, as we say, special to us. For if love is an appreciation of a person’s value qua person, then how is it that we do not love everyone? Moreover, if love is an appropriate response to the value of a person qua person, then ought we to love everyone? The burden that Velleman’s account must take up is that of making sense of love’s partiality, and it is a burden, I suggest, that his account cannot carry.
Velleman acknowledges that he must distinguish love from Kantian respect, even if both love and respect are responses to the value of rational nature. The essential feature of love, Velleman argues, is that it “disarms our emotional defenses” in response to an object’s “incomparable value as a self-existent end” (365). And to have one’s emotional defenses disarmed by someone is not simply to possess an intellectual awareness that this someone is a person; it is to see the person as a human
being (371). This is how love differs from respect: whereas the immediate object of respect is the intelligible aspect of a person, the immediate object of love is “the manifest person, embodied in flesh and blood” (371). It takes the manifest qualities of a person to disarm us, even if it is his or her intelligible aspect that we love. This, then, is the sense in which we “love a person for his observable [and distinctive] features – the way he wears his hat and sips his tea [and so on]” (371). But it is not the features themselves that we love; it is the features “as an expression or symbol or reminder of his value as a person” (371). Velleman’s account thus rather neatly explains, albeit not entirely plausibly in my view, why some people are more lovable than others: some people’s empirical selves signal their rational nature more clearly than others. But the notion that people embody their rational nature in different ways, and to different degrees, still does not explain why we love only some people.
One reason why it does not is that it jars with an excellent point that Velleman makes against the commonly held, but mistaken, view that loving someone entails wanting to care for him, be with him, and so on. This view, as Velleman powerfully argues, is a “sentimental fantasy” (353). The painful but familiar truth is that one can love someone whom it drives one crazy to be around. Just think of the “cranky grandfather, smothering parent” (353) and so on. How does Velleman explain our partiality here? Even if one can accept that one’s love for difficult relatives is constituted by an arresting awareness of their value as a self-existent end, one surely cannot think that why one has come to love these people is because they are marvelous exemplars of rational nature. That would be to replace a sentimentalist fantasy with a rationalist one. Velleman’s account of why we love some and not others, then, will need to be fleshed out to make room for cases such as this.
The shape that any such account must take, however, is clear. Some people’s empirical features explain how we have come to see their personhood, and thus how we have come to love them and not others. However, even though the manifest features of people play a crucial role in our coming to love them and not others, our love of them is not a love for any of these manifest features qua manifest features. The empirical person we see is loved for the value of personhood that the empirical person before us reveals. On this view, then, the key reason that we do not love everyone is that we all imperfectly convey and imperfectly perceive personhood. Think, then, of people that you know but do not love. If love for a person is identified with seeing his or her value as a person, then the reason that you do not love these people must be because you fail to see their true value. Here Velleman’s view, and at times his tone, is similar to that of certain mystical writings about the world’s existence. Most of us most of the time, caught up in daily life, fail to look and really see the world; but, most of us, some of the time at least, have moments when the blinders come off and we look at the world in appropriate wonder. Afterwards, when one cannot get in touch with this response (the defenses are up again), one can still feel that this is how one should see the world, even if (for a variety of practical and psychological reasons) one cannot. According to Velleman, it seems, this is how we should think of our lack of universal love. Everyone deserves a response that constitutes a full apprehension of his or her value but, in practice, it is only possible to love some.
This is an explanation of why we love only some people, but it is awfully far-fetched. Is it not possible to really see someone and not love what one sees? On Velleman’s account, the answer must be “no,” for to really see someone is to love him. The fact that Velleman’s account of love makes it so hard to explain the fact that we love only some suggests that something has gone wrong with the account. And the heart of the problem, it seems to me, is the marginalization of people’s distinctive features in his conception of love. This marginalization is especially glaring if one thinks of romantic love. Now, Velleman is right in suggesting that focusing excessively on romantic love is liable to mislead, as the example of love for difficult relatives illustrates above. Indeed, one upshot of that example might be that there can be no single, overarching conception of love. Velleman does, however, mean to be offering an account of love that is broad enough to encompass love between spouses or life-partners (as he must if his account is to be responsive to the kind of Fried example that Williams exploits), so examining his account in regard to romantic love in this sense at least is fair game. And here the picture that his account delivers is very unsatisfying.
On Velleman’s account, remember, the manifest features of one’s beloved are loved for the value of personhood for which they are, as he puts it, “conduits” (371). These features are loved, then, not for their intrinsic appeal, but only insofar as they are manifestations of personhood. But this is unconvincing. I am not suggesting that we treasure a person’s features in isolation from the person whose features they are. That would, indeed, make one some kind of feature-fetishist. But, by the same token, the person we treasure is not totally separable from his or her particular features. That is, I do not love my partner merely because she instantiates features of personhood (as if any features would do); I love her, at least in part, because she has these features. True, a time-honored lovers’ game consists in asking “Would you still love me if…?”, and we do feel that real love endures through fairly substantial changes in one’s beloved. But the fact that you would continue to love your partner through changes in his or her distinctive features in no way shows that your love for him or her is a love that transcends his or her distinctive features. There are cases in which love naturally and admirably continues despite massive changes in the beloved (due to serious illness or accident, for instance), but it is a mistake to think that this forces one to accept Velleman’s position. The right response is to incorporate the history of a person into his or her distinctive features. I take it that this is not merely an ad hoc response but is independently plausible. It can, of course, be hard to say what features of one’s beloved are central to one’s love or for what features we ourselves wish to be loved. A comprehensive answer, perhaps, is neither possible nor desirable. But I submit that it is nothing so thin as mere personhood.
It is a curious upshot of Velleman’s rationalistic view that it threatens to make our particular loves totally rationally arbitrary. The problem seems, at least in part, generated by the wish to situate love within the impartial outlook of morality. More specifically, it is generated by the fact that Velleman makes the reason for loving the same for all of us. Everybody has the same reason for loving everybody else (and, of course, his or her own self). And it is because we have the same reason to love everybody that loving the people that we do looks, from the standpoint of reason, totally arbitrary. But love is not that arbitrary. There are reasons that explain, and to some extent justify, one’s particular loves. 4 The key point, however, is that they are not reasons that everybody necessarily shares (simply, as it were, by being a person). In the case of romantic love, for instance, my love for my partner makes sense given her distinctive features and mine. Some of the qualities that move me simply will not be reasons, or at any rate nowhere near so salient, to others with different dispositions (or different histories). They are part of what make me partial to her. But in order to provide such straightforward and satisfying explanations of why we love only some people one must give a far greater role to people’s manifest properties than Velleman allows. When we love someone we do not love only that aspect of him or her that is found in all persons.
We are now in a position to look directly at Velleman’s discussion of instances of apparent conflict between love and morality such as those with which we began. I think that doing so shows definitively the unacceptable price that Velleman’s account of love must pay for its attempt to transcend love’s partiality.
- Velleman on the Conflict between Love and Morality
Velleman writes of the original Fried example: “I do believe that the man’s love for his wife should heighten his sensitivity to her predicament. But I cannot believe that it would leave him less sensitive to the predicament of others who are in–or perhaps alongside–the same boat” (373). Here, recall, there is a real question about what would count as being appropriately sensitive to the drowning stranger and to what extent that sensitivity can co-exist with the thoughts of a man setting out to save the person who means “everything” to him. More startlingly, however, Velleman continues:
Of course the man should save his wife in preference to strangers. But the reasons why he should save her have nothing essentially to do with love…. The grounds for preference in this case include, to begin with, the mutual commitments and dependencies of a loving relationship…. No doubt, the man also has nonmoral, self-regarding reasons for preferring to save his wife. Primary among these reasons may be that he is deeply attached to her and stands in horror at the thought of being separated from her by death. But attachment is not the same as love. (373, emphasis added).
Here Velleman is forced to make the implausible claim that the man’s reasons for saving his wife over a stranger have nothing to do with love. This is an odd claim, but we can now see why Velleman must make it. His reason for saying this depends on his theory that love proper is akin to Kantian respect and, more specifically, because he makes the basis for love the same for each of us (rational nature). This, I take it, is the explanation of Velleman’s view that the “grounds of preference” for the man saving his wife cannot be derived from love, for the grounds that love provides to save his wife, coming as they must from rational nature, are equally grounds to save the stranger. But the fact that Velleman’s account of love entails that love must be excluded from the man’s grounds of preference to save his wife seems to me tantamount to a reductio of Velleman’s view.
The reasons that move and justify the man’s actions to save his wife are neither plausibly nor appealingly described as consisting solely in the fact that the two have had a relationship or by his horror at the thought of separation. A central reason for the man to save his wife is because, to use the adjective that Velleman himself uses (inadvisably perhaps given his own view), the relationship is a “loving relationship” (373)–or more straightforwardly still (and why not say it?) because he loves her. 5 To exclude love from the man’s reasons for saving his wife in particular, and thus to reduce these reasons to the quasi-contractual (“mutual commitments and dependencies”) or the “non-moral and self-regarding,” is unappealing and implausible. A realistic and attractive conception of love, surely, would suggest the man both would be and should be partial to saving his wife.
The counterfactual condition, Velleman notes, aims to show that the conflict between love and morality is “manageable” through “segregation of the conflicting parties” (341), and he thinks this concedes too much to those who claim such a conflict by implying that love is not a moral emotion. One point that arose from our discussion in section 2 was that, even if the counterfactual condition can assuage worries over the potential conflict between love and morality, it cannot make all of Williams’ concerns go away, for it does not seem realistic or attractive to think that love runs only so far as morality permits. This is why Velleman’s view is of such interest, since he suggests a conception of love according to which the very idea of a conflict between the ideals of love and morality is mistaken. Thus, seen from Velleman’s conception, it is a mistake to think of love proper as capable of outrunning impartial morality. Many who advocate the counterfactual condition also, of course, think this 6–but Velleman offers a substantive conception of love that tries to make good on this claim. 7
Velleman’s attempt to dissolve the very idea of a conflict between love and morality by making love more impartial, however, must be rejected. The attempt to do so forces him to abandon some of our deepest considered judgments about love. Thus the gains achieved for impartial morality come at too high a cost. Indeed, I am inclined to think that any ideal of love that is comprehensively shaped by the demand that it mesh neatly with impartial morality is likely to be insufficiently deep or robust. Hence, I suggest, the attempt to dissolve the potential conflict between love and morality by shaping love around impartial morality ought to be rejected. 8 Love, as all the novelists tell us, is much deeper and much tougher than that. 9
Baron, Marcia. “The Alleged Moral Repugnance of Acting from Duty.” The Journal of Philosophy 81 (1984): 216-217.
Callcut, Daniel. Bernard Williams and the End of Morality. Johns Hopkins UP 2003.
Fried, Charles. An Anatomy of Values. Cambridge, Mass.: Harvard UP 1970.
Velleman, J. D. “Love as a Moral Emotion.” Ethics 109 (Jan 1999): 338-374.
Williams, Bernard. “Persons, character and morality.” Moral Luck. Cambridge: Cambridge UP, 1981.
- Or, in Marcia Baron’s more general formulation: “Part of what it is to act from duty is to act with a counterfactual condition always at hand (though not always in one’s thoughts): one would not do this if it were morally counter-recommended” (216-217). For a related point, see Peter Railton’s “Alienation, Consequentialism and the Demands of Morality” in Consequentialism and its Critics, ed. S. Scheffler (Oxford: Oxford University Press, 1988) 105.
- Neither, of course, does love have built in the condition that one will do anything for the sake of one’s beloved. And Williams clearly is not advocating such an implausibly Romantic view. As he says in relation to friendship, just because “there is some friendship with which his life is much involved [it does not follow] that a man must prefer any possible demand of that over other, impartial, moral demands. That would be absurd…” (17).
- I discuss the latter option in chapter two of my doctoral thesis, Bernard Williams and the End of Morality (Johns Hopkins UP, 2003).
- “Justify” unfortunately suggests the idea that there is a standing need, absent any specific challenge or criticism, to defend one’s love. Perhaps then the notion of reasons endorsing particular relationships is better.
- Furthermore, this helps explain his horror at the thought of being separated by her death. Such a thought need not be purely “self-regarding.” (It is the thought of being separated by her death that gives the thought its particular horror.) But, of course, the man’s concern for his wife is not purely other-regarding or altruistic either. His reasons, it seems to me, fit into the important category of reasons that Williams has helpfully called “non-egoistic.”
- Such as Marcia Baron. See note 8 below.
- And, significantly, the conception of love that makes sense of the judgment that love proper is incapable of outrunning morality undermines the initial motivation to propose the counterfactual condition. For if love proper is incapable of outrunning morality then the nature of love must be such that, as Velleman says, there is no need to worry about the psychological harmony of love and morality in the first place.
- This conclusion, then, shows me to be also in disagreement with the following in Baron: “If opponents insist that such a counterfactual condition gets in the way of friendship and love, then, I want to say, so much the worse for friendship and love – thus conceived. Unless one buys into an extremely romantic notion of love and a similar notion of unconditional friendship, the conflict is bogus,” 217, note omitted. To suggest that love and friendship, attractively conceived, can conflict with the counterfactual condition does not (as Baron here seems to suggest) thereby commit one to the disastrous view that love and friendship should be conceived of as unconditional. Williams is very clear about this.
- Thanks to Jacob Ossar, Jerry Schneewind and Susan Wolf for their comments on earlier versions of this paper.