The Journal of Value Inquiry (2007) 41:59–76 DOI 10.1007/s10790-007-9061-0


Springer 2007

Friends without Favoritism MARK BERNSTEIN Purdue University, 100 N University St, West Lafayette, IN 47907-2098, USA; e-mail: [email protected]

Our moral education often begins with our parents trying to teach us that we ought not to play favorites. Failure to learn this lesson attracts rebuke; we are characterized as unfair, a criticism that carries substantial sting. Sharpening a bit and using an argot unfamiliar to our mothers and fathers, we enshrine, as part of commonsense morality, what we may call the equal consideration of interests principle that all things being equal or special circumstances aside, we ought to equally consider the like interests of all persons. Equally entrenched in our ordinary moral thinking is the notion that friendships are paradigmatic examples of such special circumstances. In fact, all else being equal, friendships not only permit participants preferentially to consider the interests of their friends over the like interests of a stranger, they require that the parties mutually extend special concern. We are obligated to favor the interests of friends over the similar interests of strangers; friends are entitled to have their interests specially considered. More formally, we have the so-called partiality of friendship principle that all things being equal, friendships require mutual preferential consideration of interests. As an intuition pump for anyone dubious of the partiality of friendship principle, let us consider the following example. John and Fran are friends; John and Sue are strangers. Fran and Sue have equal interests in relieving their identically debilitating headaches. John has access to one aspirin, the whole of which is necessary and sufficient for relieving either Fran’s pain or Sue’s. Bracketing all other morally relevant factors, the example elicits the intuition that John ought to give the aspirin to Fran and not Sue. It seems that the only viable explanation for this intuition is the belief that friendships are relationships that legitimately ground moral requirements for preferential concern. Casting the point in terms of reasons, we can say that John being a friend of Fran but not of Sue is a good reason for John favoring the interests of Fran over the like interests of Sue.

60

MARK BERNSTEIN

This thesis is revisionary. Conventional wisdom notwithstanding, the attribution of such moral power to friendships is ill-founded, and we cannot legitimately appeal to these relationships to ground the obligation to extend special concern. This consequence is not as bracing as it might initially appear. It does nothing to denigrate the significance of friendships in our lives; indeed friendships are among the most important relationships in which we participate. Fortunately, we can live in the best of both worlds; we can extend the scope of fairness while inheriting the copious benefits that friendships provide.

1. The falsity of the thesis that is argued for here could be verbally assured by a matter of definition. This is an unpromising tactic. Consulting dictionaries offers no reason for accepting the connection between ‘‘friendship’’ and ‘‘obligation to behave partially’’ as analytic. Such an attempt to undermine the thesis argued for here trades on a technical term distinct from the mundane one that peppers ordinary language. But assuming that our lexicographical evidence is idiosyncratic and that the proposed definitional equivalence was accurate, little would be resolved. While friendships would then, trivially, legitimize preferential concern, quasi-friendship’ might not, where such relationships are just like their close cousins absent their semantic intimacy. Our question then becomes whether, and in what way, quasi-friendships suffice for the moral judgment expressed above. Let us drop the offensive coinage and ask what seems on its face to be a substantive question: ‘‘What is it about friendships that justifies the truth of the above moral judgment?’’

2. Our point of departure needs to be an untendentious account of friendship for which Aristotle’s analysis still serves. Aristotle begins by characterizing a friend as someone who likes and is liked by another person where liking is understood as ‘‘wanting for someone what one thinks good, for his own sake and not for one’s own, and being inclined, so far as one can, to do things for him.’’1 To distinguish friendship from mere good will, eunoia, the well-wishing must be reciprocated. Moreover, each person must recognize that the other person is wishing him well for his own sake.2

FRIENDS WITHOUT FAVORITISM

61

Friends have affection for each other, feel comfortable being with each other, and enjoy spending time together. Friends share an intimacy allowing for exchanges of personal and sensitive thoughts and feelings that would be impossible in more casual relationships. Friends are saddened and mourn each other’s failures and losses and are gladdened and rejoice in each other’s successes and accomplishments. Instead of relegating such empathetic feelings to merely incidental concomitants to the relationship, they are more accurately conceived as having normative significance and being partly constitutive of friendships. We think it right or fitting that John enjoy the fact that Fran’s headache has been relieved and would find it puzzling if he were not pleased. Absent extenuating circumstances such as that John is simply unaware of Fran’s pain being eradicated or that his father’s illness has left him emotionally spent, we would begin to question the depth, or even the existence, of the alleged friendship between John and Fran if his spirits were not lifted after recognizing Fran’s relief. There is a limited iterative structure to the emotions of friendship. Not only does John enjoy the knowledge of Fran’s improved health, but Fran derives pleasure from knowledge of John’s enjoyment of her enhanced welfare. We feel good when our friends do well and the recognition of this by our friend is an additional source of pleasure for them. For better or worse, we are saved from bursting from pleasure by our psychology; the vivacity of our satisfaction tends to diminish with each succession, and is a rare couple whose propensities for enjoyment allow them any measurable increase after a recursion or two. This slightly supplemented Aristotelian account makes friendship a relationship that necessarily comprises volitional, cognitive, behavioral, and affective elements. From the facts that we want what is best for our friends for their own sake, realize that our friends have similar desires for our welfare, are disposed to act to enhance their well-being, and are emotionally influenced by their successes and frustrations in ways that we would be affected if they were our own, there is yet no obviously discernible ground for a moral requirement for friends to mutually extend preferential consideration. There is as yet no legitimizing ground for the moral judgment expressed above no reason, to be more particular, for John being duty-bound to give Fran, and not Sue, the aspirin. This is not to deny the distinct point that the constitutive facts about friendship provide a basis for relating with friends differently from the manner in which one relates with strangers. Having affection, for example, for a friend but not a stranger is naturally expressed by wanting to spend more time and desiring more up-to-date information on the friend than the stranger. But that we would prefer to engage in particular activities with

62

MARK BERNSTEIN

friends than we would with strangers does not imply that we preferentially consider my friend’s interests over the like interests of strangers, let alone that we are morally required to extend favoritism to our friends. More support for the view that friendships demand partial behavior may come from concentrating on the popular notion that friendships provide uniquely fecund environs for significant life-enhancing goods. Impartialists should register their agreement that, for nearly all of us, there are goods, particular to friendships, that are ingredients to a happy, flourishing life. Mundane activities such as going to a movie, eating dinner, attending a baseball game, tend to be far more pleasurable when done with a friend than when either performed alone or in the company of strangers. Friends are useful in ways that strangers are not. Having found ourselves without sufficient funds one night at the River Cafe´, we might call a friend for help, who would show up later with a hundred dollar bill and a knowing smile. No doubt a similar reaction would not have accompanied a plea to someone with whom we had only nodding acquaintance with, let alone someone whose name was randomly selected from a Brooklyn phone book. Friendships are the fertile grounds for trust and intimacy. We can speak openly and freely with friends even about the most closeted of subjects. Since friends care about us for our own sakes, our concerns about being met with ridicule or embarrassment are mitigated. In addition to being opportunities of catharsis, such engagements are prime opportunities for getting to know and understand ourselves better. Presumably self-knowledge is itself a good and is instrumental in securing other goods. By understanding ourselves better we get an enhanced idea of what our preferences are, how to go about modifying them if we wish, and improving ourselves in the light of norms that we find appropriate. At least in the best of friendships we think of our friends as admirable people, with good sense and fine character. That people like them voluntarily enter into friendships with us ameliorates our feelings of self-worth and self-esteem. Finally, we should not dismiss the plausibility of friendships owning intrinsic value. The emotional ties in friendships are arguably valuable as ends in themselves, and for persons such as Aristotle who conceive people in social and teleological terms, friendships are simply fitting relationships in which humans should partake; in exercising their capacities to form friendships, persons are exemplifying part of what is good for their species. There are two primary ways of trying to justify the above moral judgment on the basis that friendships, perhaps uniquely, provide such abundant goods. We might be tempted to think that special concern for a friend is called for as a means of repaying a friend for all the goods she has provided during the course of the friendship. On this so-called

FRIENDS WITHOUT FAVORITISM

63

repayment model, the extension of special concern is an appropriate means of extinguishing a debt one friend has accumulated from the other. The second attempt conceives preferential concern as an apt expression of gratitude for goods supplied by friends. Partiality cannot be grounded solely on the fact that we stand in a debtor or creditor relationship with friends. Since we frequently stand in this same relationship with strangers, we would be obligated to extend special concern to them as well. There would, then, be no significance to the fact that people are friends and the work done to show why it is that we ought to preferentially consider the interests of friends over the like interests of strangers would be vitiated. Advocates of the repayment model might remind us that many of the goods created in friendships are unique to, or at least best provided in, such relationships, and that when goods such as intimacy and self-esteem are provided the proper mode of repayment is partiality. It is in virtue of benefiting from such goods that we incur the very special kind of debt whose extinction requires extending preferential consideration to the benefactor. The repayment model, however, is fundamentally flawed by the mischaracterization by its proponents of the nature of the goods in question and the mode of their exchange. The proponents of this model invite us to conceive of the goods of friendship as discrete, quantifiable packets, analogous to the way we think of the units of exchange in paradigmatic creditor and debtor relationships. If Carl gives Debbie five dollars, then Debbie incurs a debt of five dollars, the satisfaction of which is required for the debt to be extinguished. Once the debt is paid off, the creditor and debtor relationships are terminated, albeit amenable to reinstatement if another loan is made from one party to another. But the goods of friendship are amorphous. Speaking of packets of self-esteem seems perverse, and even the more philosophically acceptable language of hedons and utiles, where pleasure and utility are respectively quantized, is recognized by most of us to be fictional with occasional heuristic value. Moreover, in opposition to the repayment model, exchanges of goods, far from ending friendships, are part and parcel of strengthening them. When friends provide each other with pleasure and utility, and a fortiori, with opportunities for self-development, their relationship tends to deepen and becomes more intimate. Finally, the basic conceptualization of the exchange of goods as repayments of debts is a caricature. We need to keep in mind that friendships are essentially mutual and reciprocal relationships. While John benefits from the love Fran bestows upon him, Fran is also the beneficiary of John’s friendship. It is not as though John and Fran think of themselves as continually paying off debts; they conceive of themselves as willfully and lovingly expressing their friendship. The

64

MARK BERNSTEIN

notion that friendships are, at root, relationships between accountants feverishly tallying the benefits conferred to ensure, as best as possible, that all debts are extinguished, is antithetical to our conception of such engagements. Debts of gratitude are created when goods are received from another person who is neither morally required to provide such goods nor motivated by self-interest to make such provision. Friendships, as love relationships, are fertile grounds for gratitude. The proponents of the gratitude model of friendship conceive of partiality as an appropriate expression of gratitude for the goods received by a friend. Much as a thank-you note is an expression of gratitude for a wedding gift, extending preferential consideration to someone from whom we have continually altruistically benefited is forwarded as a fitting acknowledgment of what a friend has done for us. In supplying special attention, friends communicate their appreciation for the benefits he has received. The gratitude model of friendship has significant advantages over the repayment model of friendship. No longer are we burdened with the picture of packaged benefits being constantly measured against each other. Removed as well is the idea that partial consideration should be viewed as a repayment at all, an idea that is antithetical to the core idea of love relationships. Nevertheless, serious doubts about this explanatory model linger. Gratitude is not particular to friendships. We incur debts of gratitude toward strangers who see that we ran out of gas, pull over on the highway and offer us a ride to the nearest gas station. Therefore, special consideration of the interests of friends cannot be grounded on the fact that we are obligated to communicate our feelings of gratitude. Why, for example, does not a heartfelt thank you suffice for stranger and friend alike? Whatever the mode of acknowledging our gratitude, once it is expressed the relationship is closed. A helpful stranger does not incur an obligation to respond to our thank you card, nor need we continue to show our thankfulness. But partiality in friendships does not work this way. Instead of terminating a relationship, preferential concern injects it with momentum, encouragement, and depth. Where gratefulness is a backward-looking practice, partialism in friendships has both backward and forward-looking elements. Kant believed that gratitude is a means of honoring another person and in so doing creates a moral hierarchy between benefactor and beneficiary. To Kant, gratitude places the beneficiary in an inferior position making him feel resentful and unworthy, and it is only by fully discharging his duty of gratitude to his benefactor is the beneficiary able to regain his position of equality.3 While an analysis that would have us attribute feelings of resentment and unworthiness may

FRIENDS WITHOUT FAVORITISM

65

strike us as hyperbolic, the fundamental point that in bestowing supererogatory benefits we create an inequality between benefactor and beneficiary seems plausible. This should give pause to the proponents of the gratitude model of friendship. Friendships thrive when the participants conceive of each other as roughly equivalent in terms of value or worth. When friends extend favored concern to each other, they do not think of this exchange as an attempt to recapture the equality they antecedently shared. Any suggestion that the parties stand in superior or inferior positions is a signal that the relationship has changed from friendship into something different. Survival of friendships and the alleged special concern that derives from it, far from welcoming the hierarchism brought about by gratitude, requires its dismantlement.

3. That the repayment and gratitude models fail as explanatory analogues for grounding preferential consideration in friendships should not be surprising.4 Typically, friends do not think of themselves as discharging any kind of debt when dispensing special concern upon their friends. If compelled to use the language of justification at all, friends would most likely respond that their special concern is justified simply by virtue of their friendship. This simple, direct, non-reductionist answer hints that a more profitable strategy for partialists seeking to render legitimate favored behavior among friends is to investigate the nature of the relationship itself rather than the goods that derive from it. Instead of thinking that the goods provided by friends ground partial behavior, partialists might urge that the only way to express friendship or love to another person is to favor her interests over the like interests of others.5 In implementing this tactic, partialists insist that being inclined to preferentially consider the other person’s interests is constitutive of loving or befriending someone. In an attempt to augment this conviction, partialists may revisit the above example involving aspirin, and ask whether we would consider John and Fran friends if John did not to decide to give Fran aspirin. Using such intuitions, partialists will conclude that we pre-reflectively believe that special concern is intrinsically linked to loving relationships. Since we should accept the reality of love and friendships and agree that these relationships do require expression, such a challenge offered by partialists challenge amounts to articulating how love and friend relationships can be expressed or manifested without extending mutual privileged concern. The core of a loving relationship is caring for the other

66

MARK BERNSTEIN

person purely for her own sake; lovers wish their beloveds well and do what they can to benefit them. An individual does not have this will, affection, and dispositions to behave toward a stranger. But this does not commit friends to preferentially consider the interests of their friends over the like interests of strangers. John is inclined to use time, effort, and other resources to aid Fran that he is not inclined to use to help a stranger, but this not mean that when circumstances compel John to decide whether to use his resources to help Fran or an equally needy Sue, John must decide in Fran’s favor. When placed in conflict situations moral principles that otherwise have no applicability, might become relevant. There are situations, for example, in which questions of partial action find no application. If John meets Fran who needs help and there is no competition for John’s resources, then any adequte moral theory would have John help her. If John meets Sue who needs help and there is no competition for John’s resources, then, all else being equal, he again incurs the obligation to help. Neither the equal consideration of interests principle nor the partiality of friendship principle play a role in governing John’s behavior. Let us revisit the above example involving aspirin where there is competition for John’s resources and the two above principles are relevant. There is nothing untoward in John carrying on the interior monologue: ‘‘I love Fran and do not love Sue. I therefore have obligations to aid and comfort Fran that I do not have toward Sue which I will cheerfully try to fulfill if circumstances permit. I find myself in where I have one aspirin and both Fran and Sue need it and I am deeply committed to the equal consideration of interests principle, which, at this time, is a principle that is relevant in my moral deliberations, since now I need to set priorities about dispensing my resources. I conceive of myself as being morally compelled to act in accordance with the equal consideration of interests principle when the principle applies and incapable of acting with favoritism toward Fran in this situation.’’ We can make the example more radical by having Fran share John’s commitment to equal consideration of interests principle, imagining that their respective allegiance to this precept was a result of long discussions regarding meta-ethics. This dual agreement helped forge their friendship and made it stronger. Under such circumstances, Fran would not only not expect to be shown favoritism, but would wholeheartedly expect John to continue to adhere to the principle that played such a formative and fundamental role in cementing their relationship. Instead of feeling betrayed by John’s insistence to align with this principle, she gains further admiration for John for his continued faithfulness to it, especially when there is great temptation on his part to submit to the natural temptation to help his friend. Fran realizes that it is in such trying times that our true

FRIENDS WITHOUT FAVORITISM

67

characters are best revealed and, odd as it may initially seem, recognizing that her friend is of such high moral character that he cannot submit to the powerful inclination to extend special consideration to her interests actually adds to her self-esteem and feelings of self-worth. If John would have succumbed to their friendship and acted partially toward her, Fran would have thought both less of him and of herself. Even in such sterilized and idealized situations as this, then, friendships do not require partial behavior. John and Fran can remain friends despite his principled refusal to extend preferential concern. In such circumstances the friendship is expressed not in preferential consideration of a friend’s interests but in continued allegiance to a jointly held principle that is at odds with the partiality so frequently attributed to the relationship. Friendships can survive impartial concern and may even require it to flourish.

4. Let us investigate two arguments both of whose proponents conclude that our deepest commitment cannot be to the equal consideration of interests principle; some of our relationships must comprise partiality. A negative assessment of such arguments would provide some confirmation of the thesis that we can have enduring partial friendships. Since, along with parent and child relationships, we conceive of friendships as paradigmatically partial relationships, the failure of such arguments in demonstrating that our lives must be lived partially to some extent should increase our confidence in the local claim that friendships can persist impartially. Bernard Williams suggests that exhibiting preferential concern in certain circumstances is necessary for living a purposeful life. People who exclusively exercise impartiality are doomed to a meaningless and purposeless existence. He says ‘‘somewhere one reaches the necessity that such things as deep attachments to other persons will express themselves in the world in ways which cannot at the same time embody the impartial view, and that they also run the risk of offending against it. They run that risk if they exist at all; yet unless such things exist, there will not be enough substance or conviction in a man’s life to compel his allegiance to life itself. Life has to have substance if anything is to have sense, including adherence to the impartial system; but if it has substance, then it cannot grant supreme importance to the impartial system and that system’s hold on it will be, at the limit, insecure.’’6 Owen Flanagan suggests that Williams is intent on demonstrating the irrationality of living a life guided by universal impartialism.7 Flanagan’s interpretation gains additional

68

MARK BERNSTEIN

weight from Williams’s comment that it is ‘‘unreasonable for a man to give up ... something which is a condition of his having any interest in being around in that world at all.’’8 Flanagan sees two sources of this irrationality. The first source resides in the psychological fact that personal goods develop prior to the stage of mental development necessary to form a commitment to impartiality. Concrete ends must antedate abstract ends. But this is a poor reason for believing in the irrationality of abiding by an unbridled impartialism. While it might be impossible to develop a commitment to impartiality without first adopting personal allegiances, this does not show that after some time we might reflectively and rationally abdicate such partial associations and maintain an allegiance solely to impartial morality. There is no reason to believe that, in general, just because we are psychologically compelled to one stage prior to another that we cannot rationally disavow the first stage in favor of the second stage. The second source is that some ‘‘commitments are of such importance to the agents … that these agents can have no attachment to life, no motivation for anything, without being engaged in’’ them.9 It would be irrational to dedicate our lives to something that would render them meaningless. This interpretation has far greater textual support but, without further elucidation, is liable to mislead by allowing us to infer a weaker argument than Williams intends. After all, even without a fullblown account of rationality, we can agree that a mother may rationally sacrifice purpose in her life to guarantee a healthy, meaningful life for her son. Martyrs do not merely sacrifice meaning from their own lives but give up life itself so that dearly-held causes have better chances for survival. Especially when we regard the principles for which they give their lives noble and fine, such martyrs, far from being conceived as irrational, are admired, even venerated, and are heralded in stories we tell for our children as role models. Let us reconstruct Williams’s argument as a reduction ad absurdum argument. First, we assume that we can grant supreme importance to the impartial system. If we grant supreme importance to the impartial system, then our lives would have insufficient conviction or substance to want to continue living. If our lives would have insufficient conviction or substance to want to continue living, there would be no sense in pursuing any goal or to regard anything as important to oneself, and therefore, most pertinently for our discussion, there would be no sense to adhere to the impartial system. If we grant supreme importance to the impartial system, it would be senseless to adhere to the impartial system. Granting supreme importance to the impartial system is pragmatically self-defeating. Thus our original assumption is false and we cannot grant supreme importance to the impartial system.

FRIENDS WITHOUT FAVORITISM

69

At this juncture, the notion of irrationality can be reinstituted. It is irrational is to knowingly pursue a pragmatically self-defeating goal. More specifically, it is irrational for an individual to attempt to grant supreme importance to an impartial system since success in such an attempt would, according to Williams, guarantee its failure. Living lives guided by universal impartialism would leave us without sufficient motivation to live by its precept. The irrationality in question, then, is not that an irrational person is bound to live a purposeless life if he aligns himself with an unfettered impartialism, but that in living such a meaningless life he can no longer live by the impartial viewpoint to which he had granted ultimate significance. To add concreteness to the argument Williams advances, we can consider an example that is paradigmatic of the kind he envisions. Among the many people who live in countries where there are long waiting periods for medical attention, some may be able to pull strings to schedule surgery the very next week for a son with cancer rather than wait two or three months until his turn comes up. The mother of the ill son, we may suppose, loves him very much.10 Such a parent’s deep attachment to her son is expressed by pulling strings so that her son can receive early treatment of his cancer. This partial behavior conflicts with the claim of impartiality that enjoins the mother to have her child wait for his scheduled turn for treatment. We are to imagine that the mother’s deep attachment or love for her son is what gives her life substance and meaning. From her perspective, there would be no point in continuing to live if she were to submit to the moral or impartial claim. But with no point left to live, with no purpose remaining for any action, the mother would surely be either irrational to try to adhere to the dictates of impartiality, or be psychologically unable to adhere to the mandate of impartiality. An unwavering allegiance to impartial demands is, then, either irrational or psychologically impossible. While we can imagine such a mother reacting in the way Williams envisages we surely need not do so. Why, in general, should we find the second premise of Williams’s reconstituted argument persuasive? Perhaps the best we can do to support it is to argue that only if we can express our deep attachments can life have sufficient substance or conviction for maintaining our desire to live. If we grant supreme importance to the impartial system, it prevents us from expressing our deep attachments. Therefore if we grant supreme importance to the impartial system, our lives would have insufficient conviction or substance for us to want to continue living. There is no reason to think that the first premise in this short argument is true. We should agree that we have, and that it is good that we have,

70

MARK BERNSTEIN

deep attachments to other persons and that having such relationships makes us susceptible to experiencing conflicts with the claims of impartial morality. Let us even agree that life needs substance for anything to make sense, that life is something we care enough about so that we care about how we act and why we act the way we do, and needs substance in order for us to care about listening to the precepts of impartiality. But to claim that without deep personal attachments, there is insufficient ‘‘substance or conviction in a man’s life to compel his allegiance to life itself’’ is hyperbole. Aristotle reportedly chose the path of truth instead of blind allegiance to his friend and mentor, Plato, when conflict between the two arose, and Plato saw the quest for truth and wisdom having pride of place in a rational person’s goals. Greek and Shakespearean tragedies do not prompt stares of incredulity when they raise the advisability, of living a life of honor to a person filled with friendship, love, and even family. Many of our moral heroes attain their status in part because of their rare but meritorious ability to place the claims of morality above claims originating from personal relationships.11 Even in our quotidian world populated by the mother in the above example, we find the account of Williams too blunt and restrictive. It is just a fact that some mothers are psychologically equipped to side with the impartial standpoint. The devastation that Williams predicts is scarcely inevitable. The unromantic truth is that, at least most of the time, people get over these hurdles. The anodyne for romantics is the knowledge that persons can dedicate their lives to principles despite the loss of personal love in so doing. The attitudes that we have toward such a mother who makes the moral choice are far more nuanced than Williams would have us believe. While we are disposed to judge her harshly as a woman with perverted priorities and misguided judgments of what should be most important to her, we are also inclined to assess her as highly principled and honorable. We might believe that acting impartially in her situation would be too much to ask of the typical person and that it would be too harsh to blame anyone who could not muster the will to act as she does. Our ambivalent evaluations belie simplistically characterizing her decision as irrational. Looking at the interior dialogue of such mother in also suggests a more subtle and textured situation than Williams allows. Williams would have us believe that after becoming acquainted with his argument, the intelligent, non-self-deceptive mother in such a case would realize that any attempt to abide by impartialism is futile and so irrational. But such a realization seems at odds with what would likely be an internal struggle. Such a motivational battle occurs as an intense conflict, likely to produce angst, guilt, and shame regardless of which combatant wins out. If par-

FRIENDS WITHOUT FAVORITISM

71

tiality informs her will, she may feel sickened by the thought that she subjected other, equally worthy, children with an undeserved disadvantage. We can understand why she may think that other mothers are warranted in criticizing her for helping her own son to the detriment of other children. Indeed, she might recognize that she would feel the same way if positions were reversed. If the claims of impartiality carry the day, self-berating emotions may result from the realization that she did not do everything in her power to help her son. She might question the quality of her motherhood and wonder whether she ever should have given birth to her son in the first place. Far from thinking this mental battle as something foreign, we would find a mother who did not experience the decision-making process with ambivalence and dread, as might a woman who simply realized that there is no viable option for the impartial view, as almost alien. Even the mildly sensitive among us think that she should find the decision gut-wrenching, and if she does not, it does not as Williams would have it, speak approvingly of her rational powers. It speaks disapprovingly of her lack of humanity. Christine Korsgaard champions the idea that the integrity of our personal identities suffers irreparable damage if we do not exhibit partial behavior in our personal relationships. She reminds us that we have practical identities, self-conceptions that reflect how we value ourselves, what we find worthwhile about our lives, and which actions we think worth undertaking.12 We think of ourselves, for example, as people, Jews, Blacks, college professors, and parents. These conceptions give rise to reasons to act and these reasons manifest our identity or nature; acting on these reasons demonstrates who we are. To the extent that this cluster of conceptions is flouted, our identities are fractured. Korsgaard understands personal relationships as reciprocal commitments ‘‘on the part of two people to take one another’s views, interests, and wishes into account.’’13 She tells us that this type of reciprocity leads to a Kantian notion of a unity of will where, at least in areas concerning their relationship, two persons must deliberate as one. This unified deliberation means that personal relationships are constitutive of our practical identities. Both of the partners in the relationship think of themselves as part of a couple and not an isolated individual where the only desires and interests that matter are an individual’s. In a personal relationship, each person is especially committed to being an end for each other. To Korsgaard, then, as opposed to Williams, the catastrophic cost in relinquishing partial behavior is not so much that life would lose its substance, but that life would lose its integrity. But the idea of practical identity is too flimsy a basis for the necessity of partial behavior. Just as someone can conceive of herself as white and a

72

MARK BERNSTEIN

New Yorker, the virtually endless elasticity inherent in the notion of practical identity would allow a person to think of himself as a staunch apologist of impartialism. Under this self-conception, extending favoritism to members of any specially related association would constitute a breach in his identity. Thus there is nothing intrinsic to the notion of practical identity that ensures vindication of parceling out special consideration to a set of individuals. Conflicts among a person’s practical identity are all but inevitable. We may ask if a person who conceives himself as a white Jew should grant privilege the interests to a black Jew or a white Christian. Perhaps some hierarchical structure can be appealed to, but it unlikely that such a ranking could withstand the virtually unlimited number of contexts in which conflicts might arise. Fissures in the practical identities of individuals, far from being rare occurrences, happen frequently, and, by and large, we seem to survive well. We begin to wonder whether the ominous loss of integrity that follows from conflicts among segments of our practical identities is a bit of overblown rhetoric. The theme of hyperbole returns in Korsgaard’s idealization of personal relationships. While it is true that personal relationships involve seriously considering and mutually appreciating each other’s interests and values, such reciprocity scarcely guarantees univocal points of view. A married couple might make some decisions as a team but it is exaggeration to suggest that they make all, or even most, of their decisions as one person. In fact, if such were the case, it would seem more destructive, than their respective identities. People want to be challenged and provoked, at least occasionally, and having a psychological clone of oneself as a mate or friend seems like a recipe for stagnation and boredom. Close relationships without mutual respect are likely to evaporate but when the reciprocity reaches the Kantian stage to which Korsgaard alludes, consternation rather than exhilaration seems the fitting reaction. Interpreted as a commentary on moral motivation, Korsgaard’s account seems more Hobbesian than Kantian, since with it she gestures toward the idea that close relationships are ultimately maintained for reasons of self-preservation and self-interest. But surely the caring mother in the above example is not leading her son to the head of the queue because her identity would otherwise be threatened. Her motivation is other-directed or for her son’s sake. Her dilemma is not experienced as an internal conflict between maintaining and destroying her identity, but instead as between doing something that she thinks is unfair and perhaps morally wrong and acting for someone whom she deeply loves. If she decides to place her son forward in line, she might feel whole, but we assume that the thoughts of her self’s integrity are not what disposed her

FRIENDS WITHOUT FAVORITISM

73

to act in favor of her son. As spectators, we are sympathetic to her plight, because we assume that she is not motivated by self-interest. If we were later to discover that her internal battle concerned interior seamlessness, and not motherly love, our natural empathy toward her would be drastically diminished. While it might be detrimental to Hitler’s integrity to consider the interests of Jews impartially with the like interests of his fellow Nazis, our moral assessment of the situation would probably not place great weight on Hitler’s identity being splintered. Perhaps Hitler’s identity remains intact in preferentially treating the interests of Himmler and Mengele over the millions of people in Auschwitz, Buchenwald, and Theresienstadt, but it seems odious to think of this as the right sort of justification for acting partially. It is at least arguable that it is good for Hitler to privilege his fellow Nazis at the expense of others because this preferential behavior will maintain the integrity of his identity. Nevertheless, in an equally ordinary sense this particular favoritism is not good, simpliciter. The inference to be drawn is that appeal to a split self does not, in general, count as a good reason to circumvent the demands of an impartial morality. From a moral point of view and the question in hand, let us not forget, is whether partial behavior is morally legitimate, some identities should be fractured.

5. Partiality in friendships lacks firm grounding and the more global arguments purporting to demonstrate lives absent preferential concern are either impossible or catastrophic are unpersuasive. Conceiving friendships as requiring favoritism, far from deserving its platitudinous status in our commonsense morality, has counter-intuitive results. Friendships come in degree. Some persons are better friends to us than others. Since the core ingredient in friendship is the disinterested concern for the other’s welfare, it is reasonable to suppose that being a better friend is a function of how much partial concern is lavished upon the friend in question. Although not amenable to anything approaching mathematical precision, this is a product of two factors, the extent of partiality and the quantity of concern. Ideally, we want our best friends to score highly across both vectors. They care for us predominantly for our own sakes and not for their own, and they care for us a great deal. Akin to the commonly held belief that friendships comprise partiality, we have the equally entrenched so-called better friendship principle. According to this principle, all things being equal, the interests of better

74

MARK BERNSTEIN

friends require preferential consideration over the like interests of lesser friends. Persons who are skeptical of this principle are invited to consider a variant of a previous example. John and Fran are better friends than are John and Phyllis. Fran and Phyllis have equal interests in relieving their identically debilitating headaches. John has access to one aspirin, the whole of which is necessary and sufficient for relieving either Fran’s pain or Phyllis’s. Bracketing all other morally relevant factors, the example elicits the idea that John ought to dispense the aspirin to Fran and not Phyllis. The only apparent grounding for this belief is that Fran is a better friend to John than is Sue. It is true that being a better friend is a formal notion whose content may be supplied in various ways. But this should not deter us from drawing the moral consequence from the example that John ought to give the aspirin to Fran. In fact, it is a rare individual who awaits further explanation of being a better friend before either assenting to or dissenting from the above better friendship principle. This is a powerful indication that we conceive of this principle as a condition of adequacy of any account of being a better friend. We should next accept the idea that many persons instantiate the concept of self-love and selffriendship. This idea might sound quixotic. The notions of reciprocity and mutuality that play a large role in our normal concept of friendship gain little traction when only one individual is involved. Nevertheless, when we focus on love’s core idea, that the good of a beloved is desired for her sake, a large leap in imagination is not needed to realize how common self-love is. In fact, often the danger lurking in discussions of self-love comes from the direction of people who believe that it is either ubiquitous, where everyone loves themselves, or pure, where self-love, unlike other forms of love, cannot be tainted by desires for any persons sake except that of the beloved. Aristotle insightfully recognized that not everyone is a friend to himself, bracketing the bad man who ‘‘does not seem to be amicably disposed even to himself because there is nothing in him to love.’’14 Acknowledging that there is nothing good about himself, a bad man finds himself in the unenviable situation of constantly being in the company of someone he finds unattractive. Self-love need not be unadulterated. An individual can love himself for his own sake and for the sake of another person; knowing that it makes our mothers happy might be one reason that we love ourselves. Such qualifications on selflove have the salutary effect of making self-friendship more akin to our multi-party, garden variety friendships. Let us suppose that John is a friend to himself. If he is his own best friend, then according to the proponents of the better friendship principle he ought to privilege his own interests above the like interests of all other people even all his other friends. Could John be his own best friend?

FRIENDS WITHOUT FAVORITISM

75

Although there is no necessity that he is, he surely might be. He may care about his own sake, for his own sake, more than does anyone else. If so, the proponents of this principle tell us that he ought to give privilege to his interests over the like interests of anyone else.15 To make this discussion more concrete, let us consider this slight variant of the previous example.John is a better friend to himself than he is to Fran. John and Fran have equal interests in relieving their identically debilitating headaches. John has access to one aspirin, the whole of which is necessary and sufficient for relieving either his pain or Fran’s. Bracketing all other morally relevant factors, we conclude that John is obligated to give the aspirin to himself and not to Fran. But this result is counter-intuitive. It seems misguided to think that John is acting immorally if, out of friendship, he gives Fran the aspirin instead of taking it himself. Indeed, we are inclined to admire people who willingly sacrifice their self-interests for the good of others who, impartially considered, are just as worthy. Of such mettle are saints and heroes made. We are faced with a choice. We can give up the partiality of friendship principle, the better friendship principle, the belief that self-friendships are instantiated, or some of our most deeply entrenched moral attitudes toward others. In abdicating the partiality of friendship principle, we alleviate ourselves of the most fundamental cause of our problem.16 Notes 1. See Aristotle, Rhetoric, 1380b35-1381a2, in The Basic Works of Aristotle, ed. Richard McKeon (New York: Random House, 1941), pp. 1385–1387. 2. See ibid., 1155b29-1156a5, in, McKeon, op. cit., pp. 1159–1160. 3. See Immanuel Kant, The Doctrine of Virtue in The Metaphysics of Morals, ed. Mary Gregor (Cambridge, England: Cambridge University Press, 1991), ss 31–36, pp. 247–252. 4. See Simon Keller, ‘‘Four Theories of Filial Duty,’’ The Philosophical Quarterly, 56, 2006. 5. See Shelly Kagan, The Limits of Morality (New York: Oxford University Press, 1989) pp. 367–369, and Troy Jollimore ‘‘Friendship Without Partiality,’’ Ratio 13, (2000). 6. Bernard Williams, ‘‘Persons, Character, and Morality,’’ in Moral Luck (Cambridge, England: Cambridge University Press, 1981), p. 18. 7. See Owen Flanagan Varieties of Moral Personality: Ethics and Psychological Realism (Cambridge, Mass.: Harvard University Press, 1991), pp. 68–71. 8. Williams, op. cit., p. 14. 9. Flanagan, op. cit, p. 70. 10. See Marcia Baron, ‘‘Impartiality and Friendship,’’ Ethics, (1991), p. 856.

76

MARK BERNSTEIN

11. See Susan Wolf, ‘‘Moral Saints,’’ The Journal of Philosophy 79 (1982). 12. Christine M. Korsgaard, The Sources of Normativity (Cambridge, England: Cambridge University Press, 1996), p. 101. 13. Ibid., p. 127. 14. See Aristotle, op. cit. 1166b25–26, in McKeon, op. cit., p. 1083. 15. See ibid., 1168b9–10, in McKeon, op. cit., p. 1086. 16. I dedicate this paper to Ron Miller who died 5 August 2006.