Love as valuing a relationship

Philosophical Review

Volumn 112, Issue 2, 2003, Pages 135-189

  Kolodny, Niko ^a  

^a HARVARD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33746156540     PISSN: 00318108     EISSN: 15581470     Source Type: Journal    
DOI: 10.1215/00318108-112-2-135     Document Type: Article

References (73)

1. "Love as a Moral Emotion," Ethics 109 (1999): 338-74.
2. "Autonomy, Necessity, and Love" and "On Caring," in Necessity, Volition, and Love (Cambridge: Cambridge University Press, 1999), 129-41, 155-80;
3. and "Some Mysteries of Love," Lindley Lecture (Lawrence: University of Kansas, 2001).
4. This "relationship theory" is in broad agreement with a suggestion made by Thomas Hurka, "The Justification of National Partiality," in The Morality of Nationalism, ed. Robert McKim and Jeff McMahan (Oxford: Oxford University Press, 1997), 139-57, at 150.
5. To love a person "as an individual," Hurka pro-poses, "involves loving the person for certain historical qualities, ones deriving from his or her participation with one in a shared history." I have doubts, however, about Hurka's claim that the relevant, reason-giving histories are ones of "doing good" together. First, a father who has thus far neglected his child has not had a history of doing good for it, but nonetheless has reason to love it. Second, a father who has cared for his young child has done good for it, but has not done good together with it. If the child is severely autistic, say, then the father may never do good with it. It might be replied that a history of one participant's doing good for the other is sufficient. But this would imply that the stalker who seeks to ingratiate himself by benefiting his target can succeed in dramatic fashion. He can, by doing so for long enough, give his target compelling reason to love him. Finally, the good done in a relationship often depends on the value of the relationship itself. Whether one views my friendship with someone as a history of doing good together, or instead as a history of arbitrary favoritism, would seem to depend on a prior judgment on whether that friendship provides me with valid reasons to treat my friend specially. That being said, there is a significant class of valuable relationships that consist in histories of realizing together some independently defined good. These are the "extrinsically finally valuable" relationships, such as relationships of collaboration, that I discuss in note 21, below.
6. Other intimations of the relationship theory appear in W. Newton-Smith, "A Conceptual Investigation of Love," in Philosophy and Personal Relations, ed. Alan Montefiore (London: Routledge and Kegan Paul, 1973), 113-36, at 124;
7. and Robert Brown, Analyzing Love (Cambridge: Cambridge University Press, 1987), 108-9. The relationship theory is also similar, in certain respects, to the account of "attachment" offered by John Maier, "Attachment and Identity" (unpublished typescript). Maier suggests that attachment to a particular object comes about, as a causal matter, through a history of "favoring" that object. For many of the reasons just discussed in conjunction with Hurka's theory, I do not think that normative reasons for love can be identified with histories of favoring, were one to attempt to turn Maier's account to this purpose. I suspect that there is no one single kind of history that gives us reasons for love, that in the end we must accept a kind of pluralism about valuable relationships.
8. That being said, there are some important commonalities between 'love' in some of its broader senses and 'love' in the narrow sense with which I am concerned. There is the obvious commonality that 'love' in all of these senses denotes a positive or favorable response to its object. More significantly, judgments that one has reasons to love some X are what I will call "subject nonuni- versalizable." It does not follow from the fact that I have reason to love X that everyone else has reason to love X. This is as true of love of candy apples and Jerry Lewis's movies as it is of love of particular people. By contrast, judgments that one has reasons for certain other favorable attitudes, such as admiration and respect, are subject universalizable. If I have reason to admire Lincoln or to respect human rights, then you do too. This is one way in which love is more personal, or expressive of one's individuality, than admiration or respect. See note 37, below, for further discussion of subject nonuniversalizability. I am indebted to Stephen Engstrom for this point. Finally, when we consider loving one's life's work, one's cause, or one's group, we may find an even closer commonality with loving a person. In these cases, one sees an ongoing history with one's work, cause, or group as a source of reasons for one's attitudes toward it, just as in the case of loving a person one sees an ongoing history with that person as a source of reasons for one's attitudes toward her. I return to this point in the conclusion, below.
9. For an exploration of this view, see Gabrielle Taylor, "Love," Proceedings of the Aristotelian Society 76 (1976): 147-64.
10. G. Pitcher, "Emotion," Mind 74 (1965): 326-46, at 340-42.
11. and, to a lesser extent, D.W. Hamlyn, "The Phenomena of Love and Hate," Philosophy 53 (1978): 5-20, agree that love is an emotion, but conclude that it is an exception from the general rule that emotions are responses to reasons.
12. Some might think that this concedes too much to the objection. After all, are people not blamed for being unloving parents or spouses? I suspect that in these cases people are blamed not for failing to love, but instead for failing to perform some relevant voluntary action, such as doing what a loving person would do, or taking steps to bring it about that one does love. At the very least, this is all that people can properly be blamed for, or so I would claim. If I am wrong about this, and one can properly be blamed for failing to love, then so much the worse for the objection. I am indebted to Gavin Lawrence for pointing out that the relevant notion is not involuntariness, but instead nonvolun- tariness.
13. See Frankfurt, "Some Mysteries of Love," 4.
14. Compare Elizabeth Anderson's discussion of parental love in Value in Ethics and Economics (Cambridge: Harvard University Press, 1993), 4;
15. and Alan Soble, The Philosophy of Sex and Love: An Introduction (St. Paul: Paragon House, 1998), 97.
16. See O. H. Green, "Is Love an Emotion?" in Love Analyzed, ed. Roger E. Lamb, (Boulder: Westview, 1997), 209-24, at 211;
17. and Robert Kraut, "Love De Re," Midwest Studies in Philosophy 10(1986): 413-30, at 425.
18. For critical discussions of the idea that love is constant, see Amélie Oksenberg Rorty, "The Historicity of Psychological Attitudes: Love Is Not Love Which Alters Not When It Alteration Finds," Midwest Studies in Philosophy 10(1986): 399-412;
19. and Soble, Philosophy of Sex and Love, 145-51.
20. It might be said that this overlooks a reason that I have for loving Jane rather than her double: namely, that I have led Jane, but not her double, to expect certain things of me. Therefore, I have a promissory obligation, or something like it, to love Jane, whereas I have no such obligation to love her twin. To begin with, this proposal cannot save the quality theory, since it appeals to relational facts, which lie outside the confines of the quality theory proper. At most, the quality theory could be one part of a hybrid theory that viewed some combination of nonrelational personal qualities and relational promissory obligations as grounds for love. At any rate, for reasons that I can only gesture at here, a hybrid theory of this kind would still be inadequate. First, Jane may lack expectations of me. Perhaps she-my loving wife of fifty years, with whom I shared this life and raised these children-has been ravaged by Alzheimer's disease to the point where she can no longer recognize who I am. In this situation, the hybrid theory would seem to imply that now my love might as well migrate to her twin. Second, I may not have voluntarily and intentionally encouraged Jane's expectations, in which case I am not obligated to fulfill them. Now, it is true that in friendships and romantic relationships, I usually will have voluntarily and intentionally encouraged expectations. Yet in many family relationships, this is not the case. If Jane is my older sister, then I may have done nothing to invite her expectations of me. The hybrid theory would seem to imply that in this instance, my love for Jane might as well transfer to her double. Finally, and most decisively, there simply are no promissory obligations to love. This is revealed by the familiar phenomenon of "leading someone on." In leading Jane on, I get Jane to form the expectations of me that she would have if we in fact had a genuine friendship or romantic relationship, even though I lack the love for her needed for that to be the case. No one believes that I have thereby acquired an obligation to love her, and I would be insane to think I ought to go through the motions of loving her, as a kind of second-best strategy for minimizing her disappointment. Instead, my obligation is to come clean as gently as possible and to try to make amends, to the extent that I can do so in a way that is not painfully condescending. The theoretical explanation of this commonsense view, I suspect, is that one cannot have a promissory obligation to give a response, such as love, that is beyond one's voluntary control. For further discussion of these issues, see my Relationships as Reasons (Ph.D. diss., University of California, Berkeley, 2003), chap. 1.
21. On nonsubstitutability generally, see Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), 167-68;
22. Kraut, "Love De Re," 427-29;
23. Soble, Philosophy of Sex and Love, 139-45;
24. Roger Scruton, Sexual Desire: A Philosophical Introduction (London: Weidenfeld and Nicholson, 1986), chap. 5;
25. and Thomas Nagel, "Sexual Perversion," in Mortal Questions (Cambridge: Cambridge University Press, 1979), 39-52, at 42-43.
26. For a defense of the quality view against this objection, see Roger E. Lamb, "Love and Rationality," in Love Analyzed, 23-47. The nonsubstitutability objection should be distinguished from a related objection that might be labeled "subject nonuniversalizability." The objection is that if the quality view is correct, then insofar as X reasonably loves Y because Y has Q, for all Z, Z is unreasonable if Z does not love Y because Y has Q. If Jane's having Qjustifies my loving her, then it requires everyone to love her.
27. See, for example, William Lyons, Emotion (Cambridge: Cambridge University Press, 1980), 78-80. In note 37, below, I argue that the analogous objection against reasons for attraction depends on the false assumption that having a reason to 4> implies that, other things equal, it is unreasonable not to .
28. See Green, "Is Love an Emotion?
29. The notion of love as "bestowal" invoked by Irving Singer, The Pursuit of Love (Baltimore: Johns Hopkins University Press, 1994) bears a certain affinity to the view of Frankfurt that I go on to discuss.
30. "Some Mysteries of Love," 3.
31. "On Caring," 170. All subsequent page references are given in the text.
32. This is an assumption that I make, but do not defend, throughout this paper: that giving a reason to respond to some particular in a distinctive way necessarily involves predicating some general feature of that particular. I am indebted to Calvin Normore, who has expressed doubts about this assumption, for prompting me to make it explicit.
33. Proponents of this cognitivist account include Jonathan Dancy, Practical Reality (Oxford: Oxford University Press, 2000), 35-37;
34. Warren Quinn, "Putting Rationality in its Place," in Morality and Action (Cambridge: Cambridge University Press, 1993), 228-55;
35. Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986), 140-43;
36. and Engaging Reason (Oxford: Oxford University Press, 1999), 50-56;
37. T. M. Scanlon, What We Owe to Each Other (Cambridge: Harvard University Press, 1998), 37-49;
38. and R. Jay Wallace, "Addiction as a Defect of Will" Law and Philosophy 18 (1999): 621-54.
39. For a similar interpretation of the underlying appeal of Frankfurt's invocation of second-order desires, one to which I am heavily indebted, see R. Jay Wallace, "Caring, Reflexivity, and the Structure of Volition," in Autonomes Han- deln: Beitrdge zur Philosophie von Harry G. Frankfurt, ed. Monika Betzler and Barbara Guckes (Berlin: Akademie Verlag, 2000), 215-36.
40. See also Gary Watson, "Free Agency," Journal of Philosophy 72 (1975): 205-20.
41. I am indebted to Melissa Barry and Alison Simmons for prompting me to make this explicit.
42. One might suggest that a variant of the no-reasons view could enjoy many of the advantages of the relationship theory. Love, on this variant, consists not in basic desires to, say, help Jane, but instead in a nonbasic desire to help Jane produced by the conjunction of a basic desire to help one's friends and the belief that Jane is one's friend. This would provide a way to differentiate the desires constitutive of loving Jane from mere urges and instrumental desires to help Jane. It would also solve the problem of amnesia, by explaining how forgetting that Jane is one's friend extinguishes the desire to help Jane. When combined with what we might call a "weak Humean" theory of reasons for desire, this version of the no-reasons view could even account for the inappro- priateness of love in certain cases. According to the weak Humean theory, there are no reasons for basic desires, but one can have (or lack) a reason for a nonbasic desire D insofar as D would survive (or fail to survive) proper and fully informed deliberation from one's basic desires. The combination of this variant of the no-reasons view and the weak Humean theory would hold that one has no reason to love Jane if one's belief that Jane is one's friend (that is, that she reciprocates one's attitudes) is false. But it would hold that there are no reasons for love in the sense that there are no reasons for the basic desire to help one's friends that is partly constitutive of love. This variant would still be inadequate, however, in two respects. First, it would fail to explain why the absence of love is inappropriate in certain cases, such as when a father has no basic desire to help his children. Second, it would subscribe to what I believe is a distorted view of desire and rational action.
43. This issue is discussed in greater detail in my Relationships as Reasons, chap. 3.
44. This account of valuing is indebted to Samuel Scheffler, "Projects, Relationships, and Reasons," in Reason and Value: Themes from the Moral Philosophy of Joseph Raz, ed. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael Smith (Oxford: Oxford University Press, 2004). It is also similar to Anderson, Value in Ethics and Economics. My analysis, however, stresses the idea that valuing is a matter of being disposed to have certain emotional responses, rather than of actually having emotional responses. I also claim, whereas Anderson denies, that valuing X consists not only in a susceptibility to certain emotional responses, but also in certain beliefs: first, the belief that something provides reasons for this susceptibility and, second, the belief that something provides one with reasons for action.
45. For most purposes, "final" value is equivalent to what is more often called "intrinsic" value. The reason why I avoid the more familiar language of "intrinsic" value is that it suggests a value that something has independently of anything else. Regimenting this suggestion a bit, one might say that X is intrinsically valuable insofar as X is valuable whether or not any distinct Y is valuable, and X is extrinsically valuable insofar as X is valuable only because some distinct Y is valuable. While it is true that nonfinal value implies extrinsic value, it is not true that final value implies intrinsic value. Some final values are extrinsic; some things are sources of reasons only because something else is valuable. This is often true of what are called "personal projects" in the philosophical literature: individuals' avocations or life's work. A medical researcher, for instance, might have the personal project of finding a cure for some disease. On the one hand, his project is valuable only because what it might achieve-a cure-is valuable. In this sense, his project is extrinsically valuable. On the other hand, it is natural to say that his project is a source of reasons for him, a source of reasons distinct from the value of what it might achieve. This is suggested by the fact that the project provides him with reasons to continue with his own efforts, for example, even if abandoning the project would not significantly reduce the chances that a cure would be found by someone else, because, say, there is an equally talented researcher waiting to take his place. This is the phenomenon that Frankfurt has in mind, I believe, when he writes, "certain kinds of activity-such as productive work- are inherently valuable not simply in addition to being instrumentally valuable but precisely because of their instrumental value" ("On Caring," 178)
46. See also his "On the Usefulness of Final Ends" in Necessity, Volition, and Love, 82-94;
47. and David Wiggins, "Truth, Invention, and the Meaning of Life," in Needs, Values, Truth, 2d ed.(Oxford: Blackwell, 1991), 87-137, at 132-34.
48. Some relationships, such as those constituted by a history of collaboration on some joint enterprise or by a history of occupying correlative roles within an institution, are finally extrinsically valuable in a similar way. Relationships of these kinds are valuable only insofar as the aims of the enterprise or institution are valuable. If we come to believe that these aims are worthless or evil, then the relationships that they partly constitute will come to seem to us hollow, absurd, or worse. This helps to explain how we criticize claims that relationships of certain types, such as relationships between collaborators in an oppressive oligarchy, terrorist organization, or gang, are finally valuable. To say that friendships are finally valuable is not to explain why they are sources of reasons, but instead to report it. A particular relationship is a source of reasons because it falls under a specific kind, such as friendship, not because it has some further, more general property-"final value"-that it shares with, say, you, Picasso's Guernica, and the Grand Canyon. I am indebted to Judith Jarvis Thomson for compelling me to clarify this point.
49. Wai-hung Wong points out an apparent counterexample: children can be said to love their parents long before they have the conceptual resources to believe that their relationships are sources of reasons. My response is that 'love', when ascribed to children who lack these conceptual resources, simply denotes a different kind of psychological state. This is an instance of a more general phenomenon. Certain conceptual capacities are required for many of the psychological states to which normal adults human beings are subject, such as belief, desire, and fear. To have a belief, for example, one needs the concept of truth. To the extent that brutes and children lack these capacities, they do not have states of this kind. Their sensitivity to their environment involves counterparts to these states, which we typically understand by way of analogy to them.
50. Moreover, this is not to be explained in terms of differences in how A and C are causally or constitutively positioned to "promote" or "realize" instances of R. It is not as though A and C have the same reasons to "promote" friendship in general, say, and recognize that the most effective way to do this is to concentrate on their own friendships. This is not the way in which participants see their relationships as giving rise to reasons. Compare Scanlon, What We Owe to Each Other, 88-90.
51. Compare Soble's judicious insistence on the distinction between the "basis" and "object" of love, Philosophy of Sex and Love, 95.
52. In my view, the quality theorist is entitled to a similar response. Although my ground for loving Jane is that she is beautiful, the focus of my love is Jane herself, not the quality of beauty, or this particular instantiation of beauty. Although my reason for loving Jane is that she has these accidents, what I love is Jane, not the accidents. A quality theorist need not subscribe, for example, to the alleged Platonic view that what one loves is the Form that the beloved instantiates. See Gregory Vlastos, "The Individual as an Object of Love in Plato," in Platonic Studies (Princeton: Princeton University Press, 1972), 3-34;
53. and L. A. Kosman, "Platonic Love," in Facets of Plato's Philosophy, Phronesis supp. vol. 2 (1976): 53-69.
54. "Love and its Place in Moral Discourse," in Lamb, Love Analyzed, 153-63, at 156. All subsequent page references are given in the text.
55. For example, Pettit countenances rigidly individualized thoughts that contain descriptive elements: "This, my friend, is in need" (158).
56. Compare Kraut's remark, "So we might say: a proper name is committed to its bearer, in much the same way that a lover is historically committed to the object of his love" ("Love De Re," 424).
57. "Persons, Character, and Morality," in Moral Luck (Cambridge: Cambridge University Press, 1981), 1-19, at 18.
58. Frankfurt, "Some Mysteries of Love," 2-3;
59. and Liam Murphy, Moral Demands in Nonideal Theory (Oxford: Oxford University Press, 2000), 140 n. 36. Frankfurt correctly observes that the mere legal relationship of being someone's wife seems insufficient to bear the weight that Williams appears to place on it. After all, the man and his wife might despise one another, or they might have a marriage of convenience. I take it that Williams intends us to imagine a longstanding, loving marriage. She is not simply his wife, but the woman with whom he made his life, with whom he raised these children, who stayed by his side as he fought cancer, and so on.
60. R. Jay Wallace, "Three Conceptions of Rational Agency," Ethical Theory and Moral Practice 2 (1999): 217-42, at 225. At times, Pettit himself seems sympathetic to this view: "After all, it is not as if loving someone produces behaviour in a blind, reason-free gush of passion or in an unthinking exercise of habit" (157).
61. A point emphasized by Michael Bratman, Intention, Plans, and Practical Reason (Cambridge: Harvard University Press, 1987).
62. Note that this concessive reply is sufficient for the purpose of justifying partiality in friendships and romantic relationships, which is an application of the relationship theory that I describe in the conclusion.
63. The terms 'insistent' and 'noninsistent' are from Shelly Kagan, The Limits of Morality (Oxford: Oxford University Press, 1989) 378-81. I am less skeptical than Kagan is, however, that there are noninsistent reasons.
64. This phenomenon perhaps depends on a distinctively modern conception of friendship and romantic love, to which equal standing is essential. Historically, there have been relationships of these kinds that have not assumed equal standing. The present explanation implies, as seems correct, that the kind of love that "superior" participants in these relationships had for their "inferiors" did not depend on the degree to which they respected them as equals. Of course, they might still have respected their inferiors in some other way, say, as sturdy and dependable servants (albeit coarse and childish in their way).
65. Reasons and Persons (Oxford: Oxford University Press, 1984), 305.
66. See also his discussion of the nineteenth-century Russian, 327.
67. It might be objected that there cannot be reasons for attraction, because these "reasons" do not universalize over subjects. From the fact that I am attracted to Jane, in light of her sense of humor, it does not follow that you are rationally required to be attracted to her, in light of her sense of humor. This objection is misguided, although the issue is complex. Certain qualities cannot count as reasons for anyone to be attracted a person. The weight of a person's kidneys or her social security number, for example, do nothing to render attraction to her intelligible. Nevertheless, we are fairly promiscuous about the qualities that we recognize as rendering attraction intelligible. Within these permissive limits, the reasons for attraction provided by these qualities are noninsistent reasons. Finding one set of qualities (within these limits) appealing is appropriate, but failing to find some other set (within these limits) appealing is not inappropriate. This judgment universalizes. It is open to everyone, but not required of everyone, to be attracted to this set of qualities. Moreover, one can have an insistent reason to be attracted to a particular person. Given that one is the kind of person who finds this set of qualities appealing, the fact that Jane has this set of qualities is an insistent reason to be attracted to Jane. This reason also universalizes. For everyone who finds such qualities appealing, the fact that Jane has such qualities is a reason to be attracted to her. If one is not attracted to Jane, because one falsely believes that Jane lacks the qualities that one prizes, then space opens up for a kind of normative advice: "You ought to be attracted to her; she's just your type." This reason also universalizes over other objects of attraction with the same qualities. Note that one has just as much reason to be attracted to Jane as to be attracted to twin-Jane. Attraction is not nonsubstitutable. I am indebted to Sam Scheffler for helping me to sort these issues out.
68. I owe this point to Jay Wallace.
69. Whereas I argued that the problem with Frankfurt's view is that we cannot find a list of things regarding one's beloved desiring which is sufficient for love, Velleman argues, in effect, that the chief problem with Frankfurt's view, as well as the similar views of "analytic philosophers on love," is that we cannot find a list of things regarding one's beloved desiring which is necessary for love. The problem, according to Velleman, is that for any given list of things that a loving agent is supposed to desire regarding his beloved, we could imagine a case in which X does not desire these things regarding Y, but nonetheless loves Y. He observes, with characteristic sensitivity to the complexities of human psychology, that one can love a "meddlesome aunt, cranky grandfather, smothering parent, ⋯ overcompetitive sibling," or ex-spouse without having any "desire for his or her company," and that "someone whose love was a bundle of these urges, to care and share and please and impress-such a lover would be an interfering, ingratiating nightmare" (353). One might worry that these examples tell equally against my claim that love is partly constituted by the belief that one has reasons to act in one's beloved's interests. This worry would be misplaced. The case of the ex-spouses, for example, is a case, like the one described in the previous section, in which the desire to engage in certain activities with one's beloved fades, but one's noninstrumental concern does not. A lover who is an interfering, ingratiating nightmare is simply a lover who either fails to understand fully the reasons that relationships of the relevant type provide, or acts, akratically, against that understanding.
70. Hence, if the choice that Williams describes is between saving his wife or some other single person, then the man's saving his wife could be viewed as an appropriate response to rational nature. He might simply let the selection be determined by whom he loves. By the same token, it would be an equally appropriate response to rational nature to let the selection be determined by who has a deeper voice, which might lead him to save the stranger. It is to explain why it would be uniquely appropriate to save his wife, and inappropriate to save the stranger-which seems to be what Velleman has in mind-that one needs to appeal to values other than rational nature, such as his relationship to his wife.
71. Compare Michael Smith, "Immodest Consequentialism and Character," Utilitas 13 (2001): 173-94, to which I am indebted.
72. See, for example, Lawrence A. Blum, Friendship, Altruism, and Morality (London: Routledge and Kegan Paul, 1980), chap. 5;
73. and Frankfurt, "Autonomy, Necessity, and Love," 135.