Loving Someone in Particular*

Benjamin Bagley

People loved for their beauty and cheerfulness are not loved as irreplaceable, yet
people loved for “what their souls are made of” are. Or so literary romance
implies; leading philosophical accounts, however, deny the distinction, holding
that reasons for love either do not exist or do not include the beloved’s distin-
guishing features. In this, I argue, they deny an essential species of love. To
account for it while preserving the beloved’s irreplaceability, I defend a model of
agency on which people can love each other for identities still being created,
through a kind of mutual improvisation.

Let me begin with a scene from one of the most famous—if problematic—
novels about love ever written. In Wuthering Heights, Catherine Earnshaw
consents to marry Edgar Linton, a perfectly eligible match. But she is am-
bivalent about it. So she asks Ellen Dean, her longtime servant and con-
fidante, whether she ought to have done so. The following conversation in
chapter X ðrelated from Ellen’s perspectiveÞ ensues:

“There are many things to be considered before that question can

be answered properly,” I said sententiously. “First and foremost, do
you love Mr. Edgar?”
“Who can help it? Of course I do,” she answered.
Then I put her through the following catechism: for a girl of twenty-
two, it was not injudicious.

* Thanks to Andrew Franklin-Hall, Bennett Helm, Agnieszka Jaworska, Errol Lord,

Elijah Millgram, Ram Neta, Jerry Postema, Ryan Preston-Roedder, Geoff Sayre-McCord,
Andrea Westlund, the members of the Fall 2012 Dissertation Research Seminar at UNC
Chapel Hill, and the editors of Ethics. Special thanks to Michelle Mason, Jeff Seidman, Vida
Yao, and two referees for very helpful feedback on the last revisions, and to Susan Rod-
riguez, whose trenchant editing and good sense I relied on throughout. Finally, I am
particularly indebted to Susan Wolf, whose guidance and support enabled the ideas I was
reaching for in this essay to develop.

Ethics 125 ( January 2015): 1–31

© 2015 by The University of Chicago. All rights reserved. 0014-1704/2015/12502-00XX$10.00

13460.proof.3d 1 Achorn International 10/17/2014 8:14PM

 2 Ethics January 2015
“Why do you love him, Miss Cathy?”
“Nonsense, I do—that’s sufficient.”
“By no means; you must say why.”
“Well, because he is handsome, and pleasant to be with.”
“Bad!” was my commentary.
“And because he is young and cheerful.”
“Bad, still.”
“And because he loves me.”
“Indifferent, coming there.”
“And he will be rich, and I shall like to be the greatest woman of
the neighborhood, and I shall be proud of having such a husband.”
“Worst of all! And now, say how you love him.”
“As everybody loves—You’re silly, Nelly.”
“Not at all—Answer.”
“I love the ground under his feet, and the air over his head, and
everything he touches, and every word he says—I love all his looks, and
all his actions, and him entirely, and altogether. There now!”
“And why?”
“Nay—you are making a jest of it; it is exceedingly ill-natured! It’s
no jest to me!” said the young lady, scowling, and turning her face to
the fire.
“I’m very far from jesting, Miss Catherine,” I replied. “You love
Mr. Edgar, because he is handsome, and young, and cheerful, and
rich, and loves you. The last, however, goes for nothing: you would
love him without that, probably, and with it you wouldn’t, unless he
possessed the four former attractions.”
“No, to be sure not—I should only pity him—hate him, perhaps, if
he were ugly, and a clown.”
“But there are several other handsome, rich young men in the
world; handsomer, possibly, and richer than he is. What should hin-
der you from loving them?”
“If there be any, they are out of my way—I’ve seen none like Edgar.”
“You may see some; and he won’t always be handsome, and
young, and may not always be rich.”
“He is now; and I have only to do with the present—I wish you
would speak rationally.”
“Well, that settles it—if you have only to do with the present,
marry Mr. Linton.”1

Ellen’s “catechism” strikingly anticipates the issues on which contempo- q1

rary philosophical discussions of love focus and the features that leading
accounts defend as necessary conditions for loving someone as a partic-
ular individual.2 Harry Frankfurt, for instance, insists that someone loved

1. Emily Brontë, Wuthering Heights, chap. 9; all further references to this novel are also
to this chapter.
2. As the word is used in contemporary English, many things other than persons can
be loved: animals, inanimate objects, institutions, activities, abstract ideas, deities, and so

13460.proof.3d 2 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 3
in the best sense is valued as irreplaceable: if Catherine really loved Lin-
ton, it would not be a matter of indifference to her that she love him in
particular, as opposed to anyone else with the same attractions. J. David
Velleman stresses that love should involve a special openness to beloveds
as they are in themselves, not just insofar as they serve your independent
purposes or meet some prior standard. Catherine shouldn’t love Linton
just because he pleased her, or satisfied her vanity, and it’s impossible to
see her claim to love him indiscriminately and in total as other than a
sarcastic parody of really loving attention. And Niko Kolodny emphasizes
that love should be constant: it should endure through a very wide range
of possible changes in a beloved. It shouldn’t lapse, as Catherine’s would,
when beloveds lose their looks, youth, cheer, or wealth.3 q2
Even more strikingly, however, the ideal of love the novel presents in
opposition to the defective view represented in Catherine’s initial re-
sponses is one that none of these philosophers can explain or even ac-
commodate. For the real source of Catherine’s ambivalence is that, as she
well knows, she doesn’t really love Linton at all. She really loves Heathcliff,
the darkly romantic foundling. And she loves him for a very different kind
of reason—“not because he’s handsome, Nelly, but because he’s more
myself than I am. Whatever our souls are made of, his and mine are the
same, and Linton’s is as different as a moonbeam from lightning, or frost
from fire.” Catherine’s answer raises a puzzle. Why should qualities of
Heathcliff’s soul—or, less metaphorically, of his identity or character—do
any better by the standards of Ellen’s catechism than any of the qualities
Catherine cited in Linton’s case? Aren’t the values with which Heathcliff
identifies just as repeatable, in principle, as Linton’s handsomeness or
wealth and just as liable to undergo changes that real love should survive?
Isn’t their significance to Catherine just as circumscribed by her private
interests and criteria, if not more so? Perhaps impressed by such ques-
tions, Frankfurt, Velleman, and Kolodny all defend theories on which the
qualities of one’s character and values are indeed no more suited to serve
as reasons for love than any other quality of one’s person. In this, they
represent a broad consensus among analytic philosophers on love. But I
will argue below that these philosophers are wrong and Catherine is right.

on. Though my discussion touches on love for some of these things at points, I assume as a
working hypothesis that there is a distinct, philosophically interesting species of love
essentially focused on particular persons. It is with this species of love that the following is
concerned.
3. Harry Frankfurt gives his theory its signature statement in “On Caring,” in his
Necessity, Volition, and Love ðCambridge: Cambridge University Press, 1999Þ, and its most re-
fined one in Taking Ourselves Seriously and Getting It Right ðStanford: Stanford University Press,
2006Þ. J. David Velleman presents his view in “Love as a Moral Emotion,” Ethics 109 ð1999Þ:
338–74, and elaborates it in “Beyond Price,” Ethics 118 ð2008Þ: 191–212. Niko Kolodny’s
proposal is in “Love as Valuing a Relationship,” Philosophical Review 112 ð2003Þ: 135– 89.

13460.proof.3d 3 Achorn International 10/17/2014 8:14PM

 4 Ethics January 2015
My defense of Catherine’s kind of love will proceed in two stages. In
the first, I argue, against Frankfurt, Velleman, and Kolodny, for the pos-
sibility and importance of a species of interpersonal love evaluatively
grounded in attractive qualities of the beloved. Frankfurt denies that love
is a rational response to value to begin with, Velleman argues that it is a
rationally optional response to a value that all persons ðby definitionÞ
share equally, and Kolodny argues that it is a response to the value of the
relationship you have to your beloved. But I argue, first, that there must be
reasons for love; second, that these reasons must ðat least in some casesÞ be
selective; and, finally, that these reasons must ultimately derive ðagain, at
least in some casesÞ not from the types of relationships you have to be-
loveds, but from what beloveds themselves are like. These theories, then,
leave a void that the ideal of Catherine’s love promises to fill.
Still, each theory gets something important right. Taken together,
they show that the irreplaceability, openness, and constancy on which
Ellen implicitly insists really are necessary to the best kind of love. There-
fore, Catherine’s answer can fulfill its promise only if the puzzle it raises
can be solved. In the second half of the essay, I argue that it can. It’s
possible, and plausible, to conceive of one’s identity as an agent as having
special structural features that enable it, distinctively, to support a form of
love that fully satisfies Ellen’s catechism. In a way, this turn to basic fea-
tures of agency and valuing should be unsurprising: the philosophical
questions raised by the phenomenon of loving someone as an individ-
ual turn out to be questions about the nature of individuality itself. I
explain these features by taking a simple and familiar idea literally: that
who you are is something you have to work out. Taking a cue from the
phenomenology of musical improvisation, I suggest that at least some of
the values with which you identify are ones you’re essentially in the pro-
cess of determining, such that their content depends on the ongoing se-
quence of judgments and actions you take those values to call for.
To apply this model of agency to love, I draw another analogy to jazz,
this time relating the attraction and concern constitutive of interpersonal
love to the reciprocal appreciation and responsiveness of musicians who
improvise together as partners. Musicians who improvise together as
partners recognize each other to be trying to express the same musical
idea, even though the contents of their ideas are still being worked out.
Similarly, I propose, to love someone in particular is to view that person in
the same way Catherine views Heathcliff: as creating an identity that is
somehow importantly like your own, in a way that makes your beloved
someone appropriate for you to create yourself together with. But be-
cause your reasons for love are grounded in features of your and your
beloved’s identities that are in the process of being determined, those
reasons persist throughout that process and call for essentially open-
ended forms of interested attention and emotional vulnerability. Further,

13460.proof.3d 4 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 5
they make you and your beloved irreplaceably valuable to one another,
since someone you are creating your values together with can share those
values in a way that nobody else can.4

I. LOVING SOMEONE FOR NO REASON

Perhaps the most glaring problem with Catherine’s reasons for loving
Linton is that they make him too easy to replace. What qualifies him as a
suitable beloved is simply that he is a member of the general class of
handsome, cheerful, rich young men. Any other member of that class
would have done just as well. But, as Harry Frankfurt insists: “With regard
to what we love . . . that sort of indifference to the identity of the object of
concern is out of the question. Substituting some other object for the
beloved is not an acceptable and perhaps not even an intelligible option.
The significance to the lover of what he loves is not that of an exemplar;
its importance to him is not generic, but ineluctably particular.”5 Some-
one you love as a particular individual, then, is someone you value as
irreplaceable. This means, minimally, that it must be important to you
that you love the particular person you do. Now, the simplest way to
account for this importance would be to hold that reasons for love are
perfectly particular themselves. Thus, Catherine indeed would have rea-
son to love Heathcliff but not Linton, but that reason would be primitive
and hence inexplicable. The result would be a direct ðif flat-footedÞ
interpretation of Montaigne’s famous statement of his love for his best
friend: “If a man should importune me to give a reason why I loved him, I
find it could not otherwise be expressed, than by making answer: because
it was he, because it was I.”6 But that is a nonstarter. “The beloved’s bare q3
identity,” as Kolodny explains, “cannot serve as a reason for loving her. To
say ‘She is Jane’ is simply to identify a particular with itself. It is to say
nothing about that particular that might explain why a specific response
to it is called for.”7 We might as well say love has no reasons at all. q4
Such is Frankfurt’s view. Love, he argues, “is a particular mode of
caring. It is an involuntary, nonutilitarian, rigidly focused, and—as is any
mode of caring—self-affirming concern for the existence and the good of
what is loved.” Since the “lover’s concern is rigidly focused in that there

4. In stressing the historical dimension of love ðand in attributing it to more basic

features of agencyÞ, I follow Amelie Oksenberg Rorty, “The Historicity of Psychological
Attitudes: Love Is Not Love Which Alters Not When It Alteration Finds,” Midwest Studies in
Philosophy 10 ð1987Þ: 399–412; and Alexander Nehamas, “The Good of Friendship,” Pro-
ceedings of the Aristotelian Society 90 ð2010Þ: 267–94, among others.
5. Frankfurt, “On Caring,” 166.
6. “On Friendship,” Essays of Michel de Montaigne ð1580, trans. C. Cotton and ed. W. C.
Hazlitt, 1877Þ.
7. Velleman, “Love as Valuing a Relationship,” 142.

13460.proof.3d 5 Achorn International 10/17/2014 8:14PM

 6 Ethics January 2015
can be no equivalent substitute for its object, which he loves in its sheer
particularity and not as an exemplar of some more general type,” loving
cannot be “the rationally determined outcome of even an implicit de-
liberative or evaluative process.”8
Frankfurt’s theory accounts for the beloved’s irreplaceability, but at
an unpalatably high cost. Ellen’s catechism illustrates how natural it is to
give and ask for reasons for love. Catherine’s initial refusal to give a rea-
son feels like a dodge. The problem with her later answers is that the rea-
sons she gives are bad ones, not that she is making a category mistake in
giving them at all.
Two established lines of criticism underscore this point. First, it is
simply not plausible that love consists in the attitudes Frankfurt claims it
does. There is a difference between loving someone and assuming, for no
further reason, the project of being the agent of someone’s interest.
Velleman observes that at “the thought of a close friend, my heart doesn’t
fill with an urge to do something for him, though it may fill with love.”9 I
care about my close friends and would do a lot for them if they asked me
ðand in some cases even if they didn’tÞ. But there are a lot of helpful
things I could do for my friends that I feel absolutely no desire to do, like
their laundry. It’s not that such desires are overridden by others or that
my friends would find it off-putting if I acted on them. It’s rather that
these sorts of things just aren’t what friendship is about.10
Second, there are some things it just doesn’t make sense to love.11
Suppose you were gripped by an involuntary, nonutilitarian, rigidly fo-
cused, and self-affirming concern for the existence and the good of a
random picnic table. One weekend you eat a hot dog there, but on the
drive home your thoughts keep returning to it, a vague fondness rising in
your breast. So every weekend thereafter you go back to Table 7-G to
clean it off, protect it from the elements, replace rotting beams, and so
on. This attitude is not just unusual. It’s positively perverse. Love for a
random picnic table is either irrational or unintelligible. We need to
explain why, and we need reasons for love to do it.

8. Taking Ourselves Seriously and Getting it Right, 40–41.

9. Velleman, “Love as a Moral Emotion,” 353. See also Kyla Ebels-Duggan, “Against
Beneficence: A Normative Account of Love,” Ethics 119 ð2008Þ: 142–70.
10. It might be objected that a willingness to do a ðtypicalÞ adult’s laundry is too
overbearing or infantilizing to express a genuine concern for that person’s good. But this
objection misses the point. If my friends insisted that they would not feel infantilized if I did
their laundry, this might convince me that doing so would be good for them. But it would
not incline me to do it.
11. This point comes from Troy Jollimore, in Love’s Vision ðPrinceton, NJ: Princeton
University Press, 2011Þ, 22–23. ðThe reader who doubts that a picnic table can have a good
may substitute a shrub, or small animal.Þ

13460.proof.3d 6 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 7
II. LOVING SOMEONE AS A RATIONAL AGENT

Against Frankfurt, J. David Velleman argues that the “ineluctably partic-

ular” nature of a beloved’s importance is not only compatible with love’s
being a rational response to a generally held valuable property, but is best
explained by viewing love as a response to a property that all persons
share by definition. This is one’s bare rational nature—a property equally
valuable in everyone, but whose value must be appreciated in ways par-
ticular to each instance of it.
In the Kantian framework Velleman assumes, your rational nature
is what makes you worthy of being valued as you are in yourself. The
responses it warrants come in two varieties: respect and love, “the re-
quired minimum and optional maximum responses to one and the same
value.”12 Both of these attitudes are special because the value of one’s ra-
tional nature is independent of, because prior to, the value of any prop-
erties that not all persons share equally—and hence which distinguish
particular persons from one another. This is why respect is equally owed to
everyone and why it consists, roughly, in according each individual the
basic regard to which one’s dignity as a person entitles one. The same
rational independence explains the special openness appropriate to the
individual value of a beloved.
While the demands of respect are predominantly negative—they
consist, primarily, in prohibitions against treating people in ways that ig-
nore their value as persons ðe.g., by manipulating or exploiting themÞ—
love consists in a heightened sensitivity to the significance of whatever
specific characteristics, attitudes, or interests a beloved manifests. It “arrests
our tendencies toward emotional self-protection from another person,”
Velleman writes, “tendencies to draw ourselves in and close ourselves off
from being affected by him.”13 This means love isn’t to be identified with
any specific motives or sentiments: for Velleman, “a sense of wonder at the
vividly perceived reality of another person” is the closest thing to a con-
stitutive feeling of love there is.14 Rather, it exposes lovers to a very wide
range of emotional responses—not all favorable—corresponding to the
wide range of features their beloveds might exhibit on given occasions.
They might be thrilled by the admiration of their beloveds, hurt by their
insults, and concerned about their needs, even when the same admira-
tion, insults, or needs would barely register if observed in others.
Though Velleman sometimes motivates this feature of love on phe-
nomenological grounds, he recognizes that its importance goes deeper.
It captures, he argues, the way people value their beloveds as special.

12. Velleman, “Love as a Moral Emotion,” 366.

13. Ibid., 361.
14. Velleman, “Beyond Price,” 199.

13460.proof.3d 7 Achorn International 10/17/2014 8:14PM

 8 Ethics January 2015
Being valued as special “doesn’t entail being compared favorably with
others; it rather entails being seen to have a value that forbids compari-
son. Your singular value as a person is not a value that you are singular in
possessing; it’s rather a value that entitles you to be appreciated singu-
larly, in and by yourself.”15 This implies, minimally, that if you really love q5
someone, you must be open to unanticipated developments in your own
ends in light of what that person has to show you. Your attentiveness and
vulnerability to them can’t be contingent on how well they serve your
existing ends or conform to your prior ideals, and so you can’t suppose
yourself to need any special reason for your heightened sensitivity to
whatever is significant to or about them. As an illustration, Velleman
describes watching his sons grow up:

In a quick succession of years I became deeply interested in lacrosse

and Morris dancing, poetry slams and photography, and specifically
in the accomplishments of a particular midfielder, Morris dancer,
poet, or photographer, because these were the directions that my
children had set for themselves. Of course, I eventually learned to
appreciate some of these accomplishments intrinsically: I would
realize with amazement that I was cheering as my son walloped a
schoolmate with a metal stick or that I was applauding choreography
that previously would have struck me as no more than quaint. But I
learned to appreciate these accomplishments, to begin with, because
they were the ones that my children had chosen to cultivate.16

But while Velleman is correct to stress the characteristic openness of love,

I think he gets its details wrong, because he misconceives the kind of
value to which love for an individual responds. People value their be-
loveds as incomparable, but only up to a point. They still take themselves
to have reason to love some people and not others.
On Velleman’s gloss Catherine would be perfectly correct to love
Heathcliff for his similarly constituted soul. Their “souls” are their bare
rational natures, and these are indeed exactly alike. Her mistake is just in
viewing Linton’s soul as any different. Catherine’s disposition to be vul-
nerable to Heathcliff ’s rational nature ðwhich presumably constitutes her
love for himÞ is a strictly causal matter, just an incidental quirk of her
psychology.17 But clearly that’s not how she loves Heathcliff, or how she

15. Velleman, “Love as a Moral Emotion,” 370. For another forceful defense of this
requirement, also on Kantian grounds, see Rae Langton, “Love and Solipsism,” in Love
Analyzed, ed. Roger E. Lamb ðBoulder, CO: Westview, 1997Þ.
16. Velleman, “Beyond Price,” 205–6.
17. Velleman is not explicit about the kind of psychological state in which love con-
sists, and much in his presentation can instead suggest a view of it as an occurent state of
arresting awareness. The latter interpretation, however, should be rejected on grounds of
charity. You no more cease to love someone when you are vexed or preoccupied than you
forget what they look like when you close your eyes. If this weren’t so, then either most

13460.proof.3d 8 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 9
should. Catherine doesn’t see him as someone she just happens to be
arrestingly aware of, like someone you happen to be standing next to at a
party and might as well make small talk with. Rather, she endorses her
love for him specifically.
Consider that the moral and prudential considerations in favor of
Catherine’s loving Linton instead are substantial. If she could somehow
replace her disposition to arresting awareness of Heathcliff with one
directed toward Linton, it would help her keep a promise, ease her into a
life of comfort and prominence, and orient her toward safer, more
socially acceptable, and morally improving pursuits. Given that, on Velle-
man’s view, Catherine has no special reason to love Heathcliff, why
shouldn’t she regard her disposition to arresting awareness of him as
other than a mere inconvenience? Of course, people don’t regard their
loves as dispositions to be managed at their moral and prudential con-
venience. For Catherine, suddenly ceasing to love Heathcliff is unthink-
able. The prospect would appear as a disturbing failure to appreciate his
profound significance to her. In short, it is essential to her love that she
experience it as more than merely optional.18
One might, therefore, ask why Velleman shouldn’t just jettison his
claim that love is rationally optional. It will be instructive to consider this
possibility. So modified, Velleman’s view would place love on the same
level with respect, as a rationally required response to rational nature as
such. Just as you have reason to respect everyone, it would hold, so too do
you have reason to love them. It’s just that it’s generally much harder to
love people than it is to merely respect them, and nobody is in a position
to blame you for failing.19 So those of us who fail to love as we rationally
ought, even those of us who fail radically—presumably, more or less all of
us—do so forgivably. Velleman clearly aims to avoid this view, and it is easy
to see why. It completely abandons the idea that love may be unapolo-
getically selective in any but the most superficial sense. But I have just
argued that the degree of selectivity Velleman is actually entitled to is
pretty superficial anyway. The difference is just that on the present mod-
ification one is to regard one’s psychological inability to perfectly love
everyone as a genuine rational imperfection, one we have reason to work
to overcome.20
people would love much less than they think they do, or love would be even more of a
headache than it already is. ðThis point comes from Susan Wolf’s first Phi Beta Kappa
Romanell lecture.Þ
18. Or, as Catherine herself puts it, “My love for Linton is like the foliage in the woods.
Time will change it, I’m well aware, as winter changes the trees—my love for Heathcliff
resembles the eternal rocks beneath—a source of little visible delight, but necessary.”
19. Or at least, on the religious version of this view, nobody on earth.
20. Perhaps Velleman feels the influence of this point in his suggestion that we are not
equally inclined to love everyone in part because the “human body and human behavior
are imperfect expressions of personhood, and we are imperfect interpreters” ð“Love as a
Moral Emotion,” 372Þ.

13460.proof.3d 9 Achorn International 10/17/2014 8:14PM

 10 Ethics January 2015
This answer is very honest in its way and should not be dismissed out
of hand. The idea that we are all to love one another unconditionally is
historically important and very powerful. It is fundamental to the Chris-
tian ethical tradition and may, not unrelatedly, be truest to the Kantian
spirit. So perhaps it should not be surprising if this is really where Velle-
man’s theory leads him. ðI suspect it is also the only coherent conception
of love as a truly moral emotion.Þ But the resulting ideal of love is not
only demanding, but strangely impersonal. It collapses, or at least triv-
ializes, the distinction between love for particular persons and love for
humanity as such.

III. LOVING SOMEONE AS A RELATIVE

Love for particular persons, therefore, must be a selective rational response

to valuable properties of the beloved. But not just any sort of property will
do. Linton’s attractive qualities, remember, would make him too replace-
able. Here, Niko Kolodny’s appeal to relational properties represents an
important advance.
Kolodny argues that reasons for love are grounded in the value of
one’s ongoing historical relationships to one’s beloveds. I love my brother,
for instance, because he’s my brother. We have the same parents and we
grew up together. This makes it easy to explain irreplaceability: it is im-
portant to me that I love my brother, in particular, because he is a partic-
ular person to whom I stand in the fraternal relation. Other people might
stand in the fraternal relation to someone—they might be other people’s
brothers—but that doesn’t give me reason to love them. They don’t stand
in the fraternal relation to me. ðNow, it so happens that I have two broth-
ers; Kolodny’s view implies that I thus have just as much reason to love the
one as the other. But of course that is the right implication: I do have
reason to love both equally.Þ
More specifically, Kolodny argues that relationships of certain types
are nonderivatively valuable and so constitute sources of reasons for love.
For him, loving someone consists in ðiÞ believing your relationship to that
person to be an instance of a valuable type, and thus ðiiÞ taking it to be a
reason both for being emotionally vulnerable to, and for acting in the
interest of, both your beloved and the relationship itself, in ways appro-
priate to relationships of that type, and ðiiiÞ believing that others in re-
lationships of the same type would have similar reasons for similar atti-
tudes concerning their own beloveds.21 When you love someone, then,
you value both your beloved and your relationship itself—but it’s the
relationship, in virtue of its type, that you see as the source of your rea-
sons to accord special value to each. This isn’t to say, of course, that lovers

21. Here I compress Kolodny’s formulation in “Love as Valuing a Relationship,” 150–51.

13460.proof.3d 10 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 11
don’t have reason to be interested in whatever valuable qualities their
beloveds might have. On the contrary, my love for my brothers involves, in
part, my taking a special interest in all kinds of great things about them.
But on Kolodny’s theory, my reasons for doing so aren’t grounded in the
value of the qualities themselves—those reasons would apply to anyone—
but in the value of my relationships to their particular bearers.
Note that even if the value of my fraternal relationships were only
derivative, I still would have special reason to care about my brothers,
since I still wouldn’t have these relationships with anyone else. But by
claiming that it isn’t, Kolodny elegantly accounts for the thought that
love should be constant: its reasons should endure through a suitably
wide range of changes in a beloved. I leave “suitably wide” vague on pur-
pose, but the basic idea should be intuitive. If my brothers came to have
very different valuable qualities, I’d have equally good reason to take a
special interest in those. I’d even have reason to care about my brothers
in the ðdistant!Þ possible world where they ceased to have much going
for them at all. For Kolodny, the explanation for this is simple: so long as
your beloved remains such that your relationship can somehow endure
as a valuable instance of its type, your reasons for love endure as well.
Like Velleman, Kolodny identifies a real and valuable kind of love. In
addition to familial love, his theory plausibly accounts for love of animals,
gardens, cars, institutions, and other nonpersons.22 Further, I think Ko-
lodny is right that our histories with our beloveds are ultimately what
make them irreplaceable to us. But they can’t always do so in the way he
thinks. His theory gets some of the deepest loving relationships back-
wards: we value them derivatively and only because of what our particular
beloveds are like.
As a first effort at bringing out what Kolodny’s view misses, observe
that a surprisingly wide range of relationships count as loving by his
criteria. One example is teaching. The teachers who get the most out of
their jobs normally believe themselves to stand in valuable pedagogical
relationships with their students. They take these relationships to be
reasons for acting in the interests of their students in pedagogically
appropriate ways and for pedagogically appropriate forms of emotional
vulnerability to them and ðsomewhat more subtlyÞ for corresponding
attitudes toward the relationships themselves. This isn’t to say they go in
for overbearing, Dead Poets Society- style sentimentalism—just that they are
emotionally invested enough to be pleased by their students’ progress,
troubled by their unnecessary confusion, and concerned to prevent their
pedagogical relationships from going badly or ending prematurely—for
example, with their students dropping out.

22. While Kolodny explicitly defines “relationships” in his sense as necessarily inter-
personal, I see no reason to view this limitation as more than stipulative ðibid., 148Þ.

13460.proof.3d 11 Achorn International 10/17/2014 8:14PM

 12 Ethics January 2015
Though it is arguably possible to love someone specifically as a stu-
dent, it would be a stretch to say that if you value your pedagogical
relationships in this way, you necessarily love every student with whom you
have one. What you love in this case is teaching. The students you love, if
any, are the special ones. With them, your pedagogical relationships are
valuable on another level—valuable not only as instances of a generally
valuable type, but because of specific characteristics of the students in
question. Thus, Minerva might spend an extra hour helping Neville, who
is pleasant enough and tries hard, out of her love for teaching and yet
do the same for Hermione, who is brilliant and delightful, additionally
out of pedagogical love for her. It will be a matter of indifference to her
that her pedagogical relationship is to Neville, as opposed to any of the
many other adequate Hogwarts students whom she might have taught
instead. But not so with Hermione. Yet it is exactly this distinction that
Kolodny’s view lacks the resources to draw.
It might be offered on Kolodny’s behalf that while Minerva may
indeed have a special relationship to Hermione, that relationship could
be a nonderivatively valuable relationship of a different type. Minerva
could, for instance, simply be Hermione’s friend.23 But once we see how
the distinction between nonderivatively valuable relationships and those
whose value depends on the specific person in question pertains to teach-
ing, its relevance to more intimate cases becomes apparent. It’s possible
to value a marriage—even deeply and for its own sake—to someone you
don’t love. This may be how Alexei Karenin viewed his marriage, for in-
stance, at least before he learned of Anna’s infidelity. As Anna’s husband,
Karenin would indeed have reason to take a special interest in Anna’s
attractive qualities, like her sensitivity and verve—but only because those
were the qualities that happened to be instantiated in his wife. There’d
be nothing about Anna’s sensitivity and verve as such, much less about
Anna herself, that gave those particular qualities a special claim on Kar-
enin’s attention. Had Karenin been married to someone else, her qual-
ities would have been just as lovable to him; Anna’s would have been
merely attractive.
Kolodny claims, in response, that “it doesn’t seem like a distortion to
say . . . that a wife wants to be loved by her husband, at the deepest level,
because she is the woman with whom he fell in love and made his life. . . .
Let us suppose that they had never met and had made their lives with
other people. Imagining herself in that situation, would she still want him

23. It might also be suggested that the difference in value is merely one of degree.
Hermione’s virtues might make a difference simply by enabling an especially valuable
instance of the same type of relationship Minerva has to Neville, not ðas I have supposedÞ by
adding an extra dimension of value specific to Minerva’s relationship to Hermione. But
whether or not this is plausible for teaching, it is not for marriage.

13460.proof.3d 12 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 13
24
to love her? Would it make sense to her if he did?” At first glance, this
reply seems right. But I think this is only because it subtly trades on
exactly the point Kolodny’s critic should press against him. If love is
indeed grounded in the value that certain characteristics of your beloved
have for you, it is plausible that the identity of those characteristics de-
pends on the sort of person you are. But love changes you. Had the
spouses married different people, they would have become different sorts
of people themselves. Of course they would have found different char-
acteristics lovable.
In order for Kolodny’s example to vindicate his theory, therefore, we
need to hold fixed the nonrelational properties of all parties concerned.25
Suppose you are married to a person named Smith. Your marriage has
been long and happy; you and Smith delight in and admire all kinds of
things about each other. You also happen to have a colleague, Jones,
whom you dated for a semester back in college. If you gave the matter
much thought, you’d admit to yourself that you’d likely have grown to
love each other if you stayed together and even that you still find her
fairly attractive. But you really don’t give the matter much thought. Jones
is nice enough, but you can’t imagine how anything about her could
grip you the way so much about Smith does. As far as you’re concerned
there’s no contest. It’s crazy to think that, married to Jones while re-
maining exactly the sort of person you are now and knowing as much
about the two as you do, Jones would be just as lovable to you as Smith
now is, and Smith’s radiance would be dimmed to that of a moderately
attractive acquaintance. But that implies that your reasons for loving Smith
do not derive, ultimately, from valuable properties of your marriage to
him. Rather, they derive from valuable properties of Smith himself.

IV. LOVING SOMEONE FOR A SELF LIKE YOURS

Taken together, the shortcomings of Frankfurt’s, Velleman’s, and Ko-

lodny’s views show the importance of a species of love grounded in a
selective appreciation of a beloved’s distinguishing features. Again, this is
not to say that the attitudes these philosophers defend are not important
too. Velleman articulates a powerful moral ideal, and Kolodny accurately
describes the bonds people form to family members and others with
whom they share significant histories, as well as their love for their chil-
dren and pets. But it is reasonable to want more. Like Catherine and

24. Kolodny, “Love as Valuing a Relationship,” 157.

25. Or, at least, as fixed as possible. It can be hard to draw a sharp distinction between
putatively nonrelational properties of someone’s personality and character from relational
ones like memories and ongoing concrete projects. This theme will turn out to be im-
portant later, but bracket it for now.

13460.proof.3d 13 Achorn International 10/17/2014 8:14PM

 14 Ethics January 2015
Heathcliff, the greatest lovers and truest friends love each other for what
they in particular are like.
When Catherine says she loves Heathcliff for what his soul is made
of, I assume that she is referring to his identity or character ðterms I’ll use
interchangeablyÞ, as defined by the values by which he finds it funda-
mentally worthwhile to live. Now, I haven’t yet done anything to show why
the properties that constitute selective reasons for love must, specifically,
be the beloved’s values. In fact, it might seem perfectly normal to love
someone for other things. Wealth might be mercenary, and handsome-
ness superficial, but what’s wrong with loving people for their cheerful-
ness—or, for that matter, for their intelligence, sensitivity, or sense of
humor?26
The enduring popularity of Wuthering Heights itself suggests, how-
ever, that there is something about Catherine’s kind of love that many
people find compelling. At least in rough outline it’s easy to see why. To
begin with, it seems on reflection that whether someone finds your other
qualities lovable ðrather than just interesting, sexy, or otherwise pleasingÞ
normally does depend, a great deal, on what those qualities say about
your values—on the interests or drives animating your intelligence, say, or
the outlook on life embodied by your jokes, or even the sensibility ex-
hibited in how you dress or walk.27 Further, and more basically, someone q6
who loves you selectively, for the specific values you identify with, sees you
for who you distinctively are—and finds you distinctively valuable as such.
There’s something wonderfully affirming and empowering about this—
especially when you love the other person in the same way—though it’s

26. Thus, in “Love as a Reactive Emotion,” Philosophical Quarterly 61 ð2011Þ: 673–99,

Kate Abramson and Adam Leite defend a view of love as “an affectionate attachment . . .
appropriately felt as a non-self-interested response to particular kinds of morally laudable
features of character expressed by the loved one in interaction with the lover” ð677Þ—those
“especially salient in the context of fairly intimate relationships” ð679Þ, like sensitivity and
kindness. While I agree that these qualities can matter to love, in my view their importance
is derivative, and essentially unrelated to their moral worth: like beauty, intelligence, and
vigor, they can be catalysts or background conditions for love between people who identify
with values expressed in them or in activities requiring them. This isn’t to say Abramson
and Leite are wrong about the affectionate attachment the virtues they uphold can inspire,
but even if this attitude is best understood as a kind of love—as opposed to a warm
combination of trust and gratitude—it seems far short of the kind Frankfurt, Velleman, and
Kolodny objectionably neglect. Linton, it turns out, actually has these virtues in spades, and
Catherine indeed finds herself becoming affectionate toward him as he expresses them in
interaction with her. ðObserving this, Ellen wholeheartedly approves: this, to her, is what
love should be.Þ But that affection is quickly and understandably overshadowed by her less
virtuous—but more profound—love for Heathcliff.
27. Martha Nussbaum makes the same observation, in defense of a similar thesis, in
“Love and the Individual: Romantic Rightness and Platonic Aspiration,” in her Love’s
Knowledge ðOxford: Oxford University Press, 1990Þ, 327.

13460.proof.3d 14 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 15
hard to explain what. ðI’ll offer more of an explanation below, once I’ve
said more about what Catherine’s kind of love involves.Þ
Once we’ve seen how essential irreplaceability, openness, and con-
stancy are to love, however, it can seem strange—if not paradoxical—that
the idea of being loved for your values could be so appealing. For while
these features pose a challenge to any conception of love as a form of
selective appreciation, the challenge seems especially severe in the pre-
sent case. Start with irreplaceability: while the values Catherine loves in
Heathcliff might be unusual, there’s no reason to think they’re essentially
unique to him, as would seem necessary for it to be important to her that
she love him in particular. On the contrary, they can’t be, if Catherine is
correct in taking herself to share them. And it’s with respect to the be-
loved’s values that openness and constancy seem most important. Think
of people who are disappointed when their beloved’s evaluations diverge
from their own, rather than being open to them as potential enlarge-
ments of their own perspectives. Their love—if it’s intelligible as such at
all—seems narcissistic on the part of the lover and insulting to the be-
loved. And if these putative lovers further saw the continuation of their
relationships as rationally contingent on ongoing adherence to the party
line, their interest would go from narcissism to possessiveness.
What I want to argue now, however, is that the idea of loving people
for values they share with you only faces these problems if we think of the
values in question themselves as fixed, static things. I don’t think values
are like this—not all of them, at any rate. I think instead that a beloved’s
values are loved as things that are essentially in the process of being
determined, through a kind of ongoing improvisation. To that end, I’ll
defend this conception of agency in the remaining sections, showing how
it yields an account of the nature and value of love for persons as par-
ticular individuals that vindicates both Ellen’s catechism and Catherine’s
ultimate answer to it.

V. IMPROVISATION AS A MODEL OF SELF-CREATION

Improvisation differs from other species of rational agency in that im-

provisers refine the ends they pursue—that is, the norms they are com-
mitted to—as they go. When you improvise, you act in ways you take to be
appropriate without necessarily being able to explain why. You don’t see
your actions as random, but rather as parts of a process of working out an
expressive musical performance, say—or an overall way of life—that ex-
plains why each of the actions that constitutes it is appropriate in rela-
tion to the whole. To see how this might go in the music case, consider
how Keith Richards recalls the improvisation that went into Exile on Main
Street:

13460.proof.3d 15 Achorn International 10/17/2014 8:14PM

 16 Ethics January 2015
There was no preparation. But that’s not the point; that’s rock and
roll. The idea was to make the bare bones of a riff, snap the drums
in, and see what happens. And it was the immediacy of it that in
retrospect made it even more interesting. There was no time for
much reflection, for plowing the field twice. It was “It goes like this”
and see what comes out. And this is when you realize that with a
good band, you only really need a little sparkle of an idea, and be-
fore the evening’s over it will be a beautiful thing.28

What’s exciting here is the immediacy—the spontaneity and adventure—

of playing without an antecedently fixed end. That, Richards implies, is
rock and roll. But these remarks fit uneasily with Richards’s account of
improvising with others. “There’s something beautifully friendly and
elevating about playing music with people,” he writes, precisely because
of its deep sense of common purpose:

You’re sitting with some guys and you’re playing and you go “Ooh,
yeah!” That feeling is worth more than anything. There’s a certain
moment where you realize you just left the planet for a bit and that
nobody can touch you. You’re elevated because you’re with a bunch
of guys that want to do the same thing as you. And when it works,
baby, you’ve got wings. You know you’ve been somewhere most
people will never get; you’ve been to a special place. And then you
want to keep going back and keep landing again, and when you land
you get busted. But you always want to get back there. It’s flying
without a license.29

What “same thing” could Richards and his bandmates all want to do? Given
the first passage, we may safely assume that the content of Richards’s end
is not especially detailed or determinate at the point at which he first takes
his bandmates to share it. They might agree about certain, very general
musical goals, but why should so minimal a consensus continue to unite
them going forward, as their own ends become more specific? Perhaps
they have some external reason for wanting to play together—they know
that doing so will tend to produce results pleasing to audiences or critics,
for instance—but this fails to explain why playing with just the right band,
in just the right way, should be so marvelously freeing. Richards is playing
exactly as he wants, for its own sake: he is not compromising in the name
of some further goal. To explain this, I want to explore a more direct

28. Keith Richards, Life, coauthored with James Fox ðNew York: Little, Brown, 2010Þ, 306.
29. Ibid., 105 and 97. Richards makes this observation while recounting an early gig
that included Mick Jagger and Brian Jones but neither Bill Wyman nor Charlie Watts. For
this reason I hesitate to refer to “Keith Richards and his bandmates” by the obvious proper
noun.

13460.proof.3d 16 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 17
answer: Richards really does take his bandmates to want to express the
same thing he does, even though it is as yet indeterminate what that thing
involves.
This is possible because improvised ends can ground reasons for ac-
tions and attitudes in a way that other indeterminate or underspecified
ends can’t. Because their content is fixed by the ongoing sequence of
responses improvisers actually take them to call for, improvisers can intel-
ligibly—and potentially correctly—take themselves to have reasons ðrela-
tive to their endsÞ to perform certain actions over others, even when the
prior states of their ends leave the matter unsettled. And—I will argue in
the next section, once the basic model is in place—they can likewise take
their ends to be shared by others.
As a first pass, think of a spontaneous decision to play a riff because
it strikes you, right then, as especially expressive. Decisions like this seem
to occupy an intermediate position between rationally determined con-
clusions of deliberation and one-off random acts. On the one hand, you
make them because they feel like the way to go at the time, not because
anything in your performance demands them. Relative to everything
you’ve played up to that point, any number of other riffs may have been
equally musically good, or better; you might no less reasonably have
played any of them. ðThe qualification “up to that point” is critical here,
as we’ll see in a moment.Þ On the other hand, you make them because
they feel like the way to go at the time—the particular riff you pick feels
right in some way, as more than just a matter of indifference.
We can plausibly capture this middle ground by recognizing how a
riff can strike you as especially expressive of something without your
being able to say what it’s expressive of. It’s not necessary for you to be in
the possession, or suppose yourself to be in the possession, of any facts
that would explain why that specific riff was the right thing to do, given
your end. The causal explanation of your decision to play it, as opposed
to anything else you might equally well have played, would cite all sorts
of totally incidental features of your psychology and circumstances.
ðImagining a causal explanation of Keith Richards’s decisions, the mind
boggles.Þ Yet the whole point of improvisation is that in such a case
you’re not treating your decision as incidental. Parts of a musical per-
formance aren’t expressive in isolation. Since the feel of a riff is colored
by the past playing that anticipates it and the future playing that in-
tegrates and elaborates on it, to take a riff to be expressive is implicitly to
situate it, normatively speaking, in a larger context. You regard the riff as
part of the process of expressing a specific musical idea; as you work out
the contours of that idea, you’re simultaneously working out what the
riff you took to be expressive actually helps express.
For example, think of what gets expressed in the song “Happy.” ðA
“sublime example of a song winging in from the ether,” Richards recalls,

13460.proof.3d 17 Achorn International 10/17/2014 8:14PM

 18 Ethics January 2015
made from start to finish in four hours.30Þ “Happy” is a happy song, but it’s
not just that. Its opening riff is enormously upbeat, but as it gets repeated
in the context of the vocals, the baritone, and the drumming, it quickly
takes on a momentum that makes it defiant, even pleading. What the
song ends up expressing is an immediately recognizable but very com-
plicated attitude, at once happy-go-lucky, demanding, and vulnerable. I
want to say that insofar as Richards was moved to play these opening licks
because they struck him as especially expressive, they struck him as ex-
pressive of something bigger—something that ended up being this
complicated attitude. But in order for him to work out what this bigger
thing was, he needed to keep constructing the song and fit together the
things that struck him as expressive at various points. In this, he de-
termines the content of his end through pursuing it.
That is, it’s plausible—as a general thesis about rational agency—
that when you act for a reason, rather than arbitrarily, you necessarily
presuppose that there is an explanation why the action you perform is
appropriate.31 When you walk down Mulberry Street to get to the post q7
office, you presuppose that Mulberry Street is a route to it. When you play
a certain riff in order to express your musical idea, you presuppose that
your action admits of justifying explanation in the same basic sense. The
difference is that it is as yet unsettled just what the explanation is sup-
posed to be, because it is as yet unsettled what “expressing your musical
idea” consists in. It depends on the particular responses you make over
the course of the improvisation. Or, more exactly: to improvise is to pur-
sue an end whose content depends, epistemically and ontologically, on
the actions you actually take over the course of pursuing it, such that,
relative to your end, you have reason to do something just to the extent
that it admits of justifying explanation in terms of past and future actions
you perform ðand take yourself to have reason to performÞ that are
likewise explicable themselves.
How can a justifying explanation for one response essentially de-
pend on others? The question arises, I think, because it can be tempting
to assume that to give a justifying explanation for a response is to sub-
sume it under some general rule or principle: I can explain why it is
appropriate to say “10” after “8” in a simple counting game by citing a rule
of adding two. But it is a mistake to think that all justifying explanations

30. Ibid., 308. For lyrics, commentary, and a sample of the song, see “Happy—the
Rolling Stones,” AllMusic, http://allmusic.com/song/happy-mt0007444174. Thanks to an
editor of Ethics for pressing me for clarification here.
31. For an influential presentation of the concept of an “explanation why” in the sense I
have in mind, see John Broome, “Reasons,” in Reasons and Value: Themes from the Moral
Philosophy of Joseph Raz, ed. R. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael Smith
ðOxford: Oxford University Press, 2004Þ. There, Broome argues that this concept is prior to
the concept of a normative reason, but it is obviously unnecessary to accept Broome’s con-
clusion in order to admit a necessary relation between the two.

13460.proof.3d 18 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 19
32
must be like this. It’s possible to explain why a response is appropriate q8
simply by citing how it would be appropriate to respond in other cases,
thereby making the appropriateness of the response at hand intelligible
as part of a natural pattern. In the counting game, for instance, we can
say: “Look, we know ‘8’ was appropriate after ‘6,’ ‘6’ after ‘4,’ and, ‘4’ after
‘2.’ So it’s only natural that ‘8’ should call for ‘10.’” Of course, here it’s
obvious what the general rule in force would be, but in other contexts—
including, prominently, aesthetic ones—it may not be. It may not even be
possible to state such a rule at all. I doubt it’s possible to completely
articulate the complex feelings a performance of “Happy” expresses, in
abstraction from the particular responses that constitute it. ð“If you could
say it in words,” Hopper said, “there would be no reason to paint.”Þ
But there’s no need to. We can get a rich understanding of what
“Happy” expresses by seeing how its elements cohere, how they instanti-
ate recognizable musical structures and form natural patterns. As Ger-
ald Postema puts it, “melody-thinking is holistic thinking.”33 In my view,
it is holistic in the strong sense that the ends relative to which responses
in an improvisation are presupposed to be explicable may only emerge
over time, as manifested in unfolding patterns of agency.
On this model, improvising is both like and unlike pursuing an end
with determinate content you’re trying to discover.34 It’s similar in the
sense that you can intelligibly take yourself to have reason to perform
certain actions over others and thereby refine your conception of your
end. Even when the content of your end, in the form accessible to you at
the time, leaves a choice rationally underdetermined, you can still take

32. Arguably, at least some justifying explanations cannot be like this, in light of the
puzzle about rule-following Kripke famously attributed to Wittgenstein in Wittgenstein on
Rules and Private Language ðCambridge, MA: Harvard University Press, 1982Þ. Compare
John McDowell, “Non-cognitivism and Rule-Following,” in his Mind, Value, and Reality ðCam-
bridge: Cambridge University Press, 1998Þ; and Hannah Ginsborg, “Primitive Normativity and
Skepticism about Rules,” Journal of Philosophy 108 ð2011Þ: 227–54, the latter of whom likewise
takes inspiration from aesthetics.
33. Gerald Postema, “Melody and Law’s Mindfulness of Time,” Ratio Juris 17 ð2004Þ:
203–26, 208.
34. In The Retrieval of Ethics ðOxford: Oxford University Press, 2009Þ, Talbot Brewer
makes a similar appeal to musical improvisation as a paradigm of what he calls “dialectical
activity,” which he likewise conceives as a process through which agents refine their con-
ceptions of their ends through their efforts at achieving them. However, Brewer’s theory
differs from mine in that it is explicitly and unapologetically Platonic. Whereas I use
improvisation to model a process through which agents freely create their ends for them-
selves, Brewer argues that agents gradually acquaint themselves with the ideal forms in which
activities of certain types are to be pursued. This leads to some strange results. Jazz turns out
to be an effort to apprehend and instantiate objective aesthetic ideals, rather than an act of
personal expression. And while persons are properly loved for their developing evaluative
outlooks, an “evaluative outlook is properly loved only because and to the extent that it
exemplifies the zeal for adherence to objective truths about the good that is the proper telos
of the human capacity for practical reason” ð256Þ. So much for Catherine and Heathcliff.

13460.proof.3d 19 Achorn International 10/17/2014 8:14PM

 20 Ethics January 2015
your decision in favor of a given alternative to both admit of and require
justifying explanation in terms of the content of your end. You may not be
as yet in a position to explain why a riff that strikes you as expressive is
really so, in a way that others aren’t, but echoing Richards, that’s not the
point, that’s rock and roll. By presupposing that the riff will cohere with
other things you play, you thereby commit to working out the explanation
as you go.
Improvising differs from pursuing a determinate but inaccessible
end, however, in that the content your end turns out to have depends on
what you actually take it to call for. If you happen to take other things to
be expressive after playing a given riff, that riff may still end up with a
justifying explanation in virtue of them. But it would be explicable in a
different way and so expressive of something else. ðIn another song, the
opening riff in “Happy” might have been merely cheerful.Þ It’s even
possible that it wouldn’t turned out to be expressive of anything, since to
the extent that it turned out not to cohere with the rest of your perfor-
mance ðif you ignored it completely, or ended up with a total messÞ, it
wouldn’t be explicable at all. In such a case, the presupposition you made
in deciding on the riff—that you’d end up with an explanation of its
having been appropriate—will be falsified, and your decision to play it
will therefore turn out to be a mistake.35

***

Having modeled how musicians can literally work out the musical ideas
they are trying to express through the processes of expressing them, we
can extend the model to show how people, no less literally, can work out
the values by which they are trying to live through the processes of living by
them. For the sake of a name, call the latter form of agency deep impro-
visation.36 Deep improvisation is formally similar to musical improvisation q9
but differs from it in its focus and scope. Whereas jazz improvisers de-
termine the content of their musical ideas, deep improvisers determine
the content of basic values they identify with, values that define the way of
life they find fundamentally worth leading.37 As a deep improviser, you

35. Thus randomly plunking at piano keys isn’t improvising: barring some monkey-
writing-Shakespeare sort of accident, you could only produce a disorderly jumble of notes.
36. My terminology registers a debt to Charles Taylor, who uses “deep reflection” to re-
fer to a process through which one simultaneously articulates and shapes the values with
which one identifies by critically interpreting one’s evaluations in light of one’s “deepest
unstructured sense of what is important” ð“What Is Human Agency?” in his Human Agency
and Language: Philosophical Papers 1 ½Cambridge: Cambridge University Press, 1985$, 41Þ.
37. Importantly, I’m not arguing that all the values with which one can identify are up
to one to improvise. In fact, it is plausible that at least some of them ðsuch as specifically
moral valuesÞ are not.

13460.proof.3d 20 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 21
work out who you are by working out your approach to life. When you take
some action to be worthwhile, you take it to embody the identity you’re in
the process of constructing for yourself. As in the music case, this doesn’t
mean judging the action to be required by values you now determinately
have, but it also doesn’t mean treating it as rationally equivalent to actions
you could have performed but didn’t. It means taking the putative fact
that you had reason to perform that action into account in your ongoing
practical reasoning and conduct as something that is to help explain, and
be explained by, what you have reason to do on other occasions.38 But
rather than discuss this more concretely now, let me return to the puzzle
the section started with.

VI. SHARING AN INDETERMINATE END

The puzzle, recall, was this: in valuing his bandmates as “a bunch of guys
that want to do the same thing” as he, Richards apparently takes himself
to have special reason to respond to them in certain ways in virtue of some
deep similarity in their ends. But how could he do so coherently while
recognizing these ends to be indeterminate? To answer this question, we
can extend the model of the last section to multiple agents. I’ve argued
that the indeterminacy of an improvised end does not prevent you from
intelligibly taking yourself to have reason to perform certain actions over
others: it only means that the justifying explanation of your action, if any,
depends on your responses at other times. I’ll now argue that it likewise
does not prevent you from taking yourself to have reason to regard other
improvisers as determining the same ends.
The idea of responding to someone’s end according to how it’s
developing—as opposed to what it seems to be at some determinate stage
in its development—can be tricky, so it will help to break it down. Bracket
the question of what could justify Richards in taking his bandmates to
share his end and consider how, if he was thus justified, he should take
account of this fact. In the last section, I argued that improvising is like
pursuing a determinate but partly inaccessible end in certain respects.
The kind of attitude appropriate toward agents you take to share your
end is one of them. As a general point, if you are pursuing an end whose
content is not fully accessible to you but are justified in taking some
agent to be pursuing it as well, that person becomes a valuable source

38. This doesn’t mean improvisers are forever bound to their pasts. The requirement
that you act in ways that mutually explain each other only holds among actions you in fact
have reason to perform. So you may sometimes be justified in rejecting some of your past
ðor even predictable futureÞ actions and hence in ceasing to treat them as explanatory and
to be explained. Since, however, this rejection itself requires justifying explanation ðfrom
the perspective you thereby come to inhabitÞ—and this explanation may be difficult to
come by—improvisers are also not free to be entirely capricious.

13460.proof.3d 21 Achorn International 10/17/2014 8:14PM

 22 Ethics January 2015
of practical testimony—someone whose judgments of what counts as ap-
propriate with respect to the end merit prima facie acceptance. Now, the
fact that the two of you are still determining your ends means that there
is not actually some end, existing independently of your respective activ-
ity, about which each of you provides the other with evidence. But it does
not mean that the two of you cannot intelligibly accord one another the
same kind of presumption of normative authority you each would merit
if you did share such an end. Thus, if one of Richards’s bandmates takes
some action to be appropriate relative to his own end, then—absent some
compelling reason not to—Richards should also take it to be appropriate,
as something he would have reason to do himself, had he been in the
bandmate’s position.39
In Richards’s case, this presumption of authority takes the form of
the “beautifully friendly and elevating” attitude of open-ended receptivity
and personal freedom he shares with his bandmates. This explains the
appeal: if a fellow musician responds to you with something that com-
plements your playing so well it feels like a fuller expression of the
musical idea behind it, it feels affirming and enhancing. The thought is
something like: “So that’s how to do it!—that’s what was cool about where
I’ve been going. And better still, now I know to play this”—and you re-
spond reciprocally to your partner. Yet when you do respond by trying
to play in a way that coheres with your partner’s playing, you’re not just
returning the favor. You’re doing exactly what feels like the natural next
stage in your own musical project. The result is a feedback loop of mutual,
spontaneous exchange. You express what feels right to you, your partners
are spontaneously moved by what they feel in your playing to play what
feels right to them, you experience their responses as an apt development

39. The importance of such a presumption of normative authority to love has been
noticed before. Thus, Elijah Millgram proposes that friends characteristically share “a
primitive trust in ½one another’s$ practical testimony” ð Practical Induction ½Cambridge, MA:
Harvard University Press, 1997$, 168Þ, and Kyla Ebels-Duggan argues that you should “treat ½a
beloved’s$ choice of an end as if it were evidence that the end is worthwhile” ð“Against
Beneficence,” 159Þ. Neither attempts to explain the rational basis of this presumption,
however, and I think this leads Ebels-Duggan ðand arguably Millgram, who is more tentativeÞ
to overestimate its scope. While I claim that it concerns the realization of ends lovers already
putatively share, Ebels-Duggan extends it to ends your beloved has adopted but you as yet
have not. This is plausible for lovers who share all their fundamental values, but many
relationships are less demanding. Using Ebels-Duggan’s example, suppose your friend val-
ues bird watching and you don’t. You don’t think it’s silly, but you feel no inclination to try to
appreciate it yourself. You and your friend are just different people, and bird watching isn’t
for you. Ebels-Duggan argues that in withholding judgment here you fail as a lover, since you
trivialize your beloved’s interest as “mere taste.” ð160Þ. This would be true if valuing some-
thing ðas opposed to merely desiring itÞ required judging it objectively valuable. But since it
doesn’t, someone can respect you as a valuer without either accepting your evaluations as
evidence or rejecting them as mistakes.

13460.proof.3d 22 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 23
of what you’re playing and delightedly reply to it as such, your partners
reply in turn, and so on.
Now for the first question: what could justify Richards in taking his
bandmates to share his end to begin with? The answer is simple: you are
justified in taking other improvisers to share your end just insofar as
they’re in the process of determining the same end you are—such that
the responses they perform, in pursuit of their own ends, stand in the
same mutual explanatory relationship to what you have reason to do on
particular occasions as your own responses do, and the presumption of
normative authority you accord those agents is thereby vindicated. Im-
portantly, this can be a self-fulfilling prophecy.40 Having recognized his
bandmates as provisionally authoritative, what Richards takes himself to
have reason to play will depend on what they do—and so, therefore, will
the content of his end itself. This consequence is critical in connection to
love, because it explains why people who love each other for identifying
with the same values do not just mirror each other’s values but shape
them—and ultimately why such love has the qualities of irreplaceability,
openness, and constancy it does.
On the other hand, it also implies that taking someone to share your
end may be rationally underdetermined in advance. All the facts about
Richards’s and his bandmates’ ends prior to and independent of his
recognition of them may fail to explain fully why that recognition was
called for, and had he only happened not to grant it, he might have been
right, since he might then have been in the process of determining a
different end. This may seem objectionable. Why isn’t taking someone to
share your end just arbitrary, a leap of faith?
Here we need to distinguish two senses in which a response might be
said to be arbitrary. In one sense, a response might be said to be arbitrary
in a “leap of faith” sense to the extent that it is not guided or otherwise
determined by justifying considerations in principle accessible to the
agent in advance. In the other, a response might be said to be arbitrary to
the extent that it neither admits of nor requires rational justification in
terms of the agent’s ends. These senses are easy to run together, but for
agents in the process of determining their ends, they come apart. Thus,
when we recall what made Frankfurtian and Vellemanian love objec-
tionably arbitrary, we can see that the second sense is the one that matters
here. The problems with their views did not arise strictly because they
denied that loving one person rather than another was rationally deter-
mined by the beloved’s antecedent properties. They arose because they
denied that it admitted of or required justification in terms of any such

40. Also importantly, it might not be. If someone you take to share your end turns out
to be such that you cannot coherently treat what each of you putatively has reason to do as
mutually explanatory with respect to it, your attitude will turn out be unjustified.

13460.proof.3d 23 Achorn International 10/17/2014 8:14PM

 24 Ethics January 2015
properties, antecedently determinate or not. There was nothing about
Table 7-G that could make loving it inappropriate and nothing Heathcliff
had and Linton lacked that could give Catherine special reason to love
the former. But it needn’t be antecedently determinate whether or how a
beloved’s ends are similar to yours for them to play this role. Table 7-G is
an inappropriate object of love because it has no ends at all; Catherine
may intelligibly and correctly take herself to have special reason to love
Heathcliff in virtue of their similarly developing ends. If the initial state of
those ends fails to rationally determine her love, this neither renders it
unjustified nor renders the question of its justification without applica-
tion. It only prevents her from knowing the answer to that question prior
to actually loving him. It is something she will have to work out, by im-
provising with Heathcliff and seeing what, if anything, they turn out to
share.
If the reader is still skeptical, let me offer a speculative diagnosis. It
can be natural to think that justifying reasons must be capable of guiding
prospective deliberation, or otherwise be grounded in facts that are prior
to and independent of the responses they are to justify. But I suspect this
requirement seems plausible largely because much recent work on nor-
mativity has taken moral and theoretical reasoning as paradigms, and
there are strong ðthough not uncontestedÞ substantive grounds for af-
firming it in these cases. Morality, on a natural view, aims to identify de-
liberative standards the conformity to which is to insulate us from blame,
and the vindication of empirical beliefs normally depends on their causes.
But these are not reasons to enshrine the requirement as essential to jus-
tification as such, and I see no compelling grounds for doing so.41 In fact, I q10
think its present inapplicability illustrates something that may been ob-
vious from the start: the reasons of love and art are very different from
those of morality and science.
More concretely, to the extent that your recognition of another im-
proviser as sharing your ends is antecedently rationally undetermined,
your attitude toward that person will not be very articulate, and you will
have no guarantee of its justification. In this, it’s analogous to taking a riff
to be somehow especially expressive: it consists in taking there to be
something about the other agent’s ends that resonates with yours, even if
their essential fluidity makes it impossible to definitively articulate what
that something is. But this is just as should be expected. Love is notorious
for its intrinsic riskiness, and its language, like that of music, is full of
terms for inexplicable but warranted attraction. We’ve all encountered or

41. For a classic expression of skepticism on this front, see Bernard Williams, “Moral
Luck,” in his Moral Luck ðCambridge: Cambridge University Press, 1981Þ. Note that my
view implies that justification for love may be subject to luck in much the way Williams’s
remarks about Anna Karenina suggest.

13460.proof.3d 24 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 25
heard of people who admit it to be impossible for them to describe their
reasons for loving the specific people they do ðor, at any rate, to describe
them any more clearly than Catherine or Montaigne didÞ but neverthe-
less insist that they do have such reasons. I imagine most of us philoso-
phers have felt the temptation to quietly conclude that such sentiments,
while romantic, betray a basic confusion about rationality. But if I’m right,
the romantics have been right all along about an important class of
reasons that we, for the most part, have missed.

VII. LOVING SOMEONE AS A PARTNER IN DEEP IMPROVISATION

We can say that so conceived, Richards and his bandmates value each
other as improvisational partners. In this, they share a relationship with
the same general structure as relationships between lovers. To love peo-
ple for who they are in particular, I submit, is to value them as partners in
deep improvisation.
An improvisational partnership is a type of ongoing relationship
grounded in the partners’ mutual recognition of one another as sharing
an end with respect to a given activity.42 Like relatives in Kolodny’s sense,
improvisational partners value their partnership itself as well as one an-
other. But unlike relatives in Kolodny’s sense, they need only value their
partnership derivatively, only because and insofar as they recognize one
another as warranting a presumption of authority in judgment with re-
spect to the relevant activity. And they warrant this authority only because
and insofar as they are in fact pursuing the same end. With this mutual
recognition in place, however, partners work out the end they share to-
gether, each according the other a joint authority in determining its con-
tent.
Thus, partners in musical improvisation explore a common musical
idea in their playing; students may improvise as partners in working out
an interpretive approach to a text, spouses in working out the terms of a
marriage. None of these relationships necessarily involve partnership in
deep improvisation ðin which the partners would work out together the
basic significance that their activities of musical expression, textual inter-
pretation, or marriage had in their livesÞ, but all of them can. As a special
case, partnership in deep improvisation can also be global, as it is for
Catherine and Heathcliff, such that the partners work out their entire

42. Or, equivalently, with respect to a cluster of interlocking activities. Note that while I
define improvisational partnerships as mutual ðsince this makes it easiest to explain the
value partners can have to one anotherÞ, taking someone to share an end you’re improvising
can be one-sided. This may be pretty close to how Alcibiades views Socrates in the Symposium,
for instance—at least if it’s assumed that he values him for who he is in particular, rather than
as a source of evidence about the Good. Such an attitude will share many of the features of
love for a partner in deep improvisation, but not all of them.

13460.proof.3d 25 Achorn International 10/17/2014 8:14PM

 26 Ethics January 2015
approach to life together. But it need not be. I love my grad school
officemate for much of what he sees in philosophy, but not all of it, and
while we have a very satisfying friendship, we each think the other’s po-
litical views are pretty awful.
Interpersonal love differs from other forms of improvisational part-
nership because the ends lovers share constitute fundamental values with
which they identify. This makes the interest and responsiveness war-
ranted by the similarity of their ends correspondingly more profound. To
flesh out what these attitudes involve, let’s return once more to Catherine
and Heathcliff.
Catherine and Heathcliff love each other for their common wild-
ness. When she tries to explain this to Ellen, Catherine recalls a dream of
going to heaven but being miserable there. Ellen points out, reasonably
enough, that of course she wouldn’t like it: heaven isn’t supposed to be
the sort of thing sinners would like. “This is nothing,” Catherine retorts.
“I was only going to say that heaven did not seem to be my home; and I
broke my heart with weeping to come back to earth; and the angels were
so angry that they flung me out, into the middle of the heath at the top
of Wuthering Heights, where I woke sobbing for joy. That will do to ex-
plain my secret, as well as the other.” Wuthering Heights is a home to
Catherine because of the free and vigorous way of life its rawness, beauty,
and isolation enables her to lead there. Unlike anywhere else in her life, it
provides opportunities for creative exploration and discovery, physically
robust activity, and uninhibited emotional expression—all things to be
approached in a very different spirit than the domesticated concerns in
which she and Heathcliff are otherwise expected to participate. As im-
provisers, Catherine and Heathcliff are each engaged in a process of
working out just what this wildness means to them: what precisely is to be
appreciated in being in the wilderness, and how—and how the spirit of
wildness each prizes is to be embodied in an overall approach to life.
By sympathetically engaging with what the other sees, Catherine and
Heathcliff offer each other focus and reinforcement. Catherine’s judg-
ments and actions serve as a guide for Heathcliff. If something seems
worth doing to her, he’ll see, and feel, this as a hint about what he himself
has reason to do and respond accordingly. I’d imagine most of these
instances of shared practical reasoning are small and subtle. They might
concern things like what’s to be savored in an autumn wind, or what’s
interesting about a certain bird, or how and why the curate is to be tor-
mented today. In loving Catherine as a partner in deep improvisation,
Heathcliff will be drawn to her approach to life. In viewing her judgments
as warranting a presumption of normative authority, Heathcliff will ex-
perience them as attractive, as having a rightful power to shape his own
sense of his values. Similarly, in seeing what Catherine takes to be worth-
while as to be taken into account in his own normative explanations,

13460.proof.3d 26 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 27
Heathcliff will experience her personality as calling out for attention and
understanding—that is, he’ll find it fascinating. Over time, these instances
of sympathetic engagement and exchange add up. They enable the lovers
to determine and act from more richly illustrated conceptions of value
than otherwise would have been available to them.
It is this mutual self-creation that explains the irreplaceability, open-
ness, and constancy characteristic of interpersonal love. I’ll address these
features in opposite order to that in which they were introduced. Con-
veniently, this turns out to be in ascending order of complexity.

Constancy
That partners in deep improvisation have reason to continue to love each
other through a wide range of developments in one another’s values, not
necessarily capable of being anticipated in advance, should be obvious by
this point. Someone who loves you as a partner in deep improvisation
loves you, throughout your relationship together, for a specific set of
values with which you identify. Because these are in the process of being
determined, the reasons for love they constitute endure as the content of
your values is constantly being reshaped and refined. Change is not the
exception but the rule.
Note that even Sonnet 116—probably the single most quoted paean
to constancy in English—begins by describing love as the marriage of true
minds. This suggests two pertinent observations. First, the fact that rea-
sons for love are constituted by a person’s values, rather than external
characteristics like Linton’s attractions, itself means that love for a part-
ner in deep improvisation can be expected to survive the sort of surface
changes it really, obviously should. Second, constancy is important, but so
is discernment. Of course it can be appropriate to cease to love a person,
if one or the other of you undergoes a fundamental change in character
or if the two of you do not turn out to share as much as you thought.

Openness
Partners in deep improvisation are open to each other as they are in
themselves because of the distinctive way the values they share are shaped
by their particular interactions. As I explained above, the fact that an im-
provisational partner takes some action to be appropriate is in and of
itself a prima facie reason for you to do so as well, in virtue of the pre-
sumption of authority appropriate to a partner as such. And because the
actions you take to be appropriate in improvisation determine the con-
tent of your ends, the bare fact that your beloved responds to a particular
case in a certain way can in and of itself make a difference to your values.
Thus Velleman was right to stress that you do not need any special reason
for heightened sensitivity to whatever is significant to or about your be-
loved—in any given case, that you love the person is reason enough.

13460.proof.3d 27 Achorn International 10/17/2014 8:14PM

 28 Ethics January 2015
It’s important here that a partner in deep improvisation necessarily
values your living in the way you find fundamentally worthwhile and that
this constitutes a central dimension of autonomous flourishing.43 This is
why lovers cannot sensibly be indifferent to one another’s well-being—if
they were, they’d be indifferent to the continued functioning of their
partnership—and why it makes sense to speak of Heathcliff as loving
Catherine, rather than just her activity: he loves her precisely for being
herself, for engaging in the activity that constitutes who she is.44 ðContrast
this with other improvisational activities, which may be wholly distinct
from your identity or well-being—you may, in fact, be at your best as a
musical partner when succumbing to your heroin addiction.Þ And in
loving Catherine as she is in herself, Heathcliff loves her in her sheer
particularity, since he must respond to her in her sheer particularity to
determine what he loves in her.

Irreplaceability
The most basic reason why partners in deep improvisation are irreplace-
able is simply that they are incomparable: the nature of the values in
question makes the possibility of a replacement incoherent. Someone
counts as a suitable replacement for an improvisational partner if that
person enables you to realize the same value in the relevant activity that
the original did. This may be possible in most forms of improvisation: if
you’re ultimately in it for the money or the adulation, one bandmate may
be just as good as another, even though you’d be expressing different
things with each. But it is not possible in deep improvisation, since the
value of the activities you share is itself something your partner plays an
ongoing role in determining. So any standards by which putative replace-
ments might be assessed are epistemically and ontologically posterior to
continued engagement with the original. If you had a different partner,
you’d have different standards: there’s no common basis of comparison.
Now, anyone you take to share values with which you identify will be
incomparably valuable to you in this way, even when the interest isn’t mu-
tual. But when it is mutual, lovers become irreplaceably valuable to each
other in a deeper sense. Catherine says of Heathcliff that he “compre-

43. For accounts of the relationship between well-being and the successful pursuit of
one’s ðrationalÞ ends, see, e.g., Joseph Raz, The Morality of Freedom ðOxford: Oxford Uni-
versity Press, 1986Þ; T. M. Scanlon, What We Owe to Each Other ðCambridge, MA: Harvard
University Press, 1998Þ; and Stephen Darwall, Welfare and Rational Care ðPrinceton, NJ:
Princeton University Press, 2002Þ.
44. Note that nonglobal partnerships in deep improvisation will support more limited
patterns of concern; hence the continuum between intimate companions and activity
partners or casual friends.

13460.proof.3d 28 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 29
45
hends in his own person my feelings to Edgar and myself.” Interpreted
as an improviser, she is referring to Heathcliff’s access to the values with
which she identifies: as her lover, he can access them in ways that nobody
else, in principle, can. He can interpret and develop them through the
lens of his own history while still appreciating them as she does.
Catherine works out who she is by improvising from an evaluative
currency of things like interesting birds, invigorating autumn winds, and
obnoxious curates. She takes these things to be significant in ways that are
to help explain how it will be important to her to live going forward, and
that are to be explained, in turn, by their relation to the overall way of life
they help constitute. This makes her understanding of herself and her
values essentially historical and particular. The only way for anyone else to
understand them from her perspective is to attribute the same practical
significance to particular cases that she does—to see, in the same way that
she does, those cases as contributing to, and helping to explain the na-
ture and attractiveness of, the kind of life it is important to her to lead.
But I have just argued that to be committed to attributing the same sig-
nificance to particular cases that you do is precisely what it is to love you.
So Catherine and Heathcliff understand each other the same way they
understand themselves: through a joint history of particular interactions,
constituting a common evaluative currency.
Therefore, Heathcliff’s understanding of Catherine is particular to
him because his history with her is particular to him. Heathcliff does not
only share a currency of particular cases with Catherine; he offers her a
shared evaluative perspective constituted by integrating those cases with
his own history, thereby providing Catherine with a unique dimension of
access to them. When I think of what the time I spent in a place means to
me, that is, I find I cannot answer this question without reference to what
it means to the friends I had there. But what it means to each of them
depends on the particular evaluative history he or she brings to it; as such,
a loss of access to any one of these histories would be a loss of access to a
part of myself.
Why, then, might it be so important to us to share ourselves with our
lovers? Let me finish by sketching the beginning of an answer. First, the
things our lovers bring out of us might not be things we’re capable of
bringing out ourselves. This is underscored by the fact that the best lovers

45. Catherine’s statement occurs at the beginning of a speech that has rightly worried
many critics. She goes on to proclaim that all her miseries in life have been for Heathcliff,
that he is her great thought in living, that the world would be empty without him, and
so forth. On my view, the selflessness Catherine expresses is incidental to love proper.
Heathcliff is properly lovable to Catherine because he helps her live as more fully herself,
not because he gives her something to live for. ðIt helps to remember here that she speaks
as a moody and theatrical fifteen-year-old.Þ

13460.proof.3d 29 Achorn International 10/17/2014 8:14PM

 30 Ethics January 2015
often seem like opposites: consider Elinor and Marianne, Holmes and
Watson, Kirk and Spock. These people all share certain fundamental
concerns with their partners—ones centered, respectively, on ideals of
feminine autonomy and the enjoyment of everyday beauty, the pursuit of
justice tinged with an attraction to danger and a curiosity about crimi-
nality, and boldly going where no one has gone before—but embody
these concerns very differently from them. This unity within diversity
enables the lovers to see how their own values might be realized in ways
they probably wouldn’t have recognized on their own: moved by Mar-
ianne’s indignation, Elinor might find it important to stand up to an
offense she would otherwise have passively endured; appreciative of Eli-
nor’s considered response, Marianne might better understand why her
indignation was warranted in the first place. ðRecall Velleman’s discus-
sion of his sons from Sec. II.Þ It’s even possible that one might not see how
one’s inchoate jumble of interests and concerns could ground a coherent
identity until one sees them complemented in a lover or that one’s sense
of how to live might become so entwined with one’s lover’s as to make one
lost without them.
More deeply, but more obliquely, there’s something wonderful about
someone picking up on the value you see in your approach to life and
your being immediately able to say: “Oh, so it’s not just me!” Elaborating
on Heathcliff’s comprehension of her in his own person, Catherine tells
Ellen: “I cannot express it; but surely you and everybody have a notion
that there is, or should be, an existence of yours beyond you. What were
the use of my creation if I were entirely contained here?” I’m not sure I’m
any more able to explain this idea than Catherine was. Still, there seems
to be something deeply and intrinsically desirable about communicating
to another person who you are and what, as such, is important to you.
When you work out your values with a partner, they become more than
just the terms of an isolated personal project. Rather, they become inter-
subjective standards for a way of life that can be lived in common. There’s
a sense in which values seem more real—more stable and substantial—
when they are recognized by another person and can be examined and
assessed from multiple points of view. It doesn’t matter to Catherine and
Heathcliff that their values be ones that every reasonable person could be
expected to share, or even tolerate; bracketing specifically moral con-
siderations, I don’t see that it should. But it does, and should, matter to
them that their personal values are not just personal—that their authority
be intersubjective.
This point will have to stay at the level of suggestion. Even if it’s right,
it doesn’t yet make explicit why it matters that the intersubjectivity thus
secured have the historical and particular dimension I’ve claimed to be
characteristic of love. To do so, let’s take it from the top. If it’s true that
it’s intrinsically desirable that your personal values be more than just

13460.proof.3d 30 Achorn International 10/17/2014 8:14PM

 Bagley Loving Someone in Particular 31
personal, this helps explain why it’s reasonable to want to be loved se-
lectively, since it would be reasonable to want the affirmation such love
would constitute. Such love says, in effect, that you are worthy, at least to
someone, of special interest and attention because you identify with the
specific values you do—and if the person in question is someone you love
back, you’re worthy not just to anyone, but to someone whose judgment
really counts. Compare this, again, to Frankfurtian love, which ðbecause
arationalÞ does not affirm you in this way; or Vellemanian love, which
primarily affirms you, generically, as a valuer; or Kolodnyan love, which
affirms you, also generically, as a parent, child, sibling, spouse, or friend.
But the affirmation Catherine and Heathcliff want isn’t affirmation
from a universal normative perspective. They’re too protective of their in-
dividuality for that. They want affirmation from a more deeply personal
point of view, one that recognizes their values as fundamentally their own.
Uniquely, someone who loves you as a partner in deep improvisation can
provide this more personal kind of affirmation, in loving you for an
identity that remains essentially up to you to freely and continually de-
termine. Most of us, fortunately, aren’t protective of our individuality as
violently and absolutely as Catherine and Heathcliff are. But in wanting to
be loved as distinctive individuals, I think we share the same basic con-
cern. In wanting to be loved as distinctive, we want our lovers to see, and
value us for, aspects of our characters that distinguish us from others. In
wanting to be loved as individuals, we do not want to be valued merely, as
Frankfurt put it, as exemplars of more general types, identifiable and
evaluable in abstraction from our particular, concrete, ongoing histories.
The improvisational model shows how it is possible to be loved in this way,
as persons who are both knowable and endlessly interesting and sur-
prising, with identities that escape determinate categorization but can
nevertheless be responded to with fluency and delight.

13460.proof.3d 31 Achorn International 10/17/2014 8:14PM