Content uploaded by Nick Zangwill
Author content
All content in this area was uploaded by Nick Zangwill on Nov 12, 2019
Content may be subject to copyright.
This article was downloaded by: [N Z]
On: 15 July 2013, At: 12:07
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Philosophical Explorations: An
International Journal for the
Philosophy of Mind and Action
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rpex20
Love: gloriously amoral and arational
Nick Zangwill a
a Department of Philosophy , University of Durham , Durham , UK
Published online: 12 Jul 2013.
To cite this article: Philosophical Explorations (2013): Love: gloriously amoral and arational,
Philosophical Explorations: An International Journal for the Philosophy of Mind and Action, DOI:
10.1080/13869795.2013.812739
To link to this article: http://dx.doi.org/10.1080/13869795.2013.812739
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
SYMPOSIUM
Love: gloriously amoral and arational
Nick Zangwill∗
Department of Philosophy, University of Durham, Durham, UK
I argue that an evaluational conception of love collides with the way we value love. That
way allows that love has causes, but not reasons, and it recognizes and celebrates a love
that refuses to justify itself. Love has unjustified selectivity, due to its arbitrary causes.
That imposes a non-tradability norm. A love for reasons, rational love or evaluational
love would be propositional, and it therefore allows that the people we love are
tradable commodities. A moralized conception of love is no less committed to
treating those we love as tradable commodities; it is just that they are tradable moral
commodities. An evaluative criterion of adequacy, I suggest, encourages the opposite
view – a non-rational and non-evaluational concept of love. Such a love can set up
partial obligations, which may even demand that one sacrifice one’s life. Only a love
that has causes but not reasons can have the kind of value that we think love has, and
thus it would only be rational to pursue and foster such a love.
Keywords: love; reasons; rationality; causes; emotion; morality
“I love your majesty according to my bond, no more nor less.” (Cordelia in King Lear, I, 1, 93 –94)
“He was part of my life; that’s why in a certain way I’m mourning him.” (Natascha Kampusch)
What would be a good theory of love? What criteria of adequacy should we use for asses-
sing theories of love? What are the right questions to ask? I shall spend some time consider-
ing these methodological questions in Section 1 of this paper. These considerations frame
the discussion that follows. The two sections after that argue for negative theses about love.
In Section 2, I critique David Velleman’s moral view of love (Velleman 1999, 2008). I raise
objections to it in the light of the methodological considerations laid out in the first section.
In Section 3, I broaden my target and provide a quite general anti-evaluational argument. In
Section 4, I propose a positive view of love according to which love itself has no moral
content and involves no significant kind of evaluation – moral or rational – but is a
direct emotional response to a person. On this view, inspired by David Hamlyn, love has
causes but not reasons. In particular, love is consistent with negative evaluations of the
beloved. I call this the “romantic view” of love. In Section 5, I defend the romantic
view, arguing that love is not subject to rational constraints. In this sense, there are no
reasons for love. This makes trouble for Niko Kolodny’s relationship view. Nevertheless,
I allow a way in which we can and do deliberate rationally about love, as Gary Becker
has described. Lastly, in Section 6, I briefly describe some of the causes of love.
#2013 Taylor & Francis
∗Email: nick.zangwill@durham.ac.uk
Philosophical Explorations, 2013
http://dx.doi.org/10.1080/13869795.2013.812739
Downloaded by [N Z] at 12:07 15 July 2013
1. Methodological ruminations
1.1 It might seem natural to begin thinking about love by asking the Socratic-style question
“What is love?” Having answered that question, we could then go on to ask further ques-
tions – about its value, about its implications, about its place in normative moral theory and
so on. In Plato’s Meno, Socrates recommends that we should first ask what virtue is, and
only after that can we ask whether it can be taught. Similarly, the idea would be that we
should first ask what love is, and then ask about its value.
Given a focus on the “What is love?” question, the dialectic tends to play out as follows:
a theory is proposed as an answer to the question; and that theory is then tested for exten-
sional adequacy. For example, someone proposes: “Love is the desire to benefit and be
with”. No, comes the reply: What about annoying interfering relatives who one nevertheless
loves (Velleman 1999, 353)? What about love of the dead (Badhwar 2003, 46)? What about
love at first sight? The aim is to answer the “What is love?” question with an account that it
deems uncontroversial cases of love to be love, and it deems uncontroversial cases of non-
love not to be love.
The presupposition that we should prioritize the “What is love?” question seems to fit
some aspects of the intellectual life of the topic of love, which divides into an intrinsic inter-
est in love, and an interest in it for its impact on the normative theories of consequentialism
and Kantianism, which have a hard time placing love in their theories in a plausible way.
Surely – we might think – its impact depends on its nature; so we must first consider its
nature.
I think that this methodological presupposition is incorrect. I propose instead that we
prioritize the value of love as a criterion of theoretical adequacy of a theory of what love
is. By the value of love I mean both its real value and its value to us – that is, its apparent
value.
It is this value, or at least apparent value, which forces certain substantive moral theories
either brutally to reject the value of love, or to undergo impressive contortions to accom-
modate it. Hence we have recently witnessed the spectacle of huggy-touchy-feely-Kantians,
and loyal-unto-death consequentialists! These theorists are forlornly trying to square our
intuitive view of the value of love with their theories.
The value or apparent value of love should constrain theories of love. A theory of love
according to which it would not have, or at least not seem to have, moral or other value, or
that gives it the wrong kind of value, is defective and should be rejected. Foregrounding this
criterion of adequacy means that we can be less concerned with extensional adequacy and
the game of example and counterexample.
I shall argue that foregrounding the value of love has implications for its nature – its
content and its rationality. The implications are these:
(1) Love is a mental state that lacks moral content or any evaluative content.
(2) Love lacks rationality conditions, in the sense that there are mental states that make
it rational to love; nevertheless love does rationalize other mental states.
The value of love depends on these two principles, I argue. Only if love lacks evaluative
content and is non-rational can it have the value of the sort we think it has. Furthermore,
only if love lacks evaluative content and is non-rational can it generate obligations of the
sort we think it does. Thus – to put it in an almost paradoxical way – it is rational to
value love, and respect it in our actions and thoughts, only if love itself is non-evaluational
and non-rational. That is what I hope to show.
2Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
1.2 Thus far I have been happily talking of love itself and our valuing it. But let us
reflect a little on the concept of love. Someone might ask, sceptically: Is there a significant
mental kind that the concept expresses? One worry is that the concept is a culturally local
Western parochial notion that does not lock onto an interesting real kind. But even in our
culture, can we assume that there is a common notion? Nietzsche writes, of romantic love:
...man and woman have different concepts of love; and it is one of the conditions of love in
both sexes that neither sex presupposes the same feeling and the same concept of “love” in the
other. (Nietzsche 1974, 363)
First, Nietzsche is here denying that there is a shared concept of love between the sexes, at
least if love itself is to be possible. “Mars” and “Venus” talk past each other when the
common word is used. There is a gender-specific “private language” of love. Second,
with a typically surprising and interesting inversion, Nietzsche says that this miscommuni-
cation is not to be regretted since it makes the phenomenon of love possible. Contrary to an
assumption of the entire current self-help industry and of women’s magazines, systematic
miscommunication is the precondition of the phenomenon of love between the sexes. Love,
Nietzsche thinks, depends on an illusion – an illusion of shared concepts. I think that this
need not imply that our notions are completely illusory – in the sense of being completely
inadequate to the reality – only that they are only partially adequate, and that each sex’s
concept is partially adequate in a different way.
Generalizing – what Nietzsche’s speculation illustrates is at very least that facts about
our concept or concepts of love are messy, inconclusive and fraught. Thus, the impact of
intuitive examples and counterexamples will be unclear, since that genre of philosophy
assumes stable agreed concepts.
By contrast, the evaluative criteria are cleaner. All who possess various concepts of love
praise love in one way or another. It is true that it is possible that different people value it in
very different ways, but perhaps not, even given their different concepts. If not, our evalu-
ations will provide more of a fixed point from which to theorize than our concepts.
1.3 Nevertheless, some observations about our concept of love will help us. Ascriptions
of “love” are often qualified by “filial”, “maternal”, “brotherly”, “sexual”, “friendship”, etc.
We might call this phenomenon “subscripted love”. (Oedipus got his subscripts seriously
confused!) Rather than plain “a loves b”, the logical form seems to be “a loves b in way
w”. A question is: Is there something, the love part, which is common in all these cases?
Our language suggests so. But it is not obvious. Again, the “What is love?” question
may be misleading if love is universally subscripted.
One option here is a family resemblance view: there is X-love or Y-love but no general
theory of love can be had. Such a family resemblance account should surely be on the table
as an option. Compare the arts: painting, poetry, music and so on. Is there something,
arthood, which they all have in common? It is not obvious. It is possible that “love” and
“art” are like Hilary Putnam’s example of jade (Putnam 1975, 241). There is one word cov-
ering different things with different natures, misleadingly collected together under one term.
In the cases of both love and art, cultural variation encourages this diagnosis. In the
renaissance, one university faculty might teach mathematics and music, whereas another
would teach poetry and linguistics. In the traditional Japanese culture, the “road” or
“way” collects together calligraphy, swordsmanship, the tea ceremony, ikebana (Japanese
flower art) and the art of scent. To us, these groupings seem odd and gerrymandered. To
them, our groupings under “art” seem equally odd. Similarly, perhaps, our category of
Love: Gloriously Amoral and Arational 3
Downloaded by [N Z] at 12:07 15 July 2013
love might seem gerrymandered to others. And the parochialism of our notion of love
makes this a real option.
Again, the moral is that if we reflect on the concept of love, we encounter muddy terrain.
1.4 My last methodological thought is that we should keep hate in mind when thinking
about love. There are questions to be asked, such as: Is hatred the opposite of love (Ben
Zeev 2008)? Is hate subscripted the way love is? What is the logical form of hate? Does
hate involve beliefs about its object? Is hate an evaluation? Can hate be rational? The
relation between love and hate is controversial; but thinking about hate is very likely to
help us think about love. Even if love and hate are not diametrical opposites, hate stands
in enough of an intimate relation to love that understanding one is likely to cast light on
the other.
2. Velleman’s moral theory
2.1 I want to begin with a critique of an evaluational conception of love, which will make
for a nice contrast with the positive view that I want to defend. One notable proponent of an
evaluative conception is David Velleman, who writes:
All that is essential to love, in my view, is that it disarms our emotional defences towards an
object in response to its incomparable value as a self-existent end. But when the object of
our love is a person ...we are responding to the value that he possesses by virtue of being a
person or, as Kant would say, an instance of rational nature. (Velleman 1999, 365)
...love is a response to the value of a person’s rational nature. (Velleman 1999, 365)
Velleman’s evaluative view, takes a somewhat moralized Kantian form, which I think
makes especially vivid what is problematic about the evaluative view. Love, for Velleman,
is a response of our rational nature to the value of rational nature in another person, which
all persons share equally. One obvious objection is: on this view, Why not love everyone?
On this picture, Should not we have a Jesus-like “universal love”? But such a universal love
could not really be love, since love is inherently partial, a focus on a particular person as the
particular person they are (Nozick 1989; see also Kraut 1987). Real love – we might com-
plain – is not promiscuous. Nevertheless, Velleman is embracing this apparently unintuitive
universal conception, and he might reject a dogmatic particularism about love as question-
begging. Velleman’s idea is that the universal awe-worthy rationality and dignity is shared
by all people, but only that of some people gets through to us and to our emotional vulner-
abilities because of certain of their contingent qualities.
This model is like the ancient theory that the stars are holes in the sky behind which
there is ubiquitous fire: only in certain places does the light from the fire get through to
us. Similarly, some people have contingent qualities, which explain why we love them
and not another, as a response to their unconditional value as persons; but those contingent
qualities are more like absences than presences, which facilitate our awareness of a person’s
undifferentiated universal valuable nature. On this view, we do not love a person for those
contingent properties, we only love them for what is universal in them.
I find this unintuitive, and I doubt that we would value such a love in the way that we
value love as a positive response to the particularity of others. We do not respond to the
particularities of others as absences that allow their universal valuable nature to get
through to us, in the way that we value high desert ground for receiving radio waves
from outer space, because there are fewer distortions from the atmosphere than elsewhere.
When Romeo said “Juliet is the sun”, he implied a celebration of a presence, not an absence
4Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
– a celestial body, not a hole in the sky. It is true that the value of others, which gets through
to us, on Velleman’s picture, is a presence not an absence. But that positive value is some-
thing fundamentally generic, the same in all persons. The particularity of other people, that
which makes them distinctive, plays only a facilitating role for Velleman. This seems
wrong. Love is a celebration of particularity, not of what is universal and generic. (One
can say this without saying that one loves the distinguishing properties of the particular
person who one loves.)
2.2 There are further aspects of this view that are uninviting. Consider Cordelia of Sha-
kespeare’s King Lear. (Where Velleman reaches for Freud, I reach for Shakespeare.)
For Cordelia, love is a bond. But a bondsman is a slave, one who lacks freedom in
important respects. He is one who is tied, like a prisoner. Many poets, such as Petrarch
and Dante, describe falling in love as being taken prisoner, and being bound. This is far
from Velleman’s description of love as the exercise of Kantian autonomous rationality
and a response to the exercise of Kantian autonomous rationality of another. No!: love is
rope! Cordelia contains the word “cord” in her name. And her fate is in her name. She
dies by hanging from a cord, just as she lived by her bond of love. She loves “according
to her bond” (as she says in the quotation I have placed as an epigraph of this paper). A
bond can also be like a chemical bond between two surfaces. This is all misunderstood
by King Lear, who is foolish and vain. He does not deserve Cordelia’s love. But neither
does he not deserve it. Nobody deserves love. Nobody deserves not to be loved.
It is the odious Gonerill who speaks more like Velleman (!), saying that she loves Lear
because of his priceless worth:
Sir, I love you ... Beyond what can be valued, rich or rare. (King Lear I.1)
Perhaps she is being insincere. Or perhaps not. But even if she is sincere, her love is still
flawed, as is Lear’s expectations. (Gonerill here tells the foolish and vain Lear what he
wants to hear.) For there is something wrong with a love that is a reflection of an awareness
of priceless worth. If that is love it is defective love.
Velleman thinks we want to be loved as an affirmation (Velleman 1999, 362). Simon
Keller thinks similarly, and puts the point forthrightly: “If someone I romantically love
loves me back, then that’s a reason for me to feel good about myself” (Keller 2000,
254). Both Velleman and Keller see love as the recognition of worth, whether of one’s par-
ticular and possibly idiosyncratic worth (Keller) or of one’s generic moral worth (Velle-
man). They think that in wanting to be loved, one wants one’s worth recognized. But
both Velleman and Keller share Lear’s error! Cordelia refuses to say that she loves her
father on account of his virtues or his particular or generic worth. That’s why her love is
true. It is not her honesty that is important here, by contrast with that of her sisters.
What is important is the contrast between what they feel. Velleman would have us think
of what we love as beyond price. But Cordelia’s love is true not because it is a sublime
evaluation of something that has a price beyond all prices, but because it has no evaluation
at all. As we shall see below, it is important that it can coexist with a negative evaluation.
(There is a Jewish practice according to which a husband tells his wife at the Shabbat
dinner table that she has a “price beyond rubies”. But it is unlikely that the purpose of this is
for the husband to express love for his wife. Instead, the husband is expressing an evaluation
and appreciation of his wife as a wife, which is separate from the question of love. Of
course, this is not to say that he does not love her. But why celebrate that publicly at
dinner with a monetary comparison?)
Love: Gloriously Amoral and Arational 5
Downloaded by [N Z] at 12:07 15 July 2013
Real love is Cordelian not Gonerillian. The right attitude to what we evaluate highly,
even sublimely highly, is not love but respect or perhaps awe, which is quite a different
emotion. Even if we suppose Gonerill to be honest, what she says would at most indicate
that she respects Lear, but not that she loves him. She is not just tied, like Cordelia. Love is
bondage.
2.3 If we follow Velleman and think that we love others because (1) they embody uni-
versal dignity, which (2) happens to get through to us emotionally due to their non-universal
characteristics, then we seem to be vulnerable to what has become known as the “trading-
up” objection. (Robert Nozick rightly makes the issue of trading-up central to his discussion
in Nozick 1989). If we love someone because of certain esteemed characteristics, then it
seems that we should exchange this person for another person if that second person instanti-
ates those characteristics more perfectly (assuming one cannot have both.). But it is distinc-
tive of love that it is not tradable in this way. This objection is usually thought to apply to
those who love people for their physical and personal qualities such as their beauty or intel-
ligence. But it also applies to Velleman’s moral view. All of us are equal in regard of (1). But
as far as (2), some people get through to us better than others, and if so, it seems that we
should trade-up and swap to a better object of our love.
To be fair, it is not quite that on Velleman’s view, we love others because they are a
mere means to “opening us up emotionally” and recognizing our true rational self in
recognizing their’s. Nevertheless, on his view, we love who we do because they
perform this role. Velleman seems to envisage a kind of Gricean syndrome, where
there is a mutual awareness of the recognition of the other and their value (rather like
Thomas Nagel and Roger Scruton’s view of sexual desire – Nagel 1979; Scruton
1986). For Velleman it seems that a person loves one person rather than another
because they are a good “emotional opener”, as it were. Even if that is not the lover’s
reason for loving, it is the explanation. But then, given that the value of love and the
loved person resides in the emotional opening, it is still true that the person loved can
and should be traded. An emotional opener is not so unlike a can opener in this
respect. But, surely, the value usually accorded to love depends on the irreplaceability
of its object, once love exists. (It is true that one can acquire new friends, but we
should not discard the old ones like worn-out socks!) For Velleman, the beloved is like
a therapist who opens us up emotionally and helps one recognize one’s true self. But pre-
sumably just as some therapists (like doctors, electricians and plumbers) are better than
others, so, some objects of love can perform the therapeutic task better than others. On
Vellemans’ view it is just a fact that some people get through to us more than others,
and the facts that explain these are not reasons for love. Nevertheless, given the differen-
tial facts, trading-up is still warranted in a way that is intuitively inconsistent with love’s
nature and value. If one person gets through to me more than another, unless I am to love
them both, I should be motivated to trade one for the other because of the greater value-
bombardment than one affords over the other. But this is problematic.
Velleman could reply that once we have developed a close loving relationship with a
person, then that person is very likely to be the person who is best at opening us up. There-
fore there is unlikely to be a case for moral trading-up once such a relationship is in place.
But then this saves non-tradability in the wrong way. For similar reasons it might be best to
stick with one’s therapist or doctor, or indeed one’s dentist, plumber or electrician.
(In more recent work, Velleman has appealed to the idea that love is a recognition of the
“dignity” of a person, as opposed to their “price” (Velleman 2008). But I cannot see how
this helps secure untradability. Perhaps we all have the same dignity and priceless value
as autonomous rational persons. But then some people open us up better than others.
6Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
Therefore we should trade-up. There is no reason not to swap to another dignified and pri-
celess Kantian autonomous rational self – one that is a better means to unlocking my
emotional vulnerability to the dignified and priceless Kantian selves of others.)
The trading-up objection is powerful because of the constraint that our theory should
enable us to accord love the value we do. This is a central criterion of adequacy of any
theory of love – perhaps it is even the pre-eminent criterion of adequacy. But a theory
of love that makes love’s objects tradable is defective because it violates this constraint.
Tradable love is not worth much.
However, this point needs qualification, because there is no denying that there are cir-
cumstances when the value of love is outweighed in rational deliberation. And we may
deliberate about a good object of love, and make rational choices on that basis. I
examine this further in Section 5.4. Nevertheless, once love exists, its objects are non-trad-
able. Despite the qualification, Nozick was right about that.
3. Evaluational love
3.1 Putting Velleman’s specific evaluative theory to one side, I now want to fashion a quite
general anti-evaluative argument. There are others who hold an evaluative view, such as
Niko Kolodny, who writes:
Love is a kind of valuing. (Kolodny 2005, 150)
One values both the person and the relationship. (Kolodny 2005, 150)
Love depends on a large number of sophisticated evaluative beliefs. (Kolodny 2005, 151)
My view is that any such view is misguided.
Let us begin by asking: What is the “logical form of love”? By “logical form” I mean the
analysis of the mental state of loving, and not, at least primarily, a semantic analysis of our
concept of that mental state. The most significant point is that the formal object of love is a
thing – typically a person, but it could also be a team, or a place, or a time. Perhaps it can
also take an activity as an object, such as dancing, skiing or kissing. What it does not take is
apropositional object. Here there is an important contrast with other emotions. Although
fear, anger and pride can all be directed to things, they can all take propositional objects.
We can fear, be proud, be angry that p. We can fear a bear and also fear that it will eat
me. But the logical form of love makes it clear that we cannot love that p. (In English
one can, perhaps, just about say “I love it that p”; but it sounds very strained. Maybe it
just about makes sense. But the point is not really one about English. The point is to
point to a family of intentional states that are standardly picked out as states of love,
which can only have objects as their intentional contents and that cannot have propositional
contents. The existence of such a grouping is reflected in the English use of “love” to a great
extent.)
This point is important because evaluation is propositional. An evaluation takes the
form of an application of a value predicate to a thing. Even those who deny that there
are value properties and who think of evaluation as a projection of feeling seek to construct
value propositions (Blackburn 1998). Can we not talk of valuing a thing? Is not valuing
sometimes objectual as well as propositional? No. Although we may sometimes talk of
valuing a thing, which really means thinking that a thing is valuable; and that is to
ascribe a value to a thing, which is propositional. Hence there are no real objectual evalu-
ations. Evaluation is essentially propositional.
Love: Gloriously Amoral and Arational 7
Downloaded by [N Z] at 12:07 15 July 2013
3.2 The consequence of this for a theory of love is that an evaluational account (like
Velleman’s or Kolodny’s or Keller’s) ought to give love a propositional object – that X
has a certain value, or that X has certain characteristics. If so, we should be able to love
that the thing has the evaluative characteristics. But such a love would be propositional,
which would flout the logical form of love. Love, one might say, refuses to be
propositioned!
To put the argument in a crisp form:
(1) Evaluation is propositional. We think of the objects of evaluation as having evalua-
tive properties. That is the logical form of evaluation.
(2) If love is evaluation, it has a propositional logical form.
(3) But it does not. Love cannot be held towards a propositional object. That is essen-
tial to love.
Therefore,
Love is not a form of evaluation, and it does not imply an evaluation.
Velleman might reply (in fact he did so reply when I pressed this point at a conference) that
even though love is a kind of evaluation, what one loves is the person X, and one does not
love that X has the value in question. But I cannot see why this should be so on an evalua-
tional view. It would be mysterious. If love is a response to value why cannot one love that the
person has value? And why would love not be propositional as well as objectual? Why should
love not take a propositional object once one thinks of it as a kind of evaluation? It is often
thought that pride and anger are held for reasons, and they are based on evaluative beliefs –
positive ones in the case of pride and negative ones in the case of anger. In those cases, we can
be proud or angry that the things have the positive or negative feature. Love should be like
that on the evaluative view. Love should be propositional. But it is not.
3.3 Love is not a propositional attitude but an attitude to a particular, typically a person.
But even when love is an attitude to a particular person, we do not love their differentiating
properties or love them for their differentiating properties, even though the awareness of
those properties might cause our love (contrast Keller 2000). (The only properties that
we might love people for, are their “hecceities”, their primitive thisnesses, if there are
such properties.)
A concession needs to be made to evaluational views at this point – although it is a two-
edged concession. Consider Velleman’s view that we must conceive of the beloved person
as a locus of value. Perhaps this is true in a way, but only as a precondition of love. That
does not mean that love is its precondition. This is one place where it is especially useful to
remember hate. Hate has the same precondition. Hate does not deny the rationality, dignity
and personhood of the person hated – indeed it essentially depends on it. We do not hate
animals; we hate people as people. It is a metaphor to say the hater thinks of the object of
hatred as sub-human, as vermin or whatever. Hate dignifies its objects as human beings, as
persons, just as much as love. Something has to be worth hating! Love and hate share the
same precondition, and that precondition may be a recognition of a certain value of a
person. Yet love is a lot more than that precondition. (For similar reasons Iris Murdoch’s
idea of love as “really looking” (Murdoch 1970) is deeply implausible; for Hate also
involves “really looking”, and is indispensable, for example, in a good torturer.)
3.4 To see love as a matter of value judgement in my view, misdescribes love, and in
particular it misdescribes the fact that love is independent of much of our intellectual
and cognitive mental life. Think of children’s love or the love felt by those who are
8Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
demented. This love is perfectly genuine and full, and it does not involve evaluation, respect
or awe, in the sense of ascribing a value to something in a propositional way. Their love is
not less for that. Awe is something one might feel towards Saul Kripke’s intellect, not
towards those we love (see also Badhwar 2003, Section 2.2).
Yes, the lover may speak of the value of the beloved: the lover may say “X is valuable”;
but like the student relativist, the lover will quickly add “to me”, which shows how far we
are from morality. The thing loved may have impartial value that makes it a good object of
love, and if one is searching for an appropriate object to love, one may factor those charac-
teristics into one’s deliberation, as I discuss below. But that is neither an evaluation that
stems from love nor is it a part of love. Once love is in place, a kind of evaluation may
follow, but only partial or relative evaluation.
Suppose that it was insisted that one ought to love as a response to one’s awe at another’s
Kantian dignity. But actual love is often untamed, raw, inappropriate and gloriously amoral.
That’s the point. Romantic love illustrates this but it applies just the same to other kinds of
love. Was Romeo’s love across the wrong side of the political and familial divisions a
moral response to Juliet’s universal Kantian rational self that somehow happened to get
through to Romeo despite those divisions? No! And we should not want or value very
much a love that was like that. Romantic love that is thought inappropriate, even by the par-
ticipants, reveal this valuable amorality. That love is amoral in content and all the better for it.
Similarly, the love of young children for their parents or the love of demented elderly parents
for their children has no moral content. Contrary to a major theme in Kant’s moral philosophy,
there are many valuable things that are not valuings. Love is one of them.
Love as evaluation is love devalued!
4. The romantic theory: love despite negative evaluation
4.1 I now come to the approach that I endorse, which for want of a label I call the “romantic
theory”. In his important paper “The Phenomena of Love and Hate”, David Hamlyn, I
believe, got it right when he said that love is different from other emotions precisely in
respect of its lack of evaluation (Hamlyn 1978, especially 11– 20). Hamlyn wrote:
...love [can be] independent of reasons (whether or not there is a reason in the sense of cause).
(Hamlyn 1978, 16)
...no particular beliefs or disjunction of beliefs about the beloved is necessary on the part of
the lover, nor any particular ways of seeing the beloved. (Hamlyn 1978, 13)
Passionate love may easily turn into hate without any beliefs changing at the same time and
without the change being due to any newly acquired beliefs, however much different beliefs
may follow in due course. (Hamlyn 1978, 16)
Loving someone or something is not incompatible with, for example, having no respect for
them, finding them in many ways distasteful or recognizing in them a whole series of bad qual-
ities which are not overridden by good qualities. (Hamlyn 1978, 16)
There are at least two points here – one about evaluation and one about rationality. I pursue
evaluation, or the lack of it, in this section, and the more specific issue about rational con-
straints, or the lack of them, in the next section. Most important, in this section, will be the
coexistence of love with negative evaluations.
(Harry Frankfurt follows Hamlyn on the crucial question of whether love is based on
reasons and whether it is the recognition of value in Frankfurt (1999, 2004). Frankfurt
Love: Gloriously Amoral and Arational 9
Downloaded by [N Z] at 12:07 15 July 2013
also thinks that love generates reasons (Frankfurt 2004, 37), which has much plausibility.
However, Frankfurt also wants to use love as the basis for a theory of practical reason and
value more generally, which is controversial at best. A non-rational and non-evaluational
view of love can stand without such a larger edifice being erected on it.)
4.2 On the romantic theory, one loves a person not the person’s features that may cause
one’s love. Indeed, one may not even value those features, not even as manifestations of
universally valuable personhood. And one might even think that the features that cause
one to love are regrettable flaws, as Hamlyn pointed out. One may even grow to love the
flaw. (There is a Karl Kraus’ nice poem on this theme: “Dein Fehler” or, in English,
“Your Flaw”.) The insistence that love may embrace and even celebrate flaws is one of
the hallmarks of a non-evaluative romantic conception of love.
The sort of amoral view of love that I favour, which is compatible with negative evalu-
ation, is sometimes criticized as an overly romantic conception. Velleman writes, “I think
that moral philosophers could stand to be more rather than less high-minded on the subject
[of love]” (Velleman 1999, 342). As we have seen, I resist a high-minded moralized con-
ception of love, and I recommend that we embrace a romantic view. Views of love are
often criticized for being over romantic; but – echoing Velleman – I reply that moral phi-
losophers could stand to be more rather than less romantic about this subject, which is, after
all, love! What could be more appropriate than a romantic theory of love? If we cannot have
a romantic conception of love, what can we have a romantic theory of?!
4.3 I want to consider a telling example, which fits naturally with the view I am endorsing
here and which other views will have difficulty understanding in a convincing way. The
example is that of the Austrian girl Natascha Kampusch, who was kidnapped when aged
10 and held prisoner in a basement for 8 years. A relationship of sorts between her and her
kidnapper seems to have developed, perhaps when she was about 14. She was his prisoner
but he also made sure she was educated and cared for her in some respects. When she
escaped, aged 18, in 2006, her captor killed himself. Despite what she had been through,
she turned out to be a remarkably mature young adult – quite a strong person and self-
aware (see Kampusch 2010; see especially the insightful last chapter on evil). What interests
me is the announcement she made soon after her escape and his death that surprised and
shocked many people: what she said was that she wanted to mourn her captor (see one of
the epigraphs to this paper). Many were dismayed. Oh – she was so traumatized she did
not know what she was saying. Oh – she had “Stockholm syndrome” and was somehow
zombie-like in what she said. Oh – she must have deep traumatic anger at her oppressor
that she was repressing. Oh how they moralized! In fact, they seemed more traumatized
than her at the reality of this mature, self-aware, independent-minded young adult who did
not conform to their therapeutic or political norms. In an era that prizes trauma and victim-
hood, they wanted her to be more of a traumatized victim than she was. She had of course
endured what she had endured, but she was dignified and could take her emotions for
what they were, without being and feeling what so many apparently concerned onlookers
thought she ought to be and feel. Good for her, I say! I think that it is interesting and under-
standable that she felt that she wanted to mourn. It is to be expected that two people living in
close proximity, even in those circumstances, should develop love for each other. Why not?
(And it is hardly likely that she loved him as a result of finding some good in him – perhaps
his autonomous rationality; no, she was caused to feel.) Therefore she should mourn him,
even though he was a monster. I go further: given that she loved him, to some degree and
in some way, she owed him that emotion, in spite of what he did. Or at least, the mourning
was appropriate. In her recognition of that she showed more depth than the army of moraliz-
ing journalists and therapists whom she shocked.
10 Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
4.4 How are love and caring connected on the romantic view? Any theory of love needs
to say something plausible about this.
One view would be that love just is a disposition to care. Velleman points out that one
can love without being disposed to benefit the beloved (Velleman 1999, 353). This is true.
But I can desire someone’s good without desiring to promote it myself. I might love Man-
chester United football team, and want them to win, even though I do not want to achieve
that perhaps by fixing the result by bribery. I want them to do well by themselves, through
footballing excellence. One need not be interfering. However, there remains an essential
connection between loving and caring. Love is not automatically connected with caring
for the well-being of what one loves, but it does rationalize it. Love rationalizes caring,
but love and caring are distinct.
This has the consequence that Harry Frankfurt’s view must be mistaken. Frankfurt holds
that love just is a structure of desires, which includes caring desires. But love makes a
concern for another’s well-being appropriate; it is not identical with such a concern.
Although love and caring are metaphysically distinct they are normatively connected.
Frankfurt even backs away from saying that love is an emotion, which is surprising (Frank-
furt 1999, 2004). That is right to the extent that one may love a person without having feel-
ings at every moment when one loves that person. But there might still be an internal
connection between loving a person and having feelings at some point or another, just as
pride is a feeling even though one need not be feeling it at every moment when one is
proud of something. No feelings, no love. Love is an emotion, not a desire.
Neera Badhwar raises in passing the very interesting example of love of the dead (Badhwar
2003). It seems that we sometimes love the dead but we cannot benefit them. This might be
denied. After a person’s death, one might defend his reputation, or benefit his offspring, for
example, or pursue what the dead person cared about. Still, even if we can benefit people
after their deaths, if we love the dead, we do not necessarily desire to benefit them, and the
desire to do so can seem strange. I am not sure exactly what to say about the matter of love
for the dead, which seems to me to be a profound matter, in need of further investigation.
Still, it seems that love of the living typically has the consequence that we wish those we
love well, and we also think that makes sense, however it is with the love of the dead. The
love of the dead is an emotion, and it is distinct from caring even if love and caring are
usually conjoined for the living because love for the living makes sense of caring for them.
4.5 I have emphasized the value of love but have not provided a deeper theory of exactly
why love, romantically conceived, is so valuable. But we can at least see that love has a
value that is dependent on its not being an evaluation and not having rationality conditions.
To be as valuable as we take it to be, love must not be tradable in the sense that one trades
among different objects of love. Exactly why such a non-tradable love is so valuable is a
deep further matter – a matter to be integrated into a general theory of value. One
aspect, however, of the value of amoral and arational love can be appreciated if one puts
oneself in the shoes of a person who is loved amorally and arationally. One is happy that
it is oneself who is loved – one’s particularity, one’s self – and not ones virtuous properties,
for who one is and not what one is. This point suggests a more constitutive point, which is if
love were a moral or rational emotion, like anger and pride, then although such emotions are
directed to persons not to their properties, the love is still held towards the person in virtue
of their properties. Thus there is a sense in which such an emotion of love, if such it were,
would not be a pure direct response to persons, but would involve their properties in an
essential way. Thus such a love is not a celebration of particularity, like amoral and arational
love. This is why we want to be loved with amoral and arational love, more than the kind of
love, if such it is, that holds in virtue of our virtues.
Love: Gloriously Amoral and Arational 11
Downloaded by [N Z] at 12:07 15 July 2013
5. Non-rational love
5.1 Let us now turn to consider views opposed to a romantic view. Do we have reasons for
love? Many think so (Keller 2000; Kolodny 2005). I shall take this to be the question, not
merely of whether there is something good about love or whether love is sometimes appro-
priate in some generic sense, but the more specific question is of whether love is subject to
rational constraints. Those who think that there are other mental states, such as beliefs,
desires or other emotional states, that provide reasons for love, in the sense of making it
rational; and when we love someone or something it is in part because those other
mental states rationalize love. (An evaluational theory is one theory according to which
love may be rational, but the rational theory is more general.)
Clearly many emotions are subject to rational constraints: fear, anger and pride, for
example. Actually to feel fear it is not necessary that we believe that we are in danger.
Fear and rational fear are different. Fear is irrational if we do not believe we are in
danger. It should be based on that belief. Perhaps we must at least imagine that we are in
danger if we are actually to feel fear; we must entertain the thought, not necessarily
believe it. But we ought to believe it. It is irrational not to. Similarly with anger and
pride: if a person has these emotions, then there are rational constraints on their beliefs;
and perhaps there are things that we must at least imagine in order to have these emotions.
By contrast, in my view, and following Hamlyn, nothing like this is true of love. We do
not love for reasons. Love is not subject to rational requirements. Love has causes but not
reasons. This, I think, is a deep truth about love, as Hamlyn saw; and the rationalizing/
reasons view is a mistake. (As I noted in Section 4.4, love rationalizes caring, but no
mental state rationalizes love.)
5.2 Requiring that love has reasons leads to oddities. Consider that young children love
their parents in a non-defective way. But do young children have reasons to love their
parents (in the sense of it being rational to love their parents)? Surely not. Do parents
have reason to love their children? Perhaps some parents are aware of the priceless value
of their children, or perhaps of the Kantian dignity of their emerging capacity for auton-
omous rational willing (especially if they have read Kant!). But young children have no
awareness of this kind of Kantian value in their parents nor do they have any other
reasons. That comes later, if it does. And there is nothing wrong with children’s love. Chil-
dren’s love is full and proper and as good as any love is.
The same goes for the love felt by people with dementia, such as those with Alzeimer’s
disease (usually elderly people). Such people lack many faculties. They are cognitively
defective. Dementia affects our rationality. But some of their emotions are not that much
affected by this condition. In particular, the love of those with quite advanced dementia
may be full love, up to the point when they begin to forget who the object of their love
is (when dementia leads to amnesia). On the “arational view”, it is just what one would
expect that love and hate are less affected by dementia than emotions such as fear, anger
and pride, since these emotions are subject to rational constraints in the light of a
persons’ beliefs, desires and other emotional states. As Hamlyn pointed out, love and
hate differ from other emotions in this respect.
5.3 Probably the most plausible view according to which there are reasons for love is
Niko Kolodny’s view that relationships are reasons (Kolodny 2005). Such a view avoids
many of the difficulties with the idea that the reasons for love are intrinsic properties of
the person that we love. (Moral qualities of the object of our love would be such intrinsic
properties.) Instead, on this view, our relational properties, in particular, our relationships to
people, give us reasons to love them. Kolodny thus hopes to occupy a middle ground
12 Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
between the view that the reasons of love are non-relational features, such as qualities of
character or non-relational features such as Kantian personhood, and the complete denial
that there are reasons for love, as Hamlyn and Frankfurt think. According to Kolodny,
our reason for loving a person is our relationship with that person, which is an ongoing
history that one shares with that person (Kolodny 2005, 146– 150).
Kolodny argues for the idea that there are reasons for love from the fact that we some-
times praise or criticize the presence and absence of love (Kolodny 2005, 137). But this
does not show that love is subject to rational constraints. For love may be evaluated in a
broad sense without being rationally assessable. Indeed, if I am right, we prize love only
if it is not rationally assessable. But to say that is not to give an argument against Kolodny.
Kolodny asks why amnesia destroys love (Kolodny 2005, 141–142), which is a good
question. Kolodny thinks that the view of relationships as reasons can explain this.
However, as we shall see in Section 6, the mental-causes – not-reasons view can also
explain it.
Kolodny’s relationship view does better than non-relational evaluative views, but it still
faces difficulties. The anti-propositional argument of Section 3.2 also applies to Kolodny’s
view since it too is an evaluational view and thus gives love the wrong logical form. But
there are further difficulties that his particular evaluative view faces. Assume a concept
of a “relationship” such that a relationship is not partly constituted by love. If it is not,
then we might expect love to develop in the context of such a “relationship”. But what if
it does not? Is there a sense in which we ought to love? This is hard to accept. What if
one were in two similar relationships but one only loves one of the two relata? Is that
somehow rationally inconsistent? That is also hard to accept. Parents sometimes love
some of their children much more than others. This is sad. But it is unreasonable or
irrational? Is it unfair? It is true that children might feel a complaint against their parents
that they were not loved enough or that the other siblings were loved more than they
were. But what kind of complaint do they have? Perhaps it is true that the parents ought
to have loved more. The lack of love was perhaps a moral flaw. But it does not have to
be a rational flaw, a failure to be rationally responsive to an awareness of the relationship
they were in. Love can be assessable and more or less appropriate without being more or
less rational.
Furthermore, Kolodny, like Velleman, has a trading-up problem. Kolodny thinks that if
the relationship in which one stands is the reason for love, this explains the “constancy”
(¼non-tradability) of love since the qualities of the person one loves may alter, while
the relationship remains (Kolodny 2005, 147). His view thus seems to explain why we
do not and should not love someone with the same intrinsic qualities. But the relationship
itself may alter, or there might be another person with whom one can have a better relation-
ship. If the reasons for love are relationships then if one can forge a better relationship then
one should do so and swap the objects of love (or at least add a new one). But the usual
norms of love say “no” – that one does not trade-up like this, unless other competing
values outweigh love (I say more about this below). By contrast, Kolodny allows
trading-up purely from within the domain of love. Therefore a plausible version of
love’s constancy is lost on Kolodny’s account, which means that his account fails to
fulfil the evaluative adequacy condition outlined in Section 1.
The underlying problem, it seems to me, is that Kolodny has things the wrong way
round: love gives reasons for relationships, not vice versa.
Note that in the Kampusch case, she loved her captor, but it is not clear that she values
the relationship (“He was part of my life” she said). Of course in that situation the relation-
ship had some value, when so many other things did not. But for Kampusch the relationship
Love: Gloriously Amoral and Arational 13
Downloaded by [N Z] at 12:07 15 July 2013
was a fact not a value, and it was a cause of love but did not rationalize it. She had no choice
in this relationship, and in this her situation was not so different from any family. People
have no choice over their family relationships. They may endorse them in reflection. But
love is not rationalized by our judgements about the value of relationships.
5.4 An arational view is in the clear as far as the trading-up objection goes. If love only
has causes (in shared history, habit or in attraction, for example), but not reasons, then love
is anchored in those contingent causes. If we lack reasons for love altogether, then love
itself cannot give us reasons for swapping another object for love or adding new objects
of love. Love will not mandate trading-up. Love is just a fact. (As Wittgenstein said in
another context, “It is there like my life” (Wittgenstein 1969, section 559).) Lacking
reasons, love cannot by itself generate reasons to trade its objects.
Nevertheless, while this dialectically defensive point is effective, it should not be over-
played. For we can and do trade-up as far as love goes. There is no denying this. It is a fact
of life, even if it is a somewhat sad fact of life. Not only do we sacrifice love for other
matters, we do sometimes decide that one object of love offers more than another. How
should we understand this?
In one sort of case, trading-up stems from other things valued, which are then weighed
against love. If one loves someone and trades to another object of love, it is only because of
something external to love. In many cases, other reasons can weigh against love, and love
may lose.
However, there is another “internal” way that we can deliberate about love. Even though
we do not love for reasons, it does not follow that we cannot reason about love. Even though
love has causes (for example in shared history, habit or attraction) and lacks reasons, one
might nevertheless have broadly prudential reasons to pursue certain objects of love
rather than others. Given the value of love (which is not based on reasons), we have
reason to foster and pursue it in the abstract and find some object for it. We can and do
engage in practical reasoning over love. David Schmidz discusses this insightfully, and
classifies this type of deliberation as involving “maieuetic ends” in Schmidz (1995,
chapter 3). In this sense one can rationally deliberate about who one should love, or
rather who one should let oneself fall in love with. This is similar to the way that we can
indirectly deliberate about what to believe in the sense that we deliberate about what to
try to bring ourselves to believe, even though one cannot directly decide what to believe.
We cannot decide to love any more than we can decide to believe, but one can take
steps to allow or encourage or ensure that one believes or loves one object rather than
another. For example, we might deliberate about which musical instrument to learn. One
might think that there is a shortage of viola players and that one is more likely to get to
play in an orchestra if one plays the viola. One might nevertheless go on to develop a
genuine love for the viola despite having taken a calculating decision about which instru-
ment to learn. Similarly, one might decide that one would be better off if one married a well-
paid lawyer, and one might nevertheless come to truly love the lawyer one ends up marry-
ing. However, one does not love the lawyer for the reasons for which one thought it would
be good to let oneself fall in love with him or her. Presumably when and if arranged mar-
riages are defensible, it is for this reason.
Gary Becker famously claimed that there is a marriage market that operates in many
ways like any other markets among rational economic agents (Becker 1975, Chapter 11).
One may try to maximize the attaining of one’s various goals in choosing a marriage
partner and someone whom one will love. However, having chosen, rationally, love is
caused, non-rationally; and that love generates reasons. Despite the fact that love itself is
not subject to rational requirements in the way that anger, fear and pride are, love can
14 Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
feature in one’s utility function insofar as one values and desires love in the abstract, and
also insofar as one loves a particular person.
Therefore we can and do deliberate about love, and it is not unreasonable. But once one
ends up loving a person, it is stuck there in one’s utility function. It then yields person-
specific reasons, and hence reasons that by their nature do not and cannot transfer to
others. Hence the norm of non-tradability is consistent with the fact that we deliberate
about love.
6. Love’s causes
6.1 The romantic theory says that love has causes but not reasons, in the sense that
nothing makes love rational. What kind of causes does love have? Common observation
supplies two common sorts of causes. We sometimes find ourselves in friendships or
family love or romantic love by shared history and habit; and we sometimes are said
to “fall” in romantic love, due to what is colloquially known as “chemistry”. Both pro-
cesses look more like the Humean mind of habit or passion, rather than the Kantian
mind of control, autonomy and evaluation. (As Hume might have said, habit is also
the guide to love life!)
Factors such as shared history, habit or attraction cause love, but that is not rational cau-
sation. Moreover, on the arational picture of love, the particular differentiating features of
the beloved that we are aware of are causes of love not reasons for love. They are like
cupid’s arrows. An awareness of the distinguishing features of the beloved may cause
love, in combination with various psychological factors, but that awareness does not
make love rational, nor does love depend on thinking that those features are good. We
are caused to feel. And in reflection we may be aware of this process, just as we may be
aware of the causes of our sensations. With love, we are like Kampusch, taken prisoner,
as many poets said long ago.
6.2 Does this mean that the causes of love are typically unknown to the lover, those
causes being perhaps brain processes or evolutionary histories? Not necessarily. It is true
that brain processes or evolutionary histories might be necessary to explain the phenom-
enon of love. But they do not suffice without a mentalistic story in addition. An overly
mechanistic or sub-personal account will not do justice to the nature of our encounters
with the other, whom we come to love as a consequence of those encounters. The
causes– not-reasons view needs supplementation in this respect. For although love has
causes, not reasons, it has mental causes. Not all mental causes are rational mental
causes. There is what Rosalind Hursthouse has usefully called “arational” mental causation
(Hursthouse 1991). In arational mental causation, the content of mental states is essential
for their causal role, but there is nevertheless no rationalizing relation. Another example
is thought association during daydreaming (Zangwill 2006). When we daydream, one
thought causes another but no process of reasoning is involved. The right view of love
falls midway between a rational causation view and a purely sub-doxastic view.
6.3 Kolodny observes that love is destroyed if amnesia deletes our memories of the
beloved (Kolodny 2005, 141–142). This shows not just that love must have had mental
causes but also that some of those mental causes must be stored in memory. But it does
not show that those mental causes are reasons. In thought association or daydreaming,
thinking of one thing may remind me of another thing; the first thought is a non-rational
cause of the second. Similarly, love is rationally caused in part by thoughts and experiences
and joint activities that we recall. The causes of love always include past mental events, the
memory of which is destroyed by amnesia.
Love: Gloriously Amoral and Arational 15
Downloaded by [N Z] at 12:07 15 July 2013
The role of memory in love points us in the direction of a more metaphysical point,
which is that the intentional object of love – a person – is an enduring thing, not a
person at an instant. The person one loves and with whom one has encounters, is
thought of as enduring over time. One conceives of the object of love as a reidentifiable
particular that one encounters at points in its trajectory through time. Love is caused by
those encounters. They are the mental causes of (but not reasons for) love. The shared
history that causes love may, in certain limiting cases of romantic love, amount to little
more than “love at first sight” or eye contact. But in most cases, it is much more that
this. It is no accident that the first shared mental moments tend to remain particularly treas-
ured in the minds of the romantic lovers, which they revisit from the time to time. (I agree
with Nozick’s emphasis on the historical aspect of love in Nozick 1989).
6.4 It might be said, in favour of a purely sub-doxastic view of the causes of love, which
many a song and poem attest to the idea of romantic love as a drug or an addiction to a
person. Is this literally true? Some research suggests that this is the case for sexual love
(Gallup, Burch, and Platek 2002). But the concepts of a drug or of addiction are somewhat
elastic. In Accounting for Tastes, Gary Becker allows that shopping and gambling are addic-
tions (Becker 1996). If so, romantic love too may be an addiction to a person. It would be an
addiction that springs in essential part from our psychological encounters with the person
we love, not just from sub-doxastic physical mechanisms that we are unaware of, even
though, without those sub-doxastic physical mechanisms, there would be no romantic
love. Thus while it is true that love also has non-mental causes – factors we are not
aware of, such as subliminal smell in the case of romantic love, mental causes are also
required, which are there during our psychological encounters. These are also causes of
romantic love without which it would not exist.
In the light of this, we may ask the somewhat uncontemporary question: are love potions
possible? If love necessarily has some mental causes in thoughts about the beloved, then a
love potion could only work by inducing or facilitating the relevant mental states, in the
way that alcohol might make one more susceptible to certain sorts of emotions. It can only
induce love by inducing intermediate mental states that increase susceptability. A love
potion cannot act directly to induce love. Therefore there is a sense in which a philosophical
argument can be given for the conclusion that love potions are impossible!
7. Conclusion
To conclude, let us return to the methodological considerations raised at the beginning of
the paper.
The evaluational conception of love collides with the way we value love. That way
allows that love has causes, but not reasons, and it recognizes and celebrates a love that
refuses to justify itself. Love has unjustified selectivity, due to its arbitrary causes. That
imposes a non-tradability norm.
A love for reasons, rational love or evaluational love, would be propositional, and it
therefore allows that the people we love are tradable commodities. A moralized conception
of love is no less committed to treating those we love as tradable commodities; it is just that
they are tradable moral commodities.
The evaluative criterion of adequacy, I suggest, encourages the opposite view – a non-
rational and non-evaluational conception of love. Such a love can set up partial obligations,
which may even demand that one sacrifice one’s life, as Cordelia did (Zangwill 2011).
Only a love that has causes but not reasons can have the kind of value that we think love
has, and thus it would only be rational to pursue and foster such a love.
16 Nick Zangwill
Downloaded by [N Z] at 12:07 15 July 2013
Love is amoral and arational and that is its glory.
Acknowledgements
This paper was originally written for a conference in Canberra on the work of David Velleman. It was
re-presented at a conference on love in Leuven and also at the universities of Bergen and Miami.
Many thanks to Jeanette Kennett for comments and discussion, and the audiences at these occasions
for questions and discussion.
Notes on contributor
Nick Zangwill is Professor of Philosophy at Durham University. He is the author of The Metaphysics
of Beauty, and Aesthetic Creation, as well as many papers in moral Philosophy, Philosophy of Mind,
and Metaphysics.
References
Badhwar, Neera. 2003. “Love.” In Oxford Handbook of Practical Ethics, edited by Hugh Lafolette,
42–69. Oxford: Oxford University Press.
Becker, Gary. 1975. The Economic Approach to Human Behaviour. Chicago, IL: University of
Chicago Press.
Becker, Gary. 1996. Accounting for Tastes. Cambridge, MA: Harvard University Press.
Ben Zeev, Aaron. 2008. “Hating the One You Love.” Philosophia 36 (3): 277– 283.
Blackburn, Simon. 1998. Ruling Passions. Oxford: Oxford University Press.
Frankfurt, Harry. 1999. Necessity, Volition and Love. Cambridge: Cambridge University Press.
Frankfurt, Harry. 2004. The Reasons of Love. Princeton, NJ: Princeton University Press.
Gallup, Gordon, R. L. Burch, and S. M. Platek. 2002. “Does Semen Have Antidepressant Properties?”
Archives of Sexual Behavior 31 (3): 289– 293.
Hamlyn, David. 1978. “The Phenomena of Love and Hate.” Philosophy 53 (203): 5 – 20.
Hursthouse, Rosalind. 1991. “Arational Actions.” Journal of Philosophy 88 (2): 57 – 68.
Kampusch, Natasha. 2010. 3096 Days. Harmonsworth: Penguin.
Keller, Simon. 2000. “How Do I Love Thee?: Let Me Count the Properties.” American Philosophical
Quarterly 37 (2): 163– 173.
Kolodny, Niko. 2005. “Love as a Valuing Relationship.” Philosophical Review 112: 135– 189.
Kraut, Robert. 1987. “Love De Re.” Midwest Studies in Philosophy 10: 413– 430.
Murdoch, Iris. 1970. The Sovereignty of the Good. London: Routledge.
Nagel, Thomas. 1979. “Sexual Perversion.” In Mortal Questions, 39 – 52. Cambridge: Cambridge
University Press.
Nietzsche, Frederich. 1974. The Gay Science. Translated and edited by Kaufman. New York: Random
House.
Nozick, Robert. 1989. “Love’s Bond.” In The Examined Life, 68– 86. New York: Simon and Schuster.
Putnam, Hilary. 1975. “The Meaning of Meaning.” In Philosophical Papers, vol. 2. Cambridge:
Cambridge University Press.
Schmidz, David. 1995. Rational Choice and Moral Agency. Princeton, NJ: Princeton University
Press.
Scruton, Roger. 1986. Sexual Desire. London: Widenfeld and Nicholson.
Velleman, David. 1999. “Love as a Moral Emotion.” Ethics 109 (2): 338– 374.
Velleman, David. 2008. “Beyond Price.” Ethics 118 (2): 191–212.
Wittgenstein, Ludwig. 1969. On Certainty. Oxford: Blackwell.
Zangwill, Nick. 2006. “Daydreams and Anarchy.” Philosophy and Phenomenological Research
LXXIII (2): 253– 289.
Zangwill, Nick. 2011. “Cordelia’s Bond.” Oxford Studies in Normative Ethics 1: 143 – 165.
Love: Gloriously Amoral and Arational 17
Downloaded by [N Z] at 12:07 15 July 2013
