The Idea of a Duty to Love

S. MATTHEW LIAO

624 N. Broadway, 3rd Floor, Johns Hopkins University, Baltimore, MD 21205, USA; e-mail: sliao@jhsph.edu; www.smatthewliao.com

 

Forthcoming in the Journal of Value Inquiry

1. A Commandability Objection

Can there be a duty to love someone?  The kind of love we will consider is the kind of highly intense interaction that two human beings seek that involves not only strongly valuing another person for the person’s sake and wanting to promote the person’s well-being for the person’s sake, but also desiring to be physically and psychologically close to each other and desiring that the other person reciprocates our love.[1]   This kind of interaction features in romantic love, parental love, love between friends, and the love of children for their parents.

A well-known argument against the possibility of a duty to love of this kind is the commandability objection. It is generally accepted, after the Kantian point of “ought” implies “can,” that to have a duty to do something, the action must be commandable. We must be able to bring about the action with success or, as some would say, at will.  Love is, however, not commandable, because it is an emotion and emotions are not commandable.  We cannot bring about emotions with success or at will.[2]  Therefore, there cannot be a duty to love.  Kant, a proponent of this objection, says the following:

Love is a matter of feeling, not of willing, and I cannot love because I will to, still less because I ought to (I cannot be constrained to love); so a duty to love is an absurdity.[3]

 

Similarly, Richard Taylor says,

Love and compassion are passions, not actions, are therefore subject to no terms of duties or moral obligations. . . . Love, as a feeling, cannot be commanded, even by God, simply because it is not up to anyone at any given moment how he feels about his neighbor or anything else.[4]

 

Strictly speaking, feelings, emotions, and passions are different concepts.  For example, emotions can include unfelt affects, whereas feelings by definition cannot.[5]  Emotions can also include less passionate affections whereas passions by definition cannot.  There are other differences, but within our context, we can use emotion as an umbrella term for the other concepts. 

Some philosophers have tried to respond to the commandability objection by arguing that love is not an emotion, but is instead an attitude.  For example, Joseph Raz writes that some people:

believe that there cannot be a duty to love someone.  The common reason for this supposed impossibility is that love is an emotion and the emotions cannot be commanded.  This is a misguided view of both love and the emotions.  Love is an attitude, not an emotion (though “being filled with love,” feeling “love swelling in one’s bosom,” and their like are-diverse-emotions).[6]

 

Love does involve having certain attitudes.  For example, as Mike Martin has pointed out, in a loving relationship we should have the attitude of valuing the person we love for the person’s sake.[7]  Other attitudes we should have in a loving relationship may include wanting to promote another person’s well-being for the person’s sake and wanting to be with the person. 

           However, love is not just having certain attitudes.  Love also involves having certain emotions.  The emotions associated with love are wide ranging.  They involve having, or the disposition to have, a strong sense of affection and warmth for the person we love, as well as a range of other emotions including grief, joy, anger, sadness, disappointment, and embarrassment.  To simplify the discussion, let us take a strong sense of warmth and affection to be a crucial part of the emotional aspect of love at least sometimes during the course of loving another person.  This strong sense of affection and warmth is stronger than a mere liking of the other person, and may cause us to be more excited in the presence of the person we love.  It is true that such strong emotions need not always be present in a loving relationship.  But there is typically a disposition toward having such emotions in a loving relationship.  Indeed, a friendly colleague may have all the attitudes typically associated with love such as valuing the other person and wanting to promote the other person’s well-being for the other person’s sake.  But it does not follow that the colleague loves the other person.  One way in which we could distinguish a loving relationship from such a case is the emotional aspect of love.  In a loving relationship, we would also have a strong sense of affection and warmth toward the other person, which might make us want to be with the other person more often and be sad when the other person is absent, whereas in the case where we are just a colleague of the other person, we need not feel such a strong sense of affection and warmth toward the other person. 

Given that love necessarily involves having certain emotions, the commandability objection cannot be refuted by arguing that love is an attitude.  Nevertheless, the commandability objection can be met.  The reason is that, contrary to the views of some people, emotions generally and the emotional aspect of love in particular are commandable. 

 

2. The Commandability of Emotions

It is often said that physical actions are commandable, but emotions are not commandable.  Indeed, it is often said that when we intend, for example, to raise our arm or to hold our breath, we can bring about such physical actions with success, while we cannot do this with emotions.  For this reason, it is thought that love, because it is an emotion, is therefore also not commandable.  The objection here is premised on the idea that emotions are never commandable.  While emotions are not always commandable, there are at least three ways by which we can bring about particular emotions, including the emotional aspect of love, with success, when we intend to do so.  Employing these methods will not bring us love proper such as a loving relationship.  However, by employing these methods we can bring about the emotional aspect of love, which is the aspect people have thought not to be commandable.  Showing that this aspect is commandable only removes one objection against the commandability of love.  There may be all sorts of reasons why love is not commandable in certain circumstances, as when the other person is immoral. But the claim that emotions are never commandable is too strong. 

There are a number of competing theories of emotions.  For example, there is the feeling theory of emotions proposed by William James, according to which an emotion is the feeling of certain bodily changes such as fluttering of the heart, epigastric activity, and shallow breathing, which are produced immediately by our perception of certain features of the world.[8]  On this view, different types of emotions involve feelings of distinctive sets of bodily changes.  For example, being angry is feeling certain bodily changes such as faster heart beats, and quicker breathing.  An alternative to the feeling theory of emotions is the behavioral theory of emotions, according to which emotions are dispositions to act in a certain way.[9]  On this view, a person who is angry would be predisposed to act in a certain manner such as pounding a table, picking a fight, or slamming a door.  A third alternative, which many philosophers endorse, is the cognitive theory of emotions, according to which emotions are based on beliefs, or desires, or judgments.  Robert Solomon says, for instance: “I cannot be angry if I do not believe that someone has wronged or offended me.  Accordingly, we might say that anger involves a moral judgment.”[10]  There are also combined theories such as Stuart Hampshire’s view that emotions are a mix of feelings and cognitions.  Hampshire writes: “Regret is a mode of unhappiness, or unpleasure, conjoined with a thought about the past.”[11]  Following writers such as Hampshire, we will take it that emotions are a complex of feelings, behavior, and cognitions. 

           A direct way by which we can bring about particular emotions with success, when we intend to do so, is through what we may call internal control, where we use reasons to motivate ourselves to have certain emotions.  There are at least two methods of internal control.  We can bring about a particular emotion with success by giving ourselves reasons to have particular emotions.  For example, let us suppose that we are not in a good mood on a particular day and on the particular day, we by chance are also invited to attend a close friend’s wedding.  Knowing that our close friend would not want us to be in a bad mood and would instead want us to be joyful on her wedding day, we might tell ourselves that because it is a special occasion, we should not be in a bad mood and that we should instead be joyful.  In giving ourselves a reason to be joyful and not to be in a bad mood, there is a good chance that we would not be in a bad mood, and would instead be joyful.  Other examples might include a boxer telling himself to become angry in order to prepare for a fight, or Tolstoy telling himself that he should feel grief, when attending his grandparent’s funeral.  By giving ourselves reasons to have particular emotions, we can bring about particular emotions with success, when we intend to do so.

Another method of internal control calls on us to reflect on the reasons why we tend to experience particular emotions in particular circumstances or toward particular persons.  Through reflecting on these reasons, we might then decide to continue to have particular emotions, if the emotions are supported by good reasons, or to discontinue to have particular emotions, if the emotions are not supported by good reasons.  For example, in The Sovereignty of Good, Iris Murdoch gives an example where a mother-in-law feels contempt for her daughter-in-law even though the mother-in-law outwardly acts kindly toward the daughter-in-law.  The mother-in-law then decides to reflect on the reasons why she feels contempt for the daughter-in-law, and realizes that the reason is because the mother-in-law is jealous that the daughter-in-law will threaten the mother-in-law’s relationship with her son.  After knowing the cause of her contempt for the daughter-in-law, the mother-in-law decides that her feelings of contempt for her daughter-in-law are not supported by good reasons.  The mother-in-law then begins to view the daughter-in-law without her initial prejudice by perceiving the daughter-in-law as “not vulgar but refreshingly simple, not undignified but spontaneous, not noisy but gay, not tiresomely juvenile but delightfully youthful.”[12]  Through reflecting on the reasons why she feels a certain way toward the daughter-in-law and deciding that her emotions are not well-supported, the mother-in-law is able to develop affection for the daughter-in-law.  To take another example, we might have reflected on the reasons why we have friendly emotions toward a particular person, and realize that some of the reasons are because the person appreciates us for who we are and because the person really cares about our well-being.  Perhaps, unfortunately, we become engaged in an unpleasant dispute with the person, having reflected on the reasons why we have friendly emotions toward the person, we might recall these reasons.  In doing so, there is a good chance that we would be able to continue to have friendly emotions toward the person. 

A less direct, but nevertheless viable, way by which we can bring about particular emotions with success is deliberately to place ourselves in situations in which we know that we would probably experience particular emotions.  We may call this the method of external control.  For example, if we know that we tend to feel pious when attending church services or tend to feel compassion when visiting homeless shelters, then we know that we have a reasonable chance of feeling pious or feeling compassion if we do these respective things. 

We have discussed ways of bringing about particular emotions in particular situations for particular persons.  We can also cultivate our emotional capacities such that we would be more likely to have particular emotions in appropriate circumstances.  As Aristotle says, cultivation involves habituation as well as reflection.[13]  One strategy for cultivating certain emotions is to behave as if we have particular emotions.  After some time and effort, it is likely that we would cultivate the capacities for these emotions.  For example, if we wish to cultivate our capacity for joy, we might begin by behaving as if we are joyful.  We might smile, whistle, sing, skip and hop, shout hurrah, and engage in various forms of behavior that are associated with joy.  Through engaging in these forms of behavior repeatedly over time, it is likely that we would cultivate the capacity for joy.  Indeed, Augustine observes that through enacting the behavior associated with religious rituals, we seem to increase our capacity for religious feelings:

For when men pray they do with the members of their bodies what befits suppliants—when they bend their knees and stretch their hands, or even prostrate themselves, and whatever else they do visibly, although their invisible will and the intention of their heart is known to God.  Nor does He need these signs for the human mind to be laid bare to Him.  But in this way a man excites himself to pray more and to groan more humbly and more fervently . . . although these motions of the body cannot come to be without a motion of the mind preceding them, when they have been made, visibly and externally, that invisible inner motion which caused them is itself strengthened.  And in this manner the disposition of the heart which preceded them in order that they might be made, grows stronger because they are made.[14]    

          

           Another strategy for cultivating our emotional capacities is by repeatedly using the method of external control such as by repeatedly placing ourselves in situations in which we know that we would probably experience particular emotions.  Using our previous examples, let us suppose that we know that attending church services and visiting homeless shelters often elicit emotions of piety and compassion in us.  To cultivate our capacities for such emotions, we might repeatedly visit such places.  In doing so, there is a good chance that we would cultivate these emotional capacities.   

           A third strategy for cultivating our emotional capacities is by repeatedly using the method of internal control described previously such as by repeatedly using reasons to motivate us to have certain emotions. For example, let us suppose that we believe that life is more enjoyable when we have a joyful disposition, and we believe therefore that having an enjoyable life is a good reason why we should cultivate our capacity for joy.  If we repeatedly remind ourselves of this reason, there is a good chance that we would cultivate the capacity for joy.

Finally, the method of cultivating emotional capacities involves not merely a repetition of internal and external control over time, but also deep reflection on the reasons why we tend to have particular emotions and whether we have good reasons for continuing or not continuing to have these emotions.  For example, let us suppose that a person is easily angered and upon reflecting on why he is so easily angered realizes that the reason is because his anger allows him to intimidate others into submitting to his desires.  He may decide that using anger to intimidate others is not a good way of getting things done.  He may also decide that he does not want to be the kind of person who is so easily angered.  If so, and if he repeatedly uses these reasons to motivate himself not to become so easily angered, there is a good chance that he would cultivate a disposition where he would not be so easily angered.  This suggests that to cultivate our emotional capacities, we may be required to evaluate critically some of our fundamental values, and to alter some of our ingrained character traits such as becoming more reflective than superficial, or becoming more altruistic than self-interested.  Such cultivation can be difficult.  However, the cultivation of physical abilities such as learning how to play the piano or to sail a boat is often equally difficult.  Yet people acquire these capacities nevertheless.  Moreover, in all these methods, the objective is not just to have the appearance of the emotions appropriate for the circumstance, but actually to have the genuine emotions appropriate for the circumstance.

 

3. The Commandability of the Emotional Aspect of Love

The three methods discussed above are also applicable to the emotional aspect of love.  For example, we can bring about the emotional aspect of love with success using the two methods of internal control.  Let us consider parental love.  We can give ourselves reasons to feel warmth and affection for a child.  Many reasons are possible, but a good reason is that children need this emotional aspect of love in order to develop certain capacities necessary to pursue a good life.  Let us consider romantic love.  If we have been introduced to a potential partner, we might give ourselves reasons such as the person seems very kind or the person seems interested in us as a way of getting ourselves to feel warmth and affection for the person. 

In addition, we can reflect on the reasons why we tend to feel a certain way, for example, toward a particular child.  Perhaps we do not initially like a child, and, upon reflection, we realize that our antipathy toward the child is due to the facts that the child was unplanned and that the child was born at a time when we already had too many children.  We might then recognize that this is not the fault of the child and that therefore this is not a good reason for disliking the child.  If we then begin to see the child without this initial prejudice, there is a chance that we would be able to bring about warmth and affection for the child.

Moreover, we can bring about the emotional aspect of love for another person with success through external control by deliberately placing ourselves in situations in which we would have a good chance to feel affection and warmth for the person.  For example, if we know that getting enough sleep helps us to be more affectionate and warm toward the child, then we might make sure that we have enough sleep each night so that we would be more loving toward the child.  Couples or friends who are trying to rekindle their relationship, and who know that they would have positive feelings if they visited places where they had enjoyed each other’s company, might arrange to revisit these places as a way of getting themselves to feel again the warmth and affection that they had for each other. 

Finally, we can cultivate our capacity to give affection and warmth for a person.  We can do this, for example, through behavioral inducement.  We might try to act affectionately and warmly toward another person, even if we do not initially feel these emotions for the person.  By repeatedly doing so, there is a good chance that we would cultivate the capacity to feel affection and warmth for the person.  Alternatively, we can try to cultivate this capacity through external control.  If two people who are trying to rekindle their relationship know that visiting places where they had enjoyed each other’s company would bring about positive feelings for each other, then they might repeatedly arrange such trips as a means to cultivate their affections for each other.  Lastly, we might try to cultivate this emotional capacity through deep reflection such as reflecting on the reasons why we do not feel love for a particular person, and determine whether our reasons are justified.  In doing so, there is a good chance that we would develop the emotional capacity to love that person.  Indeed, when arranged marriages were prevalent and still are in some parts of the world, couples might have employed these methods as ways of getting themselves to feel the emotions of love for their partners. 

The various methods are compatible with different theories of emotions.  For example, the method of internal control can be used to achieve the commandability of our emotions if the cognitive theory of emotions were the correct theory of emotions.  The method of external control is compatible with the behavioral theory of emotions as well as the feeling theory of emotions.  For example, when we go to church to feel more pious, we would be directly affecting the behavioral aspect of the emotion of being pious in that the church might predispose us to perform actions associated with being pious such as being quiet and praying.  On the feeling theory, we would be placing ourselves in an environment that would indirectly influence the feeling component of our emotion in that our heart might flutter more being in a church.  As we have taken emotions to be a composite of feelings, attitudes and behavior, all three methods are necessary for the commandability of emotions and the emotional aspect of love.

We should not think that the entire aspect of love is commandable if we just employ the three methods.  The dimensions of love are complex and love is not just an emotion, but involves having appropriate attitudes and behavior over a long period of time.  Nevertheless, the claim that the emotional aspect of love is never commandable is too strong.

 

4. Reasonable Success and Guaranteed Success

Some people might draw a distinction between bringing about an action with reasonable success and bringing it about with guaranteed success.  Supposing further that an action is commandable if and only if we can bring it about with guaranteed success, then emotions such as love would not be commandable, because it seems that they can at best be brought about only with reasonable, but not guaranteed, success.  Indeed, attending church services in order to feel piety and telling ourselves to be joyful at our friend’s wedding are all examples of bringing about particular emotions with reasonable, but not guaranteed, success, since we cannot guarantee that we will definitely experience particular emotions just by performing the activities suggested in these examples. 

           Guaranteed success is however not necessary for an action to be deemed commandable.  If it were, it would imply that many physical actions, which are paradigm examples of what is commandable, would not be commandable.  For instance, more complex physical actions such as walking and speaking would not be commandable, because it is a fact that we occasionally fail to succeed in such actions.  We might trip and fall and we might become tongue-tied.  In fact, even more simple physical actions such as raising our arms or holding our breaths would not be commandable, because it is a fact that we sometimes fail to succeed in the actions.  Our arms might be too tired and we might be too hyperventilated to be able to hold our breaths.  

           Some people might reply that guaranteed success can still be necessary for an action to be deemed commandable, if we add the clause provided certain reasonable background conditions exist.  For example, under ordinary circumstances, such as if our arm has not fallen asleep, if no one is holding our arm down, if our arm is not too tired, then because certain reasonable background conditions exist, we would be able to raise our arm with guaranteed success.  Under ordinary circumstances, such as if we had not just run a hundred-meter sprint, we would be able to hold our breath with guaranteed success. 

           However, the clause provided that certain reasonable background conditions exist is too vague.  On an ordinary understanding of it, even if reasonable background conditions were to obtain for a particular action, it seems that there is still a chance, however minute, that the action could fail.  For example, in the case where we try to raise our arm, even if certain ordinary background conditions obtain, such as our arm has not fallen asleep, no one is holding our arm down, and our arm is not too tired, it remains possible that we can fail to raise our arm with guaranteed success. 

           Let us suppose instead that we understand the clause provided that certain reasonable background conditions exist to mean that all possible factors that could defeat an action are rendered impotent so that an action really is guaranteed to succeed.  On this understanding, someone would have to provide an argument as to why we could not speak about guaranteed success regarding emotions such as love.  As an illustration, if all possible factors that could prevent a parent from loving a child are rendered impotent, what reasons would be left for thinking that the parent could still fail to love the child? 

           Other people might agree that we could sometimes bring about emotions including the emotional aspect of love with success using the methods discussed.  However, the notions of reasonable and guaranteed success are probability notions.  What is the probability that we can produce these emotions with success?  Would a low probability of success mean that there could not be a duty to love?

           It is difficult to offer an exact figure of the probability of success of being able to produce emotions such as the emotional aspect of love, because the probability will vary from person to person and circumstance to circumstance.  However, there is no reason to believe that, generally speaking, the probability of success would be low, because having emotions such as the emotional aspect of love is a common enough experience.  As such, it seems that we have a fairly good chance of bringing it about if we put in the required efforts.  We might take as a parallel that making free throws in basketball is a percentage shot.  The probability of success will vary among persons and in various circumstances.  However, because making such a shot is common enough, it seems that we have a fairly good chance of making it with some consistency if we practiced regularly. 

           For argument’s sake though, let us suppose that the probability of our succeeding in producing emotions such as the emotional aspect of love remains low, even after employing the three methods.  This still would not undermine the idea of a duty to love.  If we find a drowning person who requires cardiopulmonary resuscitation, and the probability of our success is low, because we get nervous and we are not very strong, we may have a duty, even so, to provide it.  Government agencies and aid workers in developing countries have a duty to reduce poverty and prevent deaths from diseases, even if the probability of their succeeding is low.    

           There could be cases where a person’s emotional capacities are so damaged that there is no possibility of the person’s acquiring the emotional aspect of love no matter what the person does.  In such cases, we would have to accept that the person does not have a duty to love, at least in an emotional way.  We should not conclude though that the duty to love would cease in these cases.  It is important to distinguish between the source of the duty and the capacities of the duty bearer.  Given that the duty stems from some facts in the world, such as the fact that a child needs to be loved in order to develop properly, the duty would remain, even if certain people lack the capacities to fulfill the duty. 

The distinction between the source of the duty and the capacities of the duty bearer is also useful in the case of someone who is trying very hard to love another person, but who nevertheless fails to have the appropriate emotions.  Supposing that there is a duty to love, we might say that the person has partially fulfilled the duty.  We might also say that the person is not blameworthy, since the person has tried his or her best.  Nevertheless, we could still say that the person has not fully fulfilled the duty, since the source of the duty remains.  For example, if Jane borrowed a thousand dollars from Ken.  She paid him some of it, tried her best to get a job, but through no fault of hers, is simply unable to pay him back in full.  From the perspective of the duty bearer, we may say that she has partially fulfilled her duty, and that given the circumstances, she is not blameworthy.  However, from the perspective of the source of the duty, it remains the case that she has not fully fulfilled her duty toward Ken.

A duty to love can be demanding, especially if we are required to change our character traits.  Also, at least in the case of romantic love, an exclusive focus on a particular person may be required, which is also very demanding.  Still, there may be situations when we have a duty to love someone even though fulfilling the duty is demanding.  For example, when a marriage is not going well, in some cases, we may have a duty to try to love our spouse again in a romantic manner.  Parents may have a duty to love their children because children need love in order to develop adequately.  These duties are not obviously absurd, and it is also not obvious that they would be undermined just because loving someone can be demanding.  Once we have made some commitments toward a beloved or a child, we may have to honor the commitments even if they involve more than minimum burdens.  Moreover, while the duties may have arisen out of prior commitments, there may be situations in which we have a duty to love someone without having made a prior commitment.  The duty of foster parents to love their adopted children may be a case in point.  Possibly more controversially, if a hermit living in a remote area finds a child, who he knows needs love in order to develop adequately, he may have a duty to love the child, even though he has not made any prior commitment toward the child.  While the above arguments do not establish that there are in fact such duties, they do make such claims coherent.

 

5. Direct and Indirect Commandability

Other people may distinguish between direct and indirect commandability and may argue that emotions and love have not been shown to be directly commandable.  It may be the case that the methods of internal and external control are more indirect ways of bringing about emotions such as love.  However, a rationale of the third method of commandability, cultivation, is to help us bring about appropriate emotions and love directly or at will in appropriate circumstances.  For example, if we try to cultivate the capacity for joy, a rationale for doing this would be so that we can bring about joy directly or at will in appropriate circumstances.  As an analogy, we might consider a professional musician who practices playing a certain piece of music so that she can bring about this music directly or at will in appropriate circumstances.

Even if emotions such as love are only indirectly commandable, duties do not always require that an action is directly commandable.  If Lester borrows five pounds from Mary, and therefore has a duty to repay Mary five pounds, a direct way of repaying Mary would be if he gives her five pounds.  But lacking five pounds, to repay Mary, he borrows five pounds from another person.  In this case, he would be taking indirect steps to fulfill his duty to Mary.  But doing so does not mean that he has not fulfilled his duty toward Mary.  If this is correct, then duties do not always require that an action is commandable directly. 

 

6. The Commandability of Attitudes

Love also involves having certain attitudes.  In addition, some people believe that emotions have attitudinal components.  Given this, it might be asked whether attitudes are commandable, since if they are not, then some aspect of love would not be commandable, and we would again face a version of the commandability objection.  A reason for raising this question is that attitudes are based on beliefs, and it seems that beliefs are not commandable, since they are based on facts, and we cannot just decide what facts to believe.[15]  For example, we cannot just decide that 2+2=5. 

           This issue need not concern us.  Even if the factual aspect of an attitude is not commandable, the valuation aspect of an attitude is commandable.  For our purpose, it is sufficient that the valuation aspect may be required as a matter of duty.  Most, if not all, attitudes are made up of facts and values.  For example, an attitude that animals that are as sentient as some human beings should be treated with equal consideration is made up of certain facts such as that such animals are as sentient as some human beings and certain values such as given equal sentience, such animals should be treated equally with some human beings.  An implication of this fact-value distinction is that two people faced with the same facts may value the facts differently.  For example, a different person faced with the same facts may hold the attitude that while we should not be cruel to such animals, we should not treat them with equal consideration as human beings.

           The valuation aspect of an attitude is commandable in the same way emotions are.  For example, we can bring about certain attitudes with success through internal control such as by giving ourselves reasons to value a fact in a certain way, or, through reflecting on the reasons why we tend to value a certain fact in a certain way.  Let us suppose we want to have a more positive attitude toward the environment.  We might tell ourselves all the positive things that a better environment can bring us such as better air, nicer landscape, and more habitable climates.  In doing so, there is a good chance that we would have a more positive attitude toward the environment.  In addition, we can also bring about certain attitudes with success through external control by deliberately placing ourselves in situations.  If we want to have more negative attitudes toward inadequate help given to people in inner cities, we can deliberately place ourselves in such places.  In doing so, there is a good chance that we would develop negative attitudes toward inadequate help to these people.  Finally, we can cultivate our capacity to have certain attitudes, for example, through behavioral inducement, where we would act as if we do have the attitude.  This is a rationale behind positive thinking programs.  We can also cultivate our capacity to have certain attitudes through repeatedly using the methods of internal and external control.

The valuation aspect of our attitude toward a person we love is what may be required as a matter of duty, because love involves valuing the persons we love for their own sake, as in loving a child.  A child is born having certain temperament and genetic make-up and there is not much we can do to alter them.  But we can still try to value the child positively.  For example, faced with a child who is learning to eat, we can decide to regard the child as happy and persevering rather than as messy and annoying.  Faced with a child who does not speak much, we can decide to regard the child as quiet and shy as opposed to unresponsive and uninterested.  As with bringing about emotions, the objective is not just to have the appearance of attitudes appropriate for the circumstance or false valuations, but actually to have genuine and correct attitudes appropriate for the circumstance.  In general, we can bring about positive attitudes for a child with success in the same way we can bring about general attitudes with success through internal and external control and through cultivation.  Even though love involves having certain attitudes, love as an attitude is also commandable.

 

7. The Pretended Love Objection

If the emotional aspect of love is commandable, what happens when a person pretends to love another person when he really does not love the other person? Would he be fulfilling the duty to love the person, supposing there is such a duty?

This case should be distinguished from the case discussed earlier where someone is genuinely trying to love another person, but fails to bring about the real emotions.  The person who pretends to love another person when he does not would not be fulfilling his duty, because the fulfillment of a duty requires that the object of the duty provided is real.  If Neal owes Olga five dollars and gives her a fake five-dollar note in return, he would not have fulfilled his duty to her because if it were ever revealed that he has given her a fake bill, he would be required to give her a real one.  In any case, it is wrong to pretend to have fulfilled the duty when we have not.  Indeed, it would be wrong for Neal to give Olga a fake bill as a means to fulfilling his duty, even if no one ever finds out. 

 

8. Is Romantic Love Special?

Even agreeing that the emotional aspect of some kinds of love, such as parental love, is commandable, some people will still be skeptical that the emotional aspect of all kinds of love is commandable.  In particular, they may believe that romantic love, the paradigm example of passion, is not susceptible to the methods discussed.

The idea that romantic love is not commandable may be a relic of the German Romanticism of the eighteenth and nineteen centuries, according to which there is a sharp boundary between reasons and emotions.  Indeed, the major intellectuals at that time such as Hume and Kant all subscribed to this distinction.  It is therefore possible that we have inherited this possible bias from German romanticism, and that we have continued to perpetuate this idea in our media, literature, and music.  Indeed, there are alternative views of emotions, according to which there is not such a sharp boundary between reasons and emotions.  A more Aristotelian view of emotions would more naturally lend itself to the conclusion that emotions, including the emotional aspect of romantic love, are commandable. However, it is also possible that this view of romantic love is not a bias, but a result of our personal experiences with falling in and out of love.  We should therefore examine more closely the phenomenology of these experiences.

When we fall in love, sometimes it may seem as if we have no control over our emotions.  In particular, it may seem as if we cannot stop loving a particular person, even if there were reasons to believe that loving the person is a bad idea.   At the same time, there may be some persons whom there seem to be no reason why we should not love, and yet we cannot seem to muster any loving emotions for them. Finally, when we fall out of love, it may seem that nothing we can do will enable us to have loving emotions for that person again.  We have simply lost interest. 

           There is no simple and straightforward explanation for these all-too-common experiences of love.  Still, whom we love or do not love does not occur at random or by accident.  We typically have expectations about our ideal beloved, which have arisen out of an interplay of societal expectations of ourselves, general societal expectations, our expectations of ourselves and our expectations of others.  For example, our expectations might have been built up from our childhood from when our parents and teachers read us stories about brave knights and beautiful princesses.  They might have further been reinforced through our peers, society, and the media.  Also, given our conception of ourselves such as whether we have a positive view or a negative view of ourselves, we might accordingly also have certain expectations as to who would be the right person for us.  Given this, when we meet a certain individual, we already have a fairly extensive checklist for deciding whether that individual is the right person for us.  It is not surprising that we fall in love when someone seems to fit all the criteria, until we find out later that the person is not as perfect as we thought.  We fail to fall in love when someone lacks certain important criteria on our list, even though that person may seem to others perfect for us in all respects.  We may not be willing to admit this, since the criteria on our list may be seemingly trivial or even immoral, such as we may dislike ugly toes or we may have prejudice against people of certain races.  In fact, our reluctance to reveal that we have such a checklist might provide an explanation as to why we would perpetuate the idea that romantic love is not commandable, since it is easier to say that we just do not fancy another person than to admit that we have an embarrassing and possibly immoral checklist.  Finally, we fall out of love either when we realize that the other person does not fit our expectations or our expectations have changed as a result of certain life events such as midlife crisis.

If we do have these expectations, then it is not surprising that the commandability of the emotional aspect of romantic love would seem difficult, since we might not always be aware of them.  But, if we do have these expectations, then romantic love should be commandable, provided that there are good reasons to believe that we should revise our expectations.  For example, if on reflection, we find that we have been conditioned to be attracted only to people from a certain social class or certain race, we might decide that this is not justified and that we should revise our expectations.   In doing so, there is a good chance that we would be able to fall in love with certain people whom we did not initially find attractive.  Alternatively, we would have a midlife crisis and believe that we have fallen out of love and become attracted to someone else, though on reflection, we realize that we are attracted to the other person only because we are worried about failing in a particular goal that is important to us and that the other person seems to be able to help us reach.  If this is the case, we might decide that this is not a good reason to fall out of love with our original partner and fall in love with someone else.  If so, there is a good chance that we would be able to fall in love with our original partner again.     

The emotional aspect of romantic love is not always commandable.  There may be people to whom we should not be attracted, such as immoral people, our bosses, and students, and there is no reason to revise our expectations.  It is also doubtful that there is a general duty to love someone romantically.  We should love another person romantically because we want to and not because we have to, especially since loving someone romantically requires exclusive focus on the person.  Nevertheless, there might be situations when we have a duty to love someone romantically.  In particular, when a marriage is not going well, we might have a duty to try to love our spouse again in a romantic manner.  Once we recognize that whom we love depends partly on built-up expectations, then it should be clear that the emotional aspect of romantic love is more commandable than people have thought.  

 

9. The Motivation Objection

Some people might argue that really to love a person, we must be motivated to do so for the person’s sake.  To have a duty to love a person means however that we would not be motivated to love the person for the person’s sake, but for the sake of the duty.  Therefore, to have a duty to love a person means that we do not really love the person.  On this view, employing the three methods discussed earlier as a way to love someone means that we do not really love that individual.  We may call this the motivation objection.    

The motivation objection has its root in critiques of Kant’s moral theory.  Kant argues that an action has moral worth only if it is done for the sake of the duty.[16]  Critics of Kant argue that at least in personal relationships, we should act out of a direct concern for the others, but that a Kantian agent cannot do this, because he is acting out of a concern for a moral principle, in particular, the duty.  In recent years, this criticism has been extended to all impartial ethical theories.  For example, Michael Stocker asks us to imagine that we are hospitalized and received a visit from Smith, who alleged that he is our friend:

 

You are very bored and restless and at loose ends when Smith comes in once again.  You are now convinced more than ever that he is a fine fellow and a real friend—taking so much time to cheer you up, traveling all the way across town, and so on.  You are so effusive with your praise and thanks that he protests that he always tries to do what he thinks is his duty, what he thinks will be best.  You at first think he is engaging in a polite form of self-depreciation, relieving the moral burden.  But the more you two speak, the more clear it becomes that he was telling the literal truth: that it is not essentially because of you that he came to see you, not because you are friends, but because he thought it his duty, perhaps as a fellow Christian or Communist or whatever, or simply because he knows of no one more in need of cheering up and no one easier to cheer up.[17]

 

Stocker argues that Smith is not really a friend, because Smith appears to be motivated to see us for the sake of some impartial rule, but not for us.  Smith, as it were, is doing the right thing but for the wrong reason.  This, according to Stocker, shows that impartial ethical theories cannot adequately account for personal ethics. 

Bernard Williams makes a similar point using a different anecdote.  He asks us to suppose that two people are drowning, one of which is the rescuer’s wife, and the rescuer can save only one of them.  Williams points out that while most people would agree that the rescuer is permitted to save his wife, impartial ethical theories have the implication that this is justified because there is some impartial rule that says that whenever a person is in a situation of this kind, the person is permitted to save his wife.  According to Williams, this is “one thought too many.”[18]  For Williams, someone who has an impartial thought in such a circumstance has failed to respect the personal-impersonal divide.  The person who saves the wife, because of the impartial rule is not doing it only for his wife’s sake, which Williams thinks he should.  Since impartial ethics by definition require us to think impartially, Williams argues that they cannot explain why it is permissible to do something without any further impartial thoughts on the matter.   

The rationale behind the motivation objection is based on three ideas, two of which are uncontroversial.  The first, uncontroversial idea is that we should do the right thing for the right reason or motive.  A right action that is done for the wrong reason or motive has less moral worth than an action done for the right reason or motive.  An illustration may be Kant’s shopkeeper who should give right change to his customers because it is the right thing to do, and not because it is good for him. 

           The second, also uncontroversial idea is that in personal relationships, we should be motivated to do things for the other person’s sake.  We would not be motivated to do something for the other person’s sake if, for example, we were motivated to perform the action out of self-interest.  If we were motivated to love another person because the relationship would benefit our career, then we would be loving someone for our own sake instead of for the other person’s sake, and this would be loving someone for the wrong reason.   

The third idea, which is controversial, is that in personal relationships, when we are motivated to do the right thing or when we are motivated to do our duty, this inevitably undermines the relationship, because as with being motivated to do things out of self-interest, we would not be motivated to do things for the other person’s sake.  We would be motivated instead to do it for the sake of an act’s being a right thing to do or for the sake of a duty.  This is the point made by Stocker and Williams.  The idea is controversial and false for the following reason.  It presupposes that being motivated to do the right thing or for the sake of a duty and being motivated to do something for the other person’s sake are always distinct motivations.  It is true that sometimes when we act for the sake of a duty, we do not act for the sake of the other person.  For example, a postman may deliver mail to us everyday just because it is his duty, and not because he is doing it for our sake.  A father who is a teacher in his daughter’s class may help his daughter with schoolwork in class because this is his duty as a teacher rather than because he is her father.  But the two motives need not always be distinct.  Sometimes, being motivated to do the right thing is being motivated to do it for the other person’s sake.  The content of the duty is just to be motivated for the other person’s sake.  For example, in the case of loving a child, being motivated to do the right thing is just being motivated to give the child love for the child’s sake.  In Stocker’s example, being motivated to do the right thing is just being motivated to visit Smith for Smith’s sake.  In Williams’s case, being motivated to do the right thing is just being motivated to save the wife’s life for her sake.  We could even claim that in these cases, there is a duty to be motivated for the other person’s sake, and people who are not thus motivated would not have fulfilled their duty.  If being motivated to do the right things and being motivated to do things for another person’s sake are just the same motivation in these case, then the motive of duty and the motive of love cannot conflict since they are one and the same.  If so, the motivation objection would not undermine the idea of a duty to love.

However, some people might believe that being motivated to do the right thing and being motivated to do things for another person’s sake are always distinct motivations.  Even so, duty and love can still be compatible.  For duty and love to be incompatible, we would have to assume that we cannot be motivated by both motives at the same time.  For example, we would have to assume that a person who is motivated to love another person for the sake of a duty cannot at the same time be motivated to love the other person for the other person’s sake.  But this assumption is false.  We can be motivated to do something for the sake of a duty, and, at the same time, for the person’s sake.  In Stocker’s example, Smith could have been motivated to see you both for the sake of a duty and because he wants to see you.  To use another example, a professional cook who receives a visit from a friend while at work can be motivated to prepare delicious dishes for his friend both because he has a duty to do so, since he is hired to do so, and because he wants to do it for his friend.[19] 

Some people might respond that in personal relationships, a person should only have one motive for action, the motive to do things for the other person’s sake.  People who act out of additional motives, including a motive of duty, would, on this view, be violating a normative requirement and the relationship would be undermined.  This might be the rationale behind Williams’s point that an additional impartial motive is one thought too many, because in personal relations, we should just have one motive.

The claim here could be that in personal relationships there is a duty to have only one motive.  Given this duty, people who are motivated to act out of more than one motive in personal relationships would be violating a normative requirement.  However, if this is the claim, then it is self-defeating. We are supposing that being motivated to do the right thing and being motivated to do things for the other person’s sake are distinct motives.  Let us suppose we do try to fulfill the duty to have only one motive, in particular, being motivated to do things for the other person’s sake, we would be acting from two motives, the motive of the duty and the motive of doing things for the other person’s sake.  But acting out of two motives violates the duty to act out of only one motive.  Hence, in this context, the claim that there is a duty to have only one motive is self-defeating. 

The claim could instead be that in personal relationships we should just not think impartially.  We should do things for the other person just because it is the other person, and not because there is some impartial rule that tells us to do so.  However, even in personal relationships it is morally necessary to think impartially.  We might consider a modified version of Williams’s example where the wife is not drowning and is only slightly injured, while the other person is still drowning.  It seems that the rescuer may be required to try to save the stranger instead of attending to his wife’s minor wounds.  If the rescuer chooses instead to attend to his wife’s minor wounds, the rescuer may be blamed for having failed in his duty to try to save a drowning person.  If this is correct, to reach such a conclusion, it seems that the rescuer would have had to have considered the situation impartially in order to recognize that his wife and the stranger are not in the same predicament.  But if it is morally necessary in this case that the rescuer considered the situation impartially so that he would arrive at a correct moral decision, in the case where the wife and the stranger are both drowning, it seems also morally necessary that the rescuer considers that situation impartially.  Indeed, it may be because, when viewed impartially, the stranger and the wife are in the same situation, except that the wife is in a special relationship with the rescuer, and because we think that this fact ought to be given some moral weight, that it seems morally permissible for the rescuer to save his wife.  If this is correct, then even in personal relationships it is morally necessary to think impartially. 

The claim that being motivated to do things for the other person’s sake should be the only motive for action in personal relationships can be cogent if we drop the requirement that being motivated for the sake of a duty and being motivated for the sake of the other person are distinct motivations.  Once this requirement is dropped, then we can cogently say that there is a duty to act out of only one motivation in personal relationships, since we can view the motive to do our duty in this case as just the motive to do things for the sake of the other person.  However, dropping this requirement means that duty and love are compatible.  Hence, on the most plausible interpretation of this claim, it does not undermine the idea of a duty to love.[20]



Notes

 

[1].    See William Newton-Smith, “A Conceptual Investigation of Love,” in Alan Montefiore ed., Philosophy and Person Relations: An Anglo-French Study (London: Routledge and Kegan Paul, 1973); Gabrielle Taylor, “Love,” Proceedings of the Aristotelian Society 76 (1976), p. 151; Robert Brown, Analyzing Love (Cambridge, England: Cambridge University Press, 1987).   

[2].    See Gilbert Ryle, “Feelings,” The Philosophical Quarterly 1, No. 3 (1951); Amélie Oksenberg Rorty ed., Explaining Emotions (Berkeley, Calif.: University of California Press, 1980).

[3].    Immanuel Kant, The Metaphysics of Morals, trans. Mary Gregor (Cambridge, England: Cambridge University Press, 1996), p. 161; Immanuel Kant, Groundwork of the Metaphysic of Morals, trans. H. J. Paton (N. Y.: Harper & Row, 1964), p. 67. 

[4].    Richard Taylor, Good and Evil (New York: Macmillan, 1970), pp. 252-253.

[5].    See Stephen Leighton, “Unfelt Feelings in Pain and Emotion,” Southern Journal of Philosophy 24, No. 1 (1986).

[6].    Joseph Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford: Oxford University Press, 1994), p. 11.

[7].    See Mike Martin, “Love’s Constancy,” Philosophy 68 (1993), p. 65.

[8].    See William James, “What Is an Emotion?”, in K. Dunlap ed., The Emotions (New York: Hafner, 1967), pp. 11-13, 15-21, 25-26.

[9].    See B. F. Skinner and J. G. Holland, The Analysis of Behavior (New York: McGraw-Hill, 1961), pp. 213-214.

[10]. Robert C. Solomon, “The Logic of Emotion,” Nous 11 (1977), p. 187. 

[11]. Stuart Hampshire, “Sincerity and Single-Mindedness,” in Freedom of the Mind and Other Essays (Princeton: Princeton University Press, 1971), p. 239.

[12]. See Iris Murdoch, The Sovereignty of Good (London: Routledge and Kegan Paul, 1970), pp. 17-18.

[13]. See Aristotle, The Nicomachean Ethics, trans. W. D. Ross (Oxford: Oxford University Press, 1980), bk II, ch. 1, 1103a33-b25. 

[14]. Augustine, De Cura Pro Mortuis, trans. H. Chadwick (Oxford: Oxford University Press, 1998), 5.7.

[15]. See Bernard Williams, “Deciding to Believe,” in Problems of the Self (Cambridge, England: Cambridge University Press, 1973).

[16]. See Kant, Groundwork of the Metaphysic of Morals, op. cit.

[17]. Michael Stocker, “The Schizophrenia of Modern Ethical Theories,” The Journal of Philosophy 73 (1976), p. 462.

[18]. Bernard Williams, “Persons, Character, and Morality,” in Moral Luck (Cambridge, England: Cambridge University Press, 1981), p. 18.

[19]   See Barbara Herman, “Rules, Motives, and Helping Actions,” Philosophical Studies 45, No. 3 (1984); Justin Oakley, Morality and the Emotions (London: Routledge, 1992).

[20].  I would like to thank Jim Griffin, Nick Bunnin, the late Geoffrey Marshall, David Archard, Joseph Shaw, Douglas Wolfe, Karin Boxer, Wibke Gruetjen, the late Bernard Williams, Justin Oakley, Julie Tannenbaum, Thomas Magnell, an anonymous referee of the Journal of Value Inquiry, and audiences at the Oxford Political Theory Seminar, the Queen’s College Symposium, and the Joint Session of the Mind Association and the Aristotelian Society in Belfast, Northern Ireland, for their helpful comments and discussions on earlier versions of this paper.  Thanks are also due to the Greenwall Fellowship Program at Johns Hopkins University and the Kennedy Institute of Ethics at Georgetown University for generous research support.