[I]nterests requiring the violation of justice have no value. Having no merit in the first place, they cannot override its claims.—John Rawls1

PARENTAL partiality consists of actions by which parents bestow greater resources and affection on their own children than on others. Because different parents are not equally able or willing to provide support for their children, parental partiality contributes to inequality amongst children. That is troubling: it seems unjust that some children should face worse prospects in life than others merely because their parents were less able or willing to support them during their childhood. This ‘distributive objection’2 impels us to consider what might be said on the other side of the argument: what is so valuable about parental partiality?

This article examines the nature of an important type of good that can arise for parents and children by virtue of their relationship. Many parents believe it is valuable that they are able to do good for their children—that they are able to advance their children's interests. What they regard as valuable is not just the fact that their children come to enjoy certain generic goods they provide for them—the piano lessons, the money, the clothes, and so on. They see value also in the parental act itself of providing those generic goods. Adapting a term from Brighouse and Swift, let me call this second-order good, general promotion.3

My aim is to describe and defend what I will call a moralized view of general promotion. According to the moralized view, in order for acts of parental partiality to realize the good of general promotion they must comply with the demands of morality, in a sense I elaborate below. The non-moralized view denies that this is necessary. Acts of general promotion that consist of stealing or cheating or murdering in order to benefit one's child are no less capable of realizing what's good about general promotion for that. That isn't the same, of course, as saying that that those acts should be permitted all things considered, for their bad consequences may outweigh the goods, including the good of general promotion, that they realize.

If the moralized view is correct, then the distributive objection to parental partiality becomes stronger, at least when we focus on acts of general promotion. This is because the moralized view entails that we should not include as a consideration in favor of permitting general promotion any supposedly special value inherent in general promotion. General promotion only has special value if it is, in any case, morally permissible; considerations about its special value should, therefore, not play a role in shaping the regulation of parental partiality. The moralized view also has implications for personal ethics. It offers parents the reassurance that they will not miss an important value when they are prevented from performing acts of general promotion that are morally impermissible. In having to keep their acts of partiality within the morally permissible range of general promotion, parents don't forgo any of the value of general promotion they might have (mistakenly) hoped they might have secured in supporting their children.

The article is structured as follows. Section I sets out some preliminary points that clarify the ensuing discussion. Section II criticizes a version of the non-moralized view recently defended by Brighouse and Swift. Section III shows that other versions of the non-moralized view do not adequately explain our intuitions about particular cases of parental partiality. Section IV elaborates what I call the ‘Scanlonian argument’ for the moralized view, according to which the special value of general promotion depends on its expressing an attitude that properly values one's child. Section V concludes with a discussion of the relevance of the moralized view.


  1. Top of page
  2. Preliminaries
  3. Unrestricted General Promotion
  4. Other Non-moralized Views
  5. The Scanlonian Argument
  6. Conclusion

This section clarifies some key terms and answers some initial concerns with the moralized view.

The Moralized View

It is helpful to distinguish two types of parental partiality. Some acts of parental partiality are acts of intimacy. These are acts in which parents and children do, and share meaningful things together. This is the class of acts that we commonly refer to as ‘quality time’; it includes talking with each other, playing together, reading a story together, and so on. Acts of general promotion are different. These are acts parents perform in order to advance their children's interests in general, such as, for example, purchasing private education or private health care for them, or establishing a trust fund for them. In this article, I shall focus on acts of general promotion.4

The moralized view is a view about the prudential value that acts of parental partiality can have for parents and children. Let me describe the kind of prudential value it addresses. First, this prudential value is non-instrumental value—it is the value acts of parental partiality have, not in virtue of their being means to good outcomes, but in their own right. Secondly, it is the prudential value acts of parental partiality have in virtue of their being performed by a parent for his or her child, or in other words, in virtue of their being performed within the context of a special relationship. In order to simultaneously refer to these two features of the type of prudential value the moralized view addresses, I shall, henceforth, speak of the distinctive value of acts of parental partiality.5

Here is an illustration of how a particular act of general promotion might have distinctive value. Consider a father who spends many hours, over many years, working overtime to pay for expensive piano lessons for his daughter. Suppose that, as a result of this, she acquires a hugely rewarding passion for music. Most of us would say that what he has done has made his, and his child's life, better for them both, and not purely because it was instrumental in bringing about a hugely rewarding passion for music in his child. It was good, in itself, for both of them that he acted in that way towards her. Furthermore, the distinctive value of his so acting crucially depended on the fact that it was he who did that for her, rather than some unrelated benefactor.

The moralized view holds that distinctively valuable acts of parental partiality are restricted to those that would otherwise be morally permissible. By ‘otherwise morally permissible’ I mean that these acts would be morally permissible independently of whether or not they had distinctive value. So, the test for whether a particular act of general promotion, A, is distinctively valuable is the following: would A be morally permissible even if we assume A isn't distinctively valuable? According to the moralized view, A can only be distinctively valuable if the answer to that question is ‘yes’.

This statement of the moralized view leaves open which moral theory we should adopt in the background when judging the moral permissibility of a given act of general promotion. There can be different versions of the moralized view, depending on which moral theory one adopts in the background. A consequentialist version of the moralized view would hold that a given act of general promotion, A, has distinctive value if and only if A would bring about the best consequences (act-consequentialist version), or would comply with rules that bring about the best consequences (rule-consequentialist version), under the assumption that it did not have distinctive value.6 A contractualist version of the moralized view would hold that a given act of general promotion has distinctive value if and only if no one could reasonably reject a rule that permitted that act.

Two Initial Concerns

Let me now respond to two concerns one might have about the moralized view right from the outset. First, it may seem to present the relationship between prudential value and morality the wrong way round. It may seem to make the prudential value of an act of parental partiality depend on the moral permissibility of that act, whereas, ordinarily, our deliberation proceeds in the reverse direction: normally, we take the prudential value of an act as an ‘input’ into our deliberations about whether that act should be deemed morally permissible, not as an ‘output’ of those deliberations.

To see why that problem shouldn't trouble us, we need to keep careful track of what exactly the moralized view says should be an input and output of our moral deliberations. Notice that the moralized view does allow us to introduce prudential value as an input into our moral deliberations. What it says we should not introduce as an input is distinctive value. Suppose, for example, you could fiddle with the coin that is to determine whether you own child or some other child should receive the last remaining place at a prestigious university (assume your child and the other child stand equally to benefit, and are equally well off to begin with). The moralized view allows that the prudential value for your child of going to a prestigious university should be an input into our moral deliberations about whether you should fiddle with the coin. What it says shouldn't be an input is only the distinctive value of the general promotion in that act (i.e., its conferring a good on your child). And it is only the latter, distinctive value, that the moralized view says should be an ‘output’ of our moral deliberations. If, as seems plausible, it would be morally impermissible to fiddle with the coin, all that follows is that there would be no distinctive value in your doing this for your child.

The second problem that may appear fatal to the moralized view is that it might seem draconian with respect to the range of acts of general promotion it will find lacking in distinctive value. One might believe that if we were to bracket the fact that there is distinctive value in my acts of general promotion, many of those acts would appear to be morally impermissible. Wouldn't the resources I use to purchase music lessons for my children, for example, be better used to help others? Maybe it would be more efficient for me to take turns with other parents to feed all of our children together, canteen style, thus freeing up man-hours we could use to help others more? If that's true, the moralized view seems to entail that there isn't any distinctive value in reading bedtime stories, or in sitting down with my children for dinner, or in most other everyday things parents do for their children.

It is important not to assume that all versions of the moralized view yield draconian results. Indeed, the problem highlighted in the previous paragraph only appears to affect an act-consequentialist version of the moralized view. It may well be true, in other words, that the moralized view entails that many everyday acts, by which parents look after their children lack distinctive value when the moralized view is held against the background assumption that the moral permissibility of an act turns on whether that act produces the best consequences. But, as we saw earlier, the moralized view can come in other versions. One can also hold the moralized view within a rule-consequentialist or a contractualist moral framework. In those latter versions, the moralized view would not entail draconian conclusions. Let me briefly explain why that is so.

There are enormously valuable goods, for parents and children alike, that depend for their realization on parents doing thousands of everyday things for and with their children. It is only when the same adults share lots of experiences with the same children—like reading bedtime stories or making dinner for them—over a long period of time that those adults and children can come to enjoy the special goods of having someone in their lives they can trust and who knows them intimately.7 Thus, while it may be true of any one act of general promotion that a parent might have done more for others had he helped them instead, a rule-consequentialist could plausibly appeal to the special goods of intimacy and trust to argue that the rule that would produce the best consequences overall would not require parents to constantly re-direct their support from their children to others.8 A contractarian could also plausibly insist, in virtue of the great importance of those same special goods, that parents and children could reasonably reject a rule that required parents to frequently re-direct his support from his child to others.

Unrestricted General Promotion

  1. Top of page
  2. Preliminaries
  3. Unrestricted General Promotion
  4. Other Non-moralized Views
  5. The Scanlonian Argument
  6. Conclusion

My criticism of the non-moralized view will proceed by casting doubt on different versions of it, one by one. All versions of that view share the negative claim that acts of parental partiality need not be, otherwise, morally permissible in order to be distinctively valuable. In this section, I start by criticizing a version of the non-moralized view that holds that any act by which a parent advances his child's interests has distinctive value. I call this view unrestricted general promotion.

Unrestricted general promotion is endorsed by Harry Brighouse and Adam Swift. They write:

a loving parent will be motivated quite generally to further her children's interests, to care that her child's life go better than it otherwise might. If so, then it looks as if any action by a parent to benefit her child realizes a familial relationship good, which might seem fatal to our project.9

The claim Brighouse and Swift rely on to justify their conclusion that a familial relationship good is realized by any action whereby a parent benefits her child is this: ‘a loving parent will be motivated quite generally to further her child's interests.’10

The objection I shall mount against unrestricted general promotion appeals to our intuitions about particular cases. Recall, to begin with, that general promotion is not the first-order good a parent transfers to a child; it is a second-order good that inheres in an act, by a parent, of transferring a first-order good to a child. We can, thus, determine whether there is any distinctive value in a given act of general promotion by adopting what I will call an agent substitution test: substitute the parent with an agent unrelated to the child, while keeping the transfer of the first-order good the same, and then determine whether the value of that act would diminish as a result. If our intuitions return the answer ‘yes’, we should conclude that there is distinctive value in that act of general promotion and vice versa. That test needs refinement in view of the following problem. We should not infer from the fact that the substitution of the parent in a single act-token makes no difference to distinctive value, that substituting the parent in all tokens of that act-type makes no difference. (That an unrelated adult provides dinner for a parent's children just once, on a particular day, may not make any difference, but substituting the parent with an unrelated adult at every dinner for the next five years, does.) The test should allow that result, so it should ask our intuitions to assess the difference in distinctive value that agent substitution makes across repeated instances of a particular act-token of general promotion.

Bearing that in mind, consider now some counterexamples to unrestricted general promotion:

  • (a)
    A parent who owns a large corporation secures a high-paid job for his son, ahead of more deserving candidates.
  • (b)
    A parent secretly alters a waiting list for a medical operation, bumping her child up the list, ahead of children with stronger needs for the operation.

Proponents of unrestricted general promotion can certainly condemn what the parents do in these examples. They can say that the stronger needs or interests of other children should lead us to condemn the acts through which the above parents help their children. Unrestricted general promotion would, nevertheless, compel us also to maintain that the parents' actions are distinctively valuable. We would have to accept that familial relationship goods can be realized through acts of nepotism and cheating and therefore, that parents and children forgo distinctive value when parents are prevented from performing these acts.

Now that isn't borne out by the agent-substitution test. Assume the agents promoting the children's interests in the above examples are entirely unrelated to the children and hold everything else equal. Intuitively, the acts in the counterexamples would not lose any distinctive value as a result of that agent substitution. Furthermore, nothing of value would be lost in substitution across repeated instances of those acts. Indeed, from the point of view of both the parents and the children, it would arguably be preferable if the benefactors in the above examples were total strangers, as opposed to the children's parents. (I would prefer it if it were a total stranger who wrongly bumped me up a medical waiting list than if it were my own father). Unrestricted general promotion, by contrast, commits us to the odd conclusion that the children should see the fact that it was their parents who performed the above actions as at least one good thing that can be said about those actions, whatever else might be said against them.11

It will be useful to note that the following reply is unavailable to Brighouse and Swift. This reply asserts that the category of ‘general promotion’ is implicitly restricted to interest-promoting acts that the family is well-placed to carry out and that, although the acts in the counterexamples do all involve parents promoting their children's interests, they are hardly interest-promoting acts the family is well-placed to carry out. (A medical operation isn't something the family is well-placed to carry out.) Implicitly restricting ‘general promotion’ in this way, would allow proponents of unrestricted general promotion to argue that the parental acts in the counterexamples are not acts of general promotion, properly understood, and that their lacking in special value, thus, does not undermine the claim that all such acts realize familial relationship goods.

The reason this reply is unavailable to Brighouse and Swift is that they regard any action by a parent to benefit her child as a proper manifestation of parental love (recall: ‘… a loving parent will be motivated quite generally to further her children's interests’) and they regard the proper manifestation of parental love as a familial relationship good.12 The acts in the counterexamples are both plainly acts by parents that further their children's interests. So, it seems that Brighouse and Swift are committed, after all, to categorizing those acts as acts that loving parents will be rightly motivated to perform for their children, and, hence, as acts that realize familial relationship goods.

Other Non-moralized Views

  1. Top of page
  2. Preliminaries
  3. Unrestricted General Promotion
  4. Other Non-moralized Views
  5. The Scanlonian Argument
  6. Conclusion

Let us now consider non-moralized views that deny that there is distinctive value in general promotion just insofar as it promotes the child's interests. Such non-moralized views still maintain that acts of parental partiality need not be otherwise morally permissible in order to be distinctively valuable. We can distinguish two such non-moralized alternatives to unrestricted general promotion. The first asserts that there is distinctive value in promoting one's child's interests only provided that other non-moral conditions are simultaneously fulfilled. The second asserts that there is no distinctive value in promoting one's child's interests because the only distinctive value that might be realized in parent-child relationships is the value of intimacy; intimacy itself, furthermore, is not moralized. I begin with the latter view.

Pure Intimacy

Is it true that the only special good that arises between parents and children is the special good of intimacy? I don't believe this view matches our intuitions about particular cases. Recall the father who works overtime so he can pay for expensive piano lessons for his daughter. The father has promoted his child's interests through actions that do not have anything to do with intimacy and yet, surely, we cannot thereby dismiss that there is special good in what he has done. If this is true, then at least some acts, other than acts that share intimacy, should be included within the class capable of realizing special goods for parents and children.13

General Promotion Through Sacrifice

What about the possibility that there is a special good in parental acts that promote the child's interests provided some other non-moral condition(s) is or are simultaneously fulfilled? Let us first consider the proposal that what a parent must also do is undergo some significant degree of sacrifice for his child. Sacrifice, I assume, is a non-moral condition: it occurs when a parent gives up something significant, or exposes himself to the risk of significant loss, for the sake of his child. It is plausible to hold that an act through which a parent promotes his child's interests—even to a great extent—is not distinctively valuable if it involves no cost for the parent. If a parent only had to press a button in order to bestow a significant good on his child, that act would not count as a proper manifestation of parental love, so one might say, or it would do so only to a very slight degree.

The question we must address is whether in order for there to be distinctive value in general promotion it is sufficient that general promotion promotes a child's interests through sacrifice. There are strong grounds to doubt that this is the case. Notice that one of the acts of general promotion in the above counterexamples does, indeed, involve sacrifice. The parent who risks being caught in the grave crime of tampering with a medical waiting list makes a sacrifice for his son who receives earlier treatment as a result (those who believe that ‘taking a risk’ doesn't amount to a sacrifice are invited to imagine that the parent's crime is detected after his son's medical operation and that he is imprisoned for the crime).

General Promotion in the Context of Intimacy

Perhaps the other necessary condition—in addition to promoting one's child interests—is that such an act of general promotion must be performed within the context of an intimate relationship in order to be distinctively valuable (I assume, again, that this condition is a non-moral one).14 A parent who promotes his child's interests from an emotional distance, that is, without otherwise showing the slightest interest in his child, might not be doing anything distinctively valuable through his act of general promotion. I think that is true. But notice that the counterexamples don't lose their force if we assume that the parents, in those examples, have close relationships with their children. The parent who tampers with the waiting list might be highly affectionate and emotionally engaged with his child, and yet we would still conclude that there is no distinctive value in what he has done. That counterexample undermines the non-moralized view that holds that there is distinctive value in acts that promote one's child's interests so long as they take place against the background of an intimate relationship.

General Promotion and Satiability

We might believe that there is a point at which parents can exhaust all the special good there is to be had in acts through which they promote their children's interests—or, in other words, that the amount of special good they can realize for their children and themselves is satiable. A good is satiable if there is a limit beyond which further increments in that good are, even in principle, impossible. One explanation for the satiability of general promotion might be this. It may be the case that what is especially valuable about general promotion inheres in the fact that it enables the parent to express to his child that he loves her.15 If, as one might presume, the parents in the counterexamples have already amply expressed their love for their children, there would be no distinctive value, so one might conclude, in the acts of general promotion they perform in the counterexamples.16

But that suggestion also fails as a way of escaping the counterexamples because even if we assumed that the parents involved in the counterexamples have not yet reached the point at which the distinctive value of general promotion is sated, we would still find what they do intuitively lacking in distinctive value. Suppose the parent who tampers with his son's waiting list has not, prior to that moment, done much for his child by way of promoting his interests, and that having recently altered his attitude to their relationship, he now wishes to make up the shortfall. Suppose, also, there is not much time left for them to be together, so that he does not have ample opportunity in the future to manifest his love for his child. We would believe, even in that case, presumably, that his act of general promotion in the counterexample lacks distinctive value.

All of the Above?

Finally, let's consider the possibility that general promotion is part of a jointly sufficient set of non-moral conditions for distinctive value that includes, beside general promotion itself, all of the above—i.e., sacrifice, intimacy-as-context, and non-sated parental love. We can imagine that all of those conditions hold when the parent pushes his child up the hospital waiting list and yet (I submit) it still doesn't make a difference. Even assuming all those conditions hold, there is no distinctive value in that act that would be lost were the parent substituted by some other agent unrelated to the child.

I believe the above non-moralized attempts at escaping the counterexamples fail because they ignore the fact that stubbornly tugs at our intuitions, namely that the parents in those counterexamples all act immorally. The procedure through which I have criticized the family of non-moralized views—by addressing its various members one by one—is admittedly hostage to the possibility that I have overlooked some further members of that family. But I believe the counterexamples we have considered cast doubt on the most promising members. If the moralized view can escape those counterexamples without incurring other problems, we may, therefore, be warranted in accepting it as the most plausible view of the distinctive value of general promotion.

The Scanlonian Argument

  1. Top of page
  2. Preliminaries
  3. Unrestricted General Promotion
  4. Other Non-moralized Views
  5. The Scanlonian Argument
  6. Conclusion

Let me now sketch an argument for why acts of general promotion are moralized. The argument builds on a discussion by Scanlon of the problem of how the ‘morality of right and wrong’, as he puts it, can have priority over the values at stake in our personal relationships, such as our relationships to friends, siblings, parents and children. Scanlon's response is that relationships of love and friendship are, in effect, moralized relationships: ‘no sacrifice of friendship is involved’, he writes, ‘when I refuse to violate the rights of strangers in order to help my friend. Compatibility with the demands of interpersonal morality is built into the value of friendship itself.’17 I want to develop an argument for the moralized view of parental partiality by clarifying and adding claims to Scanlon's argument for this moralized view of friendship.

The Argument

Scanlon's argument for why the value of friendship is moralized rests on two main claims. First, genuinely valuable friendship involves acting towards a friend in ways that are compatible with a recognition of her as a person with moral standing in her own right. A true friend does not attribute moral standing to his friend purely in virtue of his feelings of affection for her. Secondly, a person who benefits his friend by disrespecting the moral rights of others, thereby also expresses by that act an attitude toward his friend that fails to recognize her as having moral standing in her own right. Hence, so the argument concludes, nothing of distinctive value in friendship is lost when friends are constrained, in supporting each other, by the moral rights of others. Scanlon illustrates the argument with an example of someone who, in the name of his friendship, infringes the moral rights of another:

There would … be something unnerving about a ‘friend’ who would steal a kidney for you if you needed one. This is not just because you would feel guilty toward the person whose kidney was stolen, but because of what it implies about the ‘friend's’ view of your right to your own body parts: he wouldn't steal them, but that is only because he happens to like you.18

Scanlon's discussion is suggestive, but needs further elaboration if it is to provide a case for the moralized view of parental partiality. That is not because a gap needs to be bridged between the case of friendship, which Scanlon considers, and the case of parents and children considered here. Rather, Scanlon's argument needs elaboration even when applied to the case of friendship. This is so for two reasons. The first reason is that the explanation for why it is unsatisfactory that your friend should base your moral standing on the fact that he ‘happens to like you’ is incomplete. Is this only because he, thereby, gives your moral standing a contingent and, therefore, potentially transient status, thus leaving you vulnerable to a possible change of mind on his part (think about what your ‘friend’ in the above example might do to you if he stopped liking you)? This might be part of the story, but it isn't the full story, for, presumably, it would still be troubling if your friend should attribute moral standing to you purely on the basis that he happens to like you even if we assume he will like you forever. There must be some additional feature of his attitude that makes it unsatisfactory: what is it exactly?

A second question is this. It is not obvious why we should infer from the fact that your friend disrespects another person's moral rights, that your friend, therefore, bases your moral standing purely on the fact that he happens to like you. Your friend might allow himself to infringe other people's moral rights in order to benefit you because those other people lack, and you possess, some other quality that he believes is necessary for moral standing: they are not White, say, or Catholic, whereas you are.

I want to address these questions, and, in the process, broaden Scanlon's argument so that it shows more clearly, not only why friendship, but also why the relationship of parents and children, is moralized. To do so, I shall invoke a distinction once drawn by Gregory Vlastos between two bases for valuing a person. We might, on the one hand, value a person on the basis of her merit, where ‘merit’ refers to ‘all the kinds of valuable qualities or performances in respect of which persons may be graded’19 such as, for example, their wit, grace of manner, or technical skill. If we wish to grade or compare individuals, we must do so on the basis of some merit or other: we cannot grade or compare individuals as such, that is, apart from comparing their merits. The other way of valuing people values them not on the basis of their merit, but in abstraction from their merit. Vlastos refers to this distinct basis for valuing others as their individuality or their individual worth. (The term ‘individuality’ can easily mislead here, for it usually connotes the possession of some quality or set of qualities that differentiate a person from others, whereas what Vlastos wants the term to refer to is the value you have when you are considered in abstraction from your merits. It needs to be kept firmly in mind that I will stick to Vlastos' usage in what follows.)

According to Vlastos, valuing someone in abstraction from their merit is central to genuinely valuable relationships between parents and children. He writes,

Constancy of affection in the face of variations of merit is one of the surest tests of whether or not a parent does love a child. If he is fond of it only when it performs well, and turns coldly indifferent or hostile when its achievements slump, then his feeling for the child can scarcely be called love.20

Martha Nussbaum also equates valuing a child as an individual with the ideal of unconditional love: ‘the norm of unconditional love of children’, she writes, ‘may lead love to disregard the particularizing qualities of the individual, and this may be seen as a good feature of parental love.’21

The distinction between valuing individuals for their merit and valuing them for their individual worth helps us develop Scanlon's argument and apply it to the case of parents and children. Recall the first problem. Why exactly was your friend's basing your moral standing on the fact that he happens to like you unsatisfactory? The answer is that we want our friends and parents to respect and love us ‘unconditionally’, not in the sense that we want them to love us permanently, but rather, independently of our merit, that is, for our individual worth. If your friend or parent bases your moral standing purely on the fact that he happens to like you, however, this implies that he ‘respects’ you only conditionally upon your arousing certain pleasurable feelings in him.

Consider the second problem with Scanlon's argument—namely, that we should not necessarily infer from the fact that a friend benefits you by disrespecting others, that he bases your moral standing only on the fact that he happens to like you (he might instead base your moral standing on the fact that you, unlike the others, are White or Catholic). The response to this problem, we can now see, is that it doesn't undercut the fundamental claim in favor of the moralized view of loving relationships. It is true that your friend's disrespect for others may not necessarily show that he bases your moral standing purely on the fact that he happens to like you, but the relevant point is that it does show that he does not respect your individual worth. Why? In disrespecting others, your friend reveals that he does not recognize their individual worth. His act, thereby, expresses an attitude towards them that does not value them in abstraction from their merits. But then that act, in benefiting you, expresses an attitude that does not value you in abstraction from your merits either. After all, you are not, in abstraction of your merits, any different from the others.22 So even if your friend were to base your moral standing on your being White or Catholic, as opposed to your arousing feelings of affection in him, it is still the case that he fails to value you properly.

Two quick notes before I address some objections. First, note that the conclusion, here, is not that the friend in Scanlon's example does not value you. Plainly he does: in stealing a kidney for you, he has done something extraordinary to help you. Whatever else can be said about him, he surely recognizes your value in some sense. What we need to remember, however, is that to value someone for their individual worth is not just to value them in some sense: it is to value someone for the right reason. And your friend, though he undoubtedly values you, reveals by his act that he values you for the wrong reason, not for your individual worth, i.e., not for the worth you have independently of your merits, but for some quality he perceives in you—perhaps, as Scanlon suggests, the quality that you have of arousing feelings of affection in him.

Secondly, the Scanlonian argument does not aim to justify a conclusion about anything as comprehensive as your friend's or parent's character, or indeed about the nature of the entire relationship you share with your friend or parent. In particular, the argument does not aim to show that in seeking to benefit you through acts that are morally impermissible, your friend or parent, thereby, reveals himself to have a degraded character, or, thereby, reveals that the entire relationship he shares with you is worthless. The aim is to justify a conclusion only about the distinctive value of a particular act: in benefiting you by acting in a morally impermissible manner, your friend or parent has failed to act in a way that is distinctively valuable. That conclusion allows that his acting in that way is a mere lapse from an otherwise decent character on his part, or from an otherwise valuable relationship you share with him.

Four Objections

The claim that the Scanlonian argument ultimately rests on is that the basis of both our love of those who are close to us, and of our respect for others, is the same: their individual worth. That claim naturally gives rise to a first objection. The argument seems to imply, implausibly, that our emotional responses to, and our treatment of, those we love—our friends, our children—should be no different from our emotional responses to, and treatment of, any other member of the moral community. The answer is that this implication does not follow. We may reasonably respond differently to the individual worth of different persons. Consider an analogy: you can believe that all of the paintings at a particular art exhibit are valuable, while also being captivated and having a more complex reaction to one painting in particular. Explaining your different reaction to that painting needn't necessarily require you to say that that painting is more valuable than all the others, or, for that matter, that none of the others have any worth. That analogy is, admittedly, imperfect in this respect: while you might explain your different reaction to the particular painting you love by pointing to various of its differentiating qualities (e.g., its use of color), it would be inappropriate for you to point out the differentiating qualities of your child (e.g., that she is very cheerful, in order to explain why you love her and not others). (Although, note how that flaw in the analogy supports the underlying point—i.e., that one's love of one's child is not properly justified by appeal to her differentiating qualities). Still, the analogy does work in this respect: it shows us that the recognition of the same worth in each item within a class—paintings, persons—need not make it unreasonable for you to have different reactions to different members of that class. It is thus not necessarily inconsistent to believe that the individual worth of strangers should evoke in us a response of respect for them, whereas the individual worth of our friends and children should evoke the different response of love for them. What is inconsistent—and this is the only conclusion the Scanlonian argument seeks to uphold—is to believe that in acting in a way that disrespects others, one might, thereby, also be acting in a way that recognizes the individual worth of the person one loves.23

A second objection focuses on the fit between the Scanlonian argument and the moralized view of parental partiality. As currently stated, the moralized view says that acts of general promotion lack distinctive value if they are otherwise morally impermissible. The Scanlonian argument, however, shows something different, namely, that such acts lack distinctive value if they express disrespect for the intrinsic worth of others. The first italicized statement refers to a broader class of actions than the second. To see this, suppose a parent is non-culpably ignorant of the fact that his actions are morally impermissible, either because he lacks the capacity to examine the issues properly, or because they are genuinely difficult to understand. If so, although his actions are morally impermissible, they would not necessarily express an attitude of disrespect for the intrinsic worth of others. The Scanlonian argument, thus, justifies a narrower version of the moralized view of parental partiality than the one I have been discussing so far.

The relevance of this second objection should not be exaggerated. It is true that particular acts of general promotion that are morally impermissible do not necessarily express disrespect for the individual worth of others and, hence, are not necessarily lacking in distinctive value. But if we, as co-legislators of a just and well-ordered society, aspire to promote a public culture that helps us all to affirm the conception of justice under which we live, this second objection will become increasingly irrelevant. For our public culture will make the moral impermissibility of various acts of parental partiality increasingly understood among us, and it will thus be true of more and more instances of morally impermissible conduct that they do, indeed, express disrespect for others. The response that our public culture should keep citizens ignorant of the conception of justice they live under, thereby enabling them to perform morally impressible actions compatibly with showing respect for each other is, I assume, too perverse to be taken seriously.

Thirdly, one might worry that the Scanlonian argument sits awkwardly with the claim I made in Section I, that the moralized view is compatible with different moral theories. In particular, one might wonder whether the Scanlonian argument can uphold a consequentialist version of the moralized view. Note two points here. First, even if the Scanlonian argument could only uphold a contractualist version of the moralized view, that would not discredit my earlier claim that a consequentialist version of the moralized view is, in principle, available. Secondly, it is not evident that the Scanlonian argument cannot justify a consequentialist version of the moralized view. I only take the following ideas from Scanlon's discussion of friendship: (1) that a person who knowingly and avoidably violates the moral claims of others disrespects their individual worth; and (2) that this person thereby fails to properly value the person—his friend, his child—for whose benefit he disrespected those others. Arguably, one can endorse (1) and (2) while adhering to a consequentialist account of our moral claims.

I would like to consider a final objection: the Scanlonian argument may seem less plausible when it comes to less extreme forms of disrespect for others (as compared, say, to stealing someone's kidney). For example, suppose a friend acts in a mildly negligent way towards others—he fails to pay all of the taxes he legitimately owes the state—so that he can instead pay for a holiday he wants you to enjoy. You might well wish that he did not do that, but would you also worry that your friend does not respect the individual worth of others, and, furthermore, that he also holds some sort of sinister attitude towards you?

Let me make two points in reply. First, recall that the Scanlonian argument does not aim to justify a conclusion about your friend's character, or, in other words, his settled dispositions or attitudes. So it does not hold, as the objection implies it does, that a one-off instance of minor tax evasion on his part reveals a settled attitude of failing to recognize the individual worth of others, or, indeed, of you. The argument only aims to show something about the distinctive value of the particular act he has performed of benefiting you through minor tax evasion: it says that that act does not contain distinctive value as an act of friendship. Secondly, our intuitions in responding to this case might be motivated by the fact that we think that minor tax evasion might, in many ordinary circumstances, be rationalized as morally acceptable (‘there are many better off citizens who avoid paying a fair share of tax, so why shouldn't I do the same?’). But note that we are assuming a clean case in which your friend knowingly and avoidably fails to contribute a genuinely fair share of taxation. Once we keep that in mind, it seems more plausible to say that that act is disrespectful of others (though, of course, not as grossly disrespectful as stealing a kidney). You should, therefore, ask these questions: why has he done this for you? What does he see in you that could entitle you to such treatment from him? It cannot be your individual worth, for if it were, he would not have allowed himself to disrespect others in this way. It must be something else: his desire to feel good about helping someone, or the fact that he wants you to continue being his friend, or some other regrettable motivation of that kind. If this is just a lapse, it would be unfair of you to dismiss him or your relationship to him. But you (and he) should not regard this act as the distinctively valuable stuff of which friendship is made.


  1. Top of page
  2. Preliminaries
  3. Unrestricted General Promotion
  4. Other Non-moralized Views
  5. The Scanlonian Argument
  6. Conclusion

When policymakers must decide whether or not to legislate limits on individual conduct, they must consider the value of that conduct to all affected parties. Consider the act of purchasing private education for one's children. While that act secures the good for those children of a higher quality education, it may also, for a number of reasons, simultaneously set back the interests of other children who remain in the state sector.24 When weighing the goods and bads of parents' purchasing private education, should policymakers include as a further good in the equation, the good of general promotion itself? According to Brighouse and Swift, the answer to that question is ‘yes’. They argue, however, that the good of general promotion realized by purchasing private education for one's child is not great enough to justify frustrating other children's interests in fair equality of opportunity.25

In this article, I have argued that we should adopt a different response to that policy controversy (and others like it). We should not see the good of general promotion as primitively valuable, but rather as depending for its value on whether the act in question would be, otherwise, morally permissible. That response tells us that there is no distinctive value in purchasing private education for one's children.

How much the implications of the moralized view differ from a non-moralized view, such as the one adopted by Brighouse and Swift, will depend on how much primitive weight the non-moralized view assigns to the good of general promotion. Because Brighouse and Swift assign little weight to general promotion, there may be few policy disagreements between the moralized view and their view. However, there are two reasons why the moralized view is still of significance.

First, the moralized view allows us to uphold a more robust version of the distributive objection to inequality-causing acts of general promotion. If, contrary to the moralized view, it is conceded that there is some primitive value in general promotion, the debate shifts to how much, and it becomes harder to argue against those who see great value in it. Why exactly should we agree with Brighouse and Swift that there is only a small amount of primitive value in general promotion? Why not a lot? Why not enough to outweigh the interests other children have in fair equality of opportunity? The moralized view, in giving reasons for why there is no primitive value in acts of general promotion that are otherwise morally impermissible, allows a more definitive rejection of inequality-causing acts of general promotion.

Secondly, the moralized view tells parents that they must not believe they lose out on a special value in being prevented from promoting their children's interests in certain ways. That is important, for the law cannot reach into and regulate all aspects of our lives. Even an elaborate and intrusive legal system will leave areas of discretion where parents will have to decide the limits of their own partiality towards their children. In reassuring them—and they have a strong claim to be reassured—that the special value of their relationship with their children will remain undiminished when they voluntarily hold back from promoting their children's interests in certain ways, the moralized view helps an egalitarian ethos take root where the law cannot reach.

  1. 1

    A Theory of Justice, rev. ed. (Cambridge, MA: Harvard University Press, 1999), p. 28.

  2. 2

    The ‘distributive objection’ hails from Samuel Scheffler, Boundaries and Allegiances (Oxford: Oxford University Press, 2001), pp. 5664, 73–6.

  3. 3

    Harry Brighouse and Adam Swift, ‘Legitimate parental partiality’, Philosophy & Public Affairs, 37 (2009), 43–80, at pp. 5964. As will be evident, I am greatly indebted to Brighouse and Swift's discussion of general promotion.

  4. 4

    I draw the distinction between intimacy and general promotion from Brighouse and Swift, ‘Legitimate parental partiality’, pp. 5659. For a related discussion, see Colin Macleod, ‘Parental responsibilities in an unjust world’, Procreation and Parenthood, ed. D. Archard and D. Benatar (Oxford: Oxford University Press, 2010), pp. 128150 at pp. 141–2.

  5. 5

    Thus, while the distinctive value of acts of parental partiality is non-instrumental value, it is, nevertheless, extrinsic value (insofar as it depends on the relational property such acts have of taking place within a special relationship). For the distinction between instrumental and extrinsic value, see Christine Korsgaard, ‘Two distinctions in goodness’, Philosophical Review, 92 (1983), 169195.

  6. 6

    It might appear odd to hold that consequentialists can assign intrinsic value to acts, conditionally upon whether those acts comply with rules that produce the best consequences overall. But consider the following possibility. Consequentialists can endorse a theory of value according to which the items that have intrinsic value include acts performed from appropriate attitudes. If so, they can hold that it is intrinsically valuable when a parent acts from an attitude that values the promotion of his child's interests, conditionally upon those interests' being promoted consistently with rules that produce the best consequences overall. An example of a theory of value that ascribes intrinsic value to acting from appropriate attitudes is Thomas Hurka's. Central to Hurka's theory is the following ‘recursion clause’: ‘If x is intrinsically good, loving x (desiring, pursuing, or taking pleasure in x) for itself is also intrinsically good.’ See Virtue, Vice, and Value (New York: Oxford University Press, 2001), p. 13.

  7. 7

    For an account of filial duty that relies on special goods to which I am indebted, see Simon Keller, ‘Four theories of filial duty’, Philosophical Quarterly, 56 (2006), 254274, at pp. 264–74.

  8. 8

    I leave open the important question of how many instances of acts of general promotion, over what duration of time, are necessary for special goods of intimacy and trust to arise between parents and children. That question is relevant for determining the extent to which special goods can justify permissible parental partiality.

  9. 9

    Brighouse and Swift, ‘Legitimate parental partiality’, p. 60. Two points about this passage: first, although the phrase, ‘it looks as if …’, is often used to state a provisional view, Brighouse and Swift use it to state their considered view of the matter. Secondly, they define ‘familial relationship goods’ as goods that depend for their realization on a familial relationship's obtaining between the persons sharing those goods (see p. 46).

  10. 10

    I take it that Brighouse and Swift mean that an act of general promotion has distinctive value so long as it is an act in which a parent is rightly motivated as a loving parent, or, as one might also say, an act that is a proper manifestation of parental love. In other words, I assume they don't assign distinctive value to an act of parental partiality just as long as it was caused by a particular motivation (some parents motivated by love might do awful things to their children), but only insofar as it lives up to the value by which it was motivated (hence my use of the term proper manifestation of parental love).

  11. 11

    It is worth considering two questions that arise in virtue of the fact that a parent owes a ‘duty of care’ to her child. First, might a parent be required to perform the acts in the counter-examples by his duty of care to his child, and if so, might this not lead us to regard those acts as distinctively valuable after all? In reply, I believe it is difficult to conceive of the duty of care as anything other than an impartially justified duty, in which case the duty of care could not require the acts in the counter-examples, given their evident moral impermissibility. Secondly, what if a parent mistakenly believed that the acts in the counter-examples were required by his duty of care? Here, drawing on the Scanlonian argument I develop in Section IV, I would say that, assuming that the parent non-culpably mistook those acts to be a part of his duty of care, the parent does not express an attitude of disrespect towards others or her child in performing it. The act would, in that case, not necessarily be lacking in distinctive value. If the parent culpably mistook that act to be a part of his duty of care, on the other hand, it would lack distinctive value.

  12. 12

    Brighouse and Swift, ‘Legitimate parental partiality’, p. 60, my italics.

  13. 13

    What I say here is in tension with Sarah Stroud's claim that partiality should be permissible in order to facilitate shared agency. She writes: ‘we should focus more on the fact that I practice piano with my daughter every day than on the fact that I pay for her piano lessons’, in ‘Permissible partiality, projects, and plural agency’, Partiality and Impartiality, eds. B. Feltham and J. Cottingham (Oxford: Oxford University Press, 2010), pp. 131149, at p. 146 note 25. I think it is true that the former act of partiality in Stroud's example contains an added dimension of value compared to the latter act of partiality, but I don't think the latter act of partiality lacks distinctive value altogether.

  14. 14

    I draw this suggestion from Brighouse and Swift, ‘Legitimate parental partiality’, p. 63.

  15. 15

    This may be the view that Brighouse and Swift, ‘Legitimate parental partiality’, p. 61, endorse when they write that were parents prohibited from promoting their children's interests, ‘forms of feeling for particular others, including the willingness to put the well-being of a loved one before one's own, would be denied a valuable mode of expression’.

  16. 16

    In a work to which I am much indebted, Colin Macleod argues that intimacy between parent and child does not require parents to confer large amounts of resources on children. Macleod concludes that strongly limiting the permissible extent of general promotion incurs no loss in the integrity of the affective family. See Colin Macleod, ‘Liberal equality and the affective family’, The Moral and Political Status of Children, ed. D. Archard and C. MacLeod (Oxford: Oxford University Press, 2002), pp. 212230.

  17. 17

    T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), p. 165.

  18. 18

    Scanlon, What We Owe to Each Other, pp. 164165.

  19. 19

    Gregory Vlastos, ‘Justice and equality’, Equality: Selected Readings, ed. L. Pojman and R. Westmoreland (New York: Oxford University Press, 1997), pp. 120136, at p. 124.

  20. 20

    Vlastos, ‘Justice and equality’, p. 125.

  21. 21

    Martha Nussbaum, Sex and Social Justice (New York: Oxford University Press, 1999), p. 424 note 18.

  22. 22

    For a further elaboration and defense of this point to which I am indebted, see J. David VellemanLove as a moral emotion’, Ethics, 109 (1999), 338374.

  23. 23

    My argument in this paragraph is indebted to Velleman, ‘Love as a moral emotion’, pp. 366370.

  24. 24

    For a good discussion of the issue see Adam Swift, How Not to be a Hypocrite (London: Routledge, 2003).

  25. 25

    Swift and Brighouse, ‘Legitimate parental partiality’, p. 62.