Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/oxford-university-press/subjectivist-perfectionism-ZJezxbDHPZ
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2018. Published by Oxford University Press on behalf of University of Notre Dame. All rights reserved. For permissions, please email: journals.permissions@oup.com.
- ISSN
- 0065-8995
- eISSN
- 2049-6494
- DOI
- 10.1093/ajj/auy009
- Publisher site
- See Article on Publisher Site
Abstract
Abstract This paper critically discusses a neglected view in contemporary political theory. Subjectivist perfectionism combines a fully subjectivist understanding of the human good with the perfectionist claim that it is permissible for the state to promote the good, actively and intentionally, of its members. The paper clarifies the subjectivism in question and shows how it can inform state efforts to promote the good. The paper also explains how more traditional perfectionist views, which are informed by an objective account of the human good, can learn from it. The paper concludes by arguing that proponents of subjectivist perfectionism are particularly well positioned to respond to and rebut Michael Kramer’s central objection to mainstream perfectionist views in his provocative book Liberalism with Excellence. This is the charge that political perfectionists manifest an objectionable busybody mentality, one that is eschewed by his own favored form of justice-based perfectionism. Contemporary philosophers who defend perfectionism in politics are objectivists about the good life. Whatever else they think, they maintain that some ways of living are objectively better than others. Of course, as has been pointed out often, the terms “objective” and “subjective” are slippery. They are invoked in philosophical discussion to mark a range of different contrasts. Some regimentation of terms is therefore in order. Providing the needed regimentation, as we will see, brings into view some important issues. While one can be an objectivist about the good and reject all state efforts to promote the good, perfectionists standardly hold that it is permissible in principle for the state to promote, actively and intentionally, the good of its members.1 A view that is unexplored in contemporary political theory is one that rejects the first perfectionist commitment, but affirms the second. On this view, the good is fully subjective, but the state can and should promote it. This paper discusses this neglected view.2 Why discuss it? Some will think that this view, which I hereby dub “subjectivist perfectionism,” is rightly neglected. They may think that it combines the worst features of certain anti-perfectionist views, where skepticism about objective goods underpins the commitment to state neutrality,3 with what some consider to be “the quidnunc mentality”4 of those who want to enlist the state to promote the good. But I will try to show that even if the view should be rejected, we can learn from it. Although I am not a subjectivist about the good myself, the view is perennially interesting and has won the assent of some extremely able philosophers.5 Further, there is reason to think that any plausible objectivist view about the good must include a subjectivist component. Finally, as I will try to show, subjectivists can present a distinctive case for state perfectionism that has considerable force. Subjectivist state perfectionist measures are, in a manner I will explain, inherently non-alienating. For all of these reasons, the view merits attention. My discussion proceeds as follows. Section I clarifies the subjectivism to be discussed. It also explains why, and in what respect, a plausible objectivist view about the good will include a subjectivist component. Sections II and III discuss subjectivist state perfectionism. They propose some measures to promote the subjective good and discuss some of the issues these measures raise. The next two sections, IV and V, address an important objection to state perfectionism—one pressed vigorously by Matthew Kramer in his Liberalism and Excellence—and explain why subjectivist perfectionists are in a particularly good position to respond to it.6 I. Subjectivism in Focus Some subjectivists about the good are subjectivists about reasons for action generally. Their subjectivism about the good is an implication of this wider commitment. Just as any reason for action must be appropriately anchored in the agent’s motivational set, so too, and perhaps especially, must a reason that concerns an agent’s own good. Of course, some of these global subjectivists, as we might call them, may have arrived at their subjectivist commitments independently. That is, they may have first become persuaded of the truth of subjectivism about the good, and only later embraced subjectivism about reasons for action generally. Be this as it may, the view I wish to consider in this paper affirms only subjectivism about the good. Indeed, the subjectivist that I am interested in holds that there are objective reasons to promote the good, subjectively understood.7 It is these objective reasons that ground his commitment to state perfectionism. We need a sharper sense of the pivotal contrast between subjective and objective good. Subjectivists hold that a person’s good is importantly dependent on her attitudes and concerns. How should we construe this purported dependence? There are two proposals that I want to consider. The first proposal I will call the mapping view. It is well expressed in the following passage from Wayne Sumner. A subjective theory [of welfare] will map the polarity of welfare onto the polarity of attitudes, so that being well-off will depend (in some way or other) on having a favorable attitude toward one’s life (or some of its ingredients), while being badly off will require being unfavorably disposed toward it.8 Note, to start with, that it is not necessary to identify a person’s good with his welfare. Given Sumner’s concerns, it was appropriate that he focused on welfare. But, for our purposes, we can leave it an open question whether the best life for a person is the life in which his welfare is highest.9 At first pass, broadening the focus in this way appears helpful. For it seems like it enables us to sidestep, if not entirely avoid, a long-standing problem for the subjectivist about welfare. This is the problem of distinguishing the attitudes and concerns of a person that are intuitively relevant to her welfare from those that are not.10 But, on closer inspection, it is less helpful. For we are considering a subjectivist view of a person’s good, and if this does not just reduce to her welfare, then it must refer to some mix of welfare and other goods; and this, in turn, prompts the question of how to think about what constitutes the right mix. For an objectivist about the good, this question, while it may be difficult to answer, is at least well focused. He or she can think about what mix, or kinds of mixes, intuitively make sense. By contrast, for the subjectivist, the answer must be determined by the person’s own attitudes and concerns. Not just any attitude or concern is relevant. Parfit famously gives the example of a person who meets a stranger on a train and then forms a desire that the stranger fare well.11 If the person never meets the stranger, and if he plays no role in her life, then the satisfaction of this desire does not contribute to her good. One might respond that the attitudes and concerns that are relevant to a person’s good are those that are central to her evaluative outlook. Transient desires do not reflect what she really cares about. But this maneuver, assuming we have on hand an adequate account of what constitutes a person’s evaluative outlook, still allows for strange results. For instance, a person might conclude, on the basis of what he really cares most about, that the best life for him is a life in which he devotes all his energy to helping others and no energy to looking out for himself.12 We may suspect that this person is confusing the question of what is the most admirable or praiseworthy life with the question of what life is best for him. But no confusion may be present, and then we may have to accept that the person’s good lies in his selfless pursuits.13 It is a standing possibility on a subjectivist view of the good that it will leave us scratching our heads in this way. Just as, when we consider Rawls’s grass counter we find it very hard to believe that a person’s good could lie in this pointless activity, so too we will find hard to believe that the good for a person could lie in activities that fundamentally neglect what seem to be his own good. As Derek Parfit notes, in critical observation of Rawls’s subjectivist view of the good, the best life for a person, on this view, could be a life of unrelieved suffering.14 This standing possibility is not captured by the mapping view, however. For this view, although Sumner’s statement of it is a little unclear, seems to imply that there is only a necessary connection between one’s good and one’s attitudes and concerns.15 This leads Sumner to classify writers such as Joseph Raz and Ronald Dworkin as subjectivists. These writers present accounts of the good that contain a hybrid mixture of objective and subjective elements. Yet, as Sumner explains, “a theory can succeed in being objective only by assigning no essential role to subjectivity. It follows that a hybrid theory counts, overall, as subjective.”16 Both Raz and Dworkin stress that a person’s good depends on the success that he has in the projects and goals that he has adopted. Further, they both insist that for a person to lead a good life he must not be alienated from his goals and pursuits. This reference to non-alienation is the subjective element in their accounts of a person’s good. Yet the mapping view suggests a stronger dependence of good on attitudes and concerns than either of these writers would accept. For the mapping view, to recall, holds that just as being well off requires favorable attitudes towards one’s life and its ingredients, being badly off requires unfavorable attitudes toward them. However, on the hybrid view, a person can be badly off, even if she is quite satisfied with her life. Being badly off does not require, as Sumner’s statement of the mapping view claims, that one be unfavorably disposed toward one’s life. Let us amend the mapping view so that it asserts only the positive claim. On the amended view, living a good life requires that one have the appropriate attitudes towards one’s life and its projects, but one can fail to live a good life either by not having these attitudes or by failing to engage with objectively worthwhile pursuits. The amended view has considerable plausibility. It can avoid the wild examples that, or so I suggested, remain a standing possibility on subjectivist views. Nothing substantive turns on the classification issue, but I propose that the amended mapping view be understood as an objective account of the good life, one that incorporates a subjective component. I will come back to it shortly. For now, we need to consider a second proposal for characterizing the dependence of a person’s good on his attitudes and concerns. I will refer to it as the grounding view. The mapping view tethers a person’s good to his attitudes and concerns. It does not explain this dependence. This does not mean that the mapping view is inconsistent with an adequate explanation. It just means that it, on its own, does not provide an explanation. By contrast, the grounding view does so. The explanation takes this general form. Facts about what is good for a person just consist in facts about her attitudes and concerns and facts about what would further these attitudes and concerns. To be sure, few subjectivists hold that a person’s actual attitudes and concerns are authoritative in this way. The relevant attitudes and concerns are idealized. They are the attitudes and concerns that a person would have if she had full information and were free from cognitive error and irrationality.17 In this spirit, Rawls writes that the “best plan for an individual is the one that determines his real good,” where this best plan “is the one that he would adopt if he possessed full information” and committed no errors of reasoning.18 Now subjectivism about the good on the grounding view does not have to be expressed in the Rawlsian idiom of plans of life. But, for ease of exposition, I will do so in what follows. Notice that Rawls’s construction does not place restrictions on the content of the plan of life that would be chosen. Accordingly, it is an open possibility that a person would choose a plan of life that was not good for him, but was good in other ways. The person might self-consciously choose a plan of life that was good for his village, but not for himself, for example. Since our concern is what is good for a person, we need to amend Rawls’s formulation. Notice further that the best plan of life for an individual could be very different from the plan of life that he has in fact adopted. This introduces an important complication into the grounding view. I may find that I care very much about what will advance my actual plan of life, but do not care much at all about what would further the plan of life that an informed and rational version of myself would have adopted. Moreover, and importantly, there may be aspects of my present good that simply cannot be captured by the best plan of life for me. As James Griffin points out, even if my ideal self would enjoy caviar (and my ideal plan of life would include many opportunities to enjoy it), “you would do me no favor by giving me caviar now,” given my actual unrefined tastes.19 Call this the alienation problem. A standard way to address the alienation problem is to introduce an ideal advisor.20 This is someone who has full knowledge of one’s actual self and actual circumstances, and who seeks to advise one on how best to live. Working with this notion, we can reformulate Rawls’s construction of the best plan of life for an individual person. The best plan of life for a person is the plan of life that an ideal advisor would want the person to adopt if the advisor (a) were solely concerned with helping the person to lead the best life for herself (subjectively understood) and (b) were “to contemplate [the person’s] present situation from a standpoint fully and vividly informed about [the person] and her circumstances, and entirely free of cognitive error or lapses of instrumental rationality.”21 To return to Griffin’s example, an ideal advisor would not recommend giving me caviar now, but he might well advise that I undertake a course of action to cultivate my tastes so that I will come in time to enjoy it. The grounding view owes us an account of what exactly the grounding relation is. I will not say much about this. We have seen that it is an explanatory relation of a strong kind. On the Rawlsian view I have proposed, the fact that x is good for a person just consists in the fact that an ideal advisor would advise the person to adopt a plan of life that would include it.22 Importantly, the grounding relation does not imply reduction.23 The grounded fact (that x is good for a particular person) could be a normative fact, while the grounding fact (that an ideal advisor for this particular person would want him to adopt a plan of life that includes x) could be a fully naturalistic, nonnormative fact.24 Thus, the former fact might not be reducible to the latter, even if it was grounded by it. Naturally, given the tight explanatory relation it posits, the grounding view implies that there is a sufficient, and not merely necessary, connection between our good and our positive or favoring attitudes and concerns (appropriately idealized). The view thus shelters the standing possibility that it could yield intuitively wild examples, which I suggested above is a hallmark of subjectivist views of the good. This standing possibility, which might seem at first pass to be a weakness of this construal of subjectivism, is in fact a strength; for it shows that the subjectivist is serious about not introducing objective elements into his or her account of the good. In what follows, I assume that the grounding view best expresses the subjectivist claim that people’s good depends on their attitudes and concerns. Return now to the amended mapping view. This view can explain what Sumner calls “the perspectival character of prudential value.”25 Objectivist views about the human good that make no reference to our attitudes and concerns have trouble accounting for the distinction between a good human life and a good life for a human person. To be sure, some objectivists harbor skepticism toward the latter idea. The best account of a good human life, they argue, need not explain why such a life is good for the person who leads it.26 Objectivists need not embrace such skepticism, however. They can hold that there is a difference between a life that contains goodness and a life that is good for a person, that the latter notion is important to ethics and politics, and that the difference between the two notions must advert to the attitudes and concerns of the person in question. People who see no value in the activities and pursuits that give shape to their lives, or who are filled with self-hatred or doubt about the worth of their lives, do not lead lives that are good for them, even if they succeed in living lives that are good in other ways. Sumner is therefore correct to hold that a plausible view of what is good for a person must involve a subjectivist component. In caring about another person, we want her not only to engage in worthwhile activities, but also to be happy, where this means being content with, as opposed to being alienated from, her life and the course it has taken. It is possible, of course, that the skeptics are correct. The fundamental normative notion might be a good life, not a good life for a person. But our concern in this paper is with the latter idea. Working with it, we have reason to take subjectivism seriously, and to explore its relation to perfectionist state action, even if we do not in the end accept subjectivism about the good as explicated by the grounding view. II. Promoting the Subjectivist Good “Anyone with a particular conviction about the good for human beings,” Thomas Nagel once observed, “will naturally be inclined to get the power of the state behind it, not only for his own sake but out of concern for others.”27 This seems true generally, but does it hold true for subjectivists? Like objectivists about the good, they too can care about promoting the good of others. And, if subjectivists have this concern, it would seem natural for them to consider how state power could be used to further it. Still, unlike objective views, subjectivists do not have a substantive program, as it were. The good for people, on their view, is determined procedurally. It is a function of the plans of life that they would adopt, which in turn is a function of the attitudes and concerns that constitute their evaluative outlooks. Further—though, strictly speaking, this is an addition to subjectivism—the evaluative outlooks of people differ widely and often conflict. This has the consequence that if the state aims to promote Jack’s good it may set back Jane’s. None of this is surprising, of course. Let us now introduce a modest constraint on state action. If the state is going to promote the good of its members, then it should do so in a way that treats them with impartial concern. Perfectionists, on the standard view, hold that the state may favor sound over unsound conceptions of the good, and that doing so does not run afoul of the impartiality constraint. In favoring worthwhile pursuits over unworthy pursuits, the state serves the interests of all its members, since all its members have an interest in leading objectively good lives.28 But matters look very different on a subjectivist picture, for if subjectivism about the good is correct, then there are, in fact, no objectively better and worse conceptions of the good. Given subjectivism, the impartiality constraint would seem to rule out state action that promotes or favors any actual conception of the good, where an actual conception of the good consists of a schedule of substantive goods, activities and pursuits. In favoring an actual conception of the good, the state would seem to favor some of its members over others. If Jack likes pushpin and Jill likes poetry, and if there is nothing to be said in favor of one over the other, then when the state funds poetry and neglects pushpin, it looks like it is being objectionably partial to Jill. This appearance can be misleading, however. As I explain below, there are what I will call reasons of aggregation that can justify state promotion of actual conceptions of the good in a manner that satisfies the impartiality constraint. Before considering these reasons, I want to consider first a different proposal for state perfectionism, given the truth of subjectivism about the good. The proposal takes its cue from Rawls’s thin theory of the good. As is well known, the thin theory of the good picks out a set of goods that are presumptively valuable to people, whatever their actual conception of the good may be. In favoring these goods, or by ensuring their adequate provision, the state promotes the good of its members, but it purportedly does not take sides between the conceptions of the good adopted by its members. The set of goods picked out by the thin theory may not be perfectly neutral with respect to all conceptions of the good that find favor in the society in question. Early critics of Rawls seized on this point. They alleged that the thin theory was not impartial, but favored some ways of life over others. Let us put this worry to one side. The thin theory, after all, may be the most impartial, and the least nonneutral, account of the good on offer; and if the state needs to promote the good if it is to establish justice, then the thin theory may be impartially justified. There is a further point to make. Possibly, some conceptions of the good that are in conflict with the thin theory, would not be conceptions of the good that would survive exposure to full information and mistake-free reasoning. Regarding a person with such a conception of the good, an ideal advisor may recommend revision or abandonment of the conception. Let us distinguish a person’s adopted conception of the good from what we can call his guiding conception of the good. The guiding conception is the conception that an ideal advisor relies on to guide his advisee. The thin theory of the good could identify common elements, and/or all-purpose means, for promoting or advancing the good of people as expressed by their guiding, as opposed to their actual, conceptions of the good. State efforts to promote the good that were not impartial with respect to the former could be impartial with respect to the latter. And it is the latter, the guiding conception, that is determinative of a person’s good on the subjectivist view of the good that we are considering. Thus, by impartially promoting its members’ good, as expressed by their guiding conceptions, the state would satisfy the impartiality constraint. It is even possible that exposure to full information and correction for rational error would lead people to converge on one, or a small set, of conceptions of the good. More likely, such exposure and correction, would lead people to converge on a range of goods, even if they weighted these goods differently. These possibilities should not be overstated. For one thing, people’s guiding conceptions of the good must be responsive to their actual attitudes and concerns. This responsiveness is necessary to address the problem of alienation. This fact makes convergence on plans of life or on the value of goods less likely. For another thing, it is often uncertain what the best plan of life for a person would be. At best, state officials can only estimate what people would value and care about after exposure to full information and correction of rational error. Nevertheless, just as these possibilities should not be overstated, they should not be dismissed out of hand either. Consider, for example, a class of goods that I will call developmental goods. These are goods that facilitate one’s appreciation of other goods. People often will not know what plan of life is best for them. With good fortune, they may stumble upon it; but it is more likely that they will find it if they look for it and if they do so under favorable conditions. The state can adopt policies that facilitate this discovery. For instance, the state can adopt policies that foster critical reflection on plans of life, encourage experimentation, and try to ensure that all persons have access to a wide range of options. These autonomy-promoting policies would not be justified on the grounds that autonomy is an objective good, but instead on the grounds that autonomy is a developmental good that helps people to appreciate the items that figure in the plan of life that is best for them. Rawls himself called attention to the importance to people of discovering if not the best plan of life, at least plans of life that are satisfactory for them. Each person, he claimed, has a responsibility to make a rational effort to discover his good.29 Yet Rawls was insufficiently attentive to the fact that different political/institutional environments facilitate or set back people’s efforts in this regard differently. In virtue of this fact states can promote the good by sustaining or bringing about environments that enable their members to narrow the gap between their apparent and real good. Developmental goods are not necessarily neutral goods, even if they should be classified under the thin theory. If, as I have suggested, autonomy is a developmental good, then by promoting it the state will promote the good of its members. To be sure, on a subjectivist view of the good, it remains a possibility that the good of some will not be served by autonomy-promoting policies. If one’s best plan of life is inconsistent with autonomy, then state promotion of autonomy may not serve one’s good. The possibility that state efforts to promote the good will serve some, but not others, who are subject to the state’s authority raises an important issue, one to be addressed briefly below, about the fair balancing of the costs and benefits of state action. Nevertheless, even if one’s best plan of life consists in a (non)autonomous way of living, such as the life of a monk in a rigidly ordered monastery, one’s ideal advisor might still recommend critical reflection and exposure to a wide range of options for one’s actual self as necessary means for the discovery of these ideal pursuits. One’s guiding conception of the good includes the developmental goods necessary for one to appreciate one’s ideal pursuits. Consider next a further extension of the thin theory. In addition to identifying common elements or all-purpose means, one can identify common impediments to the pursuit of subjective good. These impediments include options that people would not engage with if they had full information and reasoned well, but do in fact engage with either out of weakness of will or because they mistakenly think that engagement with them furthers their real good. Richard Arneson observes: State policies recommended by subjectivist views are unlikely to satisfy neutrality of aim … This will be so if certain ways of life (for example, that centered on frequent consumption of smokeable methamphetamine) are reliably determined to be such that no person after careful reflection with full relevant information would embrace that way of life.30 Let us pause to examine this example in more detail. The claim to consider is the claim that a political/institutional environment in which the option to smoke methamphetamine is effectively foreclosed could be superior, and could be known to be superior, in terms of its impact on people’s efforts to lead good lives as judged by their ideal advisors when compared to a political/institutional environment that is the same in all respects31 except that this particular option is not foreclosed. How might this claim be challenged? First, a small point. Even if it were true that no person after careful reflection with full relevant information would embrace a way of life centered on frequent consumption of methamphetamine, it still could be true that some people would benefit from having the option to smoke this drug. With this in mind, the state might attempt to foreclose the option of engaging in a methamphetamine-centered way of life without foreclosing the option to smoke methamphetamine, but I will assume here, not implausibly, that it can only foreclose the former option by foreclosing the latter. This assumption is welcome in the present discussion, since it will help us to appreciate better some of the costs of state efforts to promote the good by closing off bad options. Consider Jack and Jill again. Jack can smoke methamphetamine on occasion without becoming addicted and without it posing a serious risk to his health. By contrast, if Jill has the option to smoke this drug, there is a high probability that she will abuse it and that she will fall into the drug-centered way of life that her ideal advisor would condemn. Critics of state perfectionism can point out that Jack can be harmed, or have his interests set back, by the elimination of this option in several different ways. The activity of smoking methamphetamine may contribute to Jack’s good. This will be true if his guiding conception of the good includes this activity. In addition, having the option to smoke this drug could contribute to his good, even if the activity of doing so does not contribute to his good.32 For having this option may (1) contribute to Jack’s autonomy, and Jack’s guiding conception of the good may consist in autonomous engagement with his ideal pursuits; or (2), and more simply, Jack’s guiding conception of the good may place a value on freedom as such, and having this option augments his freedom; or perhaps (3) Jack values, and his ideal advisor would value as well, a condition in which others trust him not to misuse his freedom and if the state were to foreclose this option it would evince distrust toward him; or possibly (4) Jack may benefit from this option being open to others, as he and others may learn from their experimentation with it.33 To be sure, Jack’s guiding conception of the good need not be self-centered. His ideal plan of life could include a solicitous regard for others in his community, including those like Jill who would suffer from the availability of this option. If so, then a concern for Jack’s good will need to take into account this countervailing consideration alongside the benefits just mentioned; and once this consideration is taken into account, the foreclosing of the option might well further both Jack and Jill’s good. Still, if the option to smoke methamphetamine ought to be foreclosed, then this will be primarily for Jill’s sake, not Jack’s. By stipulation, I have said that if the option to smoke this drug is left open, then the option to engage in the drug-centered way of life in which it figures will be left open; and this way of life is decidedly not in Jill’s interests. It does not follow strictly that it is not in Jill’s interests to have the option to engage in this way of life. For Jill, like Jack, can have the interests enumerated in (1) – (4), and these interests could outweigh the disvalue to her of having access to this option. But—given the fact that if Jill has access to the option to smoke methamphetamine, then there is a high probability that she will fall into the drug-centered way of life—it is very likely that having access to this option would not be in her interests. Indeed, when contemplating Jill’s situation, the interests enumerated in (1) – (4) do not look particularly compelling. Jill, in an autonomy-promoting society (which I have argued would be favored by subjectivist perfectionism) will have numerous options to engage with and decline to engage with. So, the loss of this particular option, as well as perhaps some others that would be bad for her to take up, if it sets back her autonomy at all, would not do so in any substantial way. The same can be said of any value that accrues to Jill in virtue of having the freedom as such to take up this option. It is possible that Jill’s ideal advisor might favor a situation in which the state trusts her not to misuse this option, even when the advisor knows she will misuse it. Jill may have a strong individualistic nature and may deeply resent the suggestion that others should not trust her to make decisions, and her advisor may need to take heed of this. Still, her advisor will likely judge that is rational for her, to paraphrase Rawls, “to protect herself against her own irrational tendencies by consenting to a scheme of penalties that may give her a sufficient motive to avoid foolish actions.”34 And if it is rational for her to so consent, then the foreclosure of the option in question will be congruent with her guiding conception of the good. Finally, while experimentation is valuable and will be encouraged by an autonomy-protecting society, the verdict on some options, such as the option to smoke methamphetamine may be in. This indeed is what is implied by the thought that the state can reliably determine that the option is not good for nearly all of its members. With these remarks, I may be guilty of belaboring the obvious. Critics of state perfectionism, however, allege that perfectionists have shown insufficient regard for the value of freedom. And, for that reason, I have thought it helpful to spell out and comment on a range of benefits that may accrue to persons from having access to options, including options to engage in activities that set back their good.35 On a subjectivist view of the good, the significance of these benefits will, of course, vary depending on the content of the guiding conceptions of the good of the people in question. Still, Arneson’s observation stands. It may be reliably determined that the vast majority of people are in Jill’s, not Jack’s position, with respect to this option. Were this to be reliably determined, would it be sufficient to justify state action that forecloses the bad option? Some will object that the state action in question wrongs Jack even if it benefits Jill. I consider this line of objection in the sequel. For now, consider the thought that the state action wrongs Jack by violating the impartiality constraint. In foreclosing the option to smoke methamphetamine, the state favors Jill’s good over Jack’s. This may be thought to be objectionably partial. But is it? Consider this memorable passage from Nozick. Why not … hold that some persons have to bear some costs that benefit other persons more, for the sake of the overall social good? But there is no social entity with a good that undergoes some sacrifice for its own good. There are only individual people, different individual people, with their own individual lives.36 A common reply to Nozick, and one that I rely on here, is that the social good can be viewed, not as an entity with its own good, but rather as a framework under which the different, and potentially conflicting, interests of persons are fairly balanced. Jack may be asked to bear a cost for the sake of Jill, but if this demand were not unfair (indeed if fairness required it), then Jack could not object that he is being treated in an objectionably partisan way. Reasons of aggregation can favor nonneutral state action, and when they do the impartiality constraint is not violated.37 This reply assumes that Jack does not have a right that would be violated by the perfectionist state action being contemplated. An adequate account of reasons of aggregation may need to appeal to background principles that fix the rights and entitlements of individuals.38 But, in exploring subjectivist perfectionism, we are presuming that individuals do not have a right to neutral treatment. The impartiality constraint that we are considering, accordingly, does not incorporate that right. The point pressed here, accordingly, is a modest one. The state can promote the good of its members impartially, even when its action intentionally favors the good of some of its members over others. This will be true when the competing interests of the affected parties are fairly balanced and no rights have been violated. III. Objective Goods and Subjective Attitudes I have been considering state action to promote the subjectivist good. Before moving on, I want to shift attention back to those objective views of the good that incorporate a subjectivist component. Naturally, on such views, state efforts to promote the good will need to facilitate engagement with objectively worthwhile goods and pursuits, but over and above these efforts, what role might the state play in furthering the subjective component of good lives? Recall that the subjective component adverts to people’s success in their adopted goals and pursuits and to their attitudes about their engagement with these goals and pursuits (as well as about the course of their lives as a whole). Call the first of these the success condition and the second the attitude condition. The proposal to be explored here is the idea that state action could have a role to play in helping people to do well with respect to both of these conditions. Regarding the success condition, it is not enough for people to settle on objectively worthwhile pursuits. If they are to live lives that are good for them, they must settle on pursuits that suit their nature and talents. They must discover the pursuits at which they can be successful. Accordingly, developmental goods that facilitate this discovery further their good. And, for this reason, the state can further the good of its members by sustaining or promoting these developmental goods. For example, and for the reasons mentioned earlier, an autonomy-supporting political environment may best enable people to hit upon the goods and activities that suit their nature. State support for development goods also can take a more targeted form. The success condition is met only if people engage with objectively worthwhile pursuits that suit them. So, to lead a good life, people need access to a set of options that include options that are both worthwhile and fitting, given their nature. Let us define a fitting good as one that is both worthwhile and suitable or enticing for some people. If opera is a fitting good, then it is both worthwhile and there at least some people for whom engagement with opera would suit them. Perhaps all objective goods are fitting goods, but fitting goods can differ in their robustness and vitality. Robustness is a function of the number of people for whom it is fitting. Vitality is a function of the percentage of people who, when exposed to the good, are motivated to engage with it.39 (A nonvital fitting good would be one that was worthwhile to engage with, and one that would suit people if they engaged with it, but leaves nearly everyone cold when exposed to it.) The state cannot support all fitting goods, but by supporting and subsidizing exposure to fitting goods that were relatively robust and relatively vital it could help its members to satisfy the success condition. Thus, exposure to robust and vital fitting goods is itself a developmental good, one that helps people come to appreciate the goods that would figure in plans of life that were good for them. It is possible, of course, that an unregulated cultural marketplace would be optimal in providing people with access to robust and vital fitting options. But there is reason for doubt on this score. The cultural marketplace answers to mass tastes, providing many options that not only lack worth, but do not even purport to be worthwhile. These options, in turn, are powerful competitors that draw many away from engagement with fitting goods.40 Turn now to the attitude condition. Several issues arise in understanding it. Suppose that a person, call her Kelly, has engaged with a range of objectively valuable pursuits—w, x, y and z. These pursuits, and her success in them, have informed her conception of the value of her life as a whole. We can ask about Kelly’s attitude toward each of these pursuits and we can ask about her attitude toward her life as a whole. Take pursuit w. Does Kelly believe that w is a worthwhile pursuit? Is her engagement with it whole-hearted, or half-hearted? Does it matter if Kelly does not believe that w is a worthwhile pursuit while she is engaged with it, but later comes to think that it was? Must her belief in the value of a pursuit be occurrent with her engagement with it for the attitude condition to be satisfied? A second cluster of questions comes on the heels of the first. If attitudes toward pursuits affect the value of engagement with the pursuits, how exactly do they do so? Dworkin writes: “my life cannot be better for me in virtue of some feature or component I think has no value.”41 On this view, if Kelly thinks that her engagement with w has no value, then it has no value. Her thinking it so makes it so. On a less austere view, appropriate attitudes would not be a condition on the value of the pursuit for the person, but instead a factor that augments the magnitude of its value for the person. Thus, if Kelly believes that her engagement with w has value, then its contribution to her good is higher than it would be if she did not have this belief. The same issue arises with the attitudes that pertain to wholehearted engagement. We also can ask about Kelly’s attitude toward her life as a whole. Possibly, Kelly believes that her life as a whole is a failure, even though she has succeeded in all her central pursuits—w, x, y and z. This belief, especially if it persisted over large stretches of her life, plausibly would make Kelly’s life less good for her. Raz writes: “The condition of whole-hearted engagement with one’s life is meant to exclude self-hatred, pathological self-doubt, and alienation from one’s life as they undermine well-being.”42 Kelly may harbor none of these negative attitudes toward her life as a whole. She may be content with it, but still think some pursuits she engaged in, say x, did not have value for her. Possibly, Kelly’s thinking this does not make it so, and her engagement with x was good for her, despite her failing to recognize this fact.43 I will not here try to answer all of these questions about the attitude condition. A plausible view, I believe, will hold that attitudes about one’s engagement with particular pursuits as well as attitudes about one’s life as a whole bear on one’s good. For present purposes, I do not need a precise statement of how, and to what degree, they do so. However, I do need to draw a distinction between what I will call value-conditioning and value-conferring attitudes. The former are attitudes that must be present for a person to engage with the activity or good in a way that is valuable. Some activities only have value if they are engaged with in the right spirit, as it were. The appropriate attitudes are necessary conditions of the value of the activity for a person. By contrast, value-conferring attitudes augment the value of an activity for the person engaging in it. They confer value on a person’s engagement with an activity over and above the value it would have for them in the absence of the attitudes. I want to ask about the quality of the value-conferring attitudes. Presumably, the attitudes are value-conferring only if they are, in some way that has to be spelled out, the genuine attitudes of the valuer. If Kelly believes z is valuable, but her belief is the product of objectionable manipulation, then this attitude is not value-enhancing for her. Should we say the same about uninformed beliefs? Suppose Kelly’s engagement with z is whole-hearted, but if she had a better understanding of its nature as well as a better understanding of the nature of alternative pursuits that were available to her, then she would cease to be attracted to it. Do her actual, poorly informed, attitudes toward z enhance the value to her of engaging with it? To answer these questions, we need an account of genuine or authentic value-conferring attitudes. Dworkin advances the following proposal: “We would not improve someone’s life, even though he endorsed the change we brought about, if the mechanisms we used to secure the change lessened his ability to consider the critical merits of the change in a reflective way.”44 This proposal suggests that the attitudes in question must be reflective, but Dworkin provides no insight into the nature of the requisite reflectiveness. The suggestion I broach here is that the notion of an ideal advisor is helpful on this score. On a fully subjectivist view of a person’s good, an ideal advisor looks to that person’s guiding conception of the good, which is responsive to the attitudes that the person would have from a standpoint fully and vividly informed about himself and his circumstances, and entirely free of cognitive error or lapses of instrumental rationality. On an objective view of a person’s good that includes a subjective component, an ideal advisor has a more complicated task. He must try to steer his advisee toward objectively valuable pursuits (and away from disvaluable ones) and, with respect to the available objectively valuable pursuits, he must help his advisee to discover those that suit her. These would be the objectively valuable pursuits that the person would desire to engage with if she were fully and vividly informed about herself and her circumstances, and was entirely free of cognitive error or lapses of instrumental rationality. On the suggestion I am broaching, value-conferring attitudes have two components. For an attitude to confer value on an activity it must be apt. If one engages in an objectively valuable activity, but one judges the activity to be worthless and one dislikes engagement with the activity for its own sake, then, whether or not the engagement with the activity makes one’s life go better for one, neither the judgment nor the dislike add additional value to one’s life.45 The value an attitude confers, however, is not simply a matter of its aptness. It is also a function of its genuineness or authenticity. Kelly might desire to engage in an activity, but not want to have this desire. Even if the desire is apt, her alienation from it detracts from its power to add value to her life. Likewise, Kelly might believe that it is good for her to engage with a certain activity, but would not believe this if she had more relevant information about the activity and her fitness for pursuing it. Here too the attitude’s power to add value to her life would be diminished. The construct of an ideal advisor captures and explains these judgments. Attitudes that result from irrationality or ignorance are inauthentic. They are not the attitudes that a person would have if she did not suffer from these defects. As a person’s actual desires correspond more closely with her ideal desires, they represent more accurately where she stands and what she values. As such, they plausibly confer more value on a person’s life than they would if they were less genuine. The suggestion here plainly needs more elaboration and defense than I have provided. The basic thought is that, if apt attitudes make our lives go better for us, then they do so in virtue of being our genuine attitudes, not attitudes that we have because of misunderstanding or mistake. And the further thought is that the subjectivist notion of an ideal advisor, once it is modified to account for objective goods, provides a promising way to conceive of or identify our genuine attitudes. If defensible, the suggestion reveals a further way by which the state could attempt to promote the good of its members by attending to the subjective component of good lives. Over and above the contribution state action can make to helping people satisfy the success condition, it can enhance their good by helping them to form genuine and apt attitudes toward the pursuits that play an important role in their lives. IV. Justice and Constraints The preceding section turned away from full-blooded subjectivism and put the spotlight on objective views of the good that incorporate a subjectivist component. Subjectivist perfectionism affirms the full-blooded view, however. As such, it would not countenance state efforts to promote the good that run counter to people’s guiding conceptions of the good.46 In this respect, subjective perfectionism may seem to have an advantage over its objectivist cousin. The perfectionist state, on the subjectivist view, never substitutes its judgment in place of the authentic or informed judgment of those over whom it rules concerning the values and goals their lives should exhibit. Subjectivist perfectionism is, in this key respect, inherently non-alienating.47 Nevertheless, even if subjectivist perfectionism is non-alienating in this way, it may still be objectionably overbearing. The subjectivist perfectionist state could be a meddlesome state, interfering in the lives of its subjects to an intolerable degree. That concern motivates the charge, recently advanced by Kramer, that perfectionists, including the subjective perfectionists we are now considering, manifest an unwholesome mentality—that of the busybody too eager to interfere in the lives of others. Is the charge fair? Perfectionist views can incorporate constraints that limit the state’s authority to promote the good, and these constraints may be sufficient to assuage the worry in question. Still, it is possible that there are sound constraints that rule out virtually all (edificatory) perfectionist state action. With this in mind, I consider, in this section, some candidate constraints. My discussion of these constraints is prompted by and framed around Kramer’s sundry critical remarks on the alleged quidnunc mentality of (edificatory) perfectionism.48 Start with a simple constraint that holds that the state should never promote the good of its members unless justice requires it to do so. Call this the justice constraint. At times Kramer appears to accept it. He writes that “insofar as perfectionism contends that governments are morally obligated and entitled to steer citizens toward improving their modes of living even when the avoidance or rectification of injustices is not at stake, it is a doctrine for quidnuncs”49 and he adds that “precisely because a distinctively edificatory-perfectionist measure goes beyond what is required of people by any applicable principles of justice, it constitutes overreaching on the part of the officials who have introduced and implemented it”50 and is therefore wrong. Further, Kramer insists that his own favored form of perfectionism—“aspirational perfectionism”—is a doctrine of justice and therefore is immune to the quidnunc objection.51 Now on some (edificatory) perfectionist views, the justice constraint will have no application. A view that maintains that justice requires the state to do all that it can to promote the good of its members makes all effective perfectionist state action a requirement of justice.52 But let us put such views to one side and consider more permissive perfectionist views that allow that some state action, including some state action to promote the good, is justice-neutral. Here the justice constraint constrains, for it excludes all perfectionist state action that is not required by justice. We should reject the justice constraint. It excludes state action that should not be excluded. To see this, consider what has come to be called “soft paternalism.” Soft paternalistic policies, such as licensing restrictions or imposed waiting periods, interfere with people’s self-regarding actions to promote their good, but they do so on the presumption that the wills of the interfered with parties are substantially nonvoluntary. Soft paternalism makes trouble for the justice constraint. If justice does not require the state to do all that it can to promote the good of its members, then it permits, but does not require, a range of state policies, including a range of soft paternalist policies. Plausibly, the state is permitted, even if it is not required by justice, to adopt such policies.53 And soft paternalist policies are justified in part on the grounds that they promote the good of those who are subject to them.54 The justice constraint thus implies that even traditional defenders of the harm to others principle, such as Mill, manifest a quidnunc mentality, as they often countenance a range of soft paternalist state policies not required by justice. This cannot be right. Mill may have been many things, but he decidedly was not a quidnunc. The harm to others principle excludes a wide range of state action that is designed to promote the good. It is also a principle for which Kramer expresses sympathy.55 With this in mind, let us factor it into the constraint that we are looking for, the constraint that reflects the worry that the perfectionist state is by its nature objectionably overbearing. Consider then the justice/harm constraint. This new constraint excludes hard paternalism, and other state efforts to promote the good against the grain of people’s wills, unless such state action is required by justice. The justice/harm constraint allows states to engage in soft paternalism and other soft measures to promote the good of their members, such as providing information to them about the nature of the options available to them—even when doing so is not a requirement of justice. Moving from the justice constraint to the justice/harm constraint has consequences for Kramer’s position. The latter, but not the former constraint, permits soft paternalistic interference that is not required by justice. It thereby fails to exclude all state action that Kramer characterizes as self-aggrandizing and self-abasing. Consistent with the justice/harm constraint, a scheme of governance could include laws and policies that prevail “upon people—coercively or enticingly—to abandon certain patterns of behavior that are not harmful to other people in any significant ways.”56 A pattern of behavior, such as dangerous mountain climbing or participation in risky research trials, could be one that is targeted by well-designed soft paternalistic measures. Further, and as a consequence, this scheme of governance would be dependent for its success on “the responsiveness of its citizens to its meddlesome ministrations”57 when no requirements of justice were at stake. Since Kramer appears to accept the legitimacy of soft paternalism (even when it is not required by justice), and since he wants to reject all self-aggrandizing and self-abasing state action, he confronts a choice. Either he must abandon his acceptance of soft paternalism and affirm the justice constraint, or modify his characterization of the kind of state action that is self-aggrandizing and self-abasing, and thereby redolent of the quidnunc mentality that he disparages. The latter course is the better option. Kramer, or someone sympathetic to his critique of edificatory perfectionism, can maintain that when the state prods and entices its citizens to act in ways that better reflect their own voluntary wills, it does not act as a busybody.58 Its action is neither self-aggrandizing nor self-abasing. But once it tries to “uplift” or “refine” the lives of those over whom it rules, once it tries to prod them to act against their own voluntary wills, then it manifests the quidnunc mentality – unless, once again, its doing so is required by justice. This indeed is likely Kramer’s considered view. What to make of it? Notice, to begin with, that the justice/harm constraint fits comfortably with the justificatory structure of subjectivist perfectionism. The subjective perfectionist, like the soft paternalist, is attuned to the very evident possibility that the actual decisions and choices of people can run counter to what they care about as revealed by their guiding conceptions of the good. And, like the soft paternalist, the subjectivist perfectionist will reject state efforts to impose conceptions of the good on its members that run counter to their evaluative outlooks. Yet, despite its attractions, the justice/harm constraint also should be rejected. It too excludes too much. Specifically, it excludes ‘modest paternalism,’ where this refers to interventions designed to help people avoid foolish or irrational actions, given what they care about or value. Modest paternalism is modest because it does not countenance interventions that run counter to a person’s settled values and preferences, but it goes beyond soft paternalism in permitting interference with a person’s voluntary self-regarding decisions. In a brief discussion of paternalism, Rawls affirms modest paternalism. “Paternalistic decisions [he writes] are to be guided by the individual’s own settled preferences and interests insofar as they are not irrational, or failing a knowledge of these, by the theory of primary goods.”59 At least some modest paternalistic policies are eminently reasonable, and these policies, or at least some of them, are plausibly not required by considerations of justice. Accordingly, they would run afoul of the justice/harm constraint. Would they thereby indict any system of governance that employed them as objectionably overbearing? Hardly. Modest paternalism, and related state measures, do not aim to impose a conception of the good on citizens. These measures instead aim to help citizens succeed in the activities and endeavors that they themselves genuinely care about and value. Let us consider a final candidate constraint, one that we can call the justice/ends constraint. This constraint excludes perfectionist state measures that cannot be justified by reference to how they satisfy and advance the settled concerns and interests of those they intend to benefit. This constraint will strike the perfectionist with an objective account of the good as unmotivated. But, for the subjectivist perfectionist, it fits neatly with his own understanding of the human good. If modest paternalism is consistent with the justice/ends constraint, then subjectivist perfectionist measures in general are as well. For these measures are also guided by “the individual’s own settled preferences and interests insofar as they are not irrational.” Indeed, subjectivist perfectionism can be viewed as an extension of the justifying idea expressed by modest paternalism. We have come full circle. Subjectivist perfectionism is nonalienating. But the worry was that even a nonalienating perfectionist state could be objectionably overbearing. We searched for a constraint that, if transgressed, would expose state perfectionism to the charge of being objectionably overbearing. From an analysis of Kramer’s animadversions on the alleged objectionable mindset of edificatory perfectionism, we arrived at the justice/ends constraint. However, this constraint does not add any limit to state meddlesomeness over and above that which already follows from the nonalienating credentials of the subjective perfectionist view. Thus the charge that the perfectionist state manifests an unwholesome quidnunc mentality appears to have no bite whatsoever when directed against the subjectivist perfectionist.60 V. Relationships and Restraint Perhaps we have been looking in the wrong place. Simply because perfectionist state action is not excluded by a constraint, it does not follow that it should be undertaken. Following Kramer, let us shift the focus from the subject to the state official. Rather than ask if subjects are treated wrongly because some constraint has been infringed, we ask whether state officials are corrupted or compromised by pursuing objectives that lie beyond their proper remit. An immediate problem arises. Don’t we need to know the constraints on state action in order to determine the remit of the state officials? Yes, but the suggestion here is that knowing the relevant constraints does not tell us everything we need to know on the matter. The point is not controversial. If a perfectionist measure were known to be ineffective, then it should not be undertaken, even if no constraint excludes it. But is there more to be said beyond this mundane observation? One line of thought, which I want to briefly explore here, attends to the relations between state officials and citizens in a modern society. The relations people have with others obviously affects the quality of their lives, enriching or impoverishing them. This is especially true of close personal relationships, but it is also true of more impersonal relationships. One’s life likely goes better if one does not have hostile or distrustful relations with one’s long term employer, for example. With this in mind, we can define the membership relation as that which obtains between state officials and the members of the state over which they govern. And we can say that the membership relation is a potential constituent of a good life for those to which it applies. It is a potential constituent for two reasons. First, and obviously, if it is of poor quality, then it may set back rather than augment the good in question; and, second, if a member has no significant interaction with, and no attitudes about, the state officials who rule over him, then, irrespective of its quality, it may have no bearing on his good. The suggestion to be explored here is that many members of a modern society have significant interaction, not necessarily personal interaction, with the state officials who govern them and that therefore proponents of state perfectionism must attend to the quality of the membership relation. In proposing measures to promote the good, they should consider, as one factor to be taken into account, how, if at all, the measures would affect the quality of the membership relation. And the further thought behind the present suggestion is that an overbearing and excessively meddlesome state could impair or degrade the relations between state officials and citizens. But what exactly would make the state overbearing and excessively meddlesome? One natural answer appeals to autonomy. If the state routinely invades the autonomy of its members, then it may be rightly judged to be officious and overbearing. Here we can say that the state action is bad for two reasons. It sets back the autonomy of its subjects, which is a perfectionist good (developmental or intrinsic), and it damages the membership relation, which is also a perfectionist good. This natural answer gives explanatory priority to autonomy, for it is the invasion of this good that primarily explains why the membership relation is damaged. Could the membership relation itself ground a reason for the state to show restraint in its pursuit of perfectionist objectives? Imagine an exemplary perfectionist state that respects the autonomy of its members and only undertakes perfectionist measures that can be competently executed. Could considerations of membership ground reasons of restraint for such a state? The answer is yes; for the membership relation is constituted in large measure by the attitudes of the involved parties. If citizens resent state action, even state action that is otherwise exemplary, then the state action could damage the membership relation. This answer, it must be said, has an air of paradox about it. Does thinking the state is overbearing actually make it so? Suppose many citizens of a modern state resent perfectionist state action because they have come to accept a false view about the proper role of the state. Shouldn’t we discount their resentment? We should, but that does not show that the resentment would not damage the membership relation. I am not suggesting that the avoidance of resentment be understood as constraint on perfectionist state action.61 Much state action, whether perfectionist or not, gives rise to resentment, whether reasonable or not. I am suggesting merely that it is a relevant consideration, one that could provide reasons for restraint on the part of state officials. Subjectivist perfectionists can accommodate all these points nicely. They can say that to the extent that the members of a state care, or would care if appropriately informed and rational, about the membership relation, it can ground reasons of restraint for state officials. Furthermore, they can say, ideal advisors must attend to the “incompletely informed and imperfectly rational” individuals whom they advise. This means that if the members of a state, in fact, resent state action, viewing it as overbearing and meddlesome, then this fact has normative significance, even if these same people would not resent the state action if they were more informed and less irrational. An ideal advisor, we have seen, must not only consider what his advisee would want if she were fully informed and rational, but also the changes that the advisee, in her present imperfect state, is capable of and the costs to her of undertaking them.62 The same considerations can be viewed from the opposite direction. If the members of a state, by and large, do not resent perfectionist state action, and would not resent it when appropriately informed and rational, then the state action would not damage the membership relation. Under these circumstances, the membership relation could not ground reasons of restraint for state officials. This, however, is intuitively the correct result. For if the perfectionist state respected the autonomy of its members, undertook no measures beyond its competence, and provoked no resentment on the part of its members, then the charge that it was nonetheless officious and overbearing would be hard to credit. Footnotes 1 In the main, when I speak of the good, I have in mind the human good, or what is good for people. On some perfectionist views the state can and should promote and protect impersonal goods as well as the personal good of its members. But in this paper I am concerned only with the state’s responsibility to promote good human lives. 2 An interesting foil to the present paper is Richard Arneson’s defense of subjectivism and neutrality in “Neutrality and Utility,” Canadian Journal of Philosophy 20 (1990): 215-40. Arneson has since rejected both subjectivism and neutrality. 3 For a clear example of such an anti-perfectionist view see Brian Barry, Justice as Impartiality (Oxford: Oxford University Press, 1995). 4 The phrase is from Matthew Kramer, Liberalism with Excellence (Oxford: Oxford University Press, 2017). 5 To name just a few: David Lewis, John Rawls, Harry Frankfurt and Peter Railton. 6 The objection in question is directed against what Kramer terms “edificatory perfectionism,” which contrasts with his favored view “aspirational perfectionism.” But subjectivist perfectionism, as I characterize it, is an instance of edificatory perfectionism. Throughout this paper, unless indicated otherwise, I use the term “perfectionism” to refer to “edificatory perfectionism.” 7 Parfit contends that this is the only kind of subjectivism about the good that is coherent. A subjectivist view of the good, he thinks, must appeal to objective reasons to care about it. Otherwise, it could not explain why anyone would have reason to advance his own good. See Derek Parfit, On What Matters, vol 1 (Oxford: Oxford University Press, 2011), 91-101. 8 Leonard Wayne Sumner, Welfare, Happiness and Ethics (Oxford: Oxford University Press, 1996), 38. (The passage is quoted by David Sobel in his critical review of Sumner’s book, a review from which I have profited. See David Sobel, From Valuing to Value [Oxford: Oxford University Press, 2016], 69-95.) 9 On the view I prefer, the best life for a person is the life (or one among a set of lives) that is most choiceworthy for him. Whether such a life is itself the life highest in welfare, the life highest in achievement, the life that is the most upright, or (more plausibly) a life that contains an appropriate mix of these different elements is an open question. A substantive account of choiceworthy lives aims to answer the question. 10 See, for example, James Griffin’s discussion of the problem in his Well-Being (Oxford: Oxford University Press, 1986), 21-23. 11 Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), 494. 12 Such a person, as I am imagining him, is not thinking that helping others is so personally rewarding that it compensates him for any loss in other goods. 13 I do not mean to imply that, on a subjectivist view, a person could not be mistaken about what will advance his own good. The point is that the subjectivist must allow that a person, such as the one here imagined, may not be mistaken in this way. 14 Parfit, On What Matters, 1: 104. 15 As Sumner makes plain on p. 38 of Welfare, Happiness, and Ethics. 16 Ibid., 54. 17 Irrationality in the procedural sense. 18 John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 366. 19 Griffin, Well-Being, 11. 20 Peter Railton, “Facts and Values,” reprinted in his Facts, Values and Norms (Cambridge: Cambridge University Press, 2003), introduced the ideal advisor idea into the subjectivist literature. It is developed by Connie S. Rosati, “Persons, Perspectives and Full Information Accounts of the Good,” Ethics 105 (1995): 296-325 and Sobel, From Valuing to Value. 21 Clause (b) draws on Railton’s formulation of the idea. 22 That is, either x is an activity or end that is part of the plan or x is a means to furthering such an activity or end. 23 See Kit Fine, “The Question of Realism,” Philosopher’s Imprint 1 (2001): 15. 24 Or a normative fact that was itself reducible to a nonnormative fact. 25 Sumner, Welfare, Happiness, and Ethics, 45. 26 See Thomas Hurka, Perfectionism (Oxford: Oxford University Press, 1993), 17-18. (As Hurka explains, one could say, on his view, that x is “good for” a person just means that x is “simply good” and that it is “a state of the person.” But this is not the common sense of the phrase.) See also Donald Regan, “Why am I my Brother’s Keeper” in Reason and Value: Themes from the Moral Philosophy of Joseph Raz, ed. R. J. Wallace, P. Pettit, S. Scheffler and M. Smith (Oxford: Oxford University Press, 2004). 27 Thomas Nagel, Equality and Partiality (Oxford: Oxford University Press, 1991), 154. 28 I am passing over a number of complications here. But, appropriately qualified, the statement in the text can stand. 29 Rawls, A Theory of Justice, 366-67. 30 Arneson, “Neutrality and Utility,” 221. 31 The “same in all respects” qualifier will strike some readers as fantastic. State action that forecloses this option would, under normal circumstances, have all sorts of other morally significant consequences. But we can imagine that the state has the requisite technology to target this option with precision and without generating untoward side effects. By so imagining, I signal agreement with Kramer (Liberalism with Excellence, 247) and Jonathan Quong, Liberalism without Perfection (Oxford: Oxford University Press, 2011), 59, and disagreement with Joseph Raz The Morality of Freedom (Oxford: Oxford University Press, 1986), 419-20, that fanciful thought experiments of this kind are legitimate aids for assessing state action. 32 As Kramer points out, and appropriately emphasizes, we should not run together the value of doing something with the value of being free to do it (Liberalism with Excellence, 200-201). 33 For instructive discussion of each of these purported benefits of having an option to do something, even when the option is not one for which it is valuable for the agent to engage with, see Kramer, Liberalism with Excellence, 200-207. 34 Rawls, A Theory of Justice, 249. 35 In doing so, I have followed Kramer’s excellent discussion in chapter 5 of Liberalism with Excellence. However, unlike Kramer, I have maintained that, even when the full value of freedom along all its dimensions is appropriately taken into account, (edificatory) perfectionist measures can be justified. I see no warrant for his strong suggestion that “the purpose of edification does not per se ever suffice to legitimize that loss of value” (233). 36 Robert Nozick, Anarchy, State and Utopia (New York: Basic Books, 1974), 32-33. 37 Some writers mistakenly identify fair aggregation with neutral treatment. See, for example, Alan Patten’s discussion of language rights in Equal Recognition (Princeton, NJ: Princeton University Press, 2014), 186-231. 38 See, for example, Thomas Scanlon’s account of aggregation in What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), 229-41. 39 For the idea of vital options see Richard Arneson, “Freedom and Desire,” Canadian Journal of Philosophy 15 (1985): 425-48. 40 Here I am paraphrasing a point pressed by Thomas Hurka, “Indirect Perfectionism,” Journal of Political Philosophy 3 (1995): 55. 41 Ronald Dworkin, Sovereign Virtue (Cambridge, MA: Harvard University Press, 2000), 268. 42 Joseph Raz, “Duties of Well-being,” in Ethics in the Public Domain (Oxford: Oxford University Press, 1994), 5. 43 See John Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980), 96-97. Finnis suggests that, at least for some goods, experiencing pleasure in them is an aspect of their reality as goods. Thus, failure to enjoy these goods would prevent one from fully participating in them. Nevertheless, one’s engagement with or participation in (objective) goods, even when “emotionally dry and subjectively unsatisfying,” remains good for one “as far as it goes.” 44 Dworkin, Sovereign Virtue, 218. 45 Aptness of a value-conferring attitude consists in having a positive attitude toward a positive good or a negative attitude toward a negative good. For discussion see Thomas Hurka, The Best Things in Life (Oxford: Oxford University Press, 2011), 119-39. I do not here discuss how, if at all, inapt value-conferring attitudes make one’s life go less well. 46 Unless, as explained earlier, such measures were necessary to fairly promote the subjectivist good of others. 47 Subjectivism perfectionism could be experienced as alienating in a certain way. A person’s view of what his ideal advisor would recommend for him could differ from the state’s view of this matter. But if the state’s view were correct, then its efforts to promote the person’s good would be oriented toward furthering what the person authentically cares about as opposed to the person’s own misunderstanding of what is important to him. 48 Kramer does not couch his discussion in terms of constraints. But he is unequivocal in his insistence that distinctively edificatory perfectionist measures are always and everywhere presumptively wrong. Liberalism with Excellence, 255. 49 Ibid., 296. 50 Ibid., 288. 51 Ibid., 341-43. 52 This is not quite accurate. Even on the demanding view of justice mentioned here, the state could still face a choice between adopting one of two perfectionist policies, where it could not adopt both and where either policy would be equally effective at promoting the good of its members and equally costly to implement. Here the state would be permitted, but not required, to adopt either policy, even if it were required to adopt one of them. 53 One reason why a permissible soft paternalist policy need not be required by justice is that its implementation could use up resources that the state permissibly could allocate for other purposes. 54 Two additional points are in order. First, you misread the soft paternalist if you think he is motivated only to protect people from making choices that they do not want to make. For the soft paternalist, it is wrong to interfere with people’s substantially nonvoluntary choices willy-nilly. (Joel Feinberg, Harm to Self [Oxford: Oxford University Press, 1986], 105 and 125.) Interference is warranted only when these self-regarding nonvoluntary choices would, or expectably would, set back their good. Second, I do not deny that Kramer accepts the moral legitimacy of soft paternalistic policies. (See Liberalism with Excellence, 233) What I claim is that acceptance of such policies contravenes the justice constraint, since soft paternalistic policies can be part of a legitimate scheme of governance designed to promote the good of its citizens, even when these policies are not required by any applicable principle of justice. 55 Kramer, Liberalism with Excellence, 275–77. 56 Ibid., 277. 57 Ibid., 278. 58 A state could engage in soft paternalism to an excessive degree. Perhaps such a state could be justly charged with being a busybody. The point is that soft paternalistic state measures as such do not betray a busybody mentality. 59 Rawls, A Theory of Justice, 249. 60 That is why I claimed at the beginning of this paper that subjectivist perfectionism is particularly well positioned to respond to it. To determine whether the charge has force when directed against objectivist versions of perfectionism would require taking a more critical look at the justice/ends constraint. Doing so, however, is unnecessary for present purposes. 61 Recall Kramer’s concern that perfectionist restrictions imposed by the state can express distrust of those subject to the restriction. (Liberalism with Excellence, 201) To the extent that this is true, and to the extent that such expressions of distrust generate resentment on the part of ordinary citizens, state perfectionist restrictions damage the membership relation. It should be noted, however, that these restrictions typically apply to classes of individuals. Each member of the class can judge that the restrictions do not express distrust of his decision-making abilities, but rather reflect an accurate assessment of the shortcomings of the decision-making abilities of the members of the class taken as a whole. Clearheaded awareness of this fact, when it is a fact, should abate feelings of resentment. 62 Railton, “Facts and Values,” 54. © The Author(s) 2018. Published by Oxford University Press on behalf of University of Notre Dame. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Journal
American Journal of Jurisprudence – Oxford University Press
Published: May 11, 2018