SEARCH

SEARCH BY CITATION

Liberal theorists of justice like John Rawls have long maintained that a theory of justice should apply primarily to the institutional mechanisms of society, and only derivatively to the behavior of individuals within institutions. Institutions of taxation, for example, may be just or unjust by the lights of a theory of justice, but such a theory should deem the behavior of individuals unjust only insofar as that behavior undermines just institutions. As Rawls puts it, “we are to comply with and to do our share in just institutions when they exist and apply to us, [and] we are to assist in the establishment of just arrangements when they do not exist.”1

Critics of this restricted conception of justice (hereafter RCJ) argue that a theory of justice should judge individual behavior directly, even when that behavior complies with just institutions. These critics have tended to focus on two kinds of behavior that they argue should fall within the subject matter of a theory of justice: the “market-maximizing” behavior of economic agents who demand incentives to exercise marketable talents in socially beneficial ways,2 and the “housework-shirking” behavior of family members who distribute power and labor unequally according to gender.3 These critics argue that RCJ implausibly places these behaviors beyond the reach of justice. Call this the “restrictiveness objection” to RCJ. A second objection to RCJ threatens to undermine RCJ from within: this criticism alleges that RCJ is arbitrary, because the theorists who embrace it lack a principled justification for restricting the subject matter of their theories to institutions while exempting the behavior of individuals within those institutions. Call this the “arbitrariness objection” to RCJ.4

My project in this article is to defend RCJ against both objections. Along the way, I consider and reject an alternative strategy for defending RCJ, but I use insights gleaned from the inadequacies of this rival strategy to build my own defense against the two objections: working from within the framework of political liberalism, I demonstrate first that a theory of justice can nonarbitrarily be restricted to the basic structure, or the institutional structure by which “the major social institutions distribute fundamental rights and duties and determine the division of advantages from social cooperation,”5 and second that such a restriction does not result in an implausibly narrow subject matter of justice. I conclude that neither objection undermines RCJ.

I do not defend RCJ as it has typically been understood, however. A crucial premise in my argument is that the delineation of the basic structure is itself a substantive normative task, the performance of which must be responsive to relevant differences among enactments of political power. I argue for a more expansive notion of legitimate political power than either critics or defenders of RCJ have tended to adopt.6 My defense of RCJ thus occupies a conceptual middle ground within the debate about the subject matter of justice: With defenders of RCJ, I maintain that a theory of justice applies directly only to the basic structure of society, such that a society with just institutions may be fully just even though housework-shirking and market-maximizing occur within it. But I agree with critics of RCJ that market-maximizing and housework-shirking should not be beyond the reach of a theory of justice. I reconcile these convictions by defending a view of political legitimacy according to which housework-shirking and market-maximizing can be targets of legitimate political interventions. While a society is not made less just by the mere occurrence of housework-shirking and market-maximizing, it can be less just for having a basic structure that enables or encourages these behaviors.

Arbitrariness, restrictiveness, and the basic structure

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

John Rawls, the best-known defender of RCJ, argues that the basic structure of society is “the primary subject of justice”:7 principles of justice “must not be confused with the principles which apply to individuals and their actions in particular circumstances” since “the two kinds of principles apply to different subjects and must be discussed separately.”8

It is not altogether clear what Rawls intends the basic structure to include. Generally, he includes just those institutions that are part of the system through which the state coercively limits the freedom of individuals.9 But when discussing his rationale for the basic structure restriction, he commits himself to what some have argued is a more expansive notion of the basic structure, which includes any major social institution that profoundly influences citizens' life prospects. Rawls claims, for example, that the “basic structure is the primary subject of justice because its effects are so profound and present from the start.”10

According to G. A. Cohen, RCJ is in tension with Rawls's stated justification for it:11 if “profundity of impact” qualifies an institution for inclusion in the basic structure, then the basic structure should include all social mechanisms that exert a profound impact on the life prospects of citizens. As Cohen points out, noninstitutional mechanisms can exert just as profound an impact on citizens as political institutions. But because these noninstitutional mechanisms are partly constituted by individual behaviors that cumulatively shape social norms, they are excluded from the subject matter of justice on RCJ.12 As an illustration, consider the informal social norms that sustain the heterosexual family as the dominant model for domestic life. Norms valorizing heterosexuality profoundly influence citizens by shaping their beliefs, the values they come to affirm, and the people they aspire to be. These norms operate through the choices and behaviors of individual citizens, but they can be just as profound in their impact as more formal political institutions. RCJ must be rejected, according to Cohen, because it restricts the purview of justice in a way that is arbitrary, given Rawls's stated rationale for it.13

While the arbitrariness objection maintains that RCJ lacks a principled justification, the restrictiveness objection alleges that RCJ is implausible for its exclusion of individual behavior from the primary purview of justice. Proponents of this objection have focused on two kinds of behavior that they claim should be part of the subject matter of justice: the behavior of “market-maximizers” who demand material incentives to exercise their talents in socially beneficial ways, and the behavior of “housework-shirkers” who allocate work and power unequally according to gender. First consider the market-maximizers. According to Cohen, a restricted theory of justice does not judge the behavior of individuals within the basic structure. Rather, a distribution of income and wealth is just so long as distributive institutions comply with principles of justice. And a just distribution remains just irrespective of the (legally permissible) behavior of individuals within those institutions. But because political institutions are blunt instruments, just institutions leave considerable space for individuals to (legally) influence the distribution of income and wealth. Assume egalitarian principles of justice. Working within just (egalitarian) institutions, talented marketeers can demand incentives that enhance their earnings, thereby disrupting equality.14 If justice judges only institutions, then a society with just institutions whose talented demand such incentives will be no less just than a more egalitarian society whose talented do not demand the incentives. But this is implausible, according to proponents of the restrictiveness objection. If individuals can help or hinder the goals of justice, then their behavior should not be beyond the reach of principles of justice.15

Now consider the housework-shirkers. According to Cohen, RCJ renders justice blind to norms that reinforce unequal sharing of paid and unpaid work and power among male and female domestic partners. Because those norms are constituted and perpetuated by the behavior of individuals within families, and because RCJ allegedly locates those behaviors beyond the legitimate reach of a theory of justice, the norms themselves appear to be beyond the reach of any theory of justice that upholds RCJ. The disadvantageous effects of strong gender norms are well documented, as are the disadvantageous effects of particular enactments of them in particular families.16 According to critics, RCJ is implausible because it renders principles of justice insufficiently attentive to these disadvantages.17

Of course, whether RCJ in fact excludes these behaviors from the reach of justice depends on what is included within the basic structure. Enactments of political power can take many different forms. The state exercises political power when it physically forces a citizen to adopt a preferred course of action, but it also exercises political power in obstructing alternative courses of action that would otherwise be available. More gently still, the state exercises political power by reducing the costs of a preferred course of action or raising the costs of nonpreferred courses, by manipulating the incentives that attach to various options in order to change the context against which citizens make choices. By charging deposits on glass bottles, the state raises the cost of throwing them out. By offering benefits like tax incentives to married couples, the state encourages the formation of stable, legally recognized partnerships.

Similarly, the state could exercise political power to discourage market-maximizing and housework-shirking. It could impose an educational curriculum that instills an egalitarian ethos,18 or provide high-quality subsidized child care to encourage a more equal sharing of domestic labor between genders.19 These interventions may turn out to be illegitimate, but this is a substantive philosophical question. In rejecting RCJ, Cohen treats the delineation of the basic structure as if it were settled pretheoretically. But the extension of the basic structure is a normative matter, and RCJ restricts justice to what it rightfully includes. As such, an assessment of RCJ must await an account of legitimate enactments of political power. Only with such an account in hand can we determine the reach of justice as determined by RCJ, and only then can we determine whether that reach is sufficiently expansive.

One of Cohen's own examples illuminates this point: Cohen asks us to imagine a sick child who desperately needs medical treatment. Because of their religious beliefs, the child's parents prefer that she not be treated. Cohen argues as follows: because the behavior of the parents clearly has a profound impact on the child, the case shows that RCJ excludes from the purview of justice some aspects of society that profoundly impact citizens' lives. But notice that this case makes Cohen's point only if the parents' behavior really falls beyond the reach of legitimate exercises of political power. We can imagine societies that recognize a legally enforceable right of children to life-sustaining medical care, regardless of the wishes of the parents. To determine whether the parents' behavior falls within the reach of justice on RCJ, then, we must ask: is it legitimate for the state to exercise power to induce the parents to accept treatment on behalf of their child? If so, then political institutions can be designed to provide that inducement, and an adequate theory of justice will judge society unjust insofar as its institutions fail to do so. Depending on the extent of legitimate political power, RCJ need not render justice blind to the behavior of individuals within society's institutions.20

Cohen's assumption that the parents' behavior is excluded from the restricted purview of justice is understandable. These “personal” interactions within families are thought to exemplify the kind of behavior that RCJ implausibly excludes from the reach of justice. Presently, I will defend a much more expansive account of legitimate political power, according to which institutions can legitimately use certain types of political power to target certain behaviors of individuals typically thought to be beyond the reach of justice. I argue, further, that under some circumstances institutions cannot legitimately abstain from doing so. In these circumstances, RCJ will deem social institutions unjust insofar as they fail to intervene to shape individual behavior in the relevant ways. Once we have on hand this account of legitimate political power, we will see that RCJ is neither arbitrary nor implausibly narrow. Before developing my account of legitimate political power, however, I want to consider a rival strategy for defusing the arbitrariness objection. I think the strategy is unsuccessful, but its difficulties are enlightening.

Does political power face a special justificatory burden?

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

Recent attempts to defend RCJ against the arbitrariness objection have alleged that enactments of political power face a “special justificatory burden”:21 in liberal societies, exercises of political power are purportedly sanctioned by citizens as a collective body,22 and those citizens are generally unable to withdraw from the jurisdiction of those institutions; thus, exercises of political power should be acceptable to all free and equal citizens. Those who pursue the special justificatory burden strategy argue that an application of principles of justice to the basic structure provides a way to meet this burden. Because the application of principles of justice to political institutions is called for by the justificatory burden those institutions face, RCJ is not arbitrary.

Political power does indeed face a justificatory burden. Still, this defense of RCJ is unsatisfactory. In order to justify a restriction of the subject matter of justice, we need to establish not just that exercises of political power face a justificatory burden, but that exercises of political power face a justificatory burden that does not apply to omissions of political power. If—as I shall argue—some omissions face the same justificatory burden that exercises of political power face, then we are left without a principled justification for excluding those individual behaviors that the basic structure leaves unregulated. Because the justificatory burden is not unique to enactments of political power relative to omissions, it will not serve as a principled justification for restricting the purview of justice to the basic structure of society.

Political power is thought to face a special justificatory burden because its exercises are legitimate only insofar as they are acceptable to free and equal citizens. Citizens are characterized as possessing two moral powers that underwrite their free and equal citizenship: the capacity for a conception of the good, and the capacity for a sense of justice. They are also characterized as having a higher-order interest in preserving those capacities.23 To protect that higher-order interest, free and equal citizens not only will accept but will in fact insist upon those exercises of political power that are necessary for the preservation of their two moral powers. If citizens' basic liberties are under threat, for example, then their capacity to form and rationally to pursue a conception of the good is jeopardized. Citizens possessed of a higher-order interest in protecting that capacity would call upon their government to enact interventions to protect their basic liberties against infringement, and to preserve social conditions conducive to the development of that capacity in the first place.

In short: political power may be a necessary means of amending social circumstances that undermine the development of free and equal citizenship. In these cases, abstaining from exercising political power is unacceptable from the perspective of free and equal citizenship. In a liberal state, then, omissions and exercises of political power face the same justificatory burden: they are illegitimate insofar as they are unacceptable to free and equal citizens.

A second condition thought to generate the special justificatory burden for political interventions is the difficulty of amending political institutions or exempting oneself from their jurisdiction. We cannot easily opt out of our society's taxation scheme, for example, or effect meaningful changes to that scheme through our individual actions. But exercises of political power are not unique in their inescapability. Coercive enactments of nonpolitical (hereafter “private”) power can be just as inescapable. Parents, for example, can exert a power over their children that is arguably less escapable and less amenable to change than the political power of the state over its citizens. Moreover, certain exercises of private power are possible only when the state abstains from exercising political power. The parents of Cohen's sick child exercise private power in refusing treatment on the child's behalf, and the child can no more escape that power than can she escape the political power of the state. But the parents' exercise of private power is possible only if the state omits a particular exercise of political power: the enactment of laws that demand life-sustaining medical treatment on behalf of sick children.

Insofar as an exercise of private power is inescapable and omissions of political power enable that exercise, the omissions inherit the moral import of inescapability attributed to exercises of political power more generally. There may well be a presumption in favor of political omissions, but this presumption cannot be grounded in the deep inescapability of political power. Because private power often also has this feature, and because private power is possible only because the state omits exercises of political power that would otherwise disrupt it, exercises of political power are not categorically less avoidable than omissions. The justificatory burden strategy does not offer a principled justification for RCJ because it fails to identify a feature of political power that is unique relative to omissions of political power.

More generally, the justificatory burden strategy fails because it is insufficiently attentive to the substantively normative project of delineating the basic structure. The political conception of the person as citizen that animates the problem of justice in the first place imposes not only negative constraints on how far the basic structure can legitimately extend, but also positive requirements on how far it must extend: it must be expansive enough to protect the development of the kind of moral personality that animates liberal conceptions of justice in the first place. Some enactments of political power are positively required, and their omission is illegitimate. Moreover, the legitimacy of enactments and omissions is grounded in the same considerations: the needs and interests of free and equal citizens. Because the same considerations that legitimize the enactments also legitimize the omissions, there is no justificatory burden that categorically distinguishes between the two and renders RCJ nonarbitrary.

In the remainder of this article, I execute an alternative strategy for vindicating RCJ. I begin by developing an account of the legitimate extension of the basic structure. I argue that the considerations that determine the legitimate extension of the basic structure also justify restricting the subject matter of justice to it: the principle for delineating the basic structure also justifies its role as the primary subject of justice. After defending RCJ against the arbitrariness objection, I turn finally to the task of defending it against the restrictiveness objection.

Political liberalism and the basic structure

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

My defense of RCJ utilizes the theoretical resources of political liberalism. We can understand political liberalism as occupying a range on one end of a spectrum of liberalisms, with comprehensive liberalism occupying a range on the other end. Comprehensive liberal theories of justice are based on comprehensive liberal conceptions of the good: conceptions that affirm liberal values like equality, autonomy, and liberty as substantive values that can legitimately and straightforwardly guide the organization of political institutions and justify political action. Comprehensive liberals maintain that liberty, autonomy, and equality are good for people, and that their goodness justifies their role in legitimizing political action.

Political liberalism, in contrast, maintains that political institutions and the principles on which they are built must be justifiable independently of any particular comprehensive conception of the good, including any comprehensive liberal conception.24 Because citizens living under free, democratic institutions will inevitably come to affirm different and incompatible conceptions of the good, any political system based on one particular conception will lose the support of those who affirm incompatible conceptions. A political system based on the substantive liberal values of autonomy and equality, for example, will fail to retain the reasoned allegiance of religious fundamentalists who reject those values as fundamental constituents of a good life. In order to ensure that society is stable for the right reasons, then, political liberalism dispenses with comprehensive value judgments, including the liberal notion that equality and autonomy are substantively good for people.25

Two theoretical commitments of political liberalism are especially relevant for our purposes. First, because the stable society is one to which citizens can give their reasoned allegiance, exercises and omissions of political power must be capable of gaining acceptance by all reasonable citizens. This commitment is spelled out in Rawls's statement of the “liberal principle of legitimacy” (hereafter, LPL): “our exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in light of principles and ideals acceptable to their common human reason.”26 To meet this burden, political interventions must not depend for their justification on value judgments peculiar to particular conceptions of the good.

Crucially, the acceptability to citizens of any particular exercise of political power is a normative matter. For an exercise of political power to be approved by LPL, it need not be acceptable to actual individuals taken as they are. Instead, it must be one that hypothetical citizens would find acceptable given a particular characterization of citizenship. Within political liberalism, citizens are characterized as reasonable and rational, free and equal, and with a higher-order interest in preserving the capacities of citizenship.27 Because not all actual citizens are in fact so constituted, determining whether exercises of political power are “acceptable to citizens” is not an empirical matter, but a substantive philosophical project.

The second theoretical commitment of political liberalism is the restricted scope of justice. Within political liberalism, the full realization of justice must leave ample space for citizens to affirm and practice nonliberal conceptions of the good in their personal lives. Though liberal values like autonomy and equality continue to play a role, the scope of these values is limited. For example, political liberalisms must respect the right of religious fundamentalists to reject such liberal values as autonomy within their personal and spiritual lives. It is by limiting the scope of justice to the basic structure that social institutions can hope to maintain the reasoned allegiance of citizens over time, whatever values and ideals those citizens embrace in their personal lives.28

Notice that these two theoretical commitments of political liberalism—the first concerning the justification of political interventions, the second concerning the scope of those interventions—are related in an important way. The justificatory commitment articulated by LPL embodies the “fundamental normative ideal” of political liberalism.29 The restricted scope commitment follows from the justificatory commitment. Political liberalism restricts the scope of political interventions and protects citizens' prerogative to enact illiberal values precisely because political liberalism's justificatory commitment—embodied by LPL—generates de facto restrictions on the legitimate reach of political power. Exercises of political power must be acceptable from the perspective of citizenship. Because citizens value their capacity for a conception of the good, they will impose a stronger burden for the justification of political interventions within areas of their lives in which the enactment of their comprehensive values—including illiberal values—is particularly important. The result is a de facto delineation of certain areas of life that tend to be beyond the reach of political intervention: certain types of intrusion tend to be proscribed by LPL, because they tend to be unacceptable by the lights of citizenship. Accordingly, behavior within these domains tends to fall outside the basic structure.

The restricted scope commitment, in other words, is not a rigid designation of certain domains of life as immune to political intrusion, but a consequence of individual verdicts rendered by LPL. Cumulatively, these verdicts generate a presumption against certain types of political intrusion. But political liberalism's delineation of the basic structure—its restricted scope—is conceptually secondary to its commitment to exercising political power only when that exercise is acceptable from the perspective of citizenship. As such, the presumption against political power in certain areas of life is overridden when the consideration which generates that presumption—LPL—approves an intrusion into that area of life.

This approval occurs when political values are implicated in defeasibly protected domains. When political values are at stake, the justification for intrusion need not rest on substantive values peculiar to particular conceptions of the good, because a legitimate, political justification is available. Suppose, as many believe, that behaviors of individuals within families enjoy a presumption against political intrusion. This presumption is generated by LPL: citizens value their capacity for a conception of the good; domestic life is a particularly important setting for the enactment of that capacity; thus, citizens will demand considerable protection against political intrusion into the home. As a matter of fact, then, LPL will deem many such intrusions illegitimate. But the presumption generated by these cumulative verdicts is overcome when a particular intrusion is approved by LPL on the basis of political values. Consider a case of domestic abuse. Because citizens are committed to the protection of the basic liberties, including the right to physical security, citizens would accept (indeed, demand) political intrusions to protect victims of abusive domestic partners. Those intrusions are therefore approved by LPL.

The basic liberties are politically valuable because political liberalism's characterization of citizenship implies that they would be valued by citizens, and on the basis of this valuation citizens would demand protections for those liberties. This discussion illustrates a framework for drawing a more general distinction between political and substantive values. Consider equality. The substantive value of equality is one commitment of comprehensively liberal conceptions of the good, but it cannot legitimately ground political intervention within political liberalism. Political equality, however, can legitimately ground political intervention. The substance of the distinction derives from political liberalism's characterization of citizenship. Political equality implies that individuals are equally entitled to the protection of a certain range of basic liberties—and to the fair value of those liberties—precisely because citizens would insist upon those provisions. Equality as a political value does not depend for its viability on the metaphysical moral equality of persons; instead, it follows from LPL and political liberalism's conceptualization of citizenship: citizens conceive of themselves as equally entitled to certain forms of equal treatment implied by their possession of and commitment to the two moral powers. In those respects, all individuals, as citizens, are equal political persons.

Equality as a substantive value may also imply that individuals are equally entitled to the protection of their basic liberties and to the material conditions for the effective exercise of those liberties. But equality as a substantive value goes beyond political equality: it entails that, as a matter of moral fact, all persons are equally morally important and ought to be treated as such. Political equality implies that people ought to be treated equally because certain forms of equal treatment are implied by the project of finding fair terms of cooperation for a democratic, pluralistic society, a project that gives rise to the conception of citizenship and LPL. Substantive equality implies that people ought to be treated equally because people really are moral equals. Substantive equality cannot permissibly be invoked to ground exercises of political power within political liberalism because not all reasonable citizens affirm the substantive equality of citizens. But accepting the political equality of persons is definitive of citizens' reasonableness. Thus, while many reasonable religious fundamentalists believe in natural differences between sexes that justify gendered power and status hierarchies, all reasonable fundamentalists accept the political equality implied by LPL and political liberalism's conceptualization of citizenship.

Just like the restricted scope of justice, then, the distinction between political and substantive values can be thought of as a heuristic that depends ultimately on political liberalism's characterization of citizenship. Equality as a political value licenses interventions not because the designation “political” carries some intrinsic import; rather, the designation tracks the verdicts rendered by LPL. We identify certain aspects of liberal values as “political” to mark those senses in which LPL approves the invocation of those values to justify exercises of political power. Even in domains in which they demand considerable protection for the expression of their personal values, citizens will sometimes deem the reasons in favor of intervention weightier than the reasons against. The reasons in favor of intervention are “political” because they figure into citizens' calculations—because the salience of the values that generate them is implied by political liberalism's characterization of citizenship. When political values are at stake, LPL can override its own presumptions against interference.

In the next section, I return to citizens' deliberations in light of competing political values. But to disarm the arbitrariness objection to RCJ, it is enough to note that within political liberalism, LPL delineates the size and shape of the basic structure: it includes only those aspects of citizens' lives that they would agree to order politically on mutually defensible grounds and all those aspects of citizens' lives that they would insist be ordered on mutually defensible grounds. The project of justice is to find the fair terms of social cooperation among free and equal citizens. If justice consists in the fair terms of cooperation among free and equal citizens, then the subject matter of justice also consists in those aspects of citizens' lives that they reasonably agree should be regulated on mutually defensible terms. Of course, critics of RCJ need not be political liberals,30 and thus may reject the account of legitimacy and the basic structure defended here. But because RCJ can be justified on principled grounds, critics cannot maintain the charge of arbitrariness. The arbitrariness objection is therefore unpersuasive.31

Housework-shirkers, market-maximizers, and the liberal principle of legitimacy

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

We might think that any defense against the restrictiveness objection situated within political liberalism is doomed to fail. Political liberalism imposes demanding standards for legitimate exercises of political power, and one of its theoretical commitments implies that political power cannot legitimately be justified on the basis of substantive conceptions of liberal values like equality. One consequence of this is that market-maximizing and housework-shirking appear implausibly to be excluded from the reach of justice on RCJ. Interventions to equalize work across genders or to equalize material well-being across social groups cannot legitimately be grounded in the claim that all persons are moral equals regardless of gender or social class background. This claim invokes the substantive equality of persons. It belongs to a (family of) comprehensive conception(s) of the good that not all citizens share, and therefore cannot legitimately ground enactments of political power. Because the standard justifications for discouraging housework-shirking and market-maximizing invoke the substantive equality of persons, RCJ appears straightforwardly to locate these behaviors beyond the reach of justice.

But appearances are deceiving. Recall that in political liberalism, the scope of justice is set by LPL. This explains why liberal values like equality and autonomy can play a restricted role within political liberalism, even though substantive conceptions of those values cannot be invoked to justify exercises of political power. The legitimate role of these values in a politically liberal society—like the scope of justice more generally—is determined by the consent of free and equal citizens. “Political” conceptions of equality and autonomy—so named because their value is implied by LPL—can permissibly be invoked to justify political interventions.

Consider the religious fundamentalist who rejects the importance of autonomy to human flourishing, and who rejects the substantive equality of all persons. The fundamentalist cannot accept interventions whose justification depends on the metaphysical conception of persons as moral equals, or on conceptualizations of the good life as one characterized by autonomous reflection on one's ends. Interventions whose justification depends on the substantive liberal values of equality and autonomy threaten to encroach objectionably on the fundamentalist's permissible lifestyle—which celebrates deep spiritual heteronomy and affirms “God-given” sex roles—without a justification for that encroachment which the fundamentalist, as a citizen, can reasonably be expected to accept. Such interventions are thus illegitimate by the lights of LPL.

But the religious fundamentalist will affirm certain aspects of autonomy as valuable. He will affirm his autonomy as a political actor in public deliberations; this autonomy is necessary for securing protection for his illiberal lifestyle and his rejection of substantive autonomy in his private life. Since his basic liberties derive from his status as an autonomous citizen capable of choosing a conception of the good, autonomous citizenship is the basis for his right to freedom of conscience. Similarly, the fundamentalist will affirm certain aspects of equality. He must protect his status as an equal political citizen if he is to protect his illiberal lifestyle against encroachment from those who celebrate more liberal values in their personal lives.

Citizens have a capacity for a conception of the good, which implies the capacity to reject the substantive moral equality of persons and the importance of autonomy. Because they value this capacity, citizens will not accept intrusions into their lives that are justified only on the assumption that illiberal lifestyles are misguided. But certain versions of liberal values are implied by citizenship itself, and are therefore acceptable to free and equal citizens. To determine how restrictively RCJ draws the purview of justice, we must ask: which exercises of political power would citizens accept, and with respect to which aspects of their lives? Without attempting to give a comprehensive answer to this question, I argue that citizens can reasonably be expected to accept a good deal more political intrusion than political liberals—and their opponents—have traditionally thought. The considerations I raise serve to undermine the restrictiveness objection by demonstrating that market-maximizing and housework-shirking are susceptible to legitimate exercises of political power, and therefore fall within the reach of principles of justice as determined by RCJ.

Recall that, for the purposes of applying LPL, citizenship must be understood normatively. We do not ask whether an exercise of political power is acceptable to actual citizens with developed conceptions of the good and loyalties to particular moral, religious, or philosophical doctrines. Rather, we ask whether the exercise of power would be acceptable to citizens as characterized according to political liberalism's conceptualization of the political person. We ask: is this exercise of political power acceptable to free and equal citizens with capacities for a conception of the good and a sense of justice and with a higher-order interest in preserving those capacities? In answering this question, we can attribute to citizens two motivating projects, both implied by political liberalism's characterization of citizenship.

First, because of their capacity for a conception of the good, citizens will be motivated to ensure that they have adequate space in which to act on that conception, and adequate protection from outside interference in doing so. In some cases, this motivating project will cause citizens to reject exercises of political power. They might reject the establishment of an official state religion, for example, in order to protect their right to practice a religion of their own choosing. In other cases, this motivating project will lead citizens to accept political interventions, or even demand them. Citizens may, for example, call upon the state to intervene in the workplace to protect against oppressive exercises of power by employers, or within the family to protect children against certain harmful parenting behaviors.

Second, because of their higher-order interest in protecting the capacities of citizenship, citizens will be motivated to ensure that conditions obtain which are conducive to the development of citizenship for all. If society failed to secure the conditions necessary for the realization of these capacities, then no particular citizen could be assured the protection of her own capacities as a citizen. Moreover, political personhood is necessary to get the very project of justice off the ground. For ideals of justice to meaningfully guide social cooperation, we must understand individuals as free and equal citizens, capable of forming and living out a conception of the good and capable of conforming their behavior to principles of justice. If individuals were not political persons in this sense, they would have no need for or interest in finding fair terms of mutual cooperation, and no capacity for complying with those terms once found. Because citizens have an interest in living in a stable and just society, they will be motivated to secure social conditions consistent with the development of citizenship for all.

This second motivating project, like the first, generates positive requirements and negative constraints on the legitimate use of political power. Negatively, the state must avoid intrusions into people's lives that can be justified only by appeal to values that citizens may reasonably reject. Positively, the state must ensure for citizens access to the necessary means for the development of their moral powers as well as adequate space in which that development can occur. This might require protections against exercises of private power. And it might require interventions to ensure access to certain social goods necessary for the effective development of citizenship, for example, to an adequate income attached to a social position from which one can derive self-respect.32 Other things equal,33 this second motivating project will lead citizens to demand enactments of political power necessary to establish and preserve those conditions.

Depending on the circumstances, different exercises of political power will be necessary to enable the realization of the two motivating projects. Consequently, the acceptability to citizens of an exercise of political power—and thus the legitimacy of that exercise—will depend on the character of the society in question. Consider an example: citizens will insist upon enactments of political power necessary to protect their political liberties from infringement and to ensure conditions under which it is materially possible to effectively exercise these liberties.34 On the basis of LPL, then, those exercises of political power are legitimate—indeed required, since citizens insist upon them. But whether political power must be exercised to protect citizens' political liberties—and to what extent it must be exercised—is context-dependent. In a society thoroughly pervaded by an ethos of sharing and community, it may well be that no coercive redistribution of income and wealth would be necessary to ensure for all citizens adequate all-purpose means to, for example, run for public office. Thus, no such redistribution would be justified on the grounds of protecting the fair value of that liberty.35 But in a society characterized by vast inequality with severe poverty and deprivation on the losing end, redistribution is straightforwardly justified as a means to the protection of the fair value of basic liberties. Interventions necessary to ensure that every citizen meets the threshold of material wealth necessary for the effective exercise of the basic liberties are required within political liberalism, because their omission is unacceptable to citizens.

The two motivating projects of citizenship will sometimes be in tension with one another.36 Because interventions that further one project may frustrate another, citizens will not automatically accept all interventions that can be justified by appeal to one of the projects. We will sometimes have to adjudicate competing considerations in determining which exercises of political power are legitimate. This need to manage trade-offs means that the legitimacy of political interventions is contingent on the type of force those interventions involve. Citizens may accept interventions that incentivize certain behaviors while rejecting the criminalization of alternative behaviors. Suppose that being raised to affirm fundamentalist religious doctrines impedes children's development of citizenship. Even so, citizens will reject legal statutes prohibiting religious fundamentalists from raising their children to affirm such doctrines, because raising one's child to share one's values is an important way to enact one's conception of the good, and citizens will want this transmission of values protected. They will reject the statutes, which would have furthered their motivating project of protecting children's development of citizenship, because the statutes are unacceptably inimical to a different motivating project: the protection of adequate space for citizens to exercise their conceptions of the good. Insofar as this trade-off is required, however, citizens will be inclined to accept gentler exercises of political power to advance the compromised motivating project, like interventions to protect children's development of citizenship that pose less of a threat to the ability of parents to share their conceptions of the good with their children. For example, citizens might accept interventions requiring that children must be educated to understand their basic rights and duties as citizens.

Exercises of political power generally impose some restrictions on some individuals, and so citizens will generally not accept a more forceful intervention where a gentler one would suffice. Accordingly, LPL imposes a presumption in favor of enacting the gentlest intervention sufficient to accomplish the objective in question.37 Because political interventions vary in their forcefulness, we cannot infer from the illegitimacy of certain interventions that the behaviors targeted by those interventions are categorically beyond the reach of justice. It would clearly be illegitimate to criminalize housework-shirking. But we cannot infer from this that household allocations of labor are beyond the reach of justice. Gentler exercises of political power may be acceptable to citizens. In the case considered above, the fundamentalist parents' transmission of their values to their children may fall within the reach of principles of justice, even though some forceful interruptions of that transmission constitute illegitimate uses of political power.

In sizing up the basic structure, then, we must keep several considerations in mind: First, citizens are motivated to ensure that conditions obtain which are conducive both to citizens' enactment of their conceptions of the good and to the development of citizenship among all individuals. These motivating projects have implications not just for enactments of political power they will reject, but additionally for enactments upon which they will insist. Second, whether or not a political intervention is acceptable to citizens depends on the circumstances in which it is to be enacted, because the interventions needed to realize citizens' motivating projects will vary depending on circumstances. Finally, whether or not a political intervention is acceptable to citizens depends on the type of political power being exercised. Relatively gentle interventions may be legitimate where more forceful interventions are not.

These considerations imply that the basic structure is a great deal more expansive than critics of RCJ assume it to be. Imagine a large community of citizens who believe that men and women are naturally unequal in their capacities and predilections, and that this inequality justifies a rigidly gendered allocation of power and labor. Due to the insularity of the community, members interact almost exclusively with other members, and illiberal values are reliably transmitted from parents to children. Under these circumstances, and given the truth of plausible empirical assumptions to be spelled out presently, citizens will accept such exercises of political power as incentives for women to work outside the home and for men to care for children. These interventions will be acceptable to citizens because, plausibly, children cannot develop a conception of themselves as equal citizens when substantive equality is never apparent to them as a feasible option.38 To be sure, citizens who affirm doctrines that deny their substantive equality may well retain conceptions of themselves as equal citizens if they experience themselves as choosing among feasible alternatives, some of which affirm their substantive equality. But those who grow up under circumstances that prevent them from recognizing substantive equality as a feasible option are unlikely to conceive of themselves as equal in any sense, including as equal citizens. Under these circumstances, political intervention is necessary to preserve the recognizability of substantive equality as a feasible option. The incentives mentioned above would therefore further citizens' motivating project of preserving social conditions under which all individuals can develop as citizens.

Of course, citizens are not motivated only to protect the capacity of all to develop as citizens. They are also motivated to protect adequate space for citizens to exercise conceptions of the good. They will therefore reject interventions that use more force than is necessary to protect the children's development of citizenship. But by incentivizing paid work among women and caring work among men, the state might establish enough heterogeneity to preserve children's ability to see substantive equality as a feasible option, and thereby preserve their capacity to develop as citizens. This intervention furthers a motivating project of citizenship without sacrificing any comparable progress toward other motivating projects. Because citizens would accept—and perhaps insist upon—this exercise of political power, it is legitimate by the lights of LPL. Gendered housework allocations are not beyond the reach of a restricted theory of justice.39

Nor does RCJ place market-maximizing behavior beyond the reach of justice. Imagine that, for generations, every “talented” citizen has demanded material incentives in exchange for the exercise of her productive talents. The result is persistent social class stratification. Citizens recognize a clear social demarcation between high-status workers who can demand high wages for their (trained) marketable talents and lower-status workers who cannot. Over time, the inequality may jeopardize the development of political personhood, as entrenched inequalities between social classes that persist across generations may plausibly undermine citizens' capacity to see themselves as political equals. In this scenario, citizens would accept exercises of political power intended to restore conditions conducive to the development of citizenship. They may, for example, accept laws requiring students to be educated in the virtues of social solidarity and fraternity in order to discourage the market-maximizing behavior that threatens to undermine citizenship. Such interventions would protect the development of citizenship at a relatively small cost in restricting the space for citizens to exercise their conceptions of the good.40

Moreover, even on RCJ, individuals incur some direct duties of justice: we are required to support just institutions and work toward their furtherance when they do not yet exist. The expanded account of the basic structure developed here may imply corresponding duties for individuals to “self-police” to ensure that their behavior not contribute to the sustaining of norms and institutions inconducive to the realization of citizenship. Granted, I have only defended the legitimacy of “gentle” interventions, and resisting incentives against housework-shirking and market-maximizing need not constitute noncompliance with the political institutions offering those incentives. A thorough investigation into what constitutes compliance with institutions enacting “gentle” political interventions is needed. But it is worth noting in the meantime that my expansive account of RCJ does more than permit political institutions to intervene to encourage certain behaviors. An expanded basic structure carries with it a correspondingly expanded set of obligations of justice requiring individuals to comply with the institutions of that basic structure. When resisting egalitarian political inducements helps perpetuate social circumstances inconducive to citizenship, that resistance might properly be regarded as noncompliance.41

Citizens will accept—and even insist upon—the most gentle interventions sufficient to preserve social conditions conducive to the development of citizenship. Under certain circumstances, these interventions include exercises of power aimed at protecting the development of citizenship by discouraging inequality-generating behaviors of individuals within households and labor markets.42 Because citizens can accept interventions that target housework-shirking and market-maximizing, institutions that aim to amend these behaviors can be a part of the basic structure as delineated by LPL. Thus, individual behaviors such as housework-shirking and market-maximizing do fall within the reach of principles of justice set by RCJ. Like the arbitrariness objection, the restrictiveness objection is unpersuasive.

The convergence with comprehensive liberalism

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

At this point, one might reasonably worry that my proposed delineation of the subject matter of justice violates the spirit—even if not the letter—of political liberalism. The point of political liberalism, after all, was to specify a set of conditions under which a diverse society can be regulated stably over time by a system of just political institutions. But the interventions whose legitimacy I have defended will be inhospitable to many conceptions of the good. Whether or not these interventions are acceptable to citizens understood in the idealized, value-laden sense of citizenship, they will not win the allegiance of all actual citizens in a pluralistic society. Those who value traditional gender roles and capitalistic acquisitiveness may resist interventions to discourage housework-shirking and market-maximizing.

Indeed, some philosophers argue that the dictates of political liberalisms are—at least within certain domains—effectively indistinguishable from the dictates of comprehensive liberalisms. Amy Gutmann, for example, argues that comprehensive and political liberalisms issue the same judgments regarding the paradigmatically controversial cases of restricting parents' authority over their children's education, because the political conception of citizenship is rich enough to justify the kinds of restrictions on parental prerogatives that comprehensive liberals have tended to support.43 Similarly, I argue that the political conception of citizenship is rich enough to justify intrusive interventions to discourage housework-shirking and market-maximizing. I frame my project as an argument for the happy conclusion that political liberals can, without sacrificing their basic theoretical commitments, endorse progressive social reforms as permissible and, in some cases, even required. Gutmann, on the other hand, uses the progressive potential of political liberalism to cast doubt on the import of the political/comprehensive distinction.

Suppose I have successfully shown that political liberalism calls for many of the same intrusions into the home and the market as comprehensive liberalisms committed to equality and autonomy as substantive values. If political liberalism does little more than offer different justifications for the same verdicts as comprehensive liberalism, does it really offer better prospects for long-term stability in a pluralistic society? And if not, what advantage does political liberalism offer over comprehensive liberalism?

Political liberals might be inclined to retain the theoretical advantage of political liberalism over comprehensive liberalism by denying the verdicts I have attributed to their theory. They may disavow the interventions I have endorsed precisely because those interventions would be rejected by some actual reasonable citizens. They may think that this rejection demonstrates that the interventions fail to be appropriately neutral among various conceptions of the good.

Indeed, the interventions I have endorsed have nonneutral consequences. But no political regime is entirely neutral in its consequences, and political liberals have long recognized that they cannot impose a standard of neutrality which forbids all political interventions that impede or promote certain ways of life.44 Inevitably, political institutions will influence the lifestyle choices and values of those living under them. What matters is that we can justify political interventions using reasons that are neutral among conceptions of the good, reasons that are acceptable from the perspective of citizenship. The interventions I have defended do not depend for their justification on controversial premises affirming the intrinsic value or choice-worthiness of particular lifestyles. Though they may effectively promote egalitarianism by discouraging market-maximizing and housework-shirking, their justification invokes only public reasons that all citizens can reasonably be expected to accept. They are therefore neutral by the lights of the only standard of neutrality that political liberalism can consistently impose.

Nor can political liberals disavow the verdicts I have attributed to their theory by criticizing the idealized, value-laden conception of citizenship my account employs. That conception is necessary for securing the most fundamental commitments of political liberalism, including the protection of the basic political liberties: the protection of these liberties against infringement remains a fundamental commitment of political liberalism, even though actual citizens who deny the intrinsic importance of civic engagement may regret that their society expends scarce resources to protect political liberties. Because some reasonable conceptions of the good subordinate the fundamental values of political liberalism to other values, actual citizens may consent to sacrificing political values when doing so furthers other causes they value more. In order to secure the central tenets of political liberalism—like the protection of basic liberties—the theory must employ the idealized conceptualization of citizenship.

Political liberalism's progressive potential is genuine, and many of its verdicts do indeed converge with those rendered by comprehensive liberalisms. Still, the extent of this convergence remains an open question, and there is some reason to think that, in the event of divergence, political liberalism will render the more conservative judgments. Because comprehensive liberalism values substantive equality, it is likely to register market-maximizing and housework-shirking as (defeasibly) susceptible to intervention in a large range of cases. But within political liberalism, it was not simply the inegalitarianism of market-maximizing and housework-shirking that justified intervention. Rather, it was the pervasiveness of the behaviors, and the political harms the behaviors threatened, given the circumstances at play. Unlike comprehensive liberalism, political liberalism registers substantively inegalitarian behaviors as susceptible to intervention only when they perpetuate norms and social circumstances inconducive to the realization of citizenship. Because some comprehensive liberalisms will find the behaviors themselves objectionable, comprehensive liberalism may authorize interventions in a broader range of cases than political liberalism. If so, then political liberalism accommodates a broader diversity of actual citizens and thereby offers more favorable prospects for stability.45

But even in the event of total convergence, political liberalism offers more favorable prospects for stability. LPL specifies conditions under which a prevailing stability is a legitimate stability:46 a stable basic structure is also legitimately stable when all exercises of political power are acceptable from the perspective of citizenship. By imposing this requirement, LPL also enhances the likelihood that stability will prevail in the first place. The requirement that political enactments be publicly justifiable imposes a corresponding presumption that those enactments be defended politically in the language of public reason. A society that abides by this presumption gives due respect to adherents of diverse ways of life, and demonstrates publicly that the basic commitments of citizenship call for certain progressive interventions. This public demonstration makes it likelier that actual citizens will accept those interventions, despite the challenge they pose to some illiberal ways of life. Moreover, the requirement of public justification ensures and announces to all that the justification for controversial interventions is one to which all citizens ought to assent, even if it leads to a conclusion they despise.

The requirement of public justification that political liberalism imposes does not ensure stability, because even reasonable people can reject interventions called for by the basic commitments of citizenship. But political philosophy remains a normative enterprise, and we should not defer entirely to the facts of human nature. Not all actual citizens will accept interventions that discourage market-maximizing and housework-shirking. But neither will all actual citizens accept interventions that protect basic liberties. Some would prefer that scarce political resources be expended instead to secure other social goods. But actual citizens should find interventions more palatable insofar as it is publicly known that those interventions are called for by political values, the endorsement of which is called for by their status as citizens. By imposing the standard of public justification embodied by LPL, political liberalism strikes a principled balance between deference to the diversity of a pluralistic society and allegiance to the values of liberalism. It ensures that, when stability prevails, it is a legitimate stability. In so doing, it enhances the likelihood that stability will prevail.47

Conclusion

  1. Top of page
  2. Arbitrariness, restrictiveness, and the basic structure
  3. Does political power face a special justificatory burden?
  4. Political liberalism and the basic structure
  5. Housework-shirkers, market-maximizers, and the liberal principle of legitimacy
  6. The convergence with comprehensive liberalism
  7. Conclusion

My account of the subject matter of justice offers something appealing to both sides. On the one hand, my defense of RCJ secures the status of justice as a restricted political ideal. Principles of justice apply, first and foremost, to the basic structure of society. On the other hand, the reach of justice delineated by RCJ is more expansive than either critics or proponents of RCJ have taken it to be, because the basic structure is more expansive than it initially appeared to be.

Even so, I suspect that neither camp will find my account fully satisfactory. My defense of RCJ legitimizes exercises of political power that Rawls would likely have found objectionable. And though I have argued that RCJ can identify institutions as unjust insofar as those institutions fail to target housework-shirking and market-maximizing behavior, my account still prevents us from categorizing that behavior itself as an injustice, something that Cohen evidently thinks a theory of justice should permit. Moreover, interventions to discourage housework-shirking and market-maximizing will be legitimate only insofar as those behaviors perpetuate norms and institutions inconducive to the development of citizenship.

But the conceptual middle ground has much to recommend it. First, my defense of RCJ is principled: it is grounded within the framework of political liberalism, and it is justified on the basis of salient considerations regarding the acceptability of political interventions to free and equal citizens. Second, the account I have defended constitutes an intuitively plausible picture of what justice judges. It can render nuanced verdicts about the legitimacy of political interventions that are sensitive to the type of power being exercised, the purpose of its exercise, and the circumstances at hand.48 One upshot is that housework-shirking and market-maximizing fall within the reach of principles of justice. Under certain circumstances, certain exercises of political power can legitimately be used to discourage those behaviors. By showing how such exercises of power can be legitimate, my approach accommodates plausible intuitions about what justice judges, while retaining the restriction of justice to the basic structure.

Footnotes
  1. 1

    John Rawls, A Theory of Justice, rev. ed. (Cambridge, Mass.: Harvard University Press, 1999), pp. 293294. See also ibid., pp. 99, 154, 415. Consider Rawls's difference principle, which maintains that institutions are to be arranged in such a way as to optimize the position of the least advantaged. Among the various possible ways of arranging society's institutions, the difference principle approves that arrangement within which the worst off are better off than under any other feasible arrangement. But the difference principle does not dictate that within society's institutions, individuals must behave in ways that benefit the least advantaged. On the contrary, individuals are required only to support and comply with institutions approved by the difference principle, and to work toward the formation of such just institutions insofar as they do not yet exist.

  2. 2

    G. A. Cohen, “Incentives, Inequality, and Community,” in The Tanner Lectures on Human Values, vol. 13, ed. G. Peterson (Salt Lake City: University of Utah Press, 1992), pp. 261329; Cohen, “The Pareto Argument for Inequality,” Social Philosophy and Policy 12 (1995): 160185; Cohen, “Where the Action Is: On the Site of Distributive Justice,” Philosophy & Public Affairs 26 (1997): 330; Cohen, “If You're an Egalitarian, How Come You're So Rich?Journal of Ethics 4 (2000): 126; Liam B. Murphy, “Institutions and the Demands of Justice,” Philosophy & Public Affairs 27 (1998): 251291.

  3. 3

    S. A. Lloyd, “Family Justice and Social Justice,” Pacific Philosophical Quarterly 75 (1994): 353371; Blain Neufeld, “Coercion, the Basic Structure, and the Family,” Journal of Social Philosophy 40 (2009): 3754; Martha Nussbaum, “The Feminist Critique of Liberalism,” in Sex and Social Justice (New York: Oxford University Press, 1999), chap. 2, pp. 5580; Nussbaum, “Rawls and Feminism,” in The Cambridge Companion to Rawls, ed. Samuel Freeman (Cambridge: Cambridge University Press, 2000); S. M. Okin, Justice, Gender, and the Family (New York: Perseus Books, 1989); Okin, “Political Liberalism, Justice, and Gender,” Ethics 105 (1994): 2343; Okin, “Forty Acres and a Mule for Women: Rawls and Feminism,” Politics, Philosophy & Economics 4 (2005): 233248.

  4. 4

    The following are representative of those who oppose RCJ: Cohen, “Incentives, Inequality, and Community”; Cohen, “Where the Action Is”; Cohen, “If You're an Egalitarian”; Cohen, “The Pareto Argument for Inequality”; Cohen, Rescuing Justice and Equality (Cambridge, Mass.: Harvard University Press, 2008); and Murphy, “Institutions and the Demands of Justice.” Michael Titelbaum concurs with Cohen that constraints on individual behavior are “required on Rawls's own terms,” although he disagrees with Cohen on the content of those constraints. Titelbaum, “What Would a Rawlsian Ethos of Justice Look Like?Philosophy & Public Affairs 36 (2008): 289322, at p. 295. In this article, I focus primarily on Cohen's formulation of the two objections. Various defenses of RCJ have been attempted. See Andrew Williams, “Incentives, Inequality, and Publicity,” Philosophy & Public Affairs 27 (1998): 225247; Jonathan Wolff, “Fairness, Respect, and the Egalitarian Ethos,” Philosophy & Public Affairs 27 (1998): 97122; David Estlund, “Liberalism, Equality, and Fraternity in Cohen's Critique of Rawls,” Journal of Political Philosophy 6 (1998): 99112; Thomas Pogge, “On the Site of Distributive Justice: Reflections on Cohen and Murphy,” Philosophy & Public Affairs 29 (2000): 137169; Joshua Cohen, “Taking People as They Are?Philosophy & Public Affairs 30 (2001): 363386; Norman Daniels, “Democratic Equality: Rawls's Complex Egalitarianism,” in Freeman , The Cambridge Companion to Rawls; Kok-Chor Tan, “Justice and Personal Pursuits,” Journal of Philosophy 101 (2004): 331362; Samuel Scheffler, “Is the Basic Structure Basic?” in The Egalitarian Conscience: Essays in Honour of G. A. Cohen, ed. Christine Sypnowich (Oxford: Oxford University Press, 2006); and Neufeld, “Coercion, the Basic Structure, and the Family.” I address Scheffler's and Neufeld's defenses directly later in this article. For Cohen's reply to Williams, see Cohen, Rescuing Justice and Equality, pp. 344371. For his reply to Pogge, see ibid., pp. 394403.

  5. 5

    Rawls, A Theory of Justice, p. 6. I use Rawls's term “basic structure” because I endorse the restriction of justice to the institutional mechanisms by which major social institutions exercise power over citizens; but, depending on one's interpretation of Rawls's usage, he and I may diverge regarding the extension of that term. This will become clearer in due course.

  6. 6

    Other theorists have argued that the basic structure extends further than is typically thought. See Estlund, “Liberalism, Equality, and Fraternity”; Cohen, “Taking People as They Are?”; Scheffler, “Is the Basic Structure Basic?”; and Tan, “Justice and Personal Pursuits.” Cohen at times seems open to this defense of RCJ. Cohen, Rescuing Justice and Equality, p. 12. Elsewhere, he appears to reject this approach. Ibid., p. 126 n20. My article is distinctive in offering a systematic investigation into RCJ's capacity to countenance political projects aimed at changing the behavior of individuals by modifying institutions. I consider projects to change market-maximizing behavior and housework-shirking behavior, and investigate the constraints within which such projects constitute legitimate exercises of political power.

  7. 7

    Rawls, A Theory of Justice, p. 6.

  8. 8

    Ibid., p. 47.

  9. 9

    For textual support of this interpretation, see Rawls, Political Liberalism (New York: Columbia University Press, 1993), pp. 282283.

  10. 10

    Rawls, A Theory of Justice, p. 7, italics added. This is just one of Rawls's reasons for restricting the purview of justice. For a discussion of the others, see Scheffler, “Is the Basic Structure Basic?” For his part, Cohen appears to ignore Rawls's other justifications for RCJ: “Given … his stated rationale for exclusive focus on the basic structure—and what other rationale could there be for calling it the primary subject of justice?—Rawls is in a dilemma.” Cohen, Rescuing Justice and Equality, p. 137. (Cohen is here referring to the profundity of effect rationale offered in the quotation to which this note is attached.)

  11. 11

    Cohen, “Where the Action Is,” pp. 1819. Okin also makes this point. See Okin, “Review of Political Liberalism,” American Political Science Review 87 (1993): 10101011. For a slightly different rendering of the arbitrariness objection, see Murphy, “Institutions and the Demands of Justice.” For a response more specifically directed at Murphy, see Thomas Pogge, “On the Site of Distributive Justice: Reflections on Cohen and Murphy,” Philosophy & Public Affairs 29 (2000): 137169.

  12. 12

    In Cohen's own words, “Why should we care so disproportionately about the coercive basic structure, when the major reason for caring about it … is also a reason for caring about informal structure and patterns of personal choice? To the extent that we care about coercive structure because it is fateful with regard to benefits and burdens, we must care equally about the ethos that sustains gender inequality and inegalitarian incentives.” Cohen, Rescuing Justice and Equality, p. 138. See also Cohen, “Where the Action Is,” pp. 2122. Baynes and Williams also endorse the arbitrariness objection to RCJ. See Kenneth Baynes, “Ethos and Institutions: On the Site of Distributive Justice,” Journal of Social Philosophy 37 (2006): 182196; and Williams, “Incentives, Inequality, and Publicity.”

  13. 13

    Notice, however, that one could defend RCJ against the arbitrariness objection without accepting Rawls's stated justification for that restriction. RCJ is arbitrary only if no principled justification can be given for it. I will argue in due course that some principled justification can be given. Because a principled justification is available, the failure of Rawls's proposed justification—if indeed it does fail—need not concern defenders of RCJ.

  14. 14

    In this context, “talented” is stipulated to mean that these individuals “are so positioned that [they] … command a high salary and … can vary their productivity according to exactly how high [that salary] is.” Cohen, “Where the Action Is,” pp. 67.

  15. 15

    On Cohen's account, principles of distributive justice apply not just to the basic structure but also to “people's legally unconstrained choices.” Cohen, Rescuing Justice and Equality, p. 116. Cohen concedes that there are some “agent-centered prerogatives” to pursue nonegalitarian personal projects. Ibid., p. 10. See also ibid., p. 61. But he evidently thinks this a modest concession: “In my view,” he claims, “there is hardly any serious inequality that satisfies the requirement set by the difference principle.” Ibid., p. 119. David Estlund and Kok-Chor Tan have argued that this seemingly modest concession actually commits Cohen to a much broader array of prerogatives than he wants to allow. Estlund, “Liberalism, Equality, and Fraternity”; Tan, “Justice and Personal Pursuits.” For Cohen's reply to Estlund, see Cohen, Rescuing Justice and Equality, pp. 387394. Michael Titelbaum argues, with Cohen, that a restricted conception of justice must be supplemented by a set of principles to regulate the behavior of individual agents within institutions, but quarrels over the content of those principles. Titelbaum, “What Would a Rawlsian Ethos of Justice Look Like?” p. 296.

  16. 16

    See, for example, G. Poeschl, “Social Norms and the Feeling of Justice about Unequal Family Practices,” Social Justice Research 21 (2008): 6985; H. Lothaller, G. Mikula, and D. Schoebi, “What Contributes to the (Im)balanced Division of Family Work between the Sexes?Swiss Journal of Psychology 68 (2009): 143152; R. Gittell, “Constrained Choices and Persistent Gender Inequity: The Economic Status of Working Women in a High-Income, Low-Poverty State with Lessons for Others,” American Behavioral Scientist 53 (2009): 170192; A. Ross-Smith and C. Chesterman, “ ‘Girl Disease’: Women Managers' Reticence and Ambivalence towards Organizational Advancement,” Journal of Management & Organization 15 (2009): 582595; R. Breen and L. P. Cooke, “The Persistence of the Gendered Division of Domestic Labour,” European Sociological Review 21 (2005): 4357; E. Kluwer, J. Heesink, and E. van de Vliert, “Marital Conflict about the Division of Household Labor and Paid Work,” Journal of Marriage & Family 58 (1996): 958969; E. Kluwer, J. Heesink, and E. van de Vliert, “The Division of Labor in Close Relationships: An Asymmetrical Conflict Issue,” Personal Relationships 7 (2000): 263282; and E. Kluwer, “Responses to Gender Inequality in the Division of Family Work: The Status Quo Effect,” Social Justice Research 11 (1998): 337357.

  17. 17

    Proponents of RCJ allow some behavior of family members to count as a legitimate target of political intervention. In explicating his most considered position on the matter, Rawls clarifies that the behavior of individuals within families does fall within the purview of justice insofar as that behavior constitutes a violation of the rights of another family member: “Political principles … do impose essential constraints on the family as an institution and so guarantee the basic rights and liberties, and the freedom and opportunities, of all its members. This they do … by specifying the basic rights of equal citizens who are the members of families. The family as part of the basic structure cannot violate these freedoms. Since wives are equally citizens with their husbands, they have all the same basic rights, liberties, and opportunities as their husbands; and this, together with the correct application of the other principles of justice, suffices to secure their equality and independence.” John Rawls, “The Idea of Public Reason Revisited,” in Collected Papers, ed. Samuel Freeman (Cambridge, Mass.: Harvard University Press, 1999), pp. 573615, at p. 597. But critics of RCJ remain unsatisfied. None of the basic liberties demands equal access to social goods within domestic partnerships, and so (non-rights-violating) cases of housework-shirking remain unchallenged by Rawls's treatment of the family. The exclusion of housework-shirking from the subject matter of justice is implausible, these critics say, given the profound disadvantages caused by the norms sustaining the gendered division of labor.

  18. 18

    For Cohen's first use of the term “ethos” to refer to a “culture of justice,” see Cohen, “Incentives, Inequality, and Community,” p. 315.

  19. 19

    Other gender egalitarian political interventions include pay equity and affirmative hiring practices, and incentives for employers that offer paid parental leave (with mechanisms in place to encourage men to take that leave).

  20. 20

    In an argument against incentive inequalities, Cohen seems to acknowledge this possibility of arranging social institutions to change personal behavior within labor markets: “Habits can change,” he says, and, continuing in a footnote, “if not always at the level of the individual, then certainly at the social level, through reformed structures of education.” Cohen, Rescuing Justice and Equality, p. 50.

  21. 21

    Variations of this strategy are pursued by Neufeld, “Coercion, the Basic Structure, and the Family”; and Scheffler, “Is the Basic Structure Basic?” Neufeld and Scheffler both refer to RCJ as a restriction of justice to the “coercive” apparatus of society, where coercion is understood broadly to include any exercise of political power. In this they follow Rawls: “Political power is always coercive power applied by the state and its apparatus of enforcement.” Rawls, Justice as Fairness (Cambridge, Mass.: Harvard University Press, 2001), p. 40; Rawls, Political Liberalism, p. 68.

  22. 22

    Rawls, Political Liberalism, p. 136.

  23. 23

    Ibid., pp. 142143.

  24. 24

    The distinction between comprehensive and political liberalism is often described in terms of the role that judgments of truth play in justifying or publicly defending political action. In Rawls's words, a political conception of justice “does without the concept of truth,” whereas a comprehensive conception invokes the truth of liberal values to justify political action. Rawls, Political Liberalism, p. 129, italics in original. But Joshua Cohen argues that judgments of truth can legitimately play a role within a political conception of justice, and develops a “political” conception of truth that he thinks serves this purpose. Joshua Cohen, “Truth and Public Reason,” Philosophy & Public Affairs 37 (2009): 242. I attempt here to set this controversy aside. By distinguishing between comprehensive and political value judgments not in terms of the truth assertions those judgments imply but instead in terms of the content of the judgments themselves, I have attempted to remain neutral regarding the role of truth judgments in political liberalism.

  25. 25

    In Rawls's words, political liberalism attempts to uncover “the fair terms of social cooperation between citizens characterized as free and equal yet divided by profound doctrinal conflict.” Rawls, Political Liberalism, p. xxv.

  26. 26

    Ibid., p. 137. LPL is derived from the criterion of reciprocity, political liberalism's fundamental normative ideal. Ibid., p. xliv.

  27. 27

    Ibid., pp. 142143.

  28. 28

    Assuming those values and ideals are reasonable. In political liberalism, a conception of the good counts as reasonable so long as it does not call for the violation of basic liberties and so long as it is compatible with the various burdens citizens face as citizens in a society marked by reasonable pluralism.

  29. 29

    Rawls, Political Liberalism, p. xliv.

  30. 30

    Cohen is a case in point.

  31. 31

    Moreover, there is a prima facie reason to think that RCJ might cohere after all with the justification Rawls gives for it, namely, that the “profound effect” of the basic structure justifies its status as the primary subject of justice. It is not at all implausible to think that citizens would agree that behaviors are susceptible to legitimate exercises of political power to the extent that those behaviors profoundly impact the lives of other citizens.

  32. 32

    Similarly, Rawls argues that a just society must ensure for all citizens adequate all-purpose means to make effective use of the basic liberties and opportunities. Rawls, Political Liberalism, pp. xlvi, 6.

  33. 33

    The importance of this qualifier will become clear below when I discuss the need to manage trade-offs among citizens' various motivating projects.

  34. 34

    Rawls, A Theory of Justice, p. 179; Rawls, Political Liberalism, pp. 324331.

  35. 35

    This is, of course, consistent with redistribution being justified on some other grounds.

  36. 36

    Or with other motivating projects of citizenship, assuming there are others.

  37. 37

    Similarly, citizens need not accept interventions intended to maximize the realization of citizenship above the threshold at which citizenship is genuinely attainable for all. Free and equal citizenship is a “threshold concept”: what matters, from the perspective of the political conception of justice, is not that citizens realize the virtues of political personhood as fully as possible, but that they attain them to a sufficient degree. I take it for granted here that a person whose conception of the good casts her as a subequal and who recognizes no other conceptions as feasible alternatives will not attain a sufficient conception of herself as a free and equal citizen.

  38. 38

    In this and other scenarios I consider later in the article, I make empirical assumptions to illustrate cases wherein citizens would insist upon certain interventions. I think these assumptions are plausible, but I do not defend them.

  39. 39

    The types of political interventions I defend here appear largely congenial to Rawls's discussion of the family as part of the basic structure. See Rawls, “The Idea of Public Reason Revisited.” In general, my account of legitimate political interventions may be friendlier to Rawls's own considered view than we might at first be inclined to think. See also 31.

  40. 40

    One of Cohen's own examples can be used to illustrate: The ratio of top executive salaries to production worker wages in 1988 was 6.5 to 1 in West Germany and 17.5 to 1 in the United States. Cohen argues that the German distribution was more just, but that Rawls cannot render that verdict, “since the smaller inequality that benefited the less well off in Germany was not a matter of law but of ethos.” Cohen contends that this alleged failure to recognize Germany's distribution as more just is “a grave defect in [Rawls's] conception of the site of distributive justice.” Cohen, Rescuing Justice and Equality, p. 143. Notice, however, that RCJ deems the two societies equally just only if the income inequality in the United States is unobjectionable from the perspective of citizenship. If the inequality is sufficient to compromise the conditions for the realization of citizenship, citizens would condemn it and insist on political interventions to remedy it. The prescribed interventions are likely to include educative measures to inculcate an egalitarian ethos of justice. Insofar as the political institutions in the United States failed in this regard, RCJ can render the verdict Cohen calls for: that the United States was less just in virtue of its less egalitarian distribution.

  41. 41

    It is worth noting, too, that the legitimacy of political intervention will often depend in part on the extent of self-policing. If people self-police sufficiently well that conditions conducive to citizenship are preserved, then the coercive political interventions otherwise called for on the grounds canvassed here will be unnecessary.

  42. 42

    Cohen attributes to Ronald Dworkin a proposal similar to the one I defend here: that justice might require political institutions that promote an ethos that discourages market-maximizing. Cohen, Rescuing Justice and Equality, p. 127, referring to comments Dworkin made at a seminar in Oxford. Cohen's response is puzzling. He claims, “Rawls could not have said that, to the extent that the indicated ethos-promoting policy failed, society would as a result be less just than if the policy had been more successful. Accordingly, if … Rawlsian justice requires government to promote an ethos friendly to equality, it could not be for the sake of making society more distributively just, … even though it would be for the sake of making its distribution more equal.” Ibid., p. 127. See also ibid., pp. 379381. But Cohen does not explain why society's failure in this regard could not constitute a failure of justice on a Rawlsian RCJ. The failure, after all, is institutional, and Cohen himself describes it as the failure of a policy. The fact that the policy has as its objective the development of an ethos does not disqualify the policy from inclusion in the basic structure. Institutions of the basic structure regularly perform the role of inducing individuals to behave in certain ways, including (as Cohen elsewhere acknowledges, for example, at ibid., p. 50) inducing them to develop certain dispositions or habits of behavior. For a similar rejoinder to Cohen's defense against Dworkin's objection, see Cohen, “Taking People as They Are?

  43. 43

    According to Gutmann, political values are no less potent than comprehensive values as justifications for restricting parents' prerogative to dictate what children learn and how they learn it. See Amy Gutmann, “Civic Education and Social Diversity,” Ethics 105 (1995): 557579. Will Kymlicka makes a similar point about the more general convergence of judgments between political and comprehensive liberalisms. See Will Kymlicka, Contemporary Political Philosophy: An Introduction (Oxford: Oxford University Press, 2002), esp. pp. 232240.

  44. 44

    As Rawls puts it, “it is surely impossible for the basic structure of a just constitutional regime not to have important effects and influences as to which comprehensive doctrines endure and gain adherents over time; and it is futile to try to counteract these effects and influences. … We must accept the facts of commonsense political sociology.” Rawls, Political Liberalism, p. 193. See also ibid., pp. 192195.

  45. 45

    To be sure, comprehensive liberals are likely to defer in many cases to the value of autonomy as a constraint on exercises of political power. They may, on this basis, defer to illiberal parents' interest in avoiding curricular requirements that instill in their children an egalitarian ethos of justice. More attention to the verdicts rendered in borderline cases would be helpful. But notwithstanding Gutmann's argument, we have no decisive reason to accept—and some reason to reject—the conclusion that political and comprehensive liberalisms' verdicts will always converge. Only in particular cases have they been shown to do so.

  46. 46

    That is, a stability for the right reasons, as distinct from stability for such reasons as resignation to an unjust status quo, accidental convergence of conceptions of the good, or suppression of dissent.

  47. 47

    Political liberalism has a further theoretical advantage: only by accepting LPL, the basic and distinctive theoretical commitment of political liberalism, can we deploy the defense of RCJ that I develop in this article.

  48. 48

    My account enables us to “apply principles of justice to dominant patterns of social behavior … [since] that, as it were, is where the action is.” Cohen, Rescuing Justice and Equality, p. 142.