The Priority of Liberty: An Argument from Social Equality

Abstract

John Rawls’s thesis that a certain package of basic liberties should be given lexical priority is of great interest for legal and political philosophy, but it has received relatively little defense from Rawls or his supporters. In this paper, I examine three arguments for the thesis: the first is based on the two ‘moral powers’; the second, on the social bases of self-respect; and the third, on a Kantian notion of autonomy. I argue none of these accounts successfully establishes 1) the distinct claim of lexical priority, 2) for the complete package of basic liberties (including the fair value of the political liberties), on the basis of reasons that are appropriately public. In turn, I propose an alternative argument, in support of those two claims, based on the social or ‘relational’ conception of equality.

Introduction

The lexical priority of the basic liberties is one of the most distinctive features of John Rawls’s theory of justice. While this aspect of his view has been widely critiqued, it has received less defence, from either Rawls or his supporters, than one might expect. This is surprising, both because of the vast literature on Rawls’s work, and because the issue of whether liberty is to be given some kind of special priority is of great significance for legal and political theory in general. Many theories of justice, including those of liberal, libertarian, and republican stripes, assign some package of liberties special priority.¹ If this widely shared approach is right, it is of great importance to determine which set of liberties should be given priority, what kind of priority they should be given, and why. We should want to know, then, what contribution Rawlsian theory might make to our understanding of these issues.

I believe Rawls does indeed offer useful theoretical resources for thinking about the nature and importance of the basic liberties, but that they have not yet been cast in the most plausible light. In this paper, I examine three arguments in support of the priority of liberty that have been put forward by Rawls and his supporters. The first is based on the two ‘moral powers’; the second on the social bases of self-respect; and the third on a Kantian notion of autonomy. I raise some problems for these views, arguing that none of them adequately defends 1) the distinct claim of lexical priority, for 2) the complete package of basic liberties (including the fair value of the political liberties), on the basis of reasons that are appropriately public. I should note, then, that I am interested in how the priority of liberty might be defended in the spirit of Rawls’s later work, e.g., in Political Liberalism. I will assume that given facts about the diversity of worldviews among members contemporary societies, principles of justice ought to be justified by reasons that can be endorsed from the standpoint of any ‘reasonable’ comprehensive doctrine.²

In the final part of the paper, I propose an alternative argument based on the social or ‘relational’ conception of equality. I characterize the function of the priority of liberty as one of giving people a particular kind of status as equals. The content of this status is developed in connection with the requirement of respecting persons as equal bearers of the moral powers, where this aim is distinguished from bringing about the state of affairs most favourable to the development and exercise of those powers. My account, then, plays two roles: it offers a useful way to characterize and develop social egalitarianism along Rawlsian lines; and in doing so, it provides a plausible defence of the priority of liberty.

The Priority Thesis

Rawls’s account of the priority of liberty changed over time, and examining its development elucidates how he thought the thesis could be best defended.³

In its original statement, the first principle is as follows:

• 1) Each person is to have an equal right to the most extensive basic liberty compatible with a similar liberty for others.⁴

This principle is lexically prior to the second:

• 2) Social and economic inequalities are to satisfy two conditions: 1) They are to be attached to offices and positions open to all under conditions of fair equality of opportunity; 2) They are to be to the greatest benefit of the least-advantaged members of society.⁵

The lexical priority of the basic liberties means ‘they have an absolute weight with respect to reasons of public good and of perfectionist values’.⁶ Two examples illustrate and motivate the kind of trade-offs disallowed: ‘the equal political liberties cannot be denied to certain social groups on the grounds that their having these liberties may enable them to block policies needed for economic efficiency and growth. Nor could a discriminatory selective service act be justified (in time of war) on the grounds that it is the least socially disadvantageous way to raise an army’.⁷

While these examples have intuitive force, it is not clear that they generalize to the extent required to support the lexical priority thesis. Lexical priority is a very strong claim. Why would we not permit small sacrifices of liberty if the gains with regard to other goods were large enough? After all, most would agree that interests in non-liberty goods such as income and wealth also matter (to some extent) for assessments of justice, so wouldn’t we expect there to be some kind rate of exchange between the two?

On Rawls’s own admission, the account of the priority of liberty in the first edition of A Theory of Justice was a ‘serious fault’ of the theory’.⁸ Criticism from H.L.A Hart, Onora O’Neill, and others pointed out significant issues that led Rawls to revise the initial formulation.⁹ The language of a ‘most extensive total system’ of liberty invites several worries. It implies that we should evaluate a social order according to a notion of liberty in the abstract, such that we will judge a regime favourably over another if it provides more ‘liberty as such’. But the notion of ‘liberty as such’ does not capture the importance we place on specific kinds of liberty. Moreover, for any given set of liberties, a ‘more extensive’ set can be invented by individuating options differently.

In response, Rawls changed the principle to read ‘each person is to have an equal right to a fully adequate scheme of equal basic liberties which is compatible with a similar scheme of liberties for all’.¹⁰ Rawls’s ‘adequate scheme’ is a particular package of liberties as follows: ‘freedom of thought and liberty of conscience; the political liberties and freedom of association, as well as the freedoms specified by the liberty and integrity of the person; and finally, the rights and liberties covered by the rule of law’.¹¹ Putting aside the question of which liberties are ‘basic’, we are left with the following question: just what are these liberties ‘fully adequate’ for?

The characterization of ‘full adequacy’ involves Rawls’s political conception of the person and its companion notion of social cooperation. On the latter idea, he writes ‘the notion of social cooperation is not simply that of coordinated social activity efficiently organized and guided by publicly recognized rules to achieve some overall end’.¹² Instead, he claims, that since social cooperation is always for mutual benefit, it involves two elements, which he calls the ‘Reasonable’ and the ‘Rational’. The Reasonable involves a shared understanding of fair cooperation. ‘Fair terms of cooperation articulate an idea of reciprocity and mutuality: all who cooperate must benefit, or share in common burdens, in some appropriate fashion judged by a suitable benchmark of comparison’.¹³ The Rational element of cooperation involves the thought that while social cooperators share common standards of fairness, each may differ with regard to the conception of the good that they are trying to advance. The idea of social cooperation thus attributes to people two capacities, or as Rawls calls them, two ‘moral powers’. Corresponding to the Reasonable, people are viewed as having a capacity for a conception of justice. Corresponding to the Rational, they are regarded as having a capacity for a conception of the good.¹⁴ Rawls’s argument, we will see, is based on the idea that the basic liberties are necessary to ensure ‘fully adequate’ social conditions for the development and exercise of these two moral powers.

Before proceeding, though, it is worth briefly noting how Rawls’s notion of the person-as-social-cooperator differs from alternative conceptions of the person that figure in other theories. Rawls does not view the person as homo economicus, or as solely concerned with their rational self-interest. As he acknowledges, this conception of the person would not adequately support the priority of liberty.¹⁵ Nor does Rawls adopt a conception of the person as having pre-political rights over themselves and their labour. Such a conception may underpin other theories of justice, as found in Locke and Nozick, for instance, but it is rejected by Rawls. Understanding Rawls’s view on the priority of liberty, then, requires understanding its connection to his particular political conception of the person as a bearer of the two moral powers.

Along with the view that the basic liberties are to be understood in connection with his political conception of the person, Rawls adopts what we might call a ‘rule-dependent’ conception of freedom. In rejecting the idea that people have rights to liberties that can be specified pre-politically, or that we are interested in liberty ‘as such’, Rawls holds that the liberties to be protected as a matter of justice must be defined by particular rules, regulations, or conventions. The following passage (which is much indebted to Hart’s critique) helpfully illustrates the idea:

The priority of [the basic] liberties is not infringed when they are merely regulated, as they must be, in order to be combined into one scheme as well as adapted to certain social conditions necessary for their enduring exercise… For example, rules of order are essential for regulating free discussion. Without the general acceptance of reasonable procedures of inquiry and precepts of debate, freedom of speech cannot serve its purpose. Not everyone can speak at once, or use the same public facility at the same time for different ends.¹⁶

In order to serve their purpose, the basic liberties need to be regulated and defined into a coherent scheme. The nature of this scheme is specified by the political conception of the person in possession of the two moral powers. On this picture, what may seem like an infringement of liberty by others lights, is for Rawls a regulation, or definition, of the kind of liberty that matters. Rules that facilitate scheduling and organization – e.g. Robert’s Rules of Order – do not count as infringements of liberty, but only regulations, since they are required for the basic liberties to serve their function. They serve to constitute, not downsize, the liberties they regulate.

There is room on this ‘rule-defined’ conception of liberty to regard even controversial rules as regulations and not infringements. Take for example laws of taxation or those that restrict certain forms of speech (e.g., racial slurs or defamation). On many conceptions of freedom, these cases would count as infringements of liberty, which may be all-things-considered justified in order to better promote liberty overall. On the rule-defined notion of liberty taken by Rawls, however, we may view these policies as regulating, or defining, the content of a ‘fully adequate scheme’ of liberties. For Rawls, people must enjoy a set of liberties that are fully adequate for the development and exercise of the two moral powers. And if this is the point of the basic liberties, we can notice that certain policies that ostensibly infringe on liberty – taxation necessary to allow the worst-off a decent socio-economic minimum, or prohibitions against hate speech that may undermine people’s self-respect – may be required for the basic liberties to serve their purpose. These policies may not count as infringements of liberty, then, insofar as they do not interfere with the fully adequate realization of the relevant liberties.

We can thus identify two roles played by the political conception of the person on Rawls’s view. First, as we will see, the moral powers figure centrally in the justification for the priority thesis; but they play a second role, as we’ve just seen, in delineating which liberties count as ‘basic’. Interestingly for Rawls, ‘the rights of acquisition and bequest, as well as the right to own means of production and natural resources’ are not basic liberties, because ‘they cannot be accounted for as necessary for the development and exercise of the moral powers’.¹⁷

One might wonder, however, whether the liberties Rawls does count as ‘basic’ are truly necessary for the development and exercise of the moral powers. Taken on its face, this claim seems false, since these powers seem to be realized in many societies that fall short of realizing Rawls’s first principle. Perhaps what Rawls means is that the first principle is necessary for all to develop and exercise the moral powers, or to have the best chances of doing so. The next section of the paper evaluates several ways in which this argument has been clarified and defended. But before proceeding we should note two further important clarifications of the priority of liberty thesis.

The first concerns whether the priority of liberty is a merely formal guarantee. In response to this issue, Rawls distinguishes between the basic liberties and their worth as follows:

the basic liberties are specified by institutional rights and duties that entitle citizens to do various things, if they wish, and that forbid others to interfere. The basic liberties are a framework of legally protected paths and opportunities. Of course, ignorance and poverty, and the lack of material means generally, prevent people from exercising their rights and from taking advantage of these openings. But rather than counting these and similar obstacles as restricting a person´s liberty, we count them as affecting the worth of liberty, that is, the usefulness to persons of their liberties.¹⁸

The priority of the basic liberties does not require their equal worth or value across persons, but Rawls makes an exception for the political liberties. The political liberties are required to have ‘equal’ value, in the sense people have ‘fair’ opportunity for political office and influence: ‘the worth of the political liberties to all citizens, whatever their social or economic position, must be approximately equal, or at least sufficiently equal, in the sense that everyone has a fair opportunity to hold public office and to influence the outcome of political decisions’.¹⁹ An opportunity is fair, for Rawls, when similarly talented and motivated people have an equal chance of success.

The second clarification is that the priority of liberty only applies once a society enjoys the level of economic development necessary for all to enjoy the basic liberties. If a society were too poor to provide the basic liberties, then justice may allow restricting liberty, if it were to enable means for their future provision. In this sense, Rawls allows that there are conditions under which justice does not require the lexical priority of the basic liberties. This distinguishes what he calls the ‘special’ conception of justice – when the first principle is given lexical priority – from the general conception, according to which all primary goods are governed by the second principle.²⁰

Three Arguments for the Priority Thesis

As we have seen, the first principle is not concerned with the enjoyment of liberty in the abstract, but instead involves a concern with the basic liberties as a package in connection with Rawls’s political conception of the person. This conception regards persons as in possession of two capacities or ‘moral powers’: first, a capacity to develop, exercise, and revise a conception of the good; second, a capacity to develop, exercise, and revise a conception of justice. This conception of the person is central, not just to Rawls’s account of which liberties are ‘basic’, but also to his defense of the priority of liberty. In this section, I evaluate three variations of the argument put forward by Rawls and his supporters. The first claims that the priority of liberty is necessary to protect the social conditions necessary for exercise and development of the two moral powers. The second is closely related, holding that the priority of liberty is necessary to protect the social bases of self-respect, self-respect being in turn necessary for the development and exercise of the moral powers. The third argument invokes a notion of Kantian autonomy in place of the political conception of the person.

For any defense of the priority thesis to be successful, two conditions must be met. These conditions are essential, because they define the thesis. They are as follows:

Lexicality: the argument must support the distinct claim of lexical priority. It would not be enough to show that the basic liberties have high priority. It must be shown, further, that it is never permissible to make trade-offs between the basic liberties and other goods such as income and wealth.

Completeness: the argument must support the package of basic liberties and not just one or another member. This includes the special treatment of political liberties – requiring their value to be ‘equal’ in the sense that people have fair opportunity to hold public office and influence political decisions. It would clearly not be enough to provide an argument, say, for freedom of conscience, and to simply stipulate that the argument carries over to the other basic liberties.

I am particularly interested, as I mentioned above, in how the priority of liberty might be defended in the spirit of the public reason approach. I will assume that we have reason to prefer a justification that is appropriately public in the sense that it may be endorsed from the standpoint of a wide range of worldviews.²¹ In what follows, I show that the three arguments offered by Rawls and his supporters fail to satisfy these conditions.

The Argument from the Moral Powers

As noted in passing, the idea that the basic liberties are strictly necessary for the development and exercise of the moral powers appears to be false. But this is not quite Rawls’s claim. Instead, he says the basic liberties are needed to ‘secure’ the social conditions necessary for the ‘full and effective’ exercise of the moral powers. In more detail, Rawls makes the following three claims:

• (1) The equal political liberties and freedom of thought are to secure the free and informed application of the principle of justice, by means of the full and effective exercise of a citizen’s sense of justice, to the basic structure of society…

• (2) Liberty of conscience and freedom of association are to secure the full and informed and effective application of the citizens’ powers of deliberative reason to their forming, revising, and rationally pursuing a conception of the good over a complete life.

• (3) The remaining (and supporting) basic liberties – the liberty and integrity of the person … and the rights and liberties covered by the rule of law – can be connected to the two fundamental cases by noting that they are necessary if the preceding liberties are to be properly guaranteed.²²

To assess these claims, we need an account of how exactly particular basic liberties secure or protect conditions necessary for the development and exercise of the moral powers, and what it means for these liberties to be ‘properly guaranteed’ by the others.

Although Rawls does not fully provide this account, Chris Melenovsky offers a helpful reconstruction in a recent paper.²³ On the first claim, he argues that the first two basic liberties – political liberty and freedom of thought – are necessary to secure what he calls ‘conditions of understanding’ and ‘conditions of expression’. These conditions are in turn necessary for the development and exercise of people’s capacity for a sense of justice. The ‘conditions of understanding’ refer to circumstances that allow citizens to understand the perspectives of others. Understanding the perspectives of others is required for people to act on fair terms, which is central to what it means to have a sense of justice. It is not enough to merely understand the views of others, however. Conditions of expression must also be met in that one must be able to voice one’s own view in public. Freedom of thought and the equal political liberties are required to meet these conditions, Melenovsky argues, because they protect people’s ability to form their own views and have them heard by others in the political process.

On the second claim, Melenovsky argues that freedom of conscience and association protect two further conditions: ‘comparison’ and ‘guidance’. These conditions, he claims, must be met for all to develop and exercise a capacity for a conception of the good. Conditions of comparison require that people are able to examine their conception of the good against a plurality of others in a kind of ‘marketplace of ideas’. In addition, conditions of guidance are required: one needs opportunities for joint activity since people do not develop a conception of the good in isolation. Instead, they do so in coordination with others, drawing from the resources of existing conceptions. On Melenovsky’s account, then, by protecting people’s ability to think what they like, and to freely associate with others, the conditions of comparison and guidance are met.

Finally, consider the third claim. On the one hand, Rawls and Melenovsky hold that the first four basic liberties – the equal political liberties, and freedoms of thought, conscience, and association – have special significance because of their close connection with the moral powers. On the other hand, freedom of the person and the rule of law are needed to ‘properly guarantee’ these basic liberties. This means that they protect people from forms of interference which, if unchecked, may impair their enjoyment of the other basic liberties.

This account plausibly motivates the idea that the basic liberties are important for the development and exercise of the moral powers. It remains unclear, however, exactly what it means to ‘protect’ or ‘secure’ the relevant social conditions. Rawls clearly does not think that this involves maximizing the development and exercise of the moral powers on aggregate; instead, he is concerned with the equal enjoyment of the moral powers.²⁴ But this leaves room for other interpretations. The claim could be that the priority of liberty makes it most probable or empirically likely that everyone will be able to exercise and develop the moral powers. Or it could be a matter of ‘reasonable’ expectation. Alternatively, it might mean that the priority of liberty secures these social conditions robustly for everyone, over a range of possible worlds.

On whichever interpretation, though, I contend that the priority thesis is not adequately supported by the considerations above. Insofar as there are other effective means to support the moral powers, it is unclear why trade-offs between these other measures and the basic liberties would be disallowed.

All of the conditions Melenovsky describes – understanding, expression, comparison, guidance – are supported by material resources.²⁵ Sensitivity to the viewpoints of others, self-expression, and exposure to different conceptions of the good – all are advanced by resources that provide various opportunities: e.g., to appreciate culture, learn new languages, travel, support political causes, lobby government, and so on. In some cases, material resources may be relatively more useful for the moral powers. This seems particularly true of political liberty. Given the extreme improbability of any person exercising decisive influence over democratic political outcomes,²⁶ it may make sense to give up the fair value of the political liberties in exchange for greater economic means, at least insofar as one is interested only in securing favourable social conditions for their moral powers. Put another way, suppose a society, or fiduciaries in the original position, can choose to adopt a scheme of less than equal political liberty. Perhaps the option is along the lines of Mill’s plural voting proposal, where the educated receive more votes.²⁷ Suppose, further, that this policy would create a significant surplus of material resources to be fairly distributed. It’s not at all clear why this option would be refused. Since equal political liberty seems relatively less useful than material means, giving up on the fair value of political liberty in exchange for greater material means may provide even more favourable conditions for the equal enjoyment of the moral powers.

One may respond in support of Rawls that without enjoyment of the basic liberties, material resources are of limited use for the moral powers. Extra material resources may mean little to a religious person who lacks the liberty to practice their faith. Many people would not give up their religious freedom for any amount of material compensation.²⁸ But these cases do not generalize to cover the full range of choices required by the priority of liberty. It still remains unclear why some restrictions – i.e., giving up the fair value of liberty to a degree – would not be permitted in exchange for a higher level of material resources.

In sum, then, this argument from the moral powers falls short of satisfying the two conditions, lexicality and completeness. First, it does not justify the particular claim of special priority, because it does not provide an account of why trade-offs between the basic liberties and other goods are never permitted. Second, the argument does not satisfy completeness, since it is unclear why the fair value of the political liberties is required.

The Argument from Self-Respect

A second argument, also found in Rawls’s work and developed by others, appeals to the role of the basic liberties in securing the social bases of self-respect. This argument is closely connected to the moral powers and may be best regarded as a way of amending or strengthening the considerations just discussed, as we will soon see. I will distinguish and evaluate two versions of the argument. The two variants adopt different understandings of self-respect: what I will call ‘confidence-self-respect’, on the one hand, versus ‘standing-self-respect’, on the other. These two conceptions reflect an important contrast that has been drawn in the literature on self-respect, and, some writers suggest, both conceptions are present in Rawls’s thought as well. I argue, however, that neither variant provides a fully satisfactory defence of both the lexicality and completeness conditions.

Rawls tells us that self-respect has two components: first, ‘a person’s sense of his own value, his secure conviction that […] his plan of life […] is worth carrying out’. Second, it involves ‘a confidence in one’s ability […] to fulfill one’s intentions’.²⁹ The reason self-respect understood this way is important – indeed, is ‘perhaps the most important primary good’ – is that without it we do not see our plans and projects as worthwhile. We fall into ‘apathy’ and ‘cynicism’, and so are unable to realize our moral powers.³⁰ For convenience, we can call this conception ‘confidence-self-respect’. For most of what he says, it would seem that Rawls thinks of self-respect in this way, and this is how his view is standardly characterized.³¹

It may be thought, then, that the priority of liberty is necessary to secure the social bases of confidence-self-respect. In support of this claim, Rawls and some supporters emphasize the importance of the status-conferring role of the basic liberties. Confidence-self-respect, it is said, involves a social or recognitional aspect, depending in part on respect or recognition from others.³² A person’s status as a citizen, defined by their enjoyment of the basic liberties, is particularly salient in this regard. An unequal distribution of basic liberties threatens people’s confidence-self-respect, Rawls argues, by ‘publicly establishing their inferiority as defined by the basic structure of society’.³³ By contrast, the basic liberties give everyone a ‘similar’ and ‘secure’ status, which works to mitigate harm to people’s self-respect that might otherwise result from economic inequalities permitted by the difference principle.³⁴ Conferring citizens an equal status through the assignment of the equal basic liberties ‘secures’ or ‘protects’ the social bases of self-respect in that it renders the equal enjoyment of self-respect empirically likely, or a matter of ‘reasonable’ expectation.³⁵

Does appealing to the role of the basic liberties in securing confidence-self-respect do a better job of satisfying the lexicality or completeness desiderata? With regard to lexicality, the argument would seem to suffer from the same problem as the argument from the moral powers previously considered: income and wealth also support confidence-self-respect, and so it is not clear why we would give the basic liberties priority, and not instead favour trade-offs for greater material resources. In addition, confidence-self-respect may derive from many sources, such as membership in groups such as families, religions, universities, networks of friends and colleagues, and so on. These are examples of what Rawls calls ‘non-comparing’ groups: they provide a source of confidence-self-respect for their members that is more-or-less insulated from the effects of other status hierarchies – say, in the way that being a valued member of one’s football club may provide a source of confidence-self-respect that is not undermined by exclusion from the yacht club in the other part of town.³⁶ These sources of confidence-self-respect may likely be supported and proliferated by a greater level of wealth in a society.³⁷ So it seems, then, to be an open possibility that self-respect may be better secured by permitting trade-offs between the basic liberties and economic gains.³⁸

In response, however, it might be argued that there is a stronger version of the argument from confidence-self-respect that survives scrutiny. The reply concedes that the priority of liberty is not strictly the only or most effective way to secure the social bases of confidence-self-respect. The claim, instead, is that it is the only just way, or the best way given the overall balance of reasons of justice.³⁹ One may concede Nir Eyal’s point that assigning liberty no priority and equally distributing income and wealth may indeed be the best way to secure the social bases of confidence-self-respect.⁴⁰ However, it might be claimed further, that there are other reasons of justice not to secure the social bases of confidence-self-respect in this way. Indeed, Rawls would say this policy would be inefficient and so ‘a great misfortune’.⁴¹ Suppose, as Rawls does, that permitting inequality with regard to income and wealth allows for an increase in the overall social product (for example, by providing incentives for the talented to take on difficult but socially useful work). Suppose, further, that the enlarged social product benefits everyone and gives the worst off more income and wealth than they would have under an equal distribution, as the difference principle requires. Since there are reasons of justice to give all more material means, one may argue that if there is a way to secure the social bases of confidence-self-respect that is consistent with allowing growth-producing inequality, then we should favour that solution. Such a solution exists, fortunately, in the form of conferring people a status as equal citizens through the priority of the basic liberties: ‘the basis for self-respect in a just society is not…one’s income share but the publicly affirmed distribution of fundamental rights and liberties’.⁴²

Suppose for the sake of argument that this reply satisfies the lexicality condition. Nonetheless, I doubt it satisfies the completeness condition. It remains far from clear why the particular status conferred by Rawls’s particular list of basic liberties is necessary to secure the social bases of confidence-self-respect. It seems to remain an open possibility that a different package of basic liberties could do an as-good or even better job.

Again, it seems especially hard to establish the importance of fair value of political liberty. It is doubtful whether any departure from the fair value of the political liberties would really fail to provide a secure basis for people’s confidence-self-respect. Steven Wall raises this doubt as follows: ‘for a society like the United States, if it does not provide substantial public funding for political campaigns, then the fair value of the political liberties will not be achieved’; however, he continues, ‘it stretches all credulity to suppose that if it does not do so, then this predictably will express a public message that will wound the self-respect of citizens’.⁴³ If Wall is right, then perhaps confidence-self-respect would be even better secured by giving citizens more money (or lowering taxes) instead of funding political campaigns to the extent required by the fair value the political liberties. People may simply not care enough about political equality for their confidence-self-respect to hinge on it.

More strikingly, there seems to be insufficient reason to reject as a matter of principle that confidence-self-respect could be secured under a plural voting scheme.⁴⁴ It may be, as Mill writes, that

entire exclusion from a voice in the common concern is one thing: the concession to others of a more potential voice on the ground of greater capacity for the management of joint interests is another . . . Everyone has a right not to feel insulted by being made a nobody and stamped as of no account at all. No one but a fool, and only a fool of a peculiar description, feels offended by the acknowledgement that there are others whose opinion, and even whose wish, is entitled to greater amount of consideration than his.⁴⁵

To be sure, we have reason to be much less optimistic than Mill seems to be; we may regard this as close to a wholly theoretical possibility. In most societies we are familiar with, being less educated is strongly correlated with belonging to an ascriptive category: e.g., race, ethnicity, gender, or social class. Under these conditions it seems nearly impossible that Mill’s proposal could count only as an acknowledgement that some may make wiser political choices to the benefit of all. It would seem plausible, then, as some of Rawls’s supporters have pointed out, that equal political liberties may be effective for providing a basis of confidence-self-respect for everyone.⁴⁶ This would seem especially true for societies that have a history of a one-person-one vote rule.⁴⁷ But this falls short of showing that equal political liberty is the best or most effective way of securing confidence-self-respect. We might imagine, for instance, a different kind of plural voting scheme, where the worst off (in the overall distribution of primary goods) are given more votes, and the better off are given fewer. Such a policy might provide even better support for the self-respect of the worse-off, without jeopardizing that of the better-off (they remain better-off, after all). Rawls’s suggestion above is that political equality provides a non-rivalrous basis of comparison, blocking damage to confidence-self-respect that might otherwise be caused by economic inequality. But why not go a step further and provide compensation, in the form of greater political influence, to the economically worse off? Perhaps by making everyone a ‘winner’, in one or the other domain, confidence-self-respect could be even better secured than under political equality.

I conclude, then, that the argument from confidence-self-respect fails to satisfy both the lexicality and completeness conditions. It might be argued, however, that there is an alternative conception of self-respect that offers a more compelling defense.⁴⁸ In the recent literature on self-respect, several philosophers have distinguished an important kind of self-respect that is distinct from confidence-self-respect. Call it ‘standing-self-respect’.⁴⁹ This alternative conception regards self-respect as a matter of regarding oneself as a possessor of a particular standing or set of entitlements – as a person who is owed certain forms of treatment.⁵⁰ A person who lacks standing-self-respect will not be disposed to demand of others the treatment they are owed in virtue of that standing. They will allow others to ‘walk all over them’, and they may regard that treatment as appropriate to them. It might be argued, then, that the priority of liberty is required on the basis that it provides the social basis of an important kind of standing-self-respect.

This argument, however, at least without further development, stills leaves us without a satisfactory account of the completeness condition. The category of standing-self-respect is quite general, and in need of specification. Supposing that Rawls’s list of liberties secures a particular kind of standing-self-respect, we are left with the question of why that particular kind matters. It may be argued that a less extensive package of liberties would secure the kind of standing-self respect required by justice. For example, suppose one thought that persons ought to have standing-self-respect for themselves as Kantian ends.⁵¹ This kind of self-respect would plausibly require some of the basic liberties, such as rights against harm. But it would not seem to require Rawls’s particular list including the equal political liberties. We might wonder why respect for oneself as a Kantian end requires fair opportunity for political office and influence, and it is not obvious how this account would go. Alternatively, one might argue that people ought to have a secure basis for a kind of standing-self-respect that requires liberties not included on Rawls’s list. One might insist, for example, that people ought to enjoy standing-self-respect as bearers of economic rights, such as freedom of contract. Of course, one may argue that once we properly understand the nature and importance of standing-self-respect, we will realize we ought to prioritize a different list of liberties than the one favoured by Rawls. But this account is not at all self-evident. This shows, then, that appealing to standing-self-respect in order to defend Rawls’s thesis is incomplete at best. Without begging the question, we need an account of the importance of the specific kind of standing-self-respect conferred by Rawls’s list of liberties. One way to do this would be to give an account of why persons are owed that specific kind of standing. The argument I give in section IV can be regarded as a way of working this out.

The Argument from Kantian Autonomy

Turning now to the third account, Robert Taylor argues we can satisfy the lexicality condition by appeal to a Kantian notion of autonomy that Rawls invokes in the original version of A Theory of Justice.⁵² On this account, autonomy is regarded as involving freedom from our choices being determined by natural and social contingencies.⁵³ This involves rationality, understood as the ability to exercise authority over one’s desires – resisting some, prioritizing others, and combining various ends into a coherent plan.⁵⁴ Rationality in this sense is regarded as a person’s highest-order interest. The ability to alter, revise, and coordinate one’s ends has lexical importance over any interest in pursuing any specific end. Taylor understands this as a kind of lexical priority of interests, of rationality over pursuit of particular ends, and argues it requires the lexical priority of the basic liberties. ‘If the parties were to sacrifice the basic liberties for the sake of other primary goods’, he claims, ‘they would be sacrificing their highest-order interest in rationality and its preconditions, and thereby failing to express their nature as autonomous beings’.⁵⁵

As it stands, this argument appears to appeal to a similar sort of interest appealed to in the argument from the moral powers, namely, the capacity to reflect on and endorse a conception of the good. We may worry, then, that it is not any more promising: given that income and wealth also seem to matter greatly for the interest in rationality so understood, it is not clear why we would not allow restrictions of some of the basic liberties in exchange for economic gains. Sensitive to this concern, Taylor introduces a final step in the argument. He proposes a more comprehensive notion of what it means for the priority of liberty to ‘express [people’s] nature as autonomous beings’, which appeals to other kinds of lexical priority in Kantian thought – e.g., the doctrine of the right over the good, and the claim that the good will is lexically more valuable than its complements such as talents, excellences of character, or happiness. Understanding the higher-order interest in rationality in the broader context of this more comprehensive set of Kantian commitments, Taylor argues, provides the needed justification for the distinct claim of lexical priority.

But even if this more comprehensive Kantian view can successfully argue for lexicality, it comes at a high cost: it does not appeal to public reasons, as Taylor himself notes.⁵⁶ This justification for the priority of liberty would not be endorsable by people who held incompatible conceptions of the good; e.g., those that did not ‘elevat[e]…rationality over the satisfaction of desire’.⁵⁷ For this reason, Rawls in the later work moved away from the Kantian idea of autonomy and developed the argument in terms of the political conception of the person previously discussed.⁵⁸

The Argument from Social Equality

None of the arguments put forward by Rawls and his supporters, then, provide an entirely satisfactory defence of the distinct claim of lexical priority, for the complete package of basic liberties, on the basis of reasons that are appropriately public. In what follows, I propose an alternative account that focuses on the status conferring role of the basic liberties previously discussed. On the approach developed here, however, the requirement that citizens enjoy the status as equals conferred by the basic liberties is not justified by appeal to that status being necessary to secure social conditions for the moral powers or self-respect. The requirement of equal status is instead regarded as an independent requirement of justice, as suggested by social or ‘relational’ accounts of equality, which I explain next.

The Nature and Importance of Social Equality

The social or relational conception of equality has been put forward as account of what is required for the state to express respect for its members as equals.⁵⁹ This requirement, it is argued, is best understood as specifying the kind of social relations people ought to stand in, and not only how goods such as resources, wellbeing, or capabilities ought to be distributed.⁶⁰ Much of the work on social equality has been occupied with the critique of so-called ‘distributive egalitarianism’, with luck egalitarianism as the main target. The positive account remains less developed, though there have been some recent contributions.⁶¹

The significance of Rawls’s work to this debate is not yet entirely clear. While prominent social egalitarians such as Elizabeth Anderson and Samuel Scheffler cite Rawls as a proponent of the view, it’s not obvious what, if anything, in Rawls’s view is distinctly ‘social’ or ‘relational’. It may seem that his view can be understood perfectly well in more familiar ‘distributive’ terms. Nor is it clear the extent to which we should regard Rawls as giving a satisfactory account of social equality. I aim to make progress on these issues in what follows.

In the recent literature, some suggest we view social equality in terms of the kinds of dispositions people bear toward one another. People relate as equals insofar as they possess dispositions of an egalitarian nature. On Samuel Scheffler’s proposal, for example, they are disposed to take each other’s interests into account equally.⁶² On Niko Kolodny’s account, they are ‘resolutely’ disposed not to interfere with one another.⁶³

On the account I favour, however, social equality involves a somewhat more complex relationship than the kind of interpersonal relations described by these accounts. On my view, social equality is better thought of as a relation between people and political and social institutions.⁶⁴ Relations of equality, in the relevant sense, consist in a civic status shared as equals that is conferred under those institutions. This account involves three components: one objective; another, communicative; and a third, intersubjective, outlined as follows.

Objective: This element specifies the distribution of goods that are constitutive of social equality, or necessary to confer the relevant kind of status as equal citizen. While I will not attempt to give a complete account of this issue here, I argue in what follows that the priority of the basic liberties is among the necessary conditions. Further conditions may also be necessary – which I leave open for the purposes of this paper – such as constraints on inequalities of income and wealth, equality of opportunity, and all enjoying particular capabilities.

Communicative: It is not enough that people are given these status conferring goods on any basis (e.g., their being equally deserving, intelligent, and so on). Instead, these goods are given to people on the basis that they are antecedently equals; and their distribution publicly communicates respect for persons as equals.⁶⁵

Intersubjective. The distribution of goods to people as equals sets a standard for how people ought to regard each other as equal citizens. This will shape and constrain their dispositions and there will be public awareness of the kind of treatment and regard owed to others. Although justice may not be compromised by people failing to fully live up to the ideal in their personal lives in all cases, they will, however, be constrained in political and public forums.⁶⁶

Social equality, understood this way, may be taken to be normatively important on several different grounds.⁶⁷ On my view, the concern with equality of civic status is fundamentally grounded in the requirement that the state respect its members as equals.⁶⁸ No state can be said to show respect for its members as equals if it assigns some, explicitly or implicitly, an inferior status to others in the community. Even if people may be happier, or better-off in some other respect in a hierarchical society, that society would be objectionable on the grounds that it failed to respect its members as equals.

This requirement that the state show respect for its members as equals is of course not foreign to Rawls’s theory. Indeed, Ronald Dworkin argues that it is Rawls’s fundamental concern, pointing to the following passage:

Some writers have distinguished between equality as it is invoked in connection with the distribution of certain goods, some of which will almost certainly give higher status or prestige to those who are more favored, and equality as it applies to the respect which is owed to persons irrespective of their social position. Equality of the first kind is defined by the second principle of justice…But equality of the second kind is fundamental. It is defined by the first principle of justice and by such natural duties as that of mutual respect; it is owed to human beings as moral persons.⁶⁹

This equality of respect that Rawls distinguishes from distributive equality is, however, as Dworkin points out, highly abstract. A wide range of social and political arrangements may be argued to best satisfy it, and one may wonder whether any specification will have to take the form of specifying some equal distribution of goods. The distinctive move made on the social egalitarian approach, on my understanding of the view, is to propose the ideal of equality of civic status as a particularly plausible interpretation of what the requirement of respect for persons as equals involves.

A Rawlsian Conception of Social Equality

The account developed so far leaves open the central question of what kind of status ought to be given to people as equals. It may seem that social egalitarianism is neutral between more versus less demanding conceptions of status. Suppose we were to compare a society in which everyone is equally denied freedom of speech to a society in which all enjoy that liberty. As far as we are concerned only with equality of status, the two societies would seem to be on a par, all else being equal. After all, in both cases, all enjoy freedom of speech equally, and so no person is inferior or superior to others on that basis. Nevertheless, we would judge the second society more just (all else being equal). A satisfactory account of social equality, then, must specify the content of the status to be shared as equals.

How can the content of this status be specified? The idea I want to pursue is that Rawls’s conception of the moral powers offers a helpful and plausible solution. We can develop the account of social equality by asking what kind of status communicates respect for people as equal bearers of the moral powers. In this way, the delineating role of the moral powers is maintained, but in a manner different from the accounts suggested by Rawls and his supporters.

What, exactly, is required to give people a status that expresses respect for them as equal bearers of the moral powers? One response might involve considerations considered earlier; namely, that this would require bringing about the social conditions most favourable to the equal enjoyment of the moral powers or the social bases of self-respect. If this were right, then the account I wish to propose would suffer from the same problems as the arguments previously considered. There is, however, an alternative (and more plausible) way in which we can characterize the relevant requirements of respect.

This alternative characterisation invokes a general distinction between the requirements of respect and the goal of bringing about valuable states of affairs.⁷⁰ This distinction bears on a range of values such as human life, friendship, and autonomy. In these cases, the requirements of respect diverge from what would best promote the value in question. Consider the ‘survival lottery’, where healthy people are randomly selected and killed to have their organs distributed for transplant, thereby saving several lives.⁷¹ This kind of policy may be a good way to promote human life – bringing about more healthy humans – but it would come at the cost of failing to exemplify adequate respect for persons. In a similar way, it may be that in certain cases one can bring about more instances of friendship by failing to respect one’s existing friends; say, if breaking a promise to a friend allowed one to attend a social event where they will make several new friends. In both these cases, the value in question may be promoted, but doing so involves a failure of respect. While these cases involve interpersonal trade-offs, we can imagine intrapersonal cases as well. Interfering with someone’s body – e.g. forcing amputation of Bob’s cancerous toe, against his will, in order to save his life – may promote his life at the cost of failing to show him appropriate respect. Likewise, it may be possible to promote someone’s capacity for autonomy in the long run by forcing them to make good choices; but doing so would fail to respect them as autonomous agents.

This difference between the requirements of respect and bringing about valuable states of affairs has intuitive force, and as we will see shortly, it bears on what it means to respect persons as equal bearers of the moral powers; before proceeding, though, it will be useful to further characterise the notion of respect at issue. What, in general, characterizes the requirements of respect? As a useful starting point, we can think of the kind of respect at issue as a kind of ‘recognition’ respect. Recognition respect, on Stephen Darwall’s classic formulation, consists in ‘giving appropriate consideration or recognition to some feature of its object’.⁷² This is a very abstract notion, leaving open both what counts as ‘appropriate consideration’, and what is to be taken as the relevant object. Sentience, rationality, or some other property of persons are possible objects of recognition respect, as are (arguably) non-persons such as the rule of law. For our purposes, we can regard persons as equal bearers of the moral powers as the relevant object.

This leaves us with the question of how to construe what counts as ‘appropriate consideration’, which admits of many interpretations. An act utilitarian may argue that respect for persons consists in nothing more than weighing everyone’s utilities equally while maximizing the amount of aggregate utility. But this is far from being the only or most obvious interpretation.⁷³ On many other approaches, this interpretation would fail to capture the kind of ‘appropriate consideration’ that respect requires. After all, this interpretation does not capture the intuitive difference between respecting and promoting valuable states of affairs in the examples just discussed involving human life, friendship, and autonomy. These examples suggest, instead, that the ‘appropriate consideration’ that respect involves requires constraints on interference.⁷⁴ As a general matter, I contend, one may fail to give another recognition respect by interfering with or impairing those capacities in virtue of which a person is owed respect. If I owe another respect based on their sentience, I fail to deliver it if I unnecessarily cause them pain; if I owe another respect based on their rationality, I fail to give it if I force my judgement on them.

On the account taken here, then, respect involves two components. First, a specification of the relevant respect-worthy property of persons; and second, an account of what it means to give ‘appropriate consideration’ to persons as possessors of that property. In the cases discussed, ‘appropriate consideration’ requires constraints on interfering with the relevant properties of persons. This characterisation distinguishes the requirements of respect from the goal of bringing about valuable states of affairs, since the two aims may be cross purposes – in the way that respecting a person’s life requires constraints on harvesting their organs for donation, even if doing so would bring about more instances of human life.

Having clarified the requirements of respect as distinct from bringing about valuable states of affairs, we are now in a position to see that it applies perfectly well to the case of persons as bearers of the moral powers. ‘Promoting’, ‘securing’, or ‘protecting’ the social bases of self-respect or the moral powers may be a matter of bringing about the state of affairs most favourable to their equal enjoyment, as on the arguments previously considered. By contrast, respecting persons as equal bearers of the moral powers requires constraining interference with or impairment of those capacities.

What might it mean for the state to impose constraints on interference with people’s moral powers? The natural suggestion, at this point, is that the state imposes such constraints by protecting the basic liberties. The protection of the basic liberties – the ability of its members to think and say what they like, freely associate with others, participate in the political process – allows for the unimpaired exercise of the moral powers. But this is not to say that protecting the moral powers from interference is the best way to bring about the state of affairs most favourable to their equal enjoyment. As we saw previously, this aim might be better supported by trading-off the basic liberties for gains in material goods. Instead, protection against interference serves a different aim: it communicates the appropriate kind of respect for persons; and since this respect is given publicly on the basis of the equality of persons, it confers them a status as equals.

The Conditions Satisfied

Now I will explain how the argument from social equality satisfies the lexicality and completeness conditions. As noted previously, satisfying the lexicality condition requires an account of why the basic liberties have high importance and why it is impermissible to allow trade-offs for gains in material goods. The argument from social equality provides this account where the others have failed; and while lexical priority is a very strong claim, I will offer a suggestion why it might not be as implausible as some may think, in virtue of its specific role of regulating the basic structure.

First, the basic liberties have high importance because, I have argued, they give persons a status as equals which is necessary for the state to meet the fundamental requirement of respecting its members as equals. Second, the basic liberties may not be traded-off because to infringe on them is to fail to give the respect and status appropriate to persons as equal bearers of the moral powers. Notice that if the aim is promotion of favourable conditions for the moral powers or self-respect, then there is no genuine trade-off or sacrifice between different aims of justice: restricting liberty for material gain may simply better promote those conditions. By contrast, there is a genuine sacrifice of people’s status as equals if the basic liberties are restricted for material gain. Consider again Rawls’s example ‘that the equal political liberties cannot be denied to certain social groups on the grounds that their having these liberties may enable them to block policies needed for economic efficiency and growth’. Even if we suppose further that the social groups whose liberty is restricted are compensated by a share of the increased social product, this would not be enough to give them the respect and status due to them as equal bearers of the moral powers. While this example gains some of its force from the fact that there is an unequal restriction of liberty, there would be a similar objection to an equal restriction of liberty. Suppose, for example, that everyone’s liberty to criticize the state were restricted because it would hinder efficiency and growth of the social product. Even if everyone were compensated with a share of the increased supply of material goods, this would come at the cost of the state failing to show respect for any of its members as a bearer of the moral powers. On this basis, if the basic liberties are important in virtue of their status-conferring function, there is a clear rationale for why the basic liberties must not be restricted for gains in material goods.

It is worth noting, however, that it nevertheless may be consistent with the priority of liberty thesis to restrict various basic liberties where they come into conflict with one another. As we saw earlier, ‘the institutional rules which define these liberties must be adjusted so that they fit into a coherent scheme of liberties’.⁷⁵ On the account developed here, we can ask how the various basic liberties can be regulated into a coherent scheme that gives persons a status as equal bearers of the moral powers. So, for example, certain forms of speech – e.g. racial slurs or defamation – may be restricted because they are at odds with all enjoying a set of basic liberties that is adequate for that purpose. Likewise, restrictions on particular religious practices may be justified on the same grounds. Such regulations of the basic liberties in these kinds of cases do not violate the priority of liberty; instead, they ‘adjust’ the basic liberties into a coherent scheme.

One may still wonder, however, why we should assign the basic liberties lexical, and not just high priority. Many may judge that if the gains in terms of other goods were great enough, we would have an all-things-considered reason to restrict liberty, even if that would involve a failure to give respect and status to people as equals. In other words, the relative importance of the concern with respect and status I have suggested, versus the concern we may have with people’s interests in having more material means, is open to doubt. However, if the concern with social equality along the lines characterised is taken to be of high importance, it may make sense to assign the basic liberties lexical priority, given the principle’s particular role: regulation of the basic structure. The lexical priority thesis is not intended, by contrast, as a characterization of the morally best state of affairs. This is significant, because there may be particular considerations that pertain to the former but not the latter kind of account. We may be cautious about whether the principles regulating the basic structure are prone to abuse, taking people as they are. We may want to prevent abuses of restricting liberty for other aims, even if we believe there are limit cases where it may be morally best. If, for example, one could press a button that slightly restricted their neighbour’s liberty for a day, but as a result made everyone in their society extremely well-off, one might think they morally ought to do so. But it does not follow that the priority of liberty should be rejected as a principle that regulates the basic structure. In a similar way, it may make sense to adopt absolute prohibitions against torture, as a matter of policy, even if we believe there to be limit cases where torture is all-things-considered justified.

Turning now to the completeness condition, I argue that the social egalitarian account developed here supports the package of liberties Rawls recommends; and importantly, it provides an explanation why the political liberties are required to have equal value across persons. On Rawls’s original account, as we saw, he suggests a connection between certain basic liberties and the moral powers. Freedom of thought and the political liberties secure the social conditions for a capacity for a sense of justice; freedom of conscience and freedom of association secure the social conditions for a capacity for a conception of the good. On the social egalitarian account, we’ve substituted the aim of ‘protecting’, ‘securing’, or ‘promoting’ with the aim of respecting. But the connection suggested by Rawls remains plausible. Preventing interference with people’s capacity to think and say what they like and participate in the political process expresses respect for their capacity for a sense of justice; and letting people hold whatever beliefs they like, and to associate with whomever they like, expresses respect for their capacity for a conception of the good. The remaining basic liberties – liberty of the person and the rule of law – further constrain interference with the moral powers, and so serve to express this respect as well.

In addition, the argument from social equality does better than the other arguments previously considered along a key dimension: accounting for the equal value of political liberty requirement. We saw that the arguments from the moral powers and self-respect have an especially hard time defending this aspect of completeness. This is because while equal political liberty may to some degree support self-respect or the moral powers, it is by no means necessary, and both interests may be better supported by material goods in a range of cases. By contrast, it is plausible that equal political liberty is necessary for social equality. Several philosophers have in recent work argued for this claim at length.⁷⁶ Here I want to focus on one consideration in particular. Recognizing the status-conferring role of political liberty points to a reason why we would require its equal worth, but not the equal worth of other basic liberties. This is because the status-conferring value of political liberty is strongly positional.⁷⁷ What I mean by this is that the extent to which political liberty confers status on a person strongly depends on the kind of political liberty others have.

To illustrate, consider again Mill’s plural voting scheme, in which educated people are given more votes than others. Greater influence for the educated has in this case a zero-sum aspect. Having one vote in a society in which others have more does not confer the same status as it does in a society in which all have one vote. In Mill’s scheme of plural voting, having only one vote signals that one is judged to have an inferior capacity for judgement in a way that threatens their status as an equal bearer of the moral powers. And a similar issue would hold, even if plural votes were allocated for different reasons – suppose again more votes are given to worse off socio-economic groups. While this scheme would intuitively be less objectionable than Mill’s, it would nevertheless be problematic from the standpoint of social equality. The scheme may serve to make a de facto hierarchy more explicit – the message is that that they are, after all, in a lower class than others.

By comparison, the other basic liberties are not positional in the same way. Consider an inequality with regard to the worth of a different basic liberty: freedom of conscience. Suppose two persons have this liberty but at different levels of worth. One person is relatively poor and can only attend their local church and worship in private, but the other person is wealthier and can afford to go on distant pilgrimages, to make large donations, and so on. Provided that the difference in the value of their liberties is not too great (supposing it is regulated by the difference principle, or in some other way) this inequality would not necessarily fail to express respect for people as equal bearers of the moral powers. It would not send the message that some are inferior with regard to their capacity for a conception of the good. This thought generalizes to freedom of thought and freedom of association. Provided one has adequate means to exercise these liberties, and that inequalities of their worth are kept within limits, we need not require their equal value to respect people as equal bearers of the moral powers.

If the argument of this paper is successful, then, we have given an adequate account of lexicality and completeness. But we have also done so in a suitably public way. This is not to say that the account in entirely value-free – we have appealed to the value of social equality and a specific interpretation of respect for persons as equal bearers of the moral powers. However, unlike the argument from Kantian autonomy previously discussed, this justification may find support from a wide range of conceptions of the good. It does not appeal to reasons that are bound to be controversial among those with diverse conceptions of the good, unlike the way people are bound to disagree about whether rationality is our highest-order interest. The argument from social equality is ecumenical in this sense, claiming that people’s diverse conceptions of the good are entitled to respect. People with disparate conceptions of the good can agree that as a matter of justice each person is an equal with regard to their having a capacity for a conception of the good and a sense of justice; and furthermore, each person is owed respect and a status as an equal on this basis.

Conclusion

This paper has had two complementary aims. On the one hand, I argued that social egalitarian considerations provide better support for the priority of liberty than the arguments offered by Rawls or his supporters. On the other hand, I demonstrated how Rawlsian considerations – a concern with the moral powers – are helpful for filling in the content of social egalitarianism. I have argued that the priority of liberty is a plausible necessary requirement of social equality. This contribution, of course, leaves unsettled the issue of what a fully worked out account of social equality would involve. For example, it remains to be determined exactly how to regulate the value of the basic liberties. We should want to know, for instance, whether the difference principle is satisfactory on a social egalitarian account, or whether some other principle is required. I hope to have provided, however, a plausible construal of social equality along Rawlsian lines that may be able to address these issues with further development.