EQUALITY’S DEMANDS ARE REASONABLE

Abstract

There are various egalitarian moral doctrines. They differ in the requirements they impose on institutions and social practices and on individual conduct. This essay sketches two versions of egalitarian social justice and claims that the requirements they impose should strike us as reasonable, all things considered. One is welfarist egalitarianism, a cousin of classical utilitarianism. This version requires bringing about good quality lives for people and fair (equal) distribution of this good across persons. A notable feature of welfarist egalitarianism is that it accommodates the seemingly antiegalitarian claim that it does not matter in itself how one person’s condition compares to that of another, so a fortiori it does not matter in itself whether or not one person’s condition compares to that of others in the one particular way of being equal. The other version is relational or freedom-oriented egalitarianism, which holds that we should above all ensure that people are free to live as they choose and relate as equals, without social hierarchy. In the latter half of the twentieth century, John Rawls developed a powerful articulation of relational egalitarian justice. This essay sketches the two rival egalitarianisms with a view to showing their respective moral attractiveness and to suggesting that the welfarist version has greater moral attraction.

I. Introduction

What does egalitarianism require? The answer has to be: it all depends. Egalitarianism comes in different flavors. Different versions of the doctrine will impose different demands on institutions and policies and on individual conduct. I shall consider two starkly different families of egalitarian views. One is an offshoot of John Stuart Mill’s utilitarianism, while the other draws from a tradition of thought with Jean-Jacques Rousseau and Immanuel Kant at its forefront. ¹ The first—welfare egalitarianism—holds that we should choose acts and institute policies and practices that properly balance the aims of maximizing individual well-being (that is, having a good quality life) with fairly, equally distributing well-being across persons. The second—relational or freedom-oriented egalitarianism—holds that we should above all ensure that people are free to live as they choose and relate as equals, without social hierarchy.

In this essay, I support both types of egalitarianism, mainly by characterizing them in a way that exhibits their appeal, but I also suggest that welfare egalitarianism is the more appealing of the two doctrines. Along the way, I try to disarm seemingly powerful objections against egalitarianism, especially the following two. (1) The first major objection holds that it does not intrinsically morally matter how one person’s condition compares with that of another, so a fortiori it does not matter whether or not people’s condition compares to that of others in the one particular way of being equal. (2) The second claims that any substantial egalitarianism is excessively demanding in the burdens it imposes on individuals and in the costs achieving equality would exact in terms of being able to achieve competing moral values.

II. Excessively Demanding?

Objection (2) can be dealt with quickly. There is no substantial issue, at the level of fundamental moral theory, concerning overdemandingness. Moral demands made on individuals are overdemanding if they impose more demands than those justified according to correct principles. By the same token, they are underdemanding if they shrink from imposing demands on individuals that are required according to correct principles. ²

To illustrate this point, suppose that Lockean libertarianism is morally correct. ³ In some circumstances, its requirements can be extremely demanding in the sense of being psychologically onerous and difficult to obey. For example, when a cruise ship sinks in freezing water and others have purchased entitlements to the available life preservers, the requirement to respect others’ property rights requires me to give up my life, when I could hit you and steal your life preserver. Here libertarianism requires accepting immediate death when one has available courses of action that would be life-preserving. If Lockean libertarianism correctly balances people’s possibly conflicting interests and specifies what morality permits and requires, all things considered in any possible circumstances, its dictates, even if burdensome, are acceptable.

To complain that in this or that set of circumstances what any proposed moral view requires is overdemanding (or underdemanding), is to object that the proposed view is actually not correct. “Morality rightly conceived and understood requires me to give up my marbles, but that is way too demanding,” is a whine, not a plausible objection against morality rightly understood.

Demandingness might be regarded as an input into the deliberation that identifies fundamental moral principles rather than as an all-things-considered verdict. That a morality is too demanding (or the reverse) would be one consideration among many pointing us in the right direction in our search for principles. The “too demanding” objection could be redescribed as a concern that the candidate morality would be too difficult or onerous for human persons, with our psychology, to obey or, alternatively, that obeying the candidate morality would leave us too little freedom to live as we choose. In this spirit, Allen Wood recommends Kantian ethics as moderately demanding: “[a] Kantian theory of duties does not threaten to be inhumanly demanding on us, as consequentialist or utilitarian theories of moral duty threaten to be.” ⁴

However, as you lower the requirements of duty, imposing less pain or more restriction on the freedom of those commanded, you increase the same requirements on those who would benefit from the commanded acts or omissions. If Dick is not required to help Allen, Allen must accept not getting helped. If Dick is required to harm Allen to help others, morality is demanding that Allen suffer the harm; but if you require Dick here not to harm, morality is demanding that the others who would have been helped, suffer the resulting disadvantage. Which approach should we choose? Morality is about what due consideration for people (including oneself) requires, when the interests of people are at stake and may conflict. Saying that we can make progress toward figuring out what morality requires of us by accepting that it must not be “too demanding” just gestures at the thought that morality dictates requirements on conduct that inter alia fairly resolve conflicts of interests among persons. When Wood observes that consequentialist moralities are “inhumanly demanding,” he is thus only stating that he believes such doctrines are incorrect, not pointing to a distinct reason to regard them as incorrect.

Holding a morality to be “too demanding” in the behavioral requirements it places on those bound to comply with it as a substantial binding constraint on candidate moral principles can seem plausible only if one fails to distinguish moral principles from rules and practices. A moral principle, if true, holds universally and necessarily. For human persons like us, with limited cognitive, volitional, and affective capacities, it is difficult to figure out what moral principles require of us and we may lack motivation to conform to these requirements as we understand them. Hence, there is a potential role for subordinate moral directives, socially imposed, that are easy to understand and administer and that in particular circumstances tend to elicit motivation to comply. ⁵ At this level of rules, overdemandingness is a substantial concern. According to multi-level theory, subordinate levels ideally serve as means to bringing about the greatest achievable fulfillment of the fundamental level principles. A rule at a subordinate level is overdemanding if the requirements imposed on individuals are such as to decrease overall greatest fulfillment of fundamental moral principles as compared to an alternative rule that imposes less-demanding requirements or more generous permissions. (Rules will be criticizable as underdemanding in a parallel way.) A rule is defective if, in the circumstances in which a proposed rule is to be applied, people will not be motivated to comply with it to a sufficient degree so that it functions less effectively than alternatives to bring about fulfillment of fundamental moral principles. Rules, according to multi-level theory, are means to an end and might serve the end well or poorly, but fundamental moral principles are not means to some further end that they might serve well or poorly.

III. Welfare Egalitarianism and Relational Egalitarianism

On the utilitarian track, egalitarianism might plausibly be construed as an axiological doctrine, that is, as a standard for assessing states of affairs or ways the world might go, in impartial terms, as morally valuable. Utilitarianism ranks states of affairs according to the total of individual well-being summed across persons within each state of affairs. ⁶ The metric is that lives are better lives for people. The welfare egalitarian objection to this approach is that we should care not only about boosting the total sum of good quality life, but also just as much about its fair distribution across persons. A maximal obtainable sum of good consisting of cakes and ale for the aristocrats and bread and water for the peasants is morally inferior to a somewhat smaller amount of good for the aristocrats spread more evenly across persons. Ranking possible states of affairs from best to worse is in itself neutral on all questions of morally right conduct and right choice of institutions, laws, and social practices.

Regarding the relation between axiology and doctrines of right, I shall consider views that accept a link between the two as consisting in a significant moral duty of beneficence. Beneficence is a positive duty binding each person to some significant degree to bring about better rather than worse states of the world, impartially assessed. If beneficence is the entirety of morality, we have act-consequentialism, which is the view that one morally ought always to do whatever would bring about an outcome that is no worse than the outcome of anything else one might instead have done. A morality that includes a significant duty of beneficence, though, can consort with a deontological morality of constraints and options. One could choose some acts that would bring about the best reachable outcome in one’s circumstances that are nonetheless morally forbidden on the ground that those acts would violate some moral constraint. Also, there are some acts one could choose that would fail to bring about the best reachable outcome, but that nonetheless would be permissible since morality leaves each person, in many circumstances, moral options to act as one chooses, provided one’s act would not violate any binding moral constraints and would lead to an outcome whose shortfall from the best one might instead have done would be within an acceptable range.

“Beneficence” might not be the ideal term for the omnibus moral requirement to improve the world that I have in mind, for the term “beneficence” may suggest morally optional philanthropy. I instead suppose that duties of beneficence are apt for coercive enforcement, when such enforcement would be effective in securing compliance and any enforcement penalties applied would be proportionate. ⁷

Some moral doctrines worth taking seriously deny that there are any significant duties of beneficence, at least if part of the idea of a duty rising to the level of being significant is that prevention of some violations of it warrants the use of coercive force. The family of Lockean libertarian views falls into this category of moral doctrines, so a reader who embraces a view in this family is unlikely to find anything of interest in this essay. However, the belief that there are no enforceable duties of beneficence is not for the morally squeamish. It implies, just to take one example, that it would be morally forbidden to threaten Arneson with a beating, when he could save others from drowning at small cost to himself, if he does not help out those others no matter how many lives will be lost from drowning if such a threat is withheld.

The link between egalitarianism and some enforceable moral requirement need not be by way of axiology and beneficence. One might hold deontological egalitarian views, so I shall consider also a range of deontological moral views that hold that the fundamental moral requirement of egalitarianism is that we relate as equals, that is, regard and treat one another as basic equals. Relational egalitarianism is also a big-tent doctrine housing a variety of positions. I take the core of this doctrine to be a prescription to refrain from instituting, sustaining, or participating in wrongful social hierarchy or wrongful relations of social inequality.

IV. The Upshot

After this preliminary hemming and hawing, the reader is entitled to be informed as to what main claims will be affirmed and supported in this essay. The main claim I shall urge in this essay is that a plausible, substantial egalitarian moral commitment neither imposes unbearable burdens on individual choices of conduct nor dictates the establishment of tyrannical political institutions. Sensible egalitarianism does not press its followers toward some twenty-first-century version of George Orwell’s 1984. Nor does egalitarianism vigorously pursued have any tendency to drive an egalitarian society toward a gray culture lacking the bright colors of creativity, individuality, diversity, excellence, deviancy, and weirdness. ⁸

I seek to reach this conclusion without cheating, that is, without watering down the content of egalitarianism to a platitudinous message. Egalitarianism, as defended in this essay, favors welfarism’s equality of condition and requires relationalism’s treating everyone the same in a certain respect. This thus makes the egalitarianism this essay defends a version that conservative egalitarianisms abhor.

My defense does not amount to a knockdown argument. On a terrain that has been fought over by philosophers and normative political theorists—and, for that matter, ordinary people talking in coffee shops and pubs for a very long time—decisive proof and refutation are unlikely to be found. The appeal here is to intuitive considerations that may affect our considered judgments in extended reflective equilibrium, that is, what we would believe after reflecting on relevant arguments while thinking straight and trying to find an overall set of consistent beliefs that hang together coherently. Such appeals can always be met with the rejoinder that what strikes you as intuitive strikes me as deeply counterintuitive.

V. Why Care about Equality Anyway? Equality and Priority

Some will find a debunking message in the question, “What does egalitarianism require?” A message lurking in the question is that everyone’s having the same, achieving the same, or being treated the same is neither morally nor prudentially valuable. Hence, even the smallest moral requirements of egalitarianism thought to be warranted by securing the supposed value of equality, would be considered excessive. This sweeping conclusion would hold for candidate moral requirements on institutions and practices as well as on individual conduct.

I follow an ecumenical path in resisting this debunking message. There are genuine, enforceable moral requirements of egalitarianism that fall on institutions, laws, and social practices as well as on individual conduct. There are at least two plausible views as to why equality matters that ground such moral requirements. Both maintain that it is among individual persons—that is, beings with rational agency capacities at or above a threshold level—that equal consideration and concern (for their welfare) or equal respect (for their status and freedom) holds. ⁹ One view sees equality as having intrinsic moral value only if equality is very thinly conceived, which allows it to accommodate much of what its debunkers claim to discern. The other view insists on a robust conception of equality, but nonetheless urges its appeal. There are various robust conceptions of equality, so if any one or a combination of them has appeal, we could find another way to defeat the debunking message. I mention this latter kind of view, though, only to set it aside, for the accommodating view of equality suffices. Even if one could locate a compelling argument in defense of a more robust conception of equality, it would not be needed to support the welfare egalitarian family of views that I seek to advance.

All that being said and eschewing claims of reaching firm conclusions, I suggest that we can locate a good contender by borrowing an insight from Harry Frankfurt, who famously repudiated not only economic egalitarianism, but more generally the doctrine of egalitarianism. ¹⁰ Frankfurt denies that any form of equality is valuable in itself, apart from any possible value it might in some circumstances have as a means to other goals. His position can be supported by a yet broader claim that how well you are doing as compared to others is in itself insignificant. Consider this simple argument, which captures Frankfurt’s position:

1. (1) It does not matter in itself how one person’s condition compares to the condition of any other.

2. (2) If it does not matter in itself how one person’s condition compares to the condition of any other, then it does not matter in itself how one person’s condition compares to the condition of any other in one particular way, namely, whether one gets or achieves the same position as any other.

3. (3) Therefore, it does not matter in itself how one person’s condition compares to the condition of any other in one particular way, namely, whether one gets or achieves the same position as any other.

Having dispensed with egalitarianism, Frankfurt suggests that what does matter for each person prudentially and morally, impersonally regarded, is having enough. Justice requires that each person have fair access to a sufficiently good position.

However, (1)–(3) do not suffice for rejecting egalitarianism. There is a recognizable, sensible, nonparadoxical version of egalitarianism that places no value on everyone’s having or achieving the same position and that also agrees with Frankfurt that it does not matter in itself how one person’s condition compares to that of another. Egalitarianism that rejects equality is prioritarianism, as elaborated by Derek Parfit. ¹¹ Interpreted as axiology, the priority view is a version of distribution-sensitive welfarism, the position that nothing in itself affects the value of any state of affairs other than the total of individual well-being that it contains and the degree to which well-being is fairly distributed across persons in that state of affairs. According to the priority view, to determine the fairness of a distribution of well-being across persons at a time, one needs to know the well-being of persons at other times. To determine the impartially assessed moral value of a state of affairs that will result if one carries out a certain action with a guaranteed outcome, the information one needs consists of the well-being that each individual has in that state of affairs as well as the lifetime well-being that individual will reach, compared to the lifetime well-being that each person would otherwise reach in alternative states of affairs in which an alternative action is chosen.

Prioritarianism as axiology is the claim that a gain in well-being for a person (or avoidance of a loss) is morally more valuable the greater the amount of the gain—and even greater, the worse off in absolute terms the person would otherwise be in lifetime well-being absent this benefit. When it is attached to a duty of beneficence as a component of a moral theory—at the limit, comprising the entirety of morality—prioritarianism holds that the greater the difference in priority-weighted well-being that would result from a beneficent act one could choose compared to what would result from refraining from doing it, the greater the moral reasons to choose and carry out the beneficent act.

Priority never asserts that any agent has any moral reason in any respect, let alone any pro tanto moral duty, to engage in “leveling down.” ¹² To level down is to bring about a more equal distribution across persons worsening the position of some better-off person(s) in a way that brings no gain to anyone else. Those who value equality of well-being across persons, in contrast, will say that leveling down is morally valuable in one respect by increasing the degree to which equality of well-being obtains across persons, but it is bad in another respect by reducing some people’s well-being. In welfare-economics terms, priority incorporates a commitment to the separability of persons; the contribution that any one person’s gain in well-being (or avoidance of loss) makes to the overall value of a state of affairs is not affected by the well-being position of any other person or persons.

Defending egalitarianism by defending prioritarianism, however, may seem to be abandoning the fort. How can one be an egalitarian and care nothing for equality, that is, everyone’s having or achieving the same position?

The priority view can be decomposed into separate elements, one of which is in a way straightforwardly equality-favoring, namely, the Pigou-Dalton norm. ¹³ This norm holds that a transfer of welfare without loss from a person with greater welfare to one with less, provided the transfer does not leave the person who gets the transfer at a higher welfare level than the other and provided no one else’s welfare is thereby changed, makes the resulting state of affairs an improvement.

According to the priority view, then, although equalizing well-being is not intrinsically valuable, bringing about a more equal distribution of well-being across persons always brings about an increase in total priority-weighted well-being so long as the transfer is not accompanied by a loss in well-being. There’s no mystery here. Imagine taking a non-drippy ice cream cone from one person who likes ice cream and is heading for a high level of lifetime well-being and giving that cone to another person who likes ice cream at least as much as the first person but is heading for a lower level of lifetime well-being. Transferring well-being without loss from better-off persons to worse-off persons (without affecting anyone else’s welfare) is necessarily instrumentally morally valuable, according to the priority view.

The Pigou-Dalton norm explains why it makes sense to regard priority as an egalitarian moral principle. As just explicated, priority necessarily favors equality. However, the favoring required to accept the Pigou-Dalton norm is weak. It says nothing about what to do when a transfer from better-off persons to worse-off persons involves some loss of welfare. What moral weight should be given to the size of a welfare benefit that might be gained for one person or another as compared to the appropriate weight that should be given to how absolutely badly off in lifetime well-being the possible recipient will be, absent the benefit being considered? One seeks to identify a “Goldilocks” weighting—neither too much nor too little—but what is that weighting? All we can do is consider a wide range of examples and seek weights that dictate consistent judgments that fit together after reflective scrutiny.

A prioritarian “egalitarian” position will favor upward transfers from worse-off to better-off persons if the change harms the worse-off just a little and benefits the better-off by a sufficiently large amount. The number of worse-off and better-off persons also matters. This result is an intuitive advantage, not a counterintuitive implication, for the position. “Be reasonable,” someone might say to me, if the pain pill in my possession will ease my slight headache slightly for a week but would extinguish my neighbor’s severe headache for six months, where my neighbor’s life is already going much better than mine. The reasonable choice is for me to give up my pill so that it will do much more good for my neighbor.

But someone who affirms equality of condition can also affirm that same reasonable judgment, if she also favors, to some degree, more well-being rather than less. For any prioritarian position that attaches weight to obtaining greater gains for people versus obtaining gains for people who are more worse off, one can formulate an egalitarian position that attaches weight to obtaining greater gains for people versus obtaining increases in the degree to which people become equally well off, such that the two positions will yield the same judgments as to what states of affairs that we could bring about for sure by choice of action would be morally better or worse. ¹⁴ This convergence in implications is not complete, though. There is divergence in some cases involving risky choice (where one does not know for sure the outcome that would result from one or another choice one might make).

To be sure, as Larry Temkin notes, “Equality describes a relation obtaining between people that is essentially comparative. People are more or less equal relative to one another. Extended humanitarianism [which is the same view I here call “prioritarianism”] is concerned with how people fare, but not with how they fare relative to each other.” ¹⁵ True enough. However, if one must impose an indivisible good (or bad) on one of several people and the benefit (or harm) will be the same to the individual, whoever gets it, and there are no further effects except on the person who gets the benefit, the extended humanitarian is logically committed to judging that shifting the benefit to a worse-off person (or the bad to a better-off person) results in a better state of affairs than shifting it to someone else better off (and the reverse for shifting a bad). The extended humanitarian qualifies as egalitarian in a broad sense by virtue of being necessarily, not merely contingently, committed to equalization as instrumentally valuable in these circumstances. In light of the Pigou-Dalton norm—and more broadly, prioritarianism—the idea of “non-relational egalitarianism” is odd but coherent. ¹⁶

Separability of persons matters. For example, in a one-person universe consisting of a lone Robinson Crusoe on an island, the set of equality-of-condition values that Temkin favors will cease to apply. Where there is only one person, there is no relation to others, and so the value of equality does not come into play. But the priority view still can matter. If Crusoe is choosing among risky actions that will affect his lifetime well-being, priority will recommend somewhat risk-averse choices. But we should not defend to the death separability of persons against egalitarians who reject it. Egalitarians and prioritarians have an intramural dispute to settle, but in broad terms, they should be viewed and should view themselves as close comrades. ¹⁷

To recapitulate this section, there are at least two replies to objection (1) identified in Section I above. Recall that the objection claims that it does not intrinsically morally matter how one person’s condition compares with that of another, so a fortiori it does not matter whether people’s condition compares to that of others in one particular way of being equal. The accommodating response insists that priority matters morally, even if equality in any form does not. An alternative response staves off the objection by insisting that if we rightly describe people’s condition—be that welfare, freedom, or some other mode of relationship—we will see that equality in those terms is indeed morally valuable.

VI. What Does Priority Require?

What does egalitarianism require, if the right interpretation of it is prioritarianism? For simplicity, we might assume that requirements of equality apply relative to a country and not with the same reason-giving force across the globe as a whole. This provisional consideration is a very consequential fact, if we should accept prioritarian welfare egalitarianism. Roughly speaking, if the requirements of priority apply within a country, its demands are comparatively modest. This is a consequence of the fact that, in the world today, income inequality across persons within a country is much less than between countries due to vast differences in countries’ mean incomes. ¹⁸ Prioritizing (or equalizing) income per person within each country is prioritizing among persons whose incomes are far less divergent than the incomes among persons across the entire Earth. If selfishness takes the form of being motivated to hang onto what one possesses and prompts more resistance to equalizing redistribution, the more one is called on to relinquish one’s possessions, the more selfish resistance there will be to global than to intracountry application of priority. This in turn will affect the prioritarian rules, norms, and practices we should seek to enforce. Depending on the circumstances, it will be counterproductive to press prioritarian redistribution beyond some point in the face of self-interested resistance.

At the level of fundamental moral principles, however, things look different. There is no welfare egalitarian reason to confine the scope of application of priority or equality only within the borders of each political society. ¹⁹ The same priority for the worse-off that justifies some redistribution from the very well-off to the somewhat well-off within a wealthy country will justify greater redistribution from the very well-off and the somewhat well-off to the worse-off when the world’s population of individuals is considered together. As a psychological matter, we can understand that people of similar ethnicity, culture, language, and ancestry who live under common institutions will be more prone to be generous to each other than will distant strangers who differ from them in these ways. ²⁰ But psychological proclivity is not in itself morally reason-giving. People are people, be they near and homogeneous or distant and heterogeneous.

Nonetheless, if we provisionally accept the widely accepted idea that egalitarian distributive principles are limited in scope to individuals who share common state membership, one can make a plausible case that these principles are tolerably well fulfilled in some real-world institutional arrangements, such as in Scandinavian social democracies. Consider also institutional arrangements along the lines of what John Rawls calls the “basic structure of society.” ²¹

Regarding institutional arrangements, we note that institutions cannot literally dispense individual well-being. On any plausible conception, the major components of a good life for a person almost all involve doings or activity, especially self-directed activity. Institutions can provide individuals with resources, opportunities, and liberties, that is, they can enable the achievement of individual well-being.

A second consequential consideration affecting the implications of priority is that even if there is a short list of fundamental human goods that is the same for all, there are a vast array of ways in which individuals can fashion life-plans that bring about good combinations of these goods. Also, what life-plans make sense for a given person depends on her particular circumstances, including her likely opportunities and the risks she faces, her talents and traits, proclivities, virtues and vices, and the basic bent of her personality. Many of these circumstances are such that the individual herself, although she can be mistaken, is usually better placed than other people to detect them. When society and government, as agent of society, seek to boost individual well-being, they cannot micromanage its production. They must provide general-purpose resources like education, a nurturing upbringing, access to productive employment and money, and support for good health as well as sometimes provide highly individual-specific resources that either the individual seeks or that becomes glaringly obvious she needs whether or not she herself acknowledges the need.

A third consequential consideration for this approach is that low socio-economic status (SES) tends to lower well-being and drag it below the average. This is a crude generalization, based partly on guesswork, and there are exceptions. “The best things in life are free,” as the saying goes, and there is surely something right about this. Wealthy and high-status people can spectacularly fail to attain these best things and by luck, skill, and savvy, poor and low-status people can attain them. But even if the best things are free, they have material prerequisites, which require costly resources. If you get enough of the merely good things, you can still fashion for yourself a good life. Low SES tends unavoidably to have stigma attached to it that consists in visible signs of low status. However, if the absolute levels of poverty and disadvantage that anyone must suffer are lowered, low SES stigma tends to decrease. Also, what there is of it tends to be less debilitating, especially in a culture in which a general disposition to be sympathetic (“There but for the grace of God go I”) and to help the needy exists.

A fourth relevant consideration is that a society can have at most two of the following: low taxes, generous welfare benefits for the worst-off members of society, and no perverse incentives. Social democracy opts for accepting high taxes in a market economy. This has proven to be an imperfect but reasonably successful strategy for improving people’s well-being, with special priority for the worse-off. Perverse incentives are incentives to engage in anti-social acts. If policing and, more broadly, law enforcement are inadequately funded by taxes, the temptation to steal, avoid paying taxes, or strike out against those who irritate us becomes difficult to resist. If we have generous welfare-state benefits going to the very worst-off but a sharp income-eligibility cutoff, people receiving benefits will have an incentive to stay unemployed or underemployed to avoid losing all benefits.

None of this is rocket science; it’s common lore. No doubt, many packages of policies can serve prioritarian aims reasonably successfully. One should not make a fetish of the social-democratic package, for its efficacy may erode. That being said, the fact that societies have sustained policies that arguably serve priority well indicates the feasibility of implementing this form of egalitarianism—and to do so on a national scale.

Nordic social democracy tempers capitalism with equality. There is a trade-off here between maximizing the total of individual well-being summed across persons and equalizing its distribution. Since we seek to maximize the egalitarian social welfare function over the long run, we seek policies that balance gains to presently living people with those to come.

Social democratic institutions seek to equalize not so much by having in place highly progressive tax rates, but more so by imposing high tax rates to generate large funds that are redistributed to achieve greater equality in post-tax income. Such taxes might, for example, take for state use 50 percent of each person’s pretax income, whatever the size of that income, which I count as high taxation. In contrast, progressive income taxes take for state use a larger percentage of a person’s pretax income, the larger that pretax income. The strategy behind Scandinavian social-democratic policies is progressive in its distribution of the large tax funds it collects. For example, tax funds might be used to provide more public parks in neighborhoods with poor residents than in neighborhoods with predominantly wealthy residents. Perhaps tax funds are used to make convenient, low-cost public transportation available to all, but having that option is unsurprisingly far more useful to those who cannot afford to pay for taxis or commuting by car. Public policies also promote a high level of labor-force participation and full employment to increase tax revenues, bring about gains in companionship and solidarity among workmates, and provide the justified sense of self-worth that productive employment tends to spur. Family-friendly policies might center on public provision of high-quality childcare for working parents to ease the strain of parenting and encourage women’s labor force participation, although Scandinavian countries have not done as well as some other countries at drawing women into desirable types of employment traditionally done mostly by men. Primary- and secondary-level public schools secure good learning outcomes for children of low SES parents. Compressed distribution of post-tax income is not matched by compressed wealth distribution. ²²

Market incentives might be deployed in a socialist economy, in which productive resources are for the most part publicly owned—perhaps managed by the state acting as agent of society—or in a capitalist economy, in which productive resources are for the most part privately owned by individuals and groups of individuals. The Nordic social-democracy model opts for the latter. Very roughly, a capitalist structure is justified by priority just in case allowing productive wealth to be privately owned results in greater achievement of the prioritarian principle over the long run.

VII. Priority with Global Scope

If we switch, as I have suggested we should, from regarding priority as applying to each political society in isolation from others to regarding priority as applying across all persons globally (and across time as well, to the extent we can foresee the future), the implications are unclear. Creating social democracy in wealthy countries, for all we know, might be an impediment to achieving the best feasible attainment of priority with global scope.

Working out what it would require to maximize some determinate function of priority-weighted well-being worldwide in given circumstances is a task beyond this essay. How far we could equalize wealth without reducing long-term growth in prosperity is uncertain. If we imagine that all people on Earth were willing to do whatever priority required, an initial guess would be that rich countries should transfer a lot of their wealth to members of poor countries, period. However, if a poor country is unstable and has a weak government, the country will be unlikely to benefit from large resource transfers from wealthy countries. A sudden infusion of wealth will likely spur greater civil conflict and ruinous instability. But a society with a well-functioning state can absorb a sudden infusion of wealth. Consider the discovery of large oil deposits underneath the sea over which Norway claims jurisdiction. That large natural resource windfall has not destabilized the country; the new wealth is available to benefit Norwegians now and in the future.

These examples suggest that large resource transfers to poor countries that are stable and ruled by well-functioning states need not threaten instability nor cause destructive disharmony. It would thus be theoretically possible for wealthy Norway to transfer ownership of its North Sea oil deposits to a poor country with a stable state, perhaps by way of direct ownership transfers to individual poor persons in the country. Any rich country could do the same. This is not a scenario likely to unfold, but our question here is: What would egalitarianism require? The suggested answer is that large resource transfers from rich to poor countries would be feasible, if there were political will to carry them out. According to prioritarianism, if the transfers were feasible, then they would be morally required, and the political will ought to be forthcoming.

We could also imagine concerted efforts by coalitions of nations to strive to bring about agreements between states around the globe sufficient to avert ruinous climate change. Rich countries interested in securing agreement on greenhouse gas emission reductions in poor countries while helping them develop economic wealth could show good faith by offering large resource transfers in return for agreement by the beneficiary country to use the provided resources for low-emission economic development.

There are also other possible mechanisms whereby rich countries could share wealth with poor countries. A rich country could open its borders to greatly expanded immigration from people in poor regions who seek to move and resettle permanently. Absent a political will in the potential host country to open its borders in this way, a society could provide expanded temporary employment opportunities to guestworkers from poor nations. ²³

From claims about what we collectively ought to do, nothing immediately follows about what you or I ought to do. There is not currently a massive commitment of people to implement global priority requirements. But our unwillingness to comply with priority requirements does not show that the requirements are unreasonable nor that they are not collectively binding on us. Progress toward satisfying global priority would require large resource sacrifices from better-off people, but the gains, given the huge disparities in global income and wealth, would include substantial benefits accruing to people far worse off. This would thus not be a case of throwing resources down the drain for little or no benefit. Scaling down from a collective level, you or I could individually sacrifice for similar proportionate gain. When you or I balk at the sacrifice, this looks to be a situation in which coercing us to act as priority demands would be justified, if a coercing agent were ready to hand.

VIII. Welfare versus Liberty?

One might protest that there is no limit to the magnitude of restrictions on liberty that might be inflicted on people to achieve sufficiently large priority-weighted well-being totals. The fear here is that seeking even a marginal increase in overall priority-weighted well-being suffices to justify squashing liberty. This holds even if freedom restrictions are imposed on everyone (or almost everyone). Stuffing everyone in railroad cars headed for the Gulag, with no chance of ever escaping concentration-camp-level unfreedom once one arrives at the destination, would be fine, so long as whatever well-being losses the freedom restriction imposes are offset by sufficient gains in overall well-being. ²⁴

These nightmare scenarios should not stampede us into abandoning welfare egalitarianism. One general observation is that the degree to which freedom restrictions are oppressive—and reasonably experienced and judged by people as oppressive—depends on the extent to which the freedom restrictions are justified by good moral reasons. What is oppressive is not imposition of unfreedom but imposition of unjustified unfreedom. Traffic safety laws can serve to illustrate this point. They enormously restrict one’s freedom to drive as one chooses at whatever speed one prefers on public roads. This is done to facilitate traffic flow into a particular pattern, namely, moving at reasonable speed without traffic jams or the harm of vehicular crashes. Since this immense freedom restriction by and large helps us to arrive safely at chosen destinations at a reasonable cost of time and inconvenience, we’re content with this restriction. Enormous freedom restriction can sit lightly on those restricted, if the freedom restriction is manifestly worth its cost in terms of values—including fair-distribution values—that we have good reason to uphold.

You might object that this first example does not suit the dialectical use to which it is being put, because traffic-safety rules restrict freedom in order to expand freedom. Their point is to facilitate people traveling to wherever they want to go, regardless of whether getting to where they want to go really makes their lives go better.

This objection fails to deflect the force of the example. First, traffic-safety rules aim to promote traffic safety, which certainly restricts people’s desires to drive as fast as they wish, consequences be damned. Some people on reflection would prefer a traffic regime offering far more freedom to do as one likes and less safety. Second, you would have to be cynical in your assessment of the extent to which the satisfaction of people’s ordinary desires to get where they want to go really serves their true interests, to deny that there is a strong welfarist case for traffic safety on any plausible conception of welfare. The point stands: Restriction of freedom that advances and fairly distributes people’s well-being fairly distributed is a good deal from the moral standpoint.

A second example illustrating this point is conscription to fight a just war. We should understand a just war as one that morally must be waged, not merely one that is morally permissible to wage. In such a case, a fair conscription brings about a fair distribution of the total individual sacrifice that is required to carry on the war effort to which all members of society have a duty to assist. In the circumstances, compliance with the coercive orders of the state is a required means of effectively fulfilling a moral duty that one has anyway, namely, to assist the war effort prior to the state’s issuing any commands. Being forced to join an army for the duration of a war effort enormously restricts one’s freedom to live as one chooses, to put it mildly. The sacrifice imposed may involve one’s suffering violent death at a young age. Given the huge costs that waging war imposes on those asked to fight in the war, on those fought against, and innocent bystanders, the expected gains of the war, measured in the currency of justice, must be huge.

A third example is global redistribution on the scale that global application of priority would require. The real (effective) freedom of people in wealthy nations to live as they choose would decrease and the real freedom of people in poor nations would increase. As an affluent individual in a wealthy nation, I would anticipate my valuable options decreasing, but this loss is morally more than offset by increased valuable options to those now just scraping by in comparison with the outcomes of alternative policies. If this weren’t the case, priority wouldn’t approve the transfers. If the morally best transfers aren’t politically feasible, due to political attitudes of affluent voters (or whoever are pivotal deciders), priority favors the feasible best.

Egalitarianism is disparaged on the following grounds. Equalizing people’s condition will in practice unavoidably press toward substantial sameness of condition, with everyone sharing the same way of life. Also, coercive state policies to promote substantial movement toward equality will provoke widespread dissent and opposition, the crushing of which will squash civil liberties. In reply: equalizing people’s resource holdings by way of increasing the resources available to the worse-off, even if it takes from the better-off, gives resources to those more in need of them. There’s no reason to expect this transfer to undermine multifaceted individual development overall. Also, while civil liberties would be threatened by protracted civil war, there’s no reason to expect such an outcome because prioritarian transfer of resources would not be feasible unless the vast bulk of people internalized that ideal and supported its fulfillment.

Regarding basic civil liberties, we should notice that, beyond their instrumental value for maximizing the right balance of individual welfare fairly distributed, they have another instrumental value, namely, to serve moral progress. Being fallible, we should acknowledge the possibility that our current best judgments as to what candidate fundamental moral principles are correct might be wrong. We want to sustain conditions that will help us in the future to arrive at better views—maybe priority is wrong and “schmiority” correct—if such there be. Freedom of speech, thought, and association ought to be sustained, in part, as means to future progress in moral knowledge.

IX. What Rawlsian Relational Equality Requires

There is a deep divide between welfare egalitarian doctrines and views that see the fundamental imperative of egalitarian justice to be one of the following: (1) establishing and sustaining a society of social equality that avoids bad social hierarchies of wealth and power; (2) achieving equal freedom as nondomination; or (3) equalizing the secure enjoyment of basic liberties for all, including political liberties construed as requiring equal opportunity for political influence.

In the ringing first paragraph of A Theory of Justice, John Rawls states, “Each person possesses an inviolability founded on justice that even the welfare of society as a whole cannot override.” ²⁵ This seems to promise an uncompromising stand for the worst-off, but it turns out that when you read the fine print, each person possesses rights to liberties that even the welfare of the very worst-off person cannot override. Moreover, the “welfare” of the worst-off members of society recognized in Rawlsian principles is not actually welfare, but rather an index of the social and economic primary social goods exclusive of the equal basic liberties whose protection takes strict priority and allows no trade-offs. ²⁶ Individual well-being as such is not a justice value at all, not even a minor one. According to Rawls’s second principle of justice—the “difference principle”—achieving greater resources such as income and wealth for the worst-off should be a factor affecting the design and operation of basic institutions only after we have done all that we can do, as a first priority, to secure and protect equal basic liberties.

There is another strict lexical priority nested in Rawls’s two principles of justice that limits sharing resources with the worst-off. One value has lexical priority over another when one should accept any loss, however tiny, in fulfillment of the superior value to achieve any gain, however huge, for however many people, in fulfillment of the inferior value. Rawls’s equal liberties principle, which has lexical priority over his second principle, requires that “[e]ach person has an equal claim to a fully adequate scheme of equal basic rights and liberties, which scheme is compatible with the same scheme for all; and in this scheme the equal political liberties, and only those liberties, are to be guaranteed their fair value.” ²⁷ The equal basic rights and liberties of persons are those needed for the development and exercise of their fundamental powers to play fair with others and to develop, assess, perhaps revise, and pursue a conception of their good. The second principle rounding out this theory of justice requires that “[s]ocial and economic inequalities are to satisfy two conditions: first, they are to be attached to positions and offices open to all under conditions of fair equality of opportunity; and second, they are to be to the greatest benefit of the least advantaged members of society.” ²⁸

Two strong equality of opportunity principles are lodged in Rawls’s doctrine. The fair value of the political liberties principle (FVPL) requires that all citizens with equal political talent and ambition have the same chances of being politically influential, of being elected to public office, and of affecting political decisions. The fair equality of opportunity principle (FEO) requires that all those with the same native talent and ambition for competitive success have the same chances of competitive success to attain positions that accord one greater social and economic primary social goods. Again, primary social goods are general-purpose resources and liberties that are necessary, or at least specially strategically useful, for developing and exercising the two fundamental moral powers. FVPL is lodged in the first-priority equal-liberties principle and, within the second principle, FEO has lexical priority over the difference principle.

In Rawls’s work, his moral revulsion to anything that smacks of utilitarianism goes further. For example, he notes that any inequalities in access to social and economic primary goods across persons (1) must be attached to positions that satisfy FEO and, as a second priority, (2) must work to the maximal advantage of the worst-off. But a society can fully satisfy the second principle simply by not allowing any social and economic inequalities at all, within the constraint that some such inequalities might be byproducts of complete fulfillment of equal basic liberties. The Rawlsian just society is not bound by any moral requirement to introduce policies that will bring about inequalities that increase the social and economic primary goods holdings of the worst-off under conditions in which FEO obtains. Provided the opportunity for development and exercise of the two moral powers is fully secured, any pursuit of greater economic growth and prosperity beyond that point is morally optional. So long as sufficient economic resources are created at some time, then for each succeeding generation, there is no moral reason, much less requirement, to make the economic pie bigger. ²⁹

While many questions of interpretation remain to be settled, for purposes of this essay, I want to mark the stark contrast between the family of welfare egalitarian views and Rawlsian egalitarianism. The latter is entirely an egalitarianism of freedom, not of welfare. It’s not that equality of welfare and larger rather than smaller totals of individual welfare are not high-priority justice values in Rawls’s scheme. Rather, on his view, these welfare values do not register at all as justice values. Moreover, whereas one might regard equal opportunity for welfare or real freedom to achieve individual well-being as important justice values for an egalitarianism of freedom, these welfarist freedom norms do not have any place at all in the Rawlsian ideal of justice. Rawlsian justice is thus profoundly Kantian in that its egalitarianism does not represent a compromise between the liberalism of Kant and Rousseau, on the one hand, and the liberalism of Mill, on the other. Rather, the Rawlsian ideal squashes Mill’s ideal of the just society to make room for a social-equality vision of freedom. This shows up, for example, in the fact that the full development and exercise of the fundamental moral powers requires development and exercise of the moral power to develop and critically assess and pursue one’s conception of the good, but there is no requirement that one attain any degree of fulfillment of one’s good. In principle, the fullest flourishing of equal Kantian freedom can coexist with zero welfare—or even negative well-being—for some or all. For a Kantian, this is not a bug but rather a feature of the doctrine.

I now turn to the question: What does Rawlsian egalitarianism require? What institutions must we build and what policies must we institute, according to this conception of justice? As we did with welfare egalitarianism, we consider first the requirements of Rawlsian egalitarianism for a single political society regarded in isolation and then for a planet in which people are ruled under different political societies that control different territories and resources. The contrast between the requirements of egalitarianism applied country by country and applied to the world as a whole is less consequential for a Rawlsian than for a welfare egalitarian. Justice for a Rawlsian likely requires less equalization of income and wealth across the members of separate political societies than it does for a welfare egalitarian (or prioritarian).

Rawls suggests that equal basic liberties will be familiar liberal freedoms of speech, thought, and association along with the right to rule of law, freedom from assault and harm, the freedom to own personal property and use it as one sees fit without imposing certain external costs on others, and equal democratic say over forming public policy. The last we can also describe as equal political liberties, which incorporate free speech, association, and assembly as needed for political democracy, institutions of representative democracy, and democratic accountability. FVPL requires that, with respect to these political liberties, each citizen has equal opportunity to be politically influential as specified above. This is an extremely demanding requirement, to put it mildly.

Rawls is describing an ideal case of a well-ordered society in which all accept the same correct principles of justice, the basic structure of institutions fulfills the principles, all are disposed to support the just institutions and fully comply with their requirements, and all of this is common knowledge. Rawls says this is not an unrealistic ideal, but rather a “realistic utopia”; however, we should not wait for its arrival. Given the laws of nature and the truths about human nature, Rawls does not rule out attaining that ideal. Furthermore, given where we are, there may be no way to get from here to there—and getting there might be a transitory achievement. I will thus assume a simple view about what Rawls’s principles of justice require of us in actual conditions: that we take effective steps, without trampling on anyone’s basic rights, to achieve over the long run the greatest degree of fulfillment of Rawls’s principles in our political society. Rawls’s lexical priorities provide clear guidance. We should seek as a nonnegotiable first priority to do all we can to bring about the greatest possible fulfillment of the equal basic liberties principle, including its FVPL component, and to keep pouring resources into this aim up to the point at which further expenditure of resources would not produce even a marginal gain. Only at that point should we switch gears and put any resources at all into attempts to bring about fulfillment of the lower priority FEO principle. And at that point, we must keep pouring more and more of our remaining resources into boosting the degree to which FEO is fulfilled, until further expenditure of resources would deliver no extra degree of fulfillment. Only then should we channel remaining resources available into attempts to boost fulfillment of the difference principle.

How this approach would play out in likely real circumstances depends on the degree to which the lexical priorities prove to be binding constraints. There is reason to think that they would be strongly binding constraints. That is to say, if we relaxed the priority of equal basic liberties so that it requires only a tolerably good degree of fulfillment, we could then have available enough resources to deploy them efficiently and bring about substantial fulfillment of FEO. If the priority for FEO over the difference principle were similarly relaxed, we could then perhaps still have available enough resources to deploy them efficiently so as to increase the degree to which social primary goods resources are accessible to the worst-off.

This case may be too abstract to wrap one’s mind around, so here is one with more detail. Suppose that FEO cannot be fulfilled. In such a case, we are limited to protecting equality in people’s access to money and other primary social and economic goods, even though everyone would be better off—and the worst-off made as well off as possible with great gains for all achieved—if we allowed inequalities that work to make the worst-off as well off as possible, and then the second worst-off, and so on, up to the best-off. We must thus all live in tents rather than houses, travel on foot rather than in cars and trains, forgo all access to great music and investments in technology that would improve the future for all, and so on.

On the next level, in order to gain ever-smaller increments of FVPL, we must give up any degree of fulfillment of FEO. These forgone gains might involve, for example, greatly increased accessibility to higher education for people whose parents are of low SES or improvements in the egalitarian socialization of men and women so that FEO is better fulfilled due to minimizing the hoarding of positions of advantage caused by an old boys’ network. Rawls might surmise as likely that improvements in fulfillment of his two equal-opportunity norms rise and fall together, so that trade-offs do not arise. Be that as it may, I’m focusing on the possible—and for all we know, likely—cases in which trade-offs exist.

Lexical priorities are only implausible where, on reflection, the good deemed superior is not so much superior to a lower-ranked good as to render the implications of lexical priority unacceptable. We might then ask what is involved in attaining greater equal opportunity for political influence, as Rawls conceives it. Equal opportunity for political influence is compatible with wide disparities in people’s ambitions to be influential, and hence in people’s participation in political deliberation and activity. Suppose that FVPL is achieved by changes in economic organization, tax law, and inheritance law that result in equalizing people’s wealth and income and that, in this setting (perhaps surprisingly), this equality causes people to become lethargic and apathetic about political issues, leading to a deterioration of the moral quality of political decisions. Society becomes less just along several fronts, but the lexically prior equal liberties, including FVPL, are better fulfilled. The moral quality of political decisions does not register as a value that might warrant some sacrifice of equal basic liberties fulfillment.

In contrast, a welfare egalitarian should say, regarding equal opportunity for political influence, that we should seek whatever political arrangements would best promote fulfillment of welfare egalitarian justice values over the long term. Such promotion might come about by improving the quality of political decisions, but it might come about in other ways as well. Greater FVPL fulfillment might stimulate public-spiritedness and increase citizens’ disposition on the whole to cooperate fairly with others, including in ways that result in more fair distribution of well-being or increased prospects for well-being across persons. In some circumstances, welfare egalitarianism would favor greater fulfillment of FVPL than would Rawlsian principles. This could happen if greater fulfillment of FVPL boosted the fulfillment of welfare egalitarian principles, but did so at the cost of fulfillment overall of the equal basic liberties principle. Perhaps protection of free speech becomes slightly more lax and there is less-secure fulfillment of free speech. But from a welfare egalitarian perspective, this loss is outweighed by gains in FVPL fulfillment, which in turn boosts overall welfare.

In other cases, a welfare egalitarian judgment would go the other way. For example, it could happen that equal opportunity for political influence is increased and, in this setting, those who are more politically talented and ambitious than others are induced to exercise this opportunity robustly, so that the politically talented and ambitious now wield far greater influence than they would have, had their opportunities for influence been less. Again, the upshot might be that choices of public policies become more unfair, skewed to the interest of the politically talented and ambitious. From a welfare egalitarian standpoint, the upshot of increased equal opportunity for political influence could be decreased achievement of social justice.

Another consideration is that in populous democracies, individuals’ opportunities to be politically influential will be very slight, so that differences in the degree to which equal opportunity for political influence prevails may involve moving most people’s chances of being politically influential from something like one in a billion to two in a billion. This difference may reasonably not register as important in any citizen’s overall set of values. ³⁰

There may be many devices and strategies that, in our present and likely future circumstances, would help to increase the degree to which equal opportunity for political influence prevails in a political society. However, there may be sharp limits to preventing the degree to which, when inequality of wealth exists, there is unequal opportunity for political influence. It may be, as many political theorists—including adherents of Rawlsian principles—surmise, that substantial progress toward fulfillment of equal opportunity for political influence cannot be made without substantially equalizing people’s wealth holdings. In this case, provided that equalizing wealth will have some positive effect in boosting equal opportunity for political influence and provided that there are no unintended consequences that diminish fulfillment of equal basic liberties overall, Rawlsian justice will require equalizing wealth.

Moreover, Rawlsian justice will require equalization of wealth, even when this produces only an increase in FVPL, no increase in other justice values, and when lower-priority justice values suffer decreased fulfillment. This does not even mention decreased fulfillment of the welfare egalitarian justice values that are not considered at all in the Rawlsian social justice bargaining framework.

X. Welfarist and Nonwelfarist Egalitarianisms: Concluding Remarks

The previous section emphasizes criticism of Rawlsian and, by implication, other nonwelfarist egalitarianisms, for bending too far away from utilitarianism. ³¹ But in an ecumenical spirit, I note that you can be attracted to this bending while staying within the egalitarian fold. The relational egalitarian opposes social hierarchy and seeks a society in which none has power or authority over others that is neither voluntarily accepted nor democratically regulated. Her ideal of social justice is not the promotion of well-being, but rather protection of each person’s freedom to set her own ends and pursue them with adequate means. With a fair framework for interaction in place, how well your life goes for you, from the relational egalitarian perspective, is your own business and not the proper concern of society or of government as its agent.

In contrast, the welfare egalitarian looks beyond people’s share of resources and opportunities to register the quality of the lives people end up living. Resources and opportunities are regarded as means to bringing about what ultimately matters, namely, good quality lives with the good fairly distributed across persons. There’s a vast divide in normative commitments here, to be sure, but one will find appealing some form of substantive egalitarianism on both sides of that divide. Ultimately, I have argued in this essay that the requirements of plausible versions of egalitarianism for public policy and individual conduct are reasonable. In painting a picture of a society that fulfills egalitarian ideals, I depict a society you would reasonably want to inhabit. Hugo Chavez’s Venezuela or Mao’s China, it’s not.