Equality and distributive justice

Almost everyone professes to believe in equality; the disagreements begin the moment we ask equality of what? That question, made famous by Amartya Sen's 1979 lecture of that title, organises the modern debate. To be an egalitarian is not yet to have a view: one must say which 'currency' of advantage justice requires us to equalise — welfare, resources, capabilities, opportunity — and what room to leave for individual responsibility.

Two great divides structure the field. The first is over the metric: Ronald Dworkin argues for equality of resources, against equality of welfare; Sen and Martha Nussbaum argue that what ultimately matters is people's capabilities — the real freedoms they have to do and be valuable things — not the resources they hold. The second divide is over the role of choice. The family of views known as luck egalitarianism (Dworkin, Richard Arneson, G.A. Cohen) holds that justice should neutralise the effects of brute luck — the circumstances we did not choose — while holding people responsible for their genuine choices. Against this, relational egalitarians (Elizabeth Anderson, Samuel Scheffler) argue that equality is not fundamentally about the pattern of any distribution at all, but about people standing to one another as social and political equals.

This note works through the metric debate (welfare, resources, capability), the luck-egalitarian distinction between brute and option luck and the machinery Dworkin builds on it, and the relational challenge that the whole 'distributive' picture misconceives equality's point. It is the natural sequel to Rawls — luck egalitarianism can be read as radicalising Rawls's thought that natural and social fortune is 'arbitrary from a moral point of view' — and it picks up Sen's capability critique of Rawls's primary goods, and Cohen's critique of the difference principle, that you met there.

The debate is a child of A Theory of Justice. Rawls had measured advantage in primary goods and had grounded the difference principle in the thought that the distribution of natural talents is morally arbitrary. Both moves provoked a generation of refinement and dissent.

The metric question was opened by Amartya Sen's Tanner Lecture 'Equality of What?' (1979), which argued that neither utility nor Rawlsian primary goods is the right measure, and proposed instead a focus on human capabilities. The responsibility question was sharpened by Ronald Dworkin's two-part article 'What is Equality?' (Philosophy & Public Affairs, 1981), whose Part 1 rejected equality of welfare and Part 2 defended equality of resources, introducing the now-standard distinction between brute luck and option luck. Richard Arneson ('Equality and Equal Opportunity for Welfare', 1989) and G.A. Cohen ('On the Currency of Egalitarian Justice', 1989) generalised the underlying idea into what would later be called luck egalitarianism: the view that it is bad — unfair — for some to be worse off than others through no fault or choice of their own.

The label 'luck egalitarianism' was in fact coined by a critic, Elizabeth Anderson, in 'What Is the Point of Equality?' (1999), which mounted the most influential attack on the whole programme and revived a relational conception of equality — equality as a matter of how people stand to one another, not of how much they each have. Samuel Scheffler ('What is Egalitarianism?', 2003) developed the relational view further. The field today is a three-cornered contest between resourcist, capability, and relational understandings of what equality requires.

Equality of what? Three currencies

The first question is the metric or 'currency' of egalitarian justice — the thing justice asks us to distribute equally (or to attend to in distributing).

• Equality of welfare holds that what matters is people's well-being, happiness, or preference-satisfaction. Its fatal difficulty is expensive tastes: if we equalise welfare, a person who has cultivated a craving for plovers' eggs and pre-phylloxera claret must be given more resources than someone content with simple fare, which seems unfair to the cheaper-to-please. Equalising welfare appears to reward extravagant preferences.
• Equality of resources (Dworkin) holds that justice equalises the means — the external resources and, suitably handled, the personal endowments — people have, and then lets welfare fall where individuals' own choices take it. People are held responsible for their tastes and ambitions, but compensated for disadvantageous circumstances.
• Equality of capability (Sen, Nussbaum) holds that neither resources nor welfare is fundamental: what matters is people's capabilities — the real freedoms they have to achieve valuable 'functionings' (being nourished, healthy, educated, able to take part in community life). Resources are mere means whose value depends on what each person can convert them into, and conversion varies enormously.

Brute luck and option luck

The second organising idea is responsibility, captured by Dworkin's distinction between two kinds of luck. Option luck is how deliberate, avoidable gambles turn out — risks one chose to run and could have declined (a bet, a speculative investment). Brute luck is how risks that are not deliberate gambles fall out — being born with a disability, or with few marketable talents, or into poverty. The luck-egalitarian thought is that justice should compensate for brute bad luck but not for the results of option luck: it is unfair to be worse off through circumstances one did not choose, but not unfair to bear the foreseeable costs of one's own gambles. Dworkin captures the upshot in the slogan that a just distribution is 'ambition-sensitive' (it tracks the different choices people make) but 'endowment-insensitive' (it does not track the unchosen advantages they were born with).

The responsibility cut

What unites Dworkin, Arneson, and Cohen — the luck egalitarians — is therefore not a shared metric but a shared structure: a cut between circumstance (for which we are not responsible, and whose effects justice should neutralise) and choice (for which we are responsible, and whose effects we must bear). Cohen put the point provocatively: Dworkin 'has, in effect, performed for egalitarianism the considerable service of incorporating within it the most powerful idea in the arsenal of the anti-egalitarian right: the idea of choice and responsibility'. (Dworkin, though routinely treated as the founder of this family, in fact later rejected the 'luck egalitarian' label itself, while his theory keeps its structure.) The deepest objections, examined below, attack precisely this cut — either as too harsh, or as misconceiving what equality is for.

Sen, 'Equality of What?' (1979)

The lecture that framed the metric debate. Sen argues that utilitarian equality (equal weight to utilities) and Rawlsian equality of primary goods both fail, the latter because primary goods are means whose worth depends on people's differing abilities to convert them into functionings. He introduces basic capability as the relevant space. Developed at length in Inequality Reexamined (1992) and Development as Freedom (1999).

Dworkin, 'What is Equality?' Parts 1 & 2 (1981)

Two articles in Philosophy & Public Affairs. Part 1: Equality of Welfare argues that no version of the welfare metric is defensible, principally because of expensive tastes. Part 2: Equality of Resources sets out the positive theory: the hypothetical auction with its envy test, the brute luck / option luck distinction, and the hypothetical insurance market for handling unequal talents and handicaps. Collected, with later essays, in Sovereign Virtue: The Theory and Practice of Equality (Harvard University Press, 2000).

Cohen, 'On the Currency of Egalitarian Justice' (1989)

A pivotal article ((1989) 99 Ethics 906). Cohen argues the right 'currency' is neither Dworkin's resources nor pure welfare but access to advantage — a heterogeneous mix — and that the fundamental cut is responsibility, not Dworkin's resources/welfare line; he presses the expensive-tastes case against Dworkin where the taste was not deliberately cultivated.

Anderson, 'What Is the Point of Equality?' (1999)

The major critique ((1999) 109 Ethics 287). Anderson names and attacks 'luck egalitarianism', pressing the harshness and disrespect objections, and defends democratic (relational) equality: equality as a relationship among citizens, not a pattern of holdings. Samuel Scheffler's 'What is Egalitarianism?' ((2003) 31 Philosophy & Public Affairs 5) is the companion statement of the relational view.

1. The expensive-tastes objection to equality of welfare

Dworkin's case against welfare metrics. If justice equalises welfare, then someone who has expensive tastes — who can only reach a normal level of satisfaction with caviar and vintage wine — is entitled to more resources than someone with modest tastes, simply because it costs more to make them equally happy. That seems perverse: we ordinarily think people should take responsibility for their preferences and trim their ambitions to their fair share. Equality of welfare, by subsidising costly tastes, gets responsibility exactly backwards. (Cohen complicates this: where an expensive taste was not deliberately cultivated — an involuntary craving — it looks more like bad brute luck than a choice, and the simple verdict wavers.)

2. Dworkin's auction and the envy test

To give content to equality of resources, Dworkin imagines immigrants arriving on a desert island with its resources to divide. Each is given equal purchasing power (Dworkin's 'clamshells') and the resources are sold by auction. The result is just when it passes the envy test: no one prefers anyone else's bundle of resources to their own. The auction makes shares ambition-sensitive — people bid according to the lives they want to lead — while equal starting power keeps them fair.

3. Brute luck, option luck, and the hypothetical insurance market

The auction handles chosen ambitions but not unchosen endowments: some people are born with disabilities or with few marketable talents, and the market will leave them far behind through no fault of their own. Dworkin's device for this is a hypothetical insurance market. We ask how much insurance, against the risk of such handicaps or low talent, people of average risk-aversion would have bought from behind a thin veil that hides their actual endowment; redistributive taxation then mimics the premiums and payouts of that imagined insurance. This converts what would be uninsurable brute luck into the moral equivalent of option luck, making the distribution endowment-insensitive without abandoning individual responsibility for choices.

4. The conversion problem and capability

Sen's central argument against resourcism. Even equal resources do not yield equal real freedom, because people differ in their ability to convert resources into functionings: a person with a disability, a pregnant woman, someone with a high metabolism or living in a harsh climate needs more resources to achieve the same beings and doings. To fix on resources (or primary goods) is to stop at the means and ignore the variable rate at which means become valuable ends. Hence Sen's insistence that the proper space of equality is capability — the freedom to achieve functionings — and Nussbaum's proposal of a definite list of central capabilities as the basis for constitutional entitlements.

The capability approach

Sen's question grew into a research programme. Capabilities are the real opportunities a person has to achieve valuable functionings (states of doing and being); the approach evaluates lives and societies by the freedoms people actually enjoy, not by the resources they hold or the utility they report. Sen keeps the list of important capabilities deliberately open, to be settled by public reasoning. Martha Nussbaum (Women and Human Development, 2000; Creating Capabilities, 2011) instead defends a definite list of ten central capabilities — life; bodily health; bodily integrity; senses, imagination and thought; emotions; practical reason; affiliation; other species; play; control over one's environment — which she argues every decent society must secure to a threshold for each citizen. The approach has been enormously influential beyond philosophy, shaping the UN Human Development Index and the measurement of poverty as multidimensional deprivation.

Relational (democratic) egalitarianism

The most important development on the other side is the relational turn. Elizabeth Anderson and Samuel Scheffler argue that egalitarians have been seduced by a false picture on which equality is a distributive ideal — a matter of equalising some quantity across persons. The true point of equality, they hold, is social and political: to abolish relations of domination, hierarchy, and oppression, and to establish a community in which people stand to one another as equals. On this view equality is fundamentally a property of relationships, not of distributions; distributive measures matter because, and insofar as, they are needed to sustain a society of equals (a guaranteed minimum so that no one is subject to others' will, not the elimination of every unchosen inequality). Anderson calls her version democratic equality. The relational view reframes the whole debate: the question is no longer 'equalise what?' but 'what distribution is required for citizens to relate as equals?'.

Anderson's two objections to luck egalitarianism

Anderson's 'What Is the Point of Equality?' presses two charges that every essay on this topic must engage. The harshness (or abandonment) objection: by making compensation turn on the absence of responsible choice, luck egalitarianism abandons those who suffer through their own option luck — the uninsured driver who is paralysed, the gambler who loses — leaving them, on the theory's own logic, with no claim to aid. That is a cruelty no decent egalitarianism should countenance. The disrespect (or paternalism) objection: to administer the brute/option-luck cut, the state must make and act on demeaning judgments about who is responsible for their plight and who is a pitiable victim of bad luck — sorting citizens into the deserving and the feckless, and requiring the worst-off to advertise their inferiority to qualify for help. This, Anderson argues, expresses contempt rather than equal respect, and so betrays the very value it claims to serve.

Cohen versus Dworkin on expensive tastes

A long-running internal dispute. Dworkin holds people responsible for their tastes and so excludes them from compensation. Cohen objects that some expensive tastes are not chosen — a craving one finds oneself with, and would be rid of if one could — and these are a form of bad brute luck for which the luck egalitarian should, by its own logic, compensate. The exchange forces the question of where exactly the line between choice and circumstance falls, and whether 'choice' should be understood in terms of control or of identification.

Capability versus resourcism

Against Sen, defenders of resources (Dworkin, Rawlsians) reply that the capability metric is indeterminate (which capabilities? weighted how?) and threatens liberal neutrality, since fixing a list of valuable functionings seems to impose a particular view of the good life — the charge Nussbaum's definite list invites most directly. Sen's open-list, public-reasoning answer trades determinacy for neutrality; Nussbaum's list trades neutrality for usable content. Resourcists add that holding people to a resource share better respects their responsibility for their own conception of the good.

Distributive versus relational equality

The deepest current debate is whether equality is fundamentally about distribution at all. Luck egalitarians charge that the relational view cannot explain why unchosen inequalities are objectionable even where no one dominates anyone; relational egalitarians reply that the luck view fetishises the pattern of holdings, generates the harshness and disrespect problems, and forgets that the historic point of egalitarian politics was to end caste, hierarchy, and subordination, not to run a vast cosmic insurance scheme against bad luck.

Against Rawls. Luck egalitarianism can be read as radicalising Rawls. Rawls had said the distribution of natural talents is 'arbitrary from a moral point of view', yet the difference principle still allows the talented to benefit from gifts they did not earn (so long as the worst-off gain). Cohen and the luck egalitarians press the arbitrariness point further: if unchosen advantages are unfair, justice should neutralise their effects more thoroughly, attending to the choice/luck cut rather than only to the position of the worst-off group. Sen, by contrast, attacks Rawls not on responsibility but on the metric, replacing primary goods with capability.

Resourcism versus the capability approach. Dworkin and Sen agree against welfare metrics but divide on what replaces them. Dworkin stops at resources (plus insurance for handicaps), holding people responsible for converting resources into a good life; Sen insists the conversion itself varies with persons' circumstances, so that equal resources can leave deeply unequal real freedoms. The disabled claimant is the test case: Dworkin compensates via the hypothetical insurance market, Sen via direct attention to capability.

Distributive versus relational equality. This is the sharpest contrast. Dworkin, Arneson, and Cohen treat equality as a pattern to be achieved across individuals, refined by the responsibility cut; Anderson and Scheffler treat equality as a relationship of equal standing to be secured among citizens, for which distributive measures are instrumental. Where a luck egalitarian asks 'is this inequality the product of choice or of luck?', a relational egalitarian asks 'does this inequality place some under the power of, or beneath the standing of, others?'. The two questions can recommend very different policies for the same case — most starkly for the imprudent, whom the first may abandon and the second will not.

Question: 'Equality requires that we neutralise the effects of luck but not of choice.' Discuss.

Introduction

The thesis states the core commitment of luck egalitarianism — the view, developed by Dworkin, Arneson, and Cohen, that distributive justice should compensate people for disadvantage flowing from unchosen circumstance (brute luck) while holding them responsible for disadvantage flowing from their own choices (option luck). This essay argues that the thesis captures a genuine and powerful intuition about fairness, but that, taken as the whole of equality, it fails: Elizabeth Anderson's harshness and disrespect objections show both that it reaches morally intolerable verdicts and that it misidentifies the point of equality. The better view is relational — equality is a matter of citizens standing as equals — within which the luck intuition can be honoured instrumentally without being made foundational.

The appeal of the luck/choice cut

The thesis answers to a deep conviction: it is unfair for someone to be worse off than others through no fault or choice of their own — born disabled, untalented, or poor — while it is not unfair for someone to bear the foreseeable costs of a gamble they freely chose to run. Dworkin builds elegant machinery on this: the auction with its envy test makes shares ambition-sensitive, and the hypothetical insurance market makes them endowment-insensitive, converting brute luck into the moral equivalent of insured option luck. The view's great strength, as Cohen observed, is that it incorporates responsibility into egalitarianism, and so disarms the right-wing complaint that equality ignores desert and choice. It explains, as crude equality of outcome cannot, why we may let the diligent prosper above the idle while still compensating the unlucky.

Anderson's objections

Yet the responsibility cut, made foundational, generates two serious problems. The harshness objection: if compensation depends on the absence of responsible choice, then those who suffer through their own option luck — the uninsured driver who is paralysed, the improvident who gamble and lose — have, by the theory's own logic, no claim to assistance. A view that would leave such people to their fate has abandoned the humane impulse that motivates egalitarianism in the first place. The disrespect objection: to operate the brute/option distinction, the state must continually judge who is responsible for their misfortune, sorting citizens into innocent victims and feckless authors of their own ruin, and requiring the needy to document their own incapacity or bad luck to qualify for aid. Far from treating people as equals, this expresses pity and contempt, and makes equal respect hostage to a demeaning audit of people's choices.

The relational alternative

Anderson and Scheffler trace these failures to a false picture on which equality is a distributive ideal — the equalising of some quantity, fine-tuned by responsibility. On the relational view, equality is instead a feature of social and political relationships: its point is to abolish domination, hierarchy, and subordination, and to constitute a community in which people stand to one another as equals. Distributive guarantees matter on this view — a secure social minimum, access to the conditions of functioning as a free and equal citizen — but as means to a society of equals, not as the elimination of every unchosen inequality. Crucially, the relational egalitarian helps the imprudent: their claim rests not on their blamelessness but on their standing as fellow citizens whom the community may not let sink below the threshold of equal participation. The harshness and disrespect problems dissolve, because aid no longer turns on a verdict about responsibility.

Judgment

The thesis is half-right. The intuition it codifies — that unchosen disadvantage has a special claim on us, and that people may fairly bear the costs of their genuine choices — is real, and any adequate theory must find room for it. But as the foundation of equality it is both too harsh, abandoning those whose misfortunes are self-incurred, and mistaken about equality's purpose, which is the abolition of social hierarchy rather than the cosmic correction of luck. The relational view is superior precisely because it can accommodate the luck intuition where it belongs — as one consideration bearing on what a society of equals owes its members — without licensing the abandonment of the imprudent or the indignity of a responsibility-audit. We should therefore neutralise the effects of some bad luck, not because luck as such is the enemy of equality, but because a community of equals does not let its members fall, however they came to be falling. Equality is, in the end, a relationship before it is a distribution.

1. Treating 'equality' as one thing. The whole subject begins with Sen's question 'equality of what?'. An answer that does not specify a currency — welfare, resources, capability — and defend it has not engaged the debate.

2. Confusing equality of resources with equality of welfare. Dworkin defends the former against the latter; the expensive-tastes objection is his argument for the switch. Attributing equality of welfare to Dworkin reverses his view.

3. Muddling brute and option luck. Option luck is the outcome of deliberate, declinable gambles; brute luck is how non-chosen risks fall. Luck egalitarianism compensates the second, not the first — and Dworkin's hypothetical insurance market is the device that converts uninsurable brute luck into the equivalent of option luck.

4. Forgetting the slogan. Dworkin's aim is a distribution that is 'ambition-sensitive' (tracks choices) but 'endowment-insensitive' (does not track unchosen advantages). Quoting it shows you grasp the structure.

5. Missing that 'luck egalitarianism' is a critic's term. Anderson coined the label in attacking the view; the harshness and disrespect objections are hers, and a strong script presents them as the central challenge, not an afterthought.

6. Equating the capability approach with equality of resources. Sen's conversion argument is precisely that equal resources need not yield equal capability. Treating capability as a species of resourcism erases his central point.

7. Assuming equality must be distributive. The relational turn (Anderson, Scheffler) denies it: equality is about how people relate as equals, with distribution instrumental. The best answers stage the distributive-versus-relational contest rather than assuming the distributive frame.

The practice questions for this topic — with model answers and mark schemes across foundation, standard, and challenge tiers — are set out below. Attempt each under timed conditions before revealing the model answer, then mark your work against the scheme.

The primary articles and the leading literature are listed below. Read Sen's 'Equality of What?' and Anderson's 'What Is the Point of Equality?' first, then Dworkin's two-part 'What is Equality?' (or the relevant chapters of Sovereign Virtue) for the resourcist machinery.