Curriculum

Appearance

Theories of Justice (Advanced) ​

Learning outcomes ​

  • When a real distributive question pulls outside the standard setup, like climate burdens across generations, reparations for historic wrongs, duties to foreigners, or how the disabled are treated, you'll know which simplifying assumption of the canonical theory is doing the work, and which alternative framework is built for the case the canonical one drops.

What the canonical setup brackets ​

Rawls's Theory of Justice (1971) framed his argument around a hypothetical society of free, equal, healthy, contemporaneous adults choosing principles from behind a veil of ignorance. The bracketing is deliberate. Justice theory at the advanced level is largely the work of letting the bracketed complications back in: future people, the dead, foreigners, the disabled, children, the sick. Each one tests whether Rawls's framework can stretch, needs supplementation, or has to be replaced.

Future generations. Most of the harms and benefits of present action will fall on people not yet born. Contractarian theories struggle because future people cannot consent. Utilitarianism struggles because Derek Parfit's non-identity problem shows that almost any policy choice changes which future people exist, so the people benefited by, say, climate action are not the same people who would have existed without it. Comparisons across non-overlapping populations run into paradoxes: Parfit's Repugnant Conclusion is that, on total utilitarianism, a vastly larger population whose lives are barely worth living must outrank any smaller population of flourishing lives. The contemporary debate runs through Parfit's Reasons and Persons (1984).

Historic injustice. What does justice require for wrongs done to ancestors, when both perpetrators and direct victims are dead? The classic cases concern slavery, colonialism, indigenous dispossession, and the Holocaust. Reparations arguments split between forward-looking views (current claims grounded in current inequalities traceable to past wrongs) and backward-looking views (claims grounded in the wrongness of the original act, surviving to descendants). Jeremy Waldron's "Superseding Historic Injustice" (1992) presses the awkward case for the backward-looking side: changes in circumstances since the original wrong can render the descendant's claim weaker than the descendant's standing claim to current resources. The empirical question of traceability (can current outcomes be linked to specific past wrongs?) interacts with the philosophical one.

Global justice. Most theorising assumes a single bounded society. What do we owe people outside our borders? Cosmopolitans argue the principles that ground justice within societies extend across them: Peter Singer on the duty to relieve distant suffering, Thomas Pogge on the affluent world's complicity in poverty through the global economic order, Charles Beitz on extending Rawls's veil to the international level. Rawls himself in The Law of Peoples (1999) moved the other way, to a state-centric position with thinner duties across borders, treating "peoples" rather than individuals as the contracting parties. The practical literature works through climate, immigration, and the international economic order.

Disability. Contractarian setups assume rational agents capable of negotiating the social contract. Many people do not fit that description in the relevant ways, temporarily (children, the very ill) or permanently (severe cognitive disability). Martha Nussbaum's Frontiers of Justice (2006) argues this is not a small adjustment but a fundamental challenge to contractarianism. She develops the capability approach (with Amartya Sen) as the alternative: justice secures a threshold of central human capabilities regardless of whether the person can negotiate for them.

Children. Children are not full members of the political community in the standard sense (they cannot vote, contract, or consent), but justice clearly applies to them. The hard work is allocating authority between parents, the developing child, and the state, given that adults are the ones doing the deciding for someone whose own preferences are still forming. Adam Swift and Harry Brighouse's Family Values (2014) defends the family as a distinctive site of relationship goods that justify some parental partiality, while pushing back on inheritances and elite schooling that pass advantage on without that justification. Children's interests now also compete with future-generation interests in policy domains like climate and pensions.

Health care. Health care is hard to distribute justly because need is concentrated, costs are high, and the relevant baseline (what counts as health) shifts with medical technology. Norman Daniels's Just Health Care (1985) extended Rawls: health is a precondition for the fair equality of opportunity Rawls already endorses, so the same framework grounds duties of health provision. The contemporary cost-effectiveness apparatus (the QALY, NICE in the UK) operationalises some of this, though whether it does so consistently with Rawlsian principles is contested: QALY weighting penalises the chronically ill or disabled by giving lower weight to extending lives spent below full health.

Luck and the equality of what ​

Within egalitarian theory, the post-Rawls debate has centred on a question Rawls left implicit: equality of what, and how is it qualified by individual responsibility?

Luck egalitarianism, developed by Ronald Dworkin, G. A. Cohen, and Richard Arneson, holds that justice should neutralise inequalities arising from brute luck (what falls on you, like genetic endowment or the family you were born into) but allow inequalities arising from option luck (gambles you knowingly took). The framework is intuitive on the disabled-versus-able-bodied case (a brute-luck inequality, so compensable) and on the responsible-versus-reckless-gambler case (option luck, no compensation). It shares Rawls's worry that the talented should not be able to claim the full market value of their talents.

The framework runs into trouble at the edges. Adaptive preferences and unchosen background conditions blur the brute-luck/option-luck line. Worse, on a strict reading luck egalitarianism produces "harshness": a reckless gambler who loses everything has no claim against the rest of us, even when the loss is starvation. Elizabeth Anderson's "What Is the Point of Equality?" (1999) presses this attack, arguing that egalitarianism's actual point is not to compensate brute-luck losers but to secure relations of equal standing among citizens. Relational egalitarianism replaces the distributive question (who has how much?) with a social question (do we treat each other as equals?), which licenses different policy moves: a guaranteed floor that does not condition on responsibility, and constraints on hierarchies (workplace, racial, gendered) that distributive equality alone leaves untouched.

The debate remains live. Distributive luck egalitarians can absorb a sufficiency floor; relational egalitarians can absorb distributive metrics where unequal distribution sustains hierarchy. The choice between them shapes how a theory reads the same case (a poor reckless gambler, a wealthy untalented heiress) and what it would prescribe.

For the underlying Rawlsian intuition this debate inherits, that talent and effort are themselves morally arbitrary in a way that breaks the link between market reward and desert, Michael Sandel's Harvard Justice lecture "What's a Fair Start? / What Do We Deserve?" (about 55 minutes) stages the argument with concrete cases (Sandra Day O'Connor's salary against Judge Judy's), pulling apart entitlement under the rules from desert in the deeper sense. Watch this if the brute-luck framing feels formal and you want the moral-desert move dramatised before deciding what to do with it.

The capability approach as alternative starting point ​

Amartya Sen and Martha Nussbaum developed the capability approach as a partial alternative to both utilitarianism and contractarian justice. The central concept is capabilities: the real freedoms a person has to live a valuable life, distinct from both resources (Rawls's currency) and utilities (the welfarist's).

The case for capabilities over resources: the same resources convert into different real freedoms for different people. A wheelchair user and an able-bodied user with the same income do not have the same access to mobility, employment, or social participation. Resourcism misses this conversion problem; capabilities do not.

The case for capabilities over utility: preference satisfaction reflects what people want, but adaptive preferences mean people in deprivation often adjust their wants downward. A framework that takes preferences at face value can make oppression invisible. Capabilities focus on objective freedoms regardless of whether the agent currently wants to use them.

Nussbaum's version specifies a list of ten central capabilities (life, bodily health, bodily integrity, senses-imagination-thought, emotions, practical reason, affiliation, other species, play, control over one's environment). Sen's version refuses a fixed list, treating the relevant capabilities as the subject of democratic deliberation in each context. A list gives normative force at the cost of imposing a substantive conception of the good. No list preserves freedom of conception at the cost of determinacy.

Universal basic income as a test case ​

Universal basic income (UBI) is a regular cash payment from the state to every citizen, with no employment, work-history, or means test. The proposal has older roots (Thomas Paine in 1797), a twentieth-century libertarian-friendly version (Hayek and Friedman's negative income tax), and a contemporary egalitarian defence in Philippe Van Parijs (justifying UBI as the cash equivalent of equal real freedom).

UBI is a useful test case because every major framework reaches a different verdict.

FrameworkVerdict on UBI
Libertarian (Nozick)Mixed. Cuts bureaucratic intervention but is still redistributive on principles libertarians reject
Liberal egalitarian (Rawls)Conditional. Permissible if it satisfies the difference principle; risks failing Rawls's reciprocity condition if too generous to those who refuse to work
Capability (Sen, Nussbaum)Generally favourable. Raises the capability floor for all, particularly unpaid carers and the precariously employed
RepublicanFavourable. Reduces domination by employers because workers can credibly walk away, even if it does not reduce interference
MarxistMixed. Can entrench wage labour (subsidising private profit at public cost) or open exit options that shift the labour-capital balance
Welfare-state pluralistSceptical. May undermine the universalist welfare-state coalition by replacing in-kind services with cash that the better-off convert into private alternatives

The empirical record is small but growing: pilots in Finland, Kenya (GiveDirectly), and several US states (Stockton). They suggest UBI does not produce large work disincentives, improves recipient well-being, and is administratively cheaper than means-tested alternatives. They do not settle whether a UBI generous enough to do the work theorists ask of it is fiscally sustainable.

Rawls remains the dominant starting point, but the literatures on disability, intergenerational ethics, and global justice either modify Rawls past recognition (Nussbaum) or start somewhere else (Sen's capability approach, the cosmopolitans, the relational egalitarians). The working assumption is pluralist: different domains need different normative tools, and the live meta-question is which tools fit which domains.

References ​