What is comparative justice? Amartya Sen's method, explained with examples

Comparative justice is a way of thinking about justice that starts from the options actually in front of us. It asks which of them would make the world less unjust. Its best-known modern defender is the economist and philosopher Amartya Sen, who set it out in The Idea of Justice in 2009. Sen calls it realisation-focused comparison, and it is the method behind much of what radical centrism tries to do.

Two questions about justice

Political philosophers have asked two different questions. One is: what would a perfectly just society look like? Sen calls the tradition built on that question transcendental institutionalism. It runs from Hobbes, Locke, Rousseau and Kant to John Rawls and Robert Nozick, and its method is to define the principles of the ideal and then design the institutions that would embody them.

The other question is: of the alternatives we can actually choose, which is more just? This is the comparative question. Sen traces it through Adam Smith, the Marquis de Condorcet, Jeremy Bentham, Mary Wollstonecraft, Karl Marx and John Stuart Mill. These thinkers disagreed about a great deal, and each of them spent much of their energy comparing real social arrangements and arguing for specific improvements.

Sen's central claim is that the second question can be answered without answering the first. The openDemocracy review of his book by Charles Barclay Roger puts the argument clearly. A picture of perfect justice is insufficient, because knowing the ideal does not tell you how to rank two imperfect options. It is also unnecessary, because you can rank those options without it. You can compare the heights of Kilimanjaro and McKinley with no knowledge of Everest.

How comparison works

Comparative justice has a few working features worth knowing.

It looks at outcomes in people's lives. Sen judges arrangements by capabilities, the real freedoms people have to be and do what they have reason to value. Two policies that spend the same money can leave people with very different freedoms, depending on health, surroundings and social rules, so the comparison has to look at what people can actually do.

It accepts partial rankings. We will not always be able to rank every option against every other. Often we do not need to. Agreeing that a famine is worse than its absence, or that a child in school is better off than a child out of it, is enough to act. Sen treats incomplete rankings as normal and useful, a strength of the method.

It accepts several principles. In Sen's flute example, three children each have an impartial claim to one flute: one plays it best, one is the poorest, one made it. Comparative justice lets all of those reasons count and looks for the options that rank well across the reasons people can defend.

It uses social choice. Sen won the Nobel Prize in economics in 1998 for work in this field, which studies how to combine many people's judgements into a collective ranking. Arrow's impossibility theorem showed that this can fail. Sen's response, as Roger summarises it, is that many of these impasses ease when we use more information, in particular about how people's well-being compares. More evidence, better comparisons.

It reasons in public, with outsiders. Sen prefers Adam Smith's impartial spectator, who can look at a society from anywhere, to a contract made only among insiders. The judgements that survive scrutiny from many directions are the ones to trust.

An example from Spain

Spanish law already asks for comparative reasoning when the government drafts a regulation. The rules on the regulatory impact report, the Memoria del Análisis de Impacto Normativo, require ministries to justify a new rule against the alternative of no regulation and to try, as a general rule, to assess more than one alternative. Spain's fiscal watchdog, AIReF, applies the same logic in its spending reviews: its 2026 evaluation of the national minimum income compared the scheme's actual reach with what it would achieve if everyone entitled claimed it. That is comparative justice done with microdata. We look at both cases in today's other articles.

Criticism and a centrist use

Comparative justice has serious critics. Laura Valentini argued in Economics and Philosophy in 2011 that Sen's call for a paradigm shift is unnecessary, that some of his criticisms misread Rawls, and that the Rawlsian approach already delivers much of what he wants. Her point has force: to say an option is better, you need some idea of the direction in which better lies.

A radical centre can take that point on board and still use the method. From the left comes the priority of manifest, remediable injustice: the people worst off cannot wait for philosophers to agree on the ideal society, and the comparison should start with them. From the liberal and conservative tradition comes the case for reforming step by step from what exists, judged by results, with suspicion of master plans. And from neither comes Sen's social choice insight: with better information about how people actually live, a democracy can reach workable rankings even when its citizens hold conflicting principles.

Taken together, they give a simple discipline for any policy debate. Name the problem. Put the real alternatives on the table, including doing nothing. Compare what each does for people's freedoms, using evidence that others can check. Choose the best available option, and keep the comparison open, ready for the next round.