Rawls Designed Principles for Institutions and Forgot What Institutions Do to Principles

John Rawls's A Theory of Justice is the most respected work of political philosophy of the last century, and its central device is genuinely clever. Imagine you must choose the rules of a society without knowing where in it you will land: rich or poor, healthy or sick, majority or minority. Behind that veil of ignorance, Rawls argues, you would choose principles that protect the worst-off, because you might be the worst-off. Those principles are then meant to govern the basic institutions of society.

I have mentioned before that the veil is a fine idea that does not help. Here I want to make the full case, because the way it fails is instructive.

Rawls wrote principles for institutions to follow, and the thing he left out is that institutions do not follow principles. They follow their interests, and the people inside them follow theirs.

The Gap in the Argument

The veil of ignorance is a thought experiment about what you would choose. It is not a mechanism by which anything is chosen. Rawls knew this, of course; the book is about what justice requires, not about how to get it. But the book then hands its principles to the basic structure of society, to legislatures, courts, and economic arrangements, as though a principle once justified would govern the institution that received it.

Look at what actually happens when a principle is handed to an institution.

A legislature adopts a principle that inequalities are justified only when they benefit the least advantaged. Every legislator now has a career, a party, donors, a district, and a re-election in three years. Every proposal that reaches them arrives with a group that stands to gain and a group that stands to lose, and the groups have unequal means of making themselves heard. The principle does not vanish. It is cited. It is cited by both sides, each finding that its own preferred policy happens to be the one that benefits the least advantaged, and the interpretation that wins is the one backed by the stronger interest.

The principle becomes the vocabulary in which the interests argue. It does not become the thing that decides.

This is not a corruption of the institution. It is the normal operation of an institution staffed by people who have positions to protect, which is every institution there is.

Behind Which Veil

Notice also who Rawls asks to stand behind the veil: everyone, once, at the founding moment. But the people who run institutions are never behind a veil. They know exactly where they stand. The minister knows he is the minister. The judge knows she is the judge. The director of the agency knows that this ruling helps the agency and that one shrinks it.

The veil produces impartiality by removing knowledge of one's position. The moment the founding is over, everyone's position is known again, and impartiality has to come from somewhere else. Rawls has nothing to put there. He has principles, and I have just described what happens to principles.

The Physics Comparison

I keep coming back to a parallel with the history of science, and I think it holds here.

For two thousand years philosophers argued about what matter really is, and the arguments were sophisticated and went nowhere. Then people who did not care about the ultimate nature of matter started measuring it, and within two centuries they understood more than the philosophers had in two millennia. The question was not answered. It was bypassed, by people who worked on how things behave instead of what they are.

Moral and political philosophy is still in the first phase. Rawls asked what justice is, and answered brilliantly, and the answer changed nothing, because the question that matters is not what justice is. It is how to arrange an institution so that its behaviour is just when the people in it are not. I have argued elsewhere that every real moral gain in history came from the second question, and Rawls does not ask it.

What Asking It Would Look Like

The question has a shape, and it is an engineering shape.

Given that everyone inside an institution will pursue their own advantage, how do you arrange the incentives so that the advantage-seeking produces the outcome you want? Who checks whom? What is visible to whom? What does defecting cost, and who has the standing to impose the cost? How is the institution's own interest in growing and persisting kept from swallowing its purpose?

None of these are questions about principles. They are questions about structure, and they have partial answers in game theory, in the study of organisations, and in the small number of political experiments that have been run. A theory of justice that started from them would look less like a philosophy book and more like a design document, and it would be worth more.

What Survives

The veil of ignorance survives as a test, and a good one. If you cannot defend an arrangement without knowing which side of it you are on, the arrangement is suspect. That is worth having.

What does not survive is the idea that passing the test is the hard part. The hard part is the institution, and the institution is precisely what a theory of principles is not about. Rawls gave us the destination on a map and no vehicle, and philosophy has spent fifty years admiring the map.