Rawls and the Veil of Ignorance

By the 1960s political philosophy in the English-speaking world was widely considered finished. Logical positivism had declared value judgements to be expressions of attitude rather than statements of fact, which left little for a theory of justice to be about, and policy thinking had settled into utilitarian cost-benefit.

John Rawls restarted the whole field with one device. Instead of asking what arrangement produces the most welfare, ask what principles people would agree to if they did not know who in the resulting society they were going to be.

Picture the choosing situation. You are setting the rules (property, liberties, taxation, opportunity) and you do not know your class, race, sex, talents, health, or even your religion and your idea of a good life. You will be assigned a position afterwards, and you have no way to influence which. He calls this the original position, and the restriction on knowledge the veil of ignorance.

The elegance is that it does not require anyone to be good. Self-interest is enough, because self-interest with no information about which self you are turns into impartiality automatically. It is the logic of one child cutting the cake and the other choosing the slice: nobody has to be generous for the outcome to be fair.

Rawls is explicit that this is a device of representation, not a claim that anyone is ever actually ignorant, and not a historical contract. It is a way of modelling what fairness requires.

What would be chosen? Two principles, in strict order. First, each person gets the most extensive basic liberties compatible with the same for everyone. Second, social and economic inequalities are permitted only where positions are genuinely open to all, and where the inequality improves the position of the least advantaged.

That second half is the difference principle, and it is regularly misread as a demand for equality. It is not. It permits inequality, sometimes large inequality, provided it actually raises the floor. Paying a surgeon far more than a receptionist is justified not because the surgeon deserves it but because the arrangement leaves the worst-off patient better off than a flat scale would.

The reasoning behind that comes from a claim many people find hard when it is stated plainly. Ask what you did to deserve your intelligence, your health, or the family that raised you. Rawls concludes that the distribution of natural talent is arbitrary from a moral point of view, so a scheme that rewards talent has to justify itself by what it does for everyone, not by desert.

The ordering also matters more than it looks. Liberty has lexical priority: the first principle must be fully satisfied before the second is even considered. Not weighted heavily, not tradeable at all. Behind the veil you would not gamble away basic liberties for economic gain, however large, because you might turn out to be the person who needs them most.

The most famous objection is Robert Nozick's. Suppose a just distribution exists, and then a million people each freely pay a small sum to watch one basketball player. The resulting distribution is unequal and arose entirely from voluntary acts. Preserving any pattern, Nozick argues, means continuously interfering with free choices. Rawls answers that his principles apply to the basic structure of society rather than to individual transactions (the rules of the game, not each move) and whether that succeeds is one of the central arguments in the field.

The device itself has escaped the seminar room. Any review that asks whether you would accept a system as the user it treats worst is running the original position, usually without knowing whose it is.