Suppose you were asked to design the basic institutions of a society — its constitution, its tax system, its rules about who owns what — without knowing which position in that society you would occupy. You would not know your class, your caste, your sex, your talents, your religion, or even your conception of what makes a life go well. This is the thought experiment at the centre of John Rawls's theory: the parties choose principles from behind a veil of ignorance, in what he calls the original position.
The device is not a claim about history. No one imagines that a founding assembly ever sat in such conditions. It is a filter. Rawls's wager is that a great deal of what passes for moral argument is really the defence of advantages one happens to hold, and that stripping away knowledge of those advantages will leave only reasoning that anyone could accept. What survives the filter is fair, because nobody could have rigged it in their own favour.
From this position Rawls argues that rational choosers would not gamble on utilitarianism, which permits the misery of a few if aggregate welfare rises. Not knowing whether you are the few, you would insist on secure basic liberties for all, and you would permit inequalities only where they work to the benefit of the worst-off. Prudence, under uncertainty, becomes indistinguishable from justice.
The objections came quickly. Robert Nozick replied that justice cannot be read off a pattern of distribution at all; what matters is whether holdings were acquired and transferred legitimately. Communitarian critics argued that the veil manufactures a self with no history, no attachments, no community — and that such a self could not choose anything, because choosing requires already caring about something. Rawls spent much of his later work conceding ground: the original position, he came to say, was not a view from nowhere but a device internal to the political culture of a constitutional democracy, one that could not claim to persuade those standing entirely outside it.