Re-examining Moral Psychology in Rawls's Original Position — Epoche C1
The parties in John Rawls's original position are asked to choose the ground rules of a society without knowing who they will be in it, and the question this note pursues is what moves them when they choose. The standard answer is that they are prudent and nothing more: rational calculators securing the best terms they can for a self whose identity is hidden from them. The earlier version of this note argued that this cannot be the whole story, and that some moral disposition — a sense of justice, a feeling for fairness — must already be at work. The conclusion defended here is that the earlier version was right that moral psychology is indispensable to Rawls's argument and wrong about where it sits. Putting it inside the parties' motivation wrecks the argument; putting it inside what they know, and inside the test their choice must pass, is what makes the case against utilitarianism work. Rawls himself later corrected his own presentation on a related point, which is the best evidence that the question is a real one. How the parties are actually specified Precision about the set-up is needed first, since the disagreement is about a detail of it. The original position is a hypothetical situation of choice: representatives meet to agree principles for the basic structure of their society — its constitution, its legal form of property, its economy and its tax system — knowing that the agreement is final and binding. Three stipulations define them. First, they are rational in a narrow and technical sense: they have a consistent ranking of the options and take the most effective means to their ends. Nothing moral is packed into this word. Second, they are mutually disinterested : each takes no interest in the interests of the others. This is not egoism in the ordinary sense: they are not assumed to want more than others or to take pleasure in others' losses, and envy in particular is excluded. They are simply indifferent to one another. Third, they choose behind the veil of ignorance , which conceals their social class, natural talents, race and sex, their conception of the good, their psychological propensities, and even the generation to which they belong. What the veil does not conceal is the hinge of everything below. The parties retain all general knowledge: the principles of economics, the workings of political institutions, the basis of social organisation, and — this is the crucial item — the general laws of human psychology. They know what human beings are like in the aggregate; they do not know which of them they are. Two further pieces of machinery complete the specification. The parties are equipped with a thin theory of the good : enough of an account of what any rational person wants to make the choice determinate, without presupposing any particular way of life. What they want are primary goods — rights and liberties, powers and opportunities, income and wealth, and the social bases of self-respect — because these are useful whatever one's final ends turn out to be. And the choice is bound by formal constraints on any acceptable conception of right: the principles must be general in form, universal in application, public, capable of ordering conflicting claims, and final. Why Rawls needs the parties not to be moral Here the earlier version of this note must be corrected, and the reason is structural rather than textual. Rawls's ambition is to derive principles of justice from a choice made under conditions that are themselves fair. If the parties were already equipped with a sense of justice — a disposition to want fair terms as such — the derivation would be circular: the fairness of the outcome would be secured by the fairness already installed in the choosers, and the argument would establish nothing that was not assumed. The whole point of stripping the parties down to bare rationality is that whatever moral content appears in the conclusion must have come from the constraints on the choice situation, chiefly the veil, and not from the choosers' good will. There is a second reason, less often noticed and more decisive. Suppose the parties were benevolent, each wanting the good of all. They would still need a principle telling them how to weigh competing goods against one another when these conflict, and benevolence does not supply one. Universal benevolence is compatible with utilitarianism, with the difference principle, and with a great many other rules; it therefore cannot select among them. Assuming benevolence would not merely make the argument circular but would leave it indeterminate. Rawls in 1971 pressed this thought to a formulation he would later withdraw, describing the theory of justice as a part — perhaps the most significant part — of the theory of rational choice. In Political Liberalism (1993) he says plainly that this was an error. The moral content of the argument, he there explains, is carried by the distinction between the reasonable and the rational : the rational is the capacity to pursue one's ends effectively, which the parties have, while the reasonable is the willingness to propose and abide by fair terms of cooperation, which is modelled by the conditions under which they choose. The reasonable is prior, and it is not derived from the rational. This is the correct version of the earlier version's intuition. Something moral is indeed present in the original position; it is present in the design of the room, not in the heads of the people in it. The question the earlier version was right to press The earlier version asked why parties concerned only with themselves would not gamble on a system of higher average welfare, hoping to land among the beneficiaries. This is the strongest objection in the field, and it has an author. John Harsanyi (1955) had already constructed a veil of his own: an impartial observer who does not know which member of society he will be, and who assigns equal probability to being any of them. Such an observer, maximising e