Maximin is a rule for making choices under conditions of uncertainty (or risk), that is, when one must select from a range of options, each one of which may result in various outcomes. Maximin says that one should identify the worst possible outcome for each option, and then select the option for which the worst outcome is least bad. It thus represents an extremely risk-averse or pessimistic standard of choice. Rawls recognizes that it is “clearly . . . not, in general, a suitable guide for choices under uncertainty” (TJ 133). However, he argues that because of the highly unusual features of the original position, it can be “useful as a heuristic device” (TJ 132) to think of the parties there selecting principles of justice on that basis.

Although it is sometimes assumed that Rawls simply stipulates that the parties would accept maximin reasoning, this is not the case. As he pointed out in a 1974 article, such a stipulation “would indeed have been no argument at all” (CP 247).There are three main features of the original position that push the parties toward a very risk-averse choice, represented by maximin reasoning. First, the veil of ignorance prevents the parties from assigning probabilities to the various possible outcomes. They have no basis for estimating the likelihood that they will ind themselves in each of the various social positions when the veil is lifted. Indeed, because they are choosing fundamental principles of justice and not institutional arrangements directly, they do not have much of a basis for determining what the various social positions will be, let alone the likelihood that they will occupy any one in particular. Second, the parties are much more concerned with ensuring that certain basic interests are satisfied than they are with the prospects of additional gains above this level. Finally, the worst outcomes of some of their possible choices fall below this minimal level. That is, some of the conceptions that they might choose fail to ensure the protection of their basic interests.