Abstract

My general conclusions are that it is not rational for agents to behave morally when moral behaviour is imprudent; and that one is left with a theory of justice which is very narrow if one requires that distributive justice be generated by rational choice. ;It follows that either the scope of distributive justice is very limited , or many situations involving distributive justice are not grounded in instrumental rationality. ;A type of case I consider has been thought of as a paradigm of bargaining game. I argue that rational agents will not reach an agreement in this game. Therefore, the situation will not turn into one which involves distributive justice. However, this game is a paradigm of the type of situation for which bargaining theorists propose a fair solution. ;My approach is to assume for argument that situations involving distributive justice arise only after an agreement has been made. If rational agents under no moral constraints reach an agreement, and if the agreement is fair, then the fair agreement has a rational grounding. ;I aruge that there are very many cases in which rational agents will not reach an agreement. Since situations involving fairness or unfairness arise only after agreements are made, these cases do not fall under the scope of distributive justice. ;My conclusion in the first part is that it is not instrumentally rational to act imprudently in keeping agreements. Also, in general, it is not instrumentally rational to become an agreement-keeper. ;In the second part of my thesis I assume that agents are coerced into keeping agreements. This lets us look at the problem of what agreements prudent agents will reach if they know that it is irrational to break agreements. I argue that it is either the case that the scope of distributive justice is very limited or that very little of distributive justice can be generated from a theory of rational choice. ;However, I argue that agents who become constrained maximizers for prudential reasons are behaving irrationally when they are imprudently keeping agreements. I also consider the issue of whether it might be prudent for an agent to become a constrained maximizer even though constrained maximizers sometimes behave irrationally. I show that it might be prudent only in very special cases. ;First, I examine David Gauthier's theory of constrained maximization. If Gauthier's theory is correct, then it is rational to keep distributively just agreements. It would then be the case that a moral duty to keep fair agreements is grounded in instrumental rationality. Gauthier argues that an agent who becomes a constrained maximizer does at least as well as he would if he were prudent, and perhaps better. If his argument holds, then an agent who is initially prudent is prudent to become a constrained maximizer. A constrained maximizer keeps fair agreements even though it may be imprudent to keep them. If it is rational to be a constrained maximizer, it is rational to keep fair agreements. ;In this thesis I discuss instrumental rationality and distributive justice. The agents I consider as instrumentally rational will adopt a policy of prudence or of maximization of expected utility. I concentrate on two main areas