There are at least two ways to think of theorizing about rationality. One way is to see the ideal theory of rationality as providing us with a manual of sorts: follow these instructions, and you will always make a rational decision, or you will at least know whether the course of action that you are taking is rational or not. 1 Another way is to see theorizing about rationality as aiming not at providing us with a manual but at providing us simply with a theory—a theory that tells us when people act rationally and when they do not, so that given a God’s-eye view of a person’s circumstances, beliefs, and motives, one would be able to tell how rational or irrational said person would be in performing a certain action. These two fascinating tasks—to which I will refer as the creation of a rational agentLs manual and the creation of an account of rationality—are more different than they may look, and conceptual blunders may ensue if one does not always know which of them one pursues. Consider, for example, a person facing a complex situation. One attempts to decide whether to take up a certain risky career, whether or not to leave a marriage in favor of a new love, or (to borrow from Williams) whether to leave one’s family and move to Tahiti to become a great painter. Desperate, one wonders: Am I crazy in wanting to take the job? Would it be irrational of me to leave my husband? Should I decide to go to Tahiti and just see what happens? Now let us assume that a certain theory of rationality does not help us very much in answering these