Intentions, decisions and rationality

Abstract

Introduction Rational agents, as conceived in decision and game theory (Myerson 1991; Osborne and Rubinstein 1994), are agents who choose the actions they believe to be the best means to achieve their preferred ends. But human beings are also planning agents. That is, we have the capacity to ‘settle in advance on more or less complex plans concerning the future, and then [let] these plans guide our later conduct’ (Bratman 1987). Do decision-theoretical models do sufficient justice to this planning aspect of human agency? In particular, can one account within a decision-theoretical framework for the fact that planning agents can commit themselves by forming intentions? In van Hees and Roy (2007) we addressed this question for one particular kind of intention, namely the intention to bring about some future states of affairs, or what we now call ‘outcome intentions’. These intentions, for example, the intention ‘to be in France next week’, can be distinguished from ‘action intentions’, which refer to the future performance of some actions, for example, the intention ‘to take the plane to France’. We argued that outcome intentions can be fruitfully incorporated within a decision-theoretical framework, and presented conditions under which rationality with respect to intentions is compatible with the standard payoff-maximizing notion of rationality.