Abstract

Historically the notion of practical rationality has played two roles. One role is that of the fundamental normative term applying to actions. When theorists conceive of practical rationality in this sense, they claim that it is nonsense to ask "Why be rational?" This is because any informative answer would have to make use of a still more fundamental normative notion. The other role which rationality has historically played is connected with proper practical mental functioning. In this sense, rationality is more connected with the assessment of agents, although actions are still assessed as rational and irrational. My dissertation argues that one notion cannot play both of these roles. And though I admit that the "mental functioning" notion, which I continue to call "rationality," is a normative one, I argue that the former notion is both conceptually and normatively prior. I call this prior notion "advisability." ;I argue that the advisability of an action is relative to neither the beliefs nor the desires of the agent. And I argue that irrational actions are those which indicate that the agent is in a state of increased likelihood of doing unadvisable actions, and may be usefully criticized. My account of advisability, and of the relation between rationality and advisability, allows me to explain the following complexity in the notion of rationality. First, reasons stemming from harms to the agent provide prima facie rational requirements, regardless of whether the agent is motivated by them. Second, reasons stemming from the interests of others can only provide rational justification if the agent is motivated by them. Thus I permit a principled marriage of the insights of internalists and externalists about reasons.