Broomean Rationality: A Methodological Dilemma
Many people have posed challenges for idealistic requirements in a theory of rationality (e.g., requirements of consistency on our beliefs). A feature of such requirements is that they (by design) do not take into consideration the limitations of the group over which they are imposed. For example, ideal theories of human rationality do not take into consideration the cognitive limitations that humans have, such as the fact that updating beliefs is a temporally extended process and thus take time to satisfy. Despite such realities, the trend of modeling requirements after ideal agents persists. John Broome is one such philosopher that adopts ideal requirements: “Requirements of rationality as I formulate them generally take no account of [the limitation of our human psychology].” In the face of challenges against ideal norms that object on the grounds that we are unable to achieve them, Broome has a story he can tell to insulate himself from such challenges. He can simply say that the ideal is what sets the standard in his theorizing, and thus our inability to achieve them is not a problem. In accepting the ideal as a guide, he is within his rights to set aside considerations about what we can actually do. However, he is careful to say that such requirements generally take no account of our limitations. What follows is a critical examination of this kind of methodology. I will argue that there are no satisfactory reasons provided for the choice to account for human limitations in some requirements and not others. The upshot is this: either Broome takes absolutely no account of the limitations of human psychology in forming rational requirements, or he does. Either option results in problems that he expressly tries to avoid.