Abstract
The type of cognitive theory of emotion traditionally espoused by philosophers of mind makes two central claims. First, that the occurrence of propositional attitudes is essential to the occurrence of emotions. Second, that the identity of a particular emotional state depends upon the propositional attitudes that it involves. In this paper I try to show that there is little hope of developing a theory of emotion which makes these claims true. I examine the underlying defects of the programme, and show that several recent variants fail to repair these defects. Furthermore, even if such a theory could be developed, it would not achieve many of the things that we look to a theory of emotion for. I argue that philosophers should turn their attention to new and more promising approaches. These have been developed by various of the special sciences, while philosophy has remained enthralled by traditional, propositional attitude psychology