Abstract
Actions expressing emotions (such as caressing the clothes of one's dead friend in grief, or tearing apart a photograph out of jealousy) pose a notorious challenge to action theorists. They are thought to be intentional in that they are in some sense under the agent's control. They are not thought to be done for a reason, however, because they cannot be explained by considerations that favor them from the agent's point of view. This seems to be the case, at least, if one subscribes to the Davidsonian standard model of action explanation. So far, philosophers have had three different reactions to this challenge. Rationalists insist that such actions can be rationalized by re-interpreting them. Arationalists insist that there simply is no reasoning process moving agents in emotional states to act. A third reaction questions the intentionality of such actions altogether. All three reactions, however, share the assumption underlying the standard account: if an agent is thought to act for a reason - and hence acts intentionally - he must entertain a desire and some means-end belief reflecting his reasoning process about how to attain what he desires by acting. In this paper, I try to show that this reflective reasoning mechanism is only one way to rationalize an action. Another way is by tracing an action to an unreflective valuing stance respresenting reasons the agent has from his point of view. Emotions are attitudes that help to grasp reasons the agent has. Since emotions come with a strong motivational potential they move the agent to act expressively. But the agent typically allows himself to do so, thereby monitoring the way in which he does it. To the extent that the agent unreflectively acts on a motive that is itself representative of his point of view, his expressive actions can be regarded as rationalizable.