Abstract

According to Stoic moral psychology, emotions are cognitive responses to perceived value in the contingent world. This dissertation begins by defending a contemporary version of this descriptive theory; it then proceeds with a critique of the Stoics' normative thesis that emotions involve amorally deplorable kind of cognitive error. I distinguish two senses in which this thesis is historically put forward, and show that both are thematically pertinent. The structural variant, as I call it, is a qualified critique of the particular cognitive flaws to which emotions are liable; the fundamental argument, on the other hand, is that emotions are categorically unreliable. My goal in the rest of the dissertation is to respond to each aspect of normative Stoicism with a positive account of how it might be possible for a moral agent to avoid false emotion without doing away with emotion altogether. This means explaining how a person's emotional perception might become reliable rather than sentimental within a given moral context, and indicating what axiological conditions must generally obtain if emotion is justifiable in any case. In doing so, I draw largely upon Kierkegaard's pseudonymous and signed writings, while also engaging with other relevant philosophical texts along the way. I argue that a morality of virtue and narrative awareness is necessary for accurate emotional perception; then, I attempt to define the conditions on which a qualified value realism might be tenable. It is one thing to argue for the importance of emotion in human life, as a number of recent studies have done, but it is quite another to show how we might cultivate truthful, and avoid false, emotion. The outcome of this inquiry into the possibility of reliable emotion is a positive account of the ideal state in which one could trust oneself to be rational in being passionate