Mike Smith

In On What Matters Derek Parfit argues that facts about reasons for action are grounded in facts about values and against the view that they are grounded in facts about the desires that subjects would have after fully informed and rational deliberation. I describe and evaluate Parfit's arguments for this value-based conception of reasons for action and find them wanting. I also assess his response to Sidgwick's suggestion that there is a Dualism of Practical Reason. Parfit seems not to notice that his preferred value-based conception of reasons for action augurs strongly in favour of a view like Sidgwick's. 1.

Evaluative judgements have both belief-like and desire-like features. While cognitivists think that they can easily explain the belief-like features, and have trouble explaining the desire-like features, non-cognitivists think the reverse. I argue that the belief-like features of evaluative judgement are quite complex, and that these complexities crucially affect the way in which an agent's values explain her actions, and hence the desire-like features. While one form of cognitivism can, it turns out that non-cognitivism cannot, accommodate all of these complexities. The upshot is that that form of cognitivism can explain both features of evaluative judgements, and that non-cognitivism can explain neither.

On one influential view, a person acts autonomously, doing what she genuinely values, if she acts on a desire that is her own, which is (on this account) a matter of it being appropriately ratified at a higher level. This view faces two problems. It doesn’t generalize, as it should, to an account of when a belief is an agent’s own, and does not let one distinguish between desires (and beliefs) happening to be one's own and their being the ones a person would need to have in order to be autonomous. The paper develops an alternative unified account of what it is for desires and beliefs to be one’s one, and argues that an account of acting autonomously should pay attention not to which desires and beliefs are one’s own, but to whether they are ones the agent has reason to have and to act on.

On one influential view, a person acts autonomously, doing what she genuinely values, if she acts on a desire that is her own, which is (on this account) a matter of it being appropriately ratified at a higher level. This view faces two problems. It doesn’t generalize, as it should, to an account of when a belief is an agent’s own and does not let one distinguish between desires (and beliefs) happening to be one’s own and their being the ones a person would need to have to be autonomous. The paper develops an alternative unified account of what it is for desires and beliefs to be one’s one and argues that an account of acting autonomously should pay attention not to which desires and beliefs are one’s own but to whether they are ones the agent has reason to have and to act on.