Document Type



Master of Arts



Date of Defense


Graduate Advisor

John Brunero, Ph.D.


Eric Wiland

Anna Alexandrova


David Velleman has argued that action has a constitutive aim. This constitutive aim is analogous to the constitutive aim of belief. The constitutive aim of belief, he argues, is to track the truth or arrive at the truth. This aim sets the standard of correctness for a belief. A belief is correct if and only if it is true. So reasons for belief are considerations that point toward truth either by guaranteeing truth or significantly raising the probability of the truth. Any belief that fails to track the truth misses the mark set by the constitutive aim of belief. Like the constitutive aim of belief, the constitutive aim of action sets the standard of correctness for an action. Velleman claims that action constitutively aims at self-knowledge (Velleman, 2000, 2006, 2009). Under this conception of action, an action is correct if and only if you know what you’re doing when you perform the act. Thus reasons for acting are now given by this new constitutive aim of action. Reasons are considerations in light of which a person has a better potential grasp of knowing what they are doing. They help the agent make sense of their action. Velleman’s thesis functions both as an account of what autonomous action is and as the basis for a “constitutive aim” account of reasons. My argument will thus be divided into two parts. In part 1, I will argue that Velleman’s account of autonomous action fails. I will do this by pointing out that the argument Velleman makes against Frankfurt’s hierarchical model works against his own view. In part 2, I will critique Velleman’s view of the constitutive aim of action. I will argue that if action has a constitutive aim, it cannot be the aim that he suggests. I will show this by pointing out that Clark’s argument against Velleman’s original view still works against his newer view.