Date of Award


Degree Type


Degree Name

Doctor of Philosophy



Major Professor

David Palmer

Committee Members

Nora Berenstain, Eldon Coffman Jr., Bruce MacLennan


What makes an event count as an action? The standard answer to this question—causalism—claims that if an event is caused in the right way it counts as an action. Causal deviance objections, however, undermine the explanatory power of causalist accounts. Non-causal theories of action offer a promising alternative; however, they also raise a myriad of difficulties. Many non-causal arguments against causalism unintentionally lead to dialectical stalemates, which are methodologically undesirable and should be avoided whenever possible. I offer a theory between these two inadequate accounts that synthesizes the strengths of non-causalism with insights from agent-causal theories. I agree with traditional non-causalist that action explanations cannot be causally reduced; however, I also agree with causal theories that an extrinsic relation between the agent and the event makes the difference between mere events and actions. I call this account an “agency-first” theory of action since it neither reduces agency—as in causalism—nor does it ignore agency to focus on the intrinsic features of actions—as in non-causalism. Instead, I claim we must not lose sight of the agent when analyzing action and thus posit the non-causal, yet extrinsic, relation of essential metaphysical dependence to explain action in terms of agency without losing the distinctive character of either concept.To this end, I claim that essential metaphysical dependence explains what makes an event count as an action by explaining how actions are grounded in agency. I first set-up the dialectic between causalists and non-causalists and raise objections to both views. I then describe the essential metaphysical dependence relation in detail and defend this account from several objections. Finally, the relation of dependence is commonly thought to be transitive, which entails a final significant objection—if actions depend on agency, and agency depends on non-agential forces, then actions are not really explained by dependence on agency. I argue, however, that plausible accounts of agency’s metaphysical emergence blocks the transitivity objection. I conclude that my agency-first theory adequately addresses what makes an event count as an action, while at the same time keeping the agent in view.

Orcid ID

Files over 3MB may be slow to open. For best results, right-click and select "save as..."