Abstract
As it is appropriate to regard mental events as properties of their subject rather than as entities, so it is appropriate to treat actions as properties of the agent rather than as particulars. It is argued that the property approach to action should not be rejected because of the implausibility of the theories of Goldman and Kim; for properties need not and should not be individuated in their way. It is also argued that the question of treating actions as particulars or properties is to be settled on pragnatic grounds: it has no clear metaphysical significance. Finally it is argued that the logical form of action sentences, which Davidson endeavours to display by treating acts as particulars, can be shown with greater simplicity and plausibility on a property approach.