This article evaluates the structural conception of interests developed by Margaret Archer as part of her dualist version of critical realism. It argues that this structural analysis of interests is untenable because, first, Archer’s account of the causal influence of interests on agents is contradictory and, second, Archer fails to offer a defensible account of her claim that interests influence agents by providing reasons for action. These problems are explored in relation to Archer’s theoretical and empirical work. I argue for (...) an alternative account of interests that focuses on agents’ understandings of their interests and problems with these understandings. (shrink)