AbstractMuch work in history, anthropology, sociology, and political science has a narrative form — the events described are emplotted into stories. A number of recent critics of narrative have argued that the story form is a poor vehicle for social scientific explanation, as it often misleads us about the causal structure of the social world. Defenders of narrative typically claim that such criticisms miss the point of narrative. Even if narrative is not the best means for providing us with causal information, it can provide us with information about something else of importance such as the events' “meanings” or others' experiences. I reject such defenses of narrative, but I then offer a novel defense in their place. On the view I defend, narratives increase our understanding of the social world not by giving us some kind of special information about the social world but rather by cuing certain kinds of responses to it. I tie this conception of the epistemic function of narrative to the political role that narrative can play in correcting failures of interpersonal recognition and promoting structural change.