This article describes five political scientists' interview-based accounts of appropriate and inappropriate use of the pronouns "I" and "we" in academic writing. The informants talked about pronoun use with reference to one of their own journal articles and also by referring to other informants' texts. Beliefs about appropriate and inappropriate use varied widely, and it was emphasized that the discipline encompasses a number of subdisciplines, which helps account for these differing pronoun preferences. The insights and implications of the study are discussed, and a heuristic that combines corpus-based and interview-based approaches to the investigation of pronouns is proposed. The corpus-based part of the heuristic provides the researcher with data on typical disciplinary patterns of pronoun use, whereas the interview-based part provides accounts of informants' motivations and intentions that inform their pronoun use. It is argued that the heuristic could be adapted to investigate other linguistic features in academic writing. (Contains 2 tables.)