This article proposes the adoption of Goffman's concept of “framing” to characterize irony across its forms; the suggestion that this framing is achieved by a shift of footing reveals links between verbal irony and other forms of talk. Examination of irony in conversation shows how the shift of footing allows for detachment, enabling the ironist to make evaluations in response to perceived transgressions with reference to common assumptions. It is both the construction of an ironic turn and its placement in a sequence that make for the discernible shift of footing, and thus the visibility of the frame; with irony, conversational expectations of what constitutes a next turn are fulfilled on the level of form, but undermined on the level of content. This analysis shows the extent to which irony is affiliative, and reveals its hitherto unacknowledged subtlety of effect and range of attitude. I am most grateful to David Britain, Anita Pomerantz, Ad Putter, and Andrew Spencer for helpful discussions and useful comments on an earlier version of this article. Paul Drew, David Good, and Rachael Harris also read versions of my work, when it formed part of a Ph.D. thesis on conversational misunderstandings, and offered invaluable guidance. I am, in addition, deeply indebted to an anonymous referee from Language in Society whose observations on an earlier draft prompted me to reconsider my presentation of certain issues, particularly with regard to the work of John Haiman. Many ensuing improvements in clarity are owed to this referee. The study would have been impossible without the conversational data; the extent of my debt to those who supplied them will be evident to anyone reading this. My thanks to all.