J. L. Austin famously distinguished between 'constative' utterances that convey information and 'performative' utterances that perform actions. In this groundbreaking new book, Douglas Robinson argues that Austin's distinction can be used to understand linguistic methodologies. Robinson uses Austin's model to introduce a new distinction between 'constative' and 'performative' linguistics. Constative linguistics, Robinson suggests, includes methodologies aimed at 'freezing' language as an abstract sign system cut off from the use of language in actual speech situations. Performative linguistics, on the other hand, covers methodologies aimed at exploring how language gets used or 'performed' in those speech situations. Robinson then tests his hypothesis on the act of translation. Constative linguists of translation always face the same problem: that the translator is always another utterer of the same utterance. In his book Robinson shows that this particular problem is solved when translation is seen as a performative utterance. Drawing on a range of language scholars and theorists including Grice, Peirce, Bakhtin, Wittgenstein, Burke and Derrida, Performative Linguistics consolidates the many disparate action-approaches to language into a single coherent new paradigm for the study of language as speech act, as performance - as doing things with words.