Relevance theory attempts to provide a psychologically realistic, explicit account of communication. It makes foundational claims about both cognition in general and utterances and how they are processed in particular. The former is the cognitive principle of relevance: cognition tends to seek maximal relevance, where an input to a cognitive process is more relevant the more positive effects it has on the mind’s representations of the world, and less relevant the greater the effort required to derive them. Although on this view we have a tendency to seek the greatest possible payoff for the least possible effort, there is no general guarantee that an input to a cognitive process will be relevant. However, communication is special. Speakers want to be understood, and they therefore tailor their utterances to their audience. Relevance theory claims that this raises a defeasible expectation that the utterance will be “optimally relevant”; that is, that it is both relevant enough to be worth processing and as relevant as the speaker is willing and able to make it. (This is the communicative principle of relevance.) It further claims that this mandates the relevance-theoretic comprehension heuristic: a fast and frugal procedure dedicated to processing utterances. Relevance theory claims that what a speaker communicates falls into two classes: explicatures, or propositions that are developments of the logical form of the sentence uttered, and other propositions conveyed, which are implicatures. A further fundamental assumption of relevance theory is that linguistically encoded meaning radically underdetermines the content that a speaker intends to convey. Much research has focused on investigating this linguistic underdetermination and on developing accounts of the interpretation of particular linguistic items and types of utterances. Specific areas of research include lexical pragmatics; figurative speech, including metaphor and irony; the interpretation of discourse connectives and linguistic items that have non-truth-conditional meaning; and the interpretation of logical linguistic items such as and, if . . . then, and negation. Turning briefly to the history of the field: relevance theory is grounded in the philosopher Paul Grice’s work on meaning and conversation, and the theoretical advances of the cognitive revolution in linguistics and psychology. It was initially developed by Dan Sperber and Deirdre Wilson in the late 1970s and 1980s, and has been one of the leading pragmatic theories since then. Both Sperber and Wilson continue to be active in developing the theory. Other key contributors include Diane Blakemore, who introduced the notion of procedural meaning, and Robyn Carston, who is best known for her work on the semantics/pragmatics interface and linguistic underdeterminacy. Relevance theory has contributed considerably to the emerging fields of experimental and developmental pragmatics, and it is in dialogue with philosophy of language.