This thesis focusses on variability in audience interpretation of a television programme, and aims to problematise and investigate the reception of broadcast communication by applying the pragmatic theory of relevance (Sperber and Wilson 1966) to an empirical study of audience response. This aim is achieved using the following method: In Chapter Two I consider the scope of pragmatic theories of inferencing and conclude that relevance theory offers the only account which can both accommodate and provide the basis for an explanation of variation in interpretation. I also assert that for relevance theory to be able to show why an audience interprets a text in a specific way the cultural background of that audience has to be considered. In Chapter Three I show how existing studies of audience response which adopt a critical cultural studies approach require a more sophisticated model of communication than they currently assume if they are to realise their aim of relating audience respo nse to socio-political structures. My contention is that the inferential model proposed by Sperber and Wilson can provide such an account. Chapters Four and Five describe, and report the results of, an empirical study I carry out based on a methodology premised on relevance theory. The study consists of two separate interviews with audiences who have distinct cultural backgrounds in each of which I show a video recording of a television programme and then question the interviewees on their understanding of the text of the programme. In Chapters Six and Seven I discuss the results of the study in relation to relevance theory and media studies. The results of my study indicate that a methodology based on relevance theory can make explicit, and show the significance of, processes involved in audience interpretation of a media text which have not previously been open to analysis. Building on Sperber and Wilson's claim (1986: 15) that the context of an utterance is a psychological construct, and is a sub-set of the set of assumptions available to the hearer of a given utterance, the results make explicit (a) relevant aspects of the encyclopaedic knowledge of two distinct audiences; (b) the contexts these audiences produce in response to a television text; (c) how these contexts are related to the audience's encyclopaedic knowledge; (d) how these contexts affect the disambiguation and enrichment of information linguistically encoded in the text (e) 'The contextual implications, or interpretations, -the audience draw from a synthesis of the information encoded in text and the contexts the audiences apply. My findings are particularly pertinent for the critical cultural approach to audience studies as they indicate how it is possible to make explicit the relationship between response and cultural background by showing how the existing knowledge of an audience affects interpretation and indicating moreover how this knowledge can be related to social determinants. The results of my study also contribute to pragmatic theory in that they show how relevance theory can be used to explain why interpretation may vary.