Thirty years ago I wrote a book called Text and Context. That book deals extensively, and quite formally, with text, but much less with context – a notion that is of crucial importance in understanding how discourse is embedded in society. In my later work in Critical Discourse Studies, for instance on racism, ideology and discourse, context is extensively dealt with as a social background for discourse, but analyzed theoretically hardly at all. Traditionally, in the study of language and discourse, context is conceived of in terms of independent social variables, such as gender, class, ethnicity, age or identity, or as social conditions of text and talk.

Both formal and ethnographic studies of indexicality define contexts rather in semantic terms, for instance as referents for deictic expressions, but most of such work is limited to spatial or temporal orientations of participants.

Speech act theories have formally accounted for some of the properties of Speakers and Hearers, such as their knowledge, wishes or status, so as to formulate appropriateness conditions, but have not further pursued a systematic analysis of such contextual conditions.

Critical Discourse Analysis (CDA) is crucially interested in the social conditions of discourse, and specifically in questions of power and power abuse, but has also failed to develop more explicit theories of context as a foundation for its own critical enterprise.