In this final section of the book (Chapters 10, 11 and 12), we address pedagogic issues, commencing with an exploration of translation ‘errors’. So far in our discussion of the role of the translator as communicator, text-level errors have been mentioned often enough to justify a section of the book being devoted entirely to an examination of the topic. In this chapter, we shall leave aside mismatches of propositional meaning or breaches of the target language code (which may be due to inadequate language competence on the part of the translator) and focus on a number of problems in language use which can only be adequately accounted for as mismatches of text and context (which may be due to problems of textual competence). Although the term ‘error’ is not entirely appropriate (see further, Chapter 12), we shall, for the sake of convenience, refer to these as text-level errors, to be considered within a comprehensive model of discourse processing. The various components of the model have already been introduced, and only those aspects relevant to the analysis of errors beyond-thesentence will be looked at more closely here. Categories belonging to register membership, pragmatics and semiotics will be invoked in an attempt to explain real cases of communication breakdown in both translation and interpreting.