Journal of Pragmatics 19 (1993) 435452 North-Holland 435 The discourse marker well: A relevance-theoretical account Andreas H. Jucker* Received November 1991; revised version September 1992 In Sperber and Wilson’s relevance theory, the discourse marker well can be seen as a signpost which constrains the interpretation process and the concomitant background selection. It signifies that the most immediately accessible context is not the most relevant one for the interpretation of the impending utterance. This analysis covers four more or less distinct uses of well: (1) as a marker of insufficiency; (2) as a face-threat mitigator; (3) as a frame; and (4) as a delay device. Relevance theory, which is a general theory of human communication based on cognitive principles, offers a unified explanation across a broad range of examples. 1. Introduction Discourse markers have attracted a lot of research recently, both in papers and in book-length studies. Some studies deal with a whole range of discourse markers (Schourup 1982, Schiffrin 1987, and Watts 1989 on English; Bazza- nella 1990 on Italian; and Holker 1991 on French), while others concentrate on individual ones (Lakoff 1973, Svartvik 1980, Owen 1981, James 1983, Carlson 1984, Schiffrin 1985, Watts 1986, and Blakemore 1988b). Or they are treated within larger discourse analytical frameworks (Labov and Fanshel 1977, Owen 1983). Sometimes the term ‘discourse particle’ is used inter- changeably with the term ‘discourse marker’. Schourup (1982) for instance, uses the former term to refer to the same linguistic elements for which Schiffrin (1987) uses the latter. Sometimes the two terms are used to refer to different phenomena. ‘Discourse marker’ is used to refer to elements like well, Correspondence to: A.H. Jucker, Justus-Liebig-Universitat Giessen, FB 10, Anglistik, Otto- Behaghel-Str. 10, D-W 6300 Giessen, Germany. E-mail: andreas.jucker@anglistik.uni-giessen.dbp.de * Earlier versions of this paper were presented at the universities of Poznan (Poland), Fribourg, Neuchltel (both Switzerland) and Giessen (Germany). I thank the audiences there for stimulating discussions. I also want to thank Claudia Caffi and two anonymous referees whose trenchant criticism caused some major revisions (and, I hope, improvements) of this paper. 0378-2166/93/$06.00 0 1993 - Elsevier Science Publishers B.V. All rights reserved