A constructional perspective on the rise of metatextual discourse markers

Abstract

Cognitive linguistics seeks to account for “a speaker’s knowledge of the full range of linguistic conventions” (LANGACKER, 1987; also GOLDBERG, 2006). It is surprising therefore that little attention has been paid in cognitive linguistics to the linguistic conventions called “discourse markers” (SCHIFFRIN, 1987) or “pragmatic markers” (FRASER, 2009, et passim). Pragmatic markers include signals of attention to social relationships (well, please), beliefs (I think, in fact), and discourse management (after all, anyway). Members of the third subtype are metatextual connectors of discourse segments (“discourse markers” in Fraser’s taxonomy). I argue that because pragmatic markers in general play a major role in negotiating meaning, they are an important part of speakers’ knowledge of language. Pragmatic markers are well-known not to have truth-conditional meaning, and not to be syntactically integrated with the host clause. However, they have conventional pragmatic meanings (HANSEN, 2012; FINKBEINER, 2019). I exemplify my recent research on the historical development in English of metatextual discourse markers with a diachronic construction grammar perspective on by the way (TRAUGOTT, 2020). Focus will be on the importance of routinized, replicated contexts in change (CROFT, 2001; BYBEE, 2010).