Reliable discourse markers for contrast relations

Abstract

Using the RST annotated corpus [Carlson et al., 2003], we use simple statistics on the distribution of discourse markers or cue phrases as evidence of the three-way distinction of Contrast relations, Contrast, Antithesis and Concession, recognized in standard Rhetorical Structure Theory (RST, Mann and Thompson 1987). We also show that however, an intuitive marker of Contrast, is not actually used statistically significantly more often in Contrast relations than in Cause-Effect relations. These results highlight the need for empirically based discourse marker identification rather than the intuitive method that is the current norm.