Projection in Discourse: A data-driven formal semantic analysis

Abstract

A sentence like "Bertrand, a famous linguist, wrote a book" contains different contributions: there is a person named "Bertrand", he is a famous linguist, and he wrote a book. These contributions convey different types of information; while the existence of Bertrand is presented as given information---it is presupposed---the other contributions signal new information. Moreover, the contributions are affected differently by linguistic constructions. The inference that Bertrand wrote a book disappears when the sentence is negated or turned into interrogative form, while the other contributions survive; this is called 'projection'. In this thesis, I investigate the relation between different types of contributions in a sentence from a theoretical and empirical perspective. I focus on projection phenomena, which include presuppositions ('Bertrand exists' in the aforementioned example) and conventional implicatures ('Bertrand is a famous linguist'). I argue that the differences between the contributions can be explained in terms of information status, which describes how content relates to the unfolding discourse context. Based on this analysis, I extend the widely used formal representational system Discourse Representation Theory (DRT) with an explicit representation of the different contributions made by projection phenomena; this extension is called 'Projective Discourse Representation Theory' (PDRT). I present a data-driven computational analysis based on data from the Groningen Meaning Bank, a corpus of semantically annotated texts. This analysis shows how PDRT can be used to learn more about different kinds of projection behaviour. These results can be used to improve linguistically oriented computational applications such as automatic translation systems.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 41 citations exist upstream.