| Project Detail |
"This research project focuses on underspecification, i.e. the imprecise use of words to express meanings which are not fully encoded in the semantics of the words themselves but which rely on common ground and other linguistic cues to be interpreted. I address in particular how underspecification explains the variable use of three French discourse markers and their English equivalents, namely ""et"" / ""and"", ""mais"" / ""but"" and ""donc"" / ""so"". This project seeks to establish the contextual and cognitive constraints to the production and interpretation of underspecified discourse markers by (i) analyzing their distribution and cooccurrence patterns in spoken, written and computer-mediated English and French and (ii) experimentally assessing their impact on discourse processing. This research thus combines corpus-based and experimental methods, as advocated by cognitivists.
According to the selfish hypothesis, underspecification is a failure of recipient design and is thus expected to be more frequent in unplanned speech and computer-mediated writing, where speakers/writers are under higher cognitive pressure than in planned speech/writing. Furthermore, this project intends to show that register-sensitive patterns of co-occurrence constitute compensatory strategies designed to minimize the potentially detrimental effect of underspecification in processing complexity. Register and cooccurrence patterns are thus considered as windows onto underlying cognitive mechanisms of interpretation.
SELFISH DISCOURSE will elucidate the relationship between a discourse marker and its co(n)text, teasing out their respective semantic-pragmatic contributions to the construal of senses and coming to terms with the theoretical overlap between polysemy, multifunctionality and underspecification. By combining corpus-based and experimental methods, this project contributes to a growing trend of linguistic research and fills a gap on the cross-modal study of underspecified discourse markers."
|
