LingLunch : C. Pozniak & B. Hemforth

Céline Pozniak (Paris 8) et Barbara Hemforth (LLF)
The aboutness hypothesis: a new way to explain relative clause processing

Subject and object relative clause processing has been explained by a combination of syntactic, semantic, and pragmatic factors. We suggest that many of the phenomena (e.g. general subject preference, effects of animacy, discourse status of relative clause internal subjects) can be explained by a general principle, the Aboutness Hypothesis: a relative clause is most acceptable and easiest to process when everything contributes to making the head its optimal aboutness topic. This principle makes predictions for the role of implicit causality verbs which make subjects or objects good candidates for aboutness topics, a bias that may conflict with the role of the antecedent. In Experiment 1 & 2, we find in acceptability judgments and reading times that object relative clauses with subject-biased verbs are the least acceptable (and more difficult to process). We then show in Experiments 3 & 4 that implicit causality influences relative clause acceptability independent of syntactic factors such as intervention effects as well as semantics factors such as thematic roles. These results confirm the Aboutness Hypothesis as a new way to explain relative clause processing.