This project addresses this CRC’s BCROSS research question by conducting a crosslinguistic investigation of the formal and discursive properties of reprise utterances (RUs), such as reprise questions (also called echo questions). RUs represent an ideal test case for answering BCROSS because they perform a nondefault, clarificatory update and also often display extraordinary formal morphological, syntactic, and prosodic properties. By adopting a crosslinguistic comparative methodology, this project will determine which of these exceptional properties are directly related to the special CG update function of RUs and which arise from grammar-internal and often language-specific factors. Once a clearer profile of the form-function relation underlying clarificatory updates is obtained, this will be used to determine whether an elaborate or simple model of the CG is required to model clarification.
The project’s particular focus is reprise questions and reprise fragments. By undertaking this empirical work, we will determine at the project-wide level which formal properties of RUs are related to their CG update function, and employ our results to determine which analyses of RUs survive crosslinguistic scrutiny. In addition, we assess the validity of a “speech-act” oriented view of RUs, according to which RUs are modelled as performative CG updates. Accomplishing these theoretical tasks requires us to also tackle a number of thorny issues concerning question formation, ellipsis, and the syntax-pragmatics and syntax-prosody interfaces. We also assess the extent to which RUs differ from similar non-canonical questions, such as quiz and courtroom questions.
The project staff are Prof. Dr. James Griffiths (PI), Dr. Timo Buchholz (postdoc), and Ebrar Beşinci (PhD student). The languages that we currently expect to investigate are English, German, Turkish, Italian, Japanese, and Czech / Russian. This sample balances typological breadth and logistical feasibility. Our data source are perceptual judgments of acceptability / pragmatic felicity and speech recordings. Whenever feasible, we issue formal judgment tasks, via crowdsourcing. Smaller, informal judgment tasks are employed when crowdsourcing is impractical. Our project works closely with Project B2 on the role of prosody in conveying bias in polar questions in Romance.
