Inferring Meaning from Indirect Answers to Polar Questions: the Contribution of the Rise-Fall-Rise Contour
Marie-Catherine de Marneffe, Judith Tonhauser · 2019
Polar questions can be given direct answers (e.g., Do you want to eat? – No) and indirect answers (e.g., Do you want to eat? – I’m not hungry). Listeners infer positive or negative responses from indirect answers to polar questions with varying degrees of confidence (e.g., Clark 1979, Hirschberg 1985, Green & Carberry 1992, 1994, de Marneffe et al. 2009). For spoken language, the prosodic realization of the indirect answer has been speculated to provide a cue to the intended meaning of the indirect answer (Green & Carberry 1999, fn. 34). This paper presents an experiment designed to identify whether and how the prosodic realization of an indirect answer to a polar question influences the response that listeners infer from the indirect answer. The experiment explored American English listeners’ interpretations of indirect answers with scalar adjectives (e.g., She’s attractive) realized with a neutral contour (H* L-L%) or the rise-fall-rise contour (L*+H L-H%) in response to polar questions with semantically stronger adjectives (e.g., Is your sister beautiful?). Listeners inferred significantly more negative responses to the polar questions when the indirect answer was realized with the rise-fall-rise contour than with the neutral contour. These findings show that the prosodic realization of an indirect answer can provide a cue to the speaker’s intended meaning. The paper also discusses implications of our findings for scalar implicature generation and the meaning of the rise-fall-rise contour.