Prosodic features of polite speech

This paper uses interactional data to investigate the acoustic characteristics of polite or deferential speech in Korean. We asked fourteen Korean speakers to perform two tasks with two different interlocutors: a status superior and a friend. Consistent with previous studies of non-interactional data, deferential speech has lower pitch and shimmer, and quieter final syllables. However, divergent from previous studies, deferential speech featured higher jitter (in some locations), higher shimmer and higher H1-H2 (on one task). Through analysis of different locations in prosodic structure, we found that females used more pitch variation on final syllables in deferential speech. We argue that these mixed results show the importance of context in signalling vocal politeness, and also complexities of using interactional data. The findings advance the study of multimodal politeness beyond the analysis of experimental data.