Unspoken Intonation: The Prosody of Pro-Speech Gesture

Abstract

Meaning is not generated only through syntactic/semantic configurations of words and morphemes that have been traditionally located in the mental lexicon; it also arises from suprasegmental features such as prosody and gesture. A majority of linguistic work regarding the interaction of speech and gesture is primarily concerned with gesture which co-occur with speech (i.e., "co-speech gesture"), but there is comparatively little work done on gestures which have their own timeslots within speech (i.e., "pro-speech gesture"). Co-speech gesture has been shown to have close ties to prosodic structure, but there has been little to no insight as to pro-speech gesture and its relation to prosody. This thesis concerns itself with that relationship, including: (i) the potential uses of pro-speech gesture, (ii) strategies for the prosodic integration of pro-speech gesture into speech, and (iii) the ways speakers realize prosody during pro-speech gesture. By examining a small corpus of 49 annotated clips, I found that speakers (i) use pro-speech gesture for a wide range of reasons, including euphemism, recall, and emphasis, (ii) often integrate pro-speech gestures into prosodic phrases, going so far as to use them to host clitics, and (iii) realize prosody explicitly through vocalizations accompanying pro-speech gestures as well as implicitly through the gestures themselves. Altogether, this suggests that speakers can account for gestures incredibly early in planning, integrating them smoothly into their speech as they would any other spoken word. This further supports hypotheses such as McNeill & Duncan's (2000) "growth point hypothesis" and Kita's (2000) "information packaging hypothesis", which attest to the communicative uses of gesture and its origin alongside speech. This has further implications for prosody, writ large, in particular with respect to the role of items typically considered silent.