Vocal Production by a Language-Competent Pan paniscus

Abstract

Human spoken language and nonhuman primate vocalization systems have traditionally been regarded as qualitatively different from one another with respect to their semanticity and the way in which individuals acquire and utilize these signals. However, recent studies of the vocal behaviors of both captive and free-ranging monkeys and apes suggest that this dichotomy may not be unequivocal. We examined the vocalizations produced by a linguistically-competent adult male bonobo (Pan paniscus) named Kanzi. We analyzed his vocalizations during communicative interactions with humans in order to determine whether they vary systematically according to the semantic context in which they are produced. We determined semantic contexts based upon a vocalization's co-occurrence with predefined behavioral correlates. Spectrographic and statistical analyses revealed that acoustic structure is similar among the vocalizations that occurred within a specific semantic context and structural differences are evident between the vocalizations produced in different contexts. The results provide evidence that, during communicative interactions with humans, Kanzi modulates his vocal output on both the temporal and spectral levels.