霊長類における平静な状態で交わされる音声から発話の進化を探る

抄録

Human speech and vocal communication of nonhuman primates share many features in semantic ability such as referentiality of calls or pragmatic inference. By contrast, a large gap is found in the phonological ability; vocal learning or volitional vocal control are hardly observed in nonhuman primates. However, recent studies have shown similarity between speech and the vocal, facial communication of nonhuman primates used in low-arousal situations. Here I focused on such vocalizations, called contact calls and close calls. These vocalizations are widely seen among primate taxa. Contact calls function to keep group cohesiveness. Close calls are soft vocalizations used in face-to-face communication and function to signal the benign intent of the caller. These vocalizations exhibit the subcomponent abilities that enables human speech as follows. First, they show some plasticity in usage and acoustic features. The use of close calls becomes partner-specific as individuals mature, and this process is influenced by other group members. The acoustic features of contact calls are modified not only developmentally through interactions with parents but also temporally during vocal exchanges. Second, just as turn-taking in human conversation, timing adjustment has been found in both contact calls and close calls, indicating a vocal coordination ability in nonhuman primates. Third, the facial movements of close calls or lip-smacking have been revealed to be putative precursors of speech rhythm. Finally, some studies have shown that monkeys can acquire volitional vocal control through training. Although primates lack direct projection from the cortex to the vocal control region in the brainstem, initiation of such vocal utterance or lip-smacking involves similar brain areas as in human speech. Further investigation of vocal control during contact and close calls in social, natural settings as well as in experimental settings, would contribute to a better understanding of the evolutionary course leading to speech.