Teaching Pronunciation from the Top Down

Abstract

In this paper, a theoretical and pedagogical foundation for research efforts is provided. Pronunciation is examined from a contextual, "top-down" perspective from which segmental articulation assumes less importance than more general properties of speech such as rhythm and voice quality. Pronunciation is described as conveying many different types of messages to a hearer related to the information structure of a discourse, the speaker's attitude and mood, and other social and psychological features of the speaker or of the relationship between the speaker and hearer. Moreover, various aspects of pronunciation are shown to relate to specific gestures. The aim is to present a more descriptively enlightening and pedagogically useful characterization of second language phonology than traditional treatments, in which phonology was identified with discrete articulations and in which suprasegmental features were relegated to the periphery of language per se, i.e., to the paralinguistic and in some cases the extralinguistic domains of communication. Suggestions for teaching pronunciation are set in a context of research and theory, and a focus on the non-segmental characteristics of speech is advocated. This discussion makes reference to the use of video and computer media in pronunciation training (see Pennington forthcoming for further discussion), as well as to the use of more traditional types of audiovisual aids. The paper concludes with a set of research questions on pronunciation instruction derived from this investigation.