Prosody, tone, and intonation

Abstract

Prosody refers to all suprasegmental aspects of speech, including pitch, duration, amplitude, and voice quality, that are used to make lexical and post-lexical contrasts. Tone refers to pitch patterns that make lexical, grammatical or morphological contrasts in some languages. Intonation refers to the melodic facet of prosody, although intonation and prosody are often used interchangeably. These three domains of speech are discussed together in this chapter because they share many underlying mechanisms. A common principle that can link them all is the articulatory-functional view, which sees speech as a system that transmits communicative meanings through an articulatory encoder. From this perspective, lexical, post-lexical, affective and social and indexical meanings are all conveyed in parallel, each coded by a unique contribution to the properties of various underlying articulatory targets. The articulation process ultimately generates the acoustic patterns perceived by listeners, which contain not only all the functional contributions, but also dynamic patterns of the articulation process itself. The chapter first outlines the articulatory dynamics underlying the production of pitch, duration, and voice quality, which serve as the basis of understanding how different functional components can be realized in parallel. The functional components are then introduced in terms of their specific communicative significance, ranging from lexical contrast through tone and stress, post-lexical demarcation through focus, sentence modality and boundary marking, and affective signaling through emotive and attitudinal prosody. Also discussed is how the articulatory encoding of the communicative functions gives rise to various prosodic phenomena, including pitch accents, prosodic structure and rhythm, and how various theories differ in their accounts of these phenomena.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 35 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.