Discourse intonation: A corpus-driven study of prominence on pronouns

Not everyone agrees on what exactly is covered by the term “prosody.” Prosody is considered to be all of the suprasegmental features which include changes in pitch, “loudness, timing (including pauses) and voice quality” (Wichmann 2000: 8). According to Reed (2006: 3), prosody includes intonation, pitch, loudness, and time, with loudness and time being realized “in the form of duration, tempo, speech rate, rhythm and pause,” but excludes voice quality. The reason for excluding voice quality is because voice quality is optional, and therefore paralinguistic, in the sense that a speaker uses a change in her “normal voice quality for interactionally motivated reasons” (Reed 2006: 10). Cauldwell (2013: 325) offers a rather open-ended definition of prosody, namely a “collective term for the variations in pitch, loudness, tempo, rhythm and other features of the sound substance of speech.”. There is no a general agreement on what prosody is, and the same is true for intonation, a subset of prosody. Pickering (2012: 5437) provides both a narrow and a broad definition of intonation. The narrow definition is “the linguistically significant use of pitch movement at the suprasegmental level”; the broader definition includes “the alignment of the pitch contour with phrasal or clausal groups separated by pauses.” Cauldwell (2013: 325), taking the discourse intonation perspective, defines intonation as the “overall shape of a stretch of speech as created by a speaker's use of prominence, non-prominence, key and tone.” The intonation system in English has been investigated from different approaches, depending on linguistic schools and countries (Chun 2002). A grammatical or syntactic approach to intonation views intonation as segmental (see, for example, Chomsky and Halle 1968; Liberman and Prince 1977; Pierrehumbert 1980; Pierrehumbert and Hirschberg 1990) in which the functions of intonation are inherently linked to grammatical relationships. This approach sees rule-driven generative phonology as a natural, or inevitable, way of viewing such prosodic features based on their notion of generative grammar. It is grounded on the notion that tones align with specific syntactic structures.