|View Our Catalog||
Prosodic Markers and Utterance
Boundaries in American Sign Language Interpretation|
Prosody in Spoken Languages: An Overview
Spoken language can be analyzed as having a hierarchical structure, with sounds as the smallest perceptible unit. Sounds can be grouped into syllables, which are assembled into words. These words are then grouped with adjacent words to create phrases of varying lengths, which may be grouped to form sentences and even longer strings of discourse. These various levels of spoken language are characterized by patterns of rhythm, timing, volume, and intonation—collectively known as prosody.
A universally accepted definition of prosody has been elusive. Shattuck-Hufnagel and Turk (1996) state that a satisfactory definition of prosody must include both a description of the relevant acoustic patterns of language as well as the higher-level constituents that account for the patterns. Crystal (1969) defines prosody as sets of phonetic properties, such as frequency (f0), duration, amplitude, quality, and reduction, which have variable relationships with their co-occurring segments. At one end is an abstract definition of prosody that is not necessarily linked to its expression and, at the other, a listing of the measurable suprasegmental features of language. Cutler et al. (1997) propose a definition that encompasses both perspectives by defining prosody as “an abstract structure coupled to a particular type of realization” (p. 142).
Prosody and syntax are bound together in the linguistic system. For example, prosody regularly marks certain syntactic structures, such as parentheticals, tags, nonrestrictive relatives, and appositives. Another prosody-syntax link may be found in the role of prosody in disambiguating sentences that may be understood in more than one way (Allbritton, McKoon, & Ratcliff, 1996; Lehiste, 1972; Price, Ostendorf, Shattuck-Hafnagel, & Fong, 1991; Streeter, 1978). Further, it has been argued that prosody contributes information about connections among constituents in discourse, conveying meaning beyond what is provided through lexical and syntactic systems (Swerts & Hirschberg, 1998; Wennerstrom, 2001). Finally, prosody interacts with syntactic structures to create boundaries in discourse, which are the focus of this book.
It is evident that there are many ways to convey similar ideas in language, both syntactically and prosodically. As Hirschberg (2002) observes, research on prosody is more a matter of “finding likelihoods,” rather than simply mapping prosodic features directly onto syntax or semantics (p. 32). It is known that speakers have options for the prosodic treatment of a given syntactic structure (Price et al., 1991); therefore, syntax does not entirely determine prosody. Additional factors that may influence a speaker’s decision about what prosodic features to employ for a given utterance include the focus of an utterance (Frota, 2000; Ladd, 1986; Pierrehumbert & Beckman, 1988), whether it contains new or previously given information (Brown, 1983), the assumptions shared by the speaker and the addressee, and speaking rate.
Prosody is perhaps the most complex and difficult area to study in the linguistic system because it is intimately related to every other aspect of that system, from phonetics to discourse (Sandler, 1999b). But it is precisely this centrality in human language that has made its study so engaging. Since spoken language prosody is better understood than signed language prosody, two areas of inquiry about spoken language prosody are examined in the following sections because of their relevance to understanding how boundaries are marked in speech: (1) the role of prosody in the comprehension of boundaries in discourse, and (2) the hierarchical organization of prosodic constituents.
Utterance Boundaries and Spoken Language Prosody
How do listeners understand spoken language? The tasks necessary to achieve comprehension include perceiving, decoding, interpreting, and often coconstructing the message that is produced by a speaker. To achieve comprehension, the listener must, among other things, recognize individual words and how they are linked together syntactically and semantically, and understand how the syntax fits into a larger discourse structure (Cutler et al., 1997). In addition, it appears that, through prior exposure to language, a listener intuitively knows that prosodic phrases divide an utterance into meaningful “chunks” of information (Bolinger, 1981). Chunking has been found to be important to comprehension and perceived naturalness of language production (Sanderman & Collier, 1997). That is, to comprehend language input, one must be able to parse it into meaningful units. Speakers use both syntactic and prosodic structure to parse the language stream into units, although the relative importance of the two may be different in different contexts.
This section reviews the body of literature that addresses how prosodic features mark phrasal boundaries in spoken language discourse. The study of the production and perception of prosody in relation to phrasal and syntactic structure has provided a wealth of information about the realization of prosody and its potential to affect comprehension. There are a variety of proposals as to the organizational structure of phrasal units (for a survey, see Shattuck-Hufnagel & Turk,