The structure and function of intonational paragraphs in native and nonnative speaker instructional discourse

In the context of classroom communication, there is a premium on the clarity of the message, and instructors will typically employ multiple linguistic cues to highlight information structure. Using a model of intonation in discourse [Brazil, D. (1986). The communicative value of intonation in English. Birmingham, England: University of Birmingham, English Language Research. Brazil, D. (1997). The communicative value of intonation in English. Cambridge: Cambridge University Press.], this paper examines the use of intonational paragraphs as an organizational tool in the teaching discourse of native speaker teaching assistants and nonnative speaker international teaching assistants at a North American university. Analysis of the native speaker data presents evidence of intonational paragraphs defined by phonological criteria and used by the speakers to underscore local and global information structure. Comparative analysis of the parallel nonnative speaker data shows a considerably weaker control of intonational structure and a disturbance in prosodic composition that materially affects the comprehensibility of the discourse for native speaker hearers. The paper closes with a discussion of some of the implications of this study for international teaching assistant programs. # 2003 The American University. Published by Elsevier Ltd. All rights reserved. University faculties and graduate programs have become increasingly diverse. The numbers of international teaching assistants (ITAs) in scientific and technical fields such as engineering, mathematics, and laboratory sciences continue to grow, and the majority of US undergraduates are now likely to have important contact with international staff in their introductory courses. Despite the widespread creation of screening programs developed to assess the linguistic ability of these instructors, English for Specific Purposes 23 (2004) 19–43 www.elsevier.com/locate/esp 0889-4906/03/$30.00# 2003 The American University. Published by Elsevier Ltd. All rights reserved. doi:10.1016/S0889-4906(03)00020-6 * Tel.: +1-205-344-4567; fax: +1-205-348-1388. E-mail address: [email protected] (L. Pickering). communication failure between ITAs and their students is not uncommon, and the need for continued development of more effective educational programs remains acute. Researchers and practitioners who are focused on ‘‘the ITA problem’’ recognize the limitations of traditional pedagogical approaches and employ ‘‘integrated frameworks’’ that apply innovative work in discourse analysis to the particular issues presented by this group [see, for example, Madden & Myers (1994) Discourse and Performance of International Teaching Assistants]. Using this approach, a range of linguistic information-structuring devices has been identified and investigated in studies of L1 and L2 academic discourse including rhetorical organization, lexical discourse markers, and syntactic complexity (Dudley-Evans, 1994; Rounds, 1987; Tyler, 1992; Tyler & Davis, 1990; Tyler et al, 1988). An equally important linguistic strategy used by speakers to elucidate information structure in academic discourse is the creation of intonational paragraphs (Barr, 1990; Thompson, 2003). The term ‘‘intonational paragraph’’ refers to a unit above the level of the tone unit and equivalent to the paragraph in written discourse (Lehiste, 1979). Due to differences in analytical approaches and to the investigation of a variety of discourse genres where these structural units may manifest themselves differently, speech paragraphs have been variously labeled as ‘‘major and minor paratones’’ (Brown, Currie, & Kenworthy, 1980; Couper-Kuhlen, 1986; Yule, 1980), ‘‘pitch sequences’’ (Brazil, 1986, 1997), ‘‘sequence chains’’ (Barr, 1990), ‘‘phonological paragraphs’’ (Lehiste, 1979; Tench, 1996; Thompson, 2003), and ‘‘major tone groups’’ (Wichmann, 2000). A number of commonalities of both the structure and function of these units, however, unite these descriptions. Analysts agree that speech paragraphs are produced and interpreted by speakerhearers using the phonetic cues that appear at their boundaries. The most prominent cues include a high pitch onset (as measured by fundamental frequency [Fo]) with an accelerated rate and volume, and a low pitch close possibly accompanied by laryngealisation, a drop in volume, and a narrowing of the pitch range (Brown et al., 1980; Lehiste, 1979; Wichmann, 2000; Yule, 1980). In addition to these boundary criteria, Tench (1996) suggests that there will be a ‘‘gradual descent’’ in pitch from the first to the final tone unit within a single speech paragraph, and Wichmann (2000) reports a similar tendency toward ‘‘supradeclination’’ within units in experimental data and broadcast news items. Changes in pitch level are accompanied by systematic variation in pause lengths. In perception studies of pause boundaries, Swerts and Geluykens (1993) found that longer pauses increased listener perception of boundary strength, and Cutler, Dahan, and Donselaar (1997) report studies in which speakers consistently produced pauses to create ‘‘paragraphs’’ in narrative retellings. Both pitch resets and pause structure interact with other levels of discourse organization in the creation of speech paragraphs, most notably, topic structure. Production 1 By tone unit, I refer to the unit of intonational analysis traditionally recognized as consisting of a single intonation contour and containing a nuclear or tonic syllable. These are also referred to as sensegroups, breath-groups, tone-groups, intonation-groups, phonological phrases, or intonational phrases. For more detailed discussion, see Cruttenden (1997) and Wennerstrom (2001). 20 L. Pickering / English for Specific Purposes 23 (2004) 19–43