Organization of discourse

type of devices (single-word adverbials) and used them in a similar way (i.e., in sentence-initial positions).

To summarize, the studies reviewed above still fall under the label of textualization as they all tend to focus on lexical and/or grammatical items. However, at this stage the question is no longer one of frequency of occurrence but rather the functions these lexical and/or grammatical items fulfil in the macro-organization of discourse.

Rowley-Jolivet and Carter-Thomas (2005a), for example, examined nine oral conference presentations (as well as the nine articles corresponding articles published in the conference proceedings) by native and non-native speakers of English in terms of information structure. Focusing, more specifically, on the use of the passive voice, extraposition, inversion, and pseudo-cleft sentences, Rowley-Jolivet and Carter-Thomas attempted to provide an explanation for both inter-genre and inter-group variation by shedding “light on the articulation between genre, syntactic choices, and rhetorical appropriacy” (p. 59).

Taboada (2006), on the other hand, used Mann and Thompson’s (1988, as cited in Taboada, 2006) Rhetorical Structure Theory in order to explore whether discourse markers signal rhetorical relations. In her study, she used a corpus of task-oriented dialogues and newspaper articles and concluded that a high proportion of rhetorical relations in the texts are not signalled, although genre variations did occur between the two corpora used in the study.

Flowerdew (1992) also studied academic lectures; however, instead of exploring discourse markers, he investigated how a particular kind of speech act, namely, definitions structure the lecture and, in turn, how they are signalled gramatically and/or lexically. His results indicate that definitions tend to appear rather frequently in lectures, either with the function of “signposting the logical discourse structure” (p. 215) or assisting comprehension.

To summarize, the studies reviewed above all exhibit a distinct preference for merging rhetoric and linguistics. Rowley-Jolivet and Carter-Thomas’s (2005a) examination of information structure, Taboada’s (2006) study featuring Rhetorical Structure Theory, and Flowerdew’s (1992) work merging speech act theory and cohesion were precursors to the next development focusing on the organization of

discourse, that is, studies focusing on “general global patterns of discourse organization” (Bhatia, 2004, p. 19).

2.3.9.1.2 General global patterns of discourse organization

After the emphasis on rhetorical patterns, focus shifted to the study of structural elements within discourse. As Bhatia (2004) argues, for studies conducted within this research paradigm, “the emphasis … was on regularities of discourse organization in terms of structural elements, which were often broadly explained by reference to the rhetorical processes involved in the construction of such texts” (p. 9).

For example, Wells (1993) explored the initiation – response – follow-up (IRF) sequence, a characteristic sequence in spoken classroom discourse, dating back to the work of Sinclair and Coulthard (1975, as cited in Wells, 1993). Based on material collected in a Grade 3 classroom during one unit of study time, Wells found that the IRF sequence is an oversimplification of the structure of classroom discourse and that the task at hand plays an important role in the actual manifestations of the individual elements within the sequence.

Similarly to the majority of studies conducted on spoken genres, Olsen and Huckin (1990) and Allison and Tauroza (1995) also focused on academic lectures;

however, instead of focusing on discourse markers, their studies focused on the macrostructures of discourse and how these aid or hinder the understanding of the material by non-native speakers of English. The methodology used by the two studies was also the same: the researchers asked both native and non-native speakers of English to listen to a lecture, take notes, and subsequently summarize verbally the information. These immediate-recall summaries were recorded, transcribed, and analyzed in terms of correct vs. incorrect information. The findings by Olsen and Huckin show that the students who managed to grasp the overall macrostructure of

the point-driven text were more successful at recall. For Allison and Tauroza (1995), it was the “situation – problem – solution – evaluation discourse structure” (p. 161, emphasis in the original) that was in the centre of attention. Their results showed that

“expectations of discourse development contribute … to the difficulties [learners]

face in interpreting unexpected moves in more elaborate lecture discourse” (p. 166).

Limberg (2007) conducted an analysis on a largely neglected type of academic talk, namely, office hour interactions between students and teachers in a tertiary education setting. Similarly to Flowerdew (1992), Limberg used the analytic tools of conversation analysis (cf., Heritage, 1984, 1995) to delineate a five-phase structure to office hour interactions: (a) prefacing sequence, (b) opening, (c) outlining of academic business, (d) negotiation of academic business, and (e) closing.

To conclude, what is shared among these studies is a focus on structural elements in discourse. Wells’s (1993) investigation focusing on the IRF sequence in classroom interactions, Allison and Tauroza’s (1995) study on the problem-solution pattern, and Limberg’s (2007) speech act analysis of office hour interactions highlighted the importance that macrostructures play in constructing and interpreting discourse, an idea that was taken further by those whose main research interests lay in analysing genres.

2.3.9.1.3 Cognitive structures and rationale in genres

The last of the three developments that fall under the organization of disourse was characterized by an interest in the definition and description of genres, which resulted in the three somewhat overlapping research traditions in genre analysis, which were reviewed above: the systemic-functional school (Eggins & Martin, 1997), the new rhetoric school (Miller, 1984), and the ESP school (Bhatia, 1993, 2004;

Swales, 1990, 2004). As opposed to the studies reviewed in the previous section

which implied that compulsory and optional structural elements provided a sufficient description of discourse, in this research trend the communicative purpose of different stretches of text (Swales, 1990) and cognitive structures (Bhatia, 1993) have replaced the strong reliance on structure. Common to the studies reviewed below is the idea that genres comprise of a set of moves and steps and the one that the move-steps analyses should be supplemented with contextual information in order to investigate the rationale for the existence of the given move structure (Bhatia, 2004). The rationale for this conducting studies in this orientation is that “specialist writers seem to be fairly consistent in the way they organize their overall messages in a particular genre, and analysis of structural organization of the genre reveals preferred ways of communicating intention in specific areas of inquiry” (Bhatia, 1993, p. 29).

The study by Edwards, et al. (1997) presents an interesting case as the audience for the paper is the readership of the Journal of Surgical Research, whom the article attempts to equip with the genre knowledge necessary for successful participation in conferences. Accordingly, instead of grounding the presented move-like description in data collected and analysed systematically, merely some seemingly mutually understood requirements of medical conference presentations are presented.

In this respect, the paper shows closer resemblance to the manuals published by the professional community Bhatia (1993, 2004) recommends should be consulted upon an attempt to describe a particular genre.

In a more traditional study, Weissberg (1993) provided an analysis of the graduate seminar. Employing analytical tools and terminology from a variety of traditions (e.g., the ethnography of speaking, cf., Saville-Troike, 2003; register studies, cf., Halliday, 1978; and move analysis, cf., Swales, 1990), Weissberg provides a detailed account of an “other research process genre” (Swales, 1990, p.

177). Due to the fact, however, that the three rather distinct research traditions are not integrated into a coherent whole, the study remains largely superficial in its treatment of the genre and falls short of the criteria and recommendations set forth by Bhatia (1993, 2004).

Thompson (1994) conducted research on a part-genre of academic lectures, their introductions. Using a corpus of 18 different introductions from lectures delivered in a range of disciplines, she attempted to establish a series of Functions and Sub-functions (cf., moves and steps in Swales, 1990) as well as the sequence of these.

While Thompson was successful in delineating the two main functions of lecture introductions (i.e., “setting up the lecture framework”, p. 176, and “putting the topic in context”, p. 178) as well as the sub-functions (e.g., “show importance/relevance of topic”, “relative ‘new’ to ‘given’”, or “refer to earlier lectures”, p. 178, for the latter main function), her results did not indicate a definite preference for a given order of these elements. One possible explanation she offers for this phenomenon is based on Swales (1990), who argued that spoken genres tend to exhibit a greater amount of flexibility as far as the sequencing of the elements is concerned.

The studies reviewed above share a number of features. Firstly, they are all examples of studies that have conducted research in the socio-cognitive space (Bhatia, 2004); secondly, they all feature a description of the context that has been claimed to manifest itself in a number of ways in the text. This marked the departure from the organization of discourse to the contextualization discourse, which will be reviewed in the following sections.