Eötvös Loránd University, Budapest Faculty of Education and Psychology
PhD School of Education
(Director: Dr Éva Szabolcs, PhD habil.) PhD Programme in Language Pedagogy
(Director: Dr Péter Medgyes, DSc)
The World Universities Debating Championship Format: A Move-Based and Contextual Analysis
by
Gergely J Tamási
Supervisor: Dr Krisztina Károly, PhD habil.
Members of the Defence Committee:
Dr Kristóf Nyíri, CSc (Chair) Dr Dorottya Holló, PhD habil. (Referee) Dr Ágnes Magnuczné Godó, PhD (Referee)
Dr Éva Illés, PhD (Secretary) Dr Péter Medgyes, DSc (Member) Dr Enikő Németh T., CSc (Member)
Dr Pál Heltai, CSc (Member)
2012
Abstract
The aim of the present study was to establish a grounded theory regarding the rhetorical structure of the opening speeches that are delivered by debaters in a widely used format of academic debate, the World Universities Debating Championship (WUDC) format. Using a corpus of ten video-recorded opening speeches, four interviews with expert debaters who are familiar with the format, and handbooks and manuals on academic debate to triangulate the findings, a model is established, which outlines a three-move structure of opening speeches: (a) setting the scene, (b) justifying the plan, and (c) closing the scene, which may be further divided into a number of steps. The findings of the research are relevant for and provide useful insights into argumentation theory, argumentation pedagogy, and genre theory.
Table of Contents
1 Introduction ... 1
2 Review of literature ... 8
2.1 Research on academic debate ... 8
2.1.1 Best practices in debating ... 9
2.1.2 The results of empirical research ... 10
2.1.3 Theoretical research on argumentation and debate ... 12
2.1.4 Summary ... 15
2.2 Linguistic analyses of argumentation ... 16
2.3 Research on genre ... 19
2.3.1 Schools of genre analysis ... 22
2.3.2 Australian systemic-functional linguistics: Genre as a staged, goal- oriented social process ... 23
2.3.3 American new rhetoric: Genre as typified rhetorical actions based in recurrent situations ... 30
2.3.4 British English for Specific Purposes (ESP) school: Genre as conventionalized communicative event ... 39
2.3.5 Summary ... 49
2.3.6 Academic debate: A genre ... 52
2.3.7 The evolution of spoken genre analysis through empirical studies on spoken genres ... 58
2.3.8 Textualization of lexico-grammar ... 59
2.3.9 Organization of discourse ... 64
2.3.10 Contextualization of discourse ... 70
2.3.11 Summary ... 75
2.3.12 The preliminary study ... 76
2.4 Research niches and research question ... 78
3 Methods ... 80
3.1 Research orientation ... 80
3.2 Data collection ... 85
3.3 Data analysis ... 89
3.4 Quality in qualitative analysis ... 93
4 Results and discussion ... 96
4.1 Contextual analysis of the WUDC sub-genre of academic debate ... 96
4.1.1 Motions in the WUDC format ... 97
4.1.2 Speaker roles and responsibilities in the WUDC format ... 101
4.1.3 Points of Information (POI) ... 107
4.1.4 Tournament schedule and preparation for the rounds ... 108
4.2 Move analysis of the Prime Minster speech ... 110
4.2.1 Setting the scene (Move 1) ... 114
4.2.2 Attracting the audience’s attention (Step 1) ... 114
4.2.3 Pointing out the harm (Step 2) ... 116
4.2.4 Pointing out the significance of the harm (Step 3) ... 117
4.2.5 Offering interpretation to the motion (Step 4) ... 118
4.2.6 Establishing framework for debate (Step 5) ... 125
4.2.7 Previewing arguments (Step 6) ... 126
4.2.8 Summary ... 128
4.2.9 Justifying the plan (Move 2) ... 132
4.2.10 Providing constructive arguments (Step 1) ... 132
4.2.11 Refuting a possible attack on the Government case (Step 2) ... 142
4.2.12 Summary ... 146
4.2.13 Closing the scene (Move 3) ... 149
4.2.14 Highlighting the most important aspects of the case (Step 1) ... 149
4.2.15 Restating the motion (Step 2) ... 151
4.2.16 Summary ... 152
4.2.17 Sample analysis ... 153
4.2.18 The distribution of the moves and steps in the corpus ... 162
4.3 The effect of the context on the move structure of the speeches ... 169
4.4 The rhetorical structure of WUDC speeches in light of the literature on argumentation and debate ... 177
4.5 The rhetorical structure of WUDC speeches in light of the genre studies literature ... 181
5 Conclusion ... 185
References ... 190
Appendix 1. The final interview protocol ... 213
Appendix 2. The rhetorical structure of Prime Minister speeches ... 214
1 Introduction
Argumentation has been widely regarded as an “essential ‘tool’ for life in western, democratic societies” (Combs, 2004, p. 55). Discussion and debate are present everywhere from the society at large, where the legislative, the executive, and the judiciary are all based on argumentation, to the more closed academic circles, where scholars are expected to acquire and demonstrate skills in critical thinking, reasoning, and debate (cf., Popper, 1945, on critical rationalism). Because of this dependence on argumentation, curricula in tertiary education, mostly in the West, have featured explicit instruction in argumentation or argumentation-based genres albeit this instruction has predominantly targeted the written form of argumentation.
Although the presence of instruction in written argumentation has already enabled practitioners to hone their students’ skills in critical thinking and related skills, oral argumentation has also been called for in the classroom, both in secondary and tertiary education (Farrow, 2006). On the one hand, in tertiary education, the lack of oral argumentation is perturbing. Kim (2006) conducted research on East Asian students studying in tertiary education in the United States and argues that English for Academic Purposes (EAP) courses should include more oral work that simulate task types (e.g., debates) from content courses the students encounter. Gring (2006) argues that speeches all inherently feature argumentation. On the other hand, what applies more to English as a foreign language (EFL) classrooms, according to the Common European Framework of Reference for Languages (Council of Europe, 2001), is that students already at the Threshold (B1) level should be able to “briefly give reasons and explanations for opinions, plans, and actions” (p. 59), while by the time they reach the Vantage (B2) level, they must be able to “develop an argument
systematically with appropriate highlighting of significant points, and relevant supporting details” (p. 59). Consequently, in Hungary, the Vantage-level Matura examination in modern foreign languages, inspired by the Common European Framework of Reference for Languages (Council of Europe, 2001), now incorporates a compulsory “For and Against” (i.e., debate) task, as a result of which there has been a revived interest on the part of EFL (and other modern foreign language) instructors to introduce debating into their classrooms (Tamási, 2008a).
The aim my research is to establish a grounded theory regarding the way in which academic debaters participating in the World Universities Debating Championships (WUDC) construct their opening speeches. There are three terms that are in need of a definition here: grounded theory, academic debate, and the WUDC format of academic debate1. Firstly, grounded theory research is “a method of conducting qualitative research that focuses on creating conceptual frameworks … through building inductive [italics added] analysis from the data” (Charmaz, 2007, p.
608). What lies at the heart of this method is the conviction that “concepts and theories are constructed by researchers out of stories that are constructed by research participants who are trying to explain and make sense out of their experiences and/or lives, both to the researcher and themselves” (Corbin & Strauss, 2008, p. 10).
Secondly, Freeley and Steinberg (2005) define academic debate as the following:
Academic debate is conducted on propositions in which the advocates have an academic interest, and the debate typically is presented before a judge or an audience without direct power to render a decision on the proposition. In fact, in academic
1 The label used for the sub-genre by the parent discourse community is British Parliamentary Debate
debate the judge is instructed to disregard the merits of the proposition and to render a decision on the merits of the debate [italics added]. The purpose of the debate is to provide educational opportunities for the participants. (p. 15)
Freeley and Steinberg contrast academic debate with applied debate (i.e., debates where the audience has “the power to render a binding decision on the proposition”, p. 15, such as presidential, judicial, television, etc. debates); however, their definition might still encompass any type of debate that takes place in academic life, be them vivos, conference presentations, or panel discussions. It is important to note, although, that the present dissertation does not purport to investigate such a variety of genres, nor does it claim that the findigs will be applicable to such diverse genres. Instead, it proposes to investigate a given format of academic (educational) debate. (This point will be further discussed when arguing for the genre status of academic debates in Section 2.3.6.)
Thirdly, the WUDC format is a format of academic debate which features four teams of two speakers (two teams on the proposition side and two teams on the opposition side), with the first speaker of the proposition team beginning the debate and the second speaker of the second opposition team ending it. Each speaker delivers one 7-minute speech, the first and last minute of which is protected, that is, it may not be interrupted by members of the teams on the other side. During the intermediary 5 minutes, on the other hand, the speaker holding the floor may accept points of information, that is, “a brief statement or question to an argument claim, example, or other point that is being made by the speaker” (Meany & Shuster, 2002, p. 206). At the WUDC, teams usually receive the propositions to be debated 15 minutes before the round commences, and speakers do not have any preparation time between the
subsequent speeches. To paraphrase the aim of the research, with these definitions in mind, I have attempted to carry out an inductive, qualitative investigation on the way in which WUDC participants explain the reasons behind their methods of speech construction (cf., Charmaz, 2007; Corbin & Strauss, 2008), in an academic debate format that calls for (a) superior argumentation skills because the debates are judged based on the quality of arguments advanced as opposed to the judges’ perception of the truth value of the proposition (Freeley & Steinberg, 2005) and (b) a significant amount of spontaneity because it places time pressure on the participants (Meany &
Shuster, 2002).
Researching academic debate and, more specifically, the WUDC format, is important because (a) forensics is underresearched as a field (Croucher, 2006;
Herbeck, 1990), (b) the WUDC format is one of the most widespread formats of academic debate in the world, (c) the results would also be transferable to other formats, and (d) the research would also be relevant for oral argumentation pedagogy.
Firstly, there is a research void in forensics. As Croucher (2006) argues, research on academic debate (and forensics in general) has focused predominantly on practical essays on teaching debate and running forensics programmes, with some papers examining the future of and ethics behind forensics, yet there are very few studies that expand the field of communication theory itself (see also Herbeck, 1990). Secondly, when choosing a format to research, the WUDC format is probably the most widely utilized format of academic debate in the world. Apart from the WUDC itself, “the premier debating event in the world” (Cirlin, 2002, p. 82), the European Universities Debating Championships (EUDC) and thereby the majority of the other national and international debating events forerunning the WUDC and the EUDC have adopted this format as the standard. Thirdly, not only would the results of the analysis be
relevant for debaters following this particular format of debating, but the outcomes of the research could also be transferred to other formats. Because of the similarities in terms of speakers’ roles and responsibilities, the results will also be applicable to secondary school debate formats, such as the World Schools debate format or the Karl Popper debate format (cf., Driscoll & Zompetti, 2003). Lastly, in Hungary the study would have practical implications for those involved in argumentation pedagogy, which has lost touch with international trends for the past decade (for more details on this, see Tamási, 2008b). With only manuals of argumentation to rely on, it is important to grasp the opportunity to research debate with a view to outlining good practices and possibly shedding practices that have been passed over on the international scene. However, the outcomes of the research could also inform international argumentation pedagogy, where Dean (1990) calls for “pedagogical articles written with a clear purpose … and which are supported by existing communication research and theory” (p. 35). Therefore, researching academic debate is beneficial for a number of reasons.
Not only is it important to conduct research on academic debate, but it is also necessary to specifically study the generic characteristics of academic debate as a genre. Firstly, the study extends the scope of inquiry in the field of spoken genre analysis to include another genre in addition to academic lectures (e.g., Camiciottoli, 2004; Jung, 2006; Lee, 2009; Morell, 2004; Thompson, 2003), classroom discourse (e.g., Csomay, 2005), academic presentations (e.g., Morita, 2002), dissertation defences (e.g., Recski, 2005), and conference presentations (e.g., Edwards, McMasters, Acland, Papp, & Garrison, 1997; Webber, 2005), that is, the spoken academic genres that have been predominantly subjects to detailed inquiry previously.
More importantly for the EFL/EAP classroom, however, genre-based instruction (and
approaches building upon genre analysis, e.g., Crossley, 2007) has yielded mostly positive results. Especially, the North American school of genre studies (e.g., see Freedman & Medway, 1994b), has been instrumental in translating findings from (rhetorical) genre research into concrete classroom implications. For example, Faigley and Hansen (1985), albeit for writing, argue that if teachers are serious about providing adequate instruction for their students, instruction must be preceded by the rhetorical analysis of the genre texts. Hyon (2001) also conducted a follow-up interview study with eight participants on an EAP reading course which featured genre-based instruction. Her results indicate that explicit genre-based instruction may facilitate aspects of L2 reading and writing; therefore, it is hoped that EFL teachers may utilize the findings of the study in their daily teaching activities to help students plan and construct, and thus be more successful with, their individual long turns and ultimately develop their overall discourse competence (cf., Canale, 1983). Thirdly, researching academic debate provides a unique opportunity for the field of genre analysis, as the acquisition of the generic competence in this case is not bound up with the acquisition of professional expertise, as it is the case in other professional contexts (Bhatia, 2004). Thus, researching academic debate allows the genre analyst to focus purely on the acquisition of generic knowledge, without interference from professional expertise. Therefore, a genre analytical approach seems to be particularly suited to the analysis of academic debate and, more specifically, the WUDC academic debate format.
Lastly, out of the numerous methods that could be used to perform a genre analysis, it needs to be clarified why the grounded theory method is the most appropriate for the investigation. As it has been pointed out by Bhatia (2004), the evolution of genre theory can be characterized as a move from the focus on (a) the
“textualization of lexico-grammar” (p. 4), through the focus on (b) the “organization of discourse” (p. 4), ultimately to the focus on (c) the “contextualization of discourse”
(p. 4), that is, a “quest for thicker descriptions [italics added] of language use …, where an attempt is made to offer a grounded [italics added] description of language use” (p. 22). Although expert feedback has long been an indispensable part of genre analysis (Bhatia, 1993), as Bhatia (2008b) argues, rarely have studies working within the field taken the context of discourse into account; therefore, she calls for a more extensive use of ethnographic methods in the analysis of genre. Because the grounded theory method emphasises the importance of (a) placing the experience in the larger context and (b) “describing the process … of action” (Corbin & Strauss, 2008, p. 17), it is particularly suited for the purposes of the investigation required.
After the introduction, in Section 2, I will argue for the existence of a number of research niches that exist in previous studies and ones that my study is intended to fill. In the review, the literature on both academic debate (Section 2.1) and genre analyses (Section 2.3) will be summarized. Section 3 will contain the description of the methods used in conducting the research, with a view to establishing the replicability of the study. Section 4, in turn, will present the findings of the study and discuss how they address the research niches previously delineated. Lastly, Section 5 will summarize and conclude the dissertation, pointing out the limitations of the study as well as point out potential areas of further research.
2 Review of literature
Although one of the defining characteristics of inductive investigations is that the research itself is not guided by hypotheses and some advocates of the grounded theory method suggest that reviewing the literature should be postponed until past the theory construction stage, it is still important to conduct a thorough review of literature at the outset to be able to “participate in the current theoretical conversation” (Lempert, 2007, p. 254). As far as the kinds of literature that need to be surveyed is concerned, Bhatia (1993, 2004) points out that (a) “discussions of the … beliefs, goals, etc. of the academic community that uses the genre in question”
(Bhatia, 2004, p. 164), (b) the “tools, methods or theories of … genre analysis” (p.
164), and (c) “the linguistic analyses of the genre … or other related or similar genres” (p. 164) need to be reviewed. Therefore, these will be reviewed in turn in the following sections of the dissertation.
2.1 Research on academic debate
Although academic debates have not traditionally been subject to linguistic analyses, due to the focus of the dissertation it is still important to review the scholarship that pertains to discussions on and analyses of academic debate.
Following the Klumpp’s (1990) classification of forensic research studies, the subsequent sections will cover (a) practical, how-to articles, (b) articles reporting on the results of empirical investigations, and (c) theoretical studies in turn. Because according to Bhatia (1993, 2004), handbooks and manuals that are written and referred to by the discourse community are valuable sources of information about their respective practices, these will also be reviewed here.
2.1.1 Best practices in debating
One major trend in forensic scholarship has been motivated by the urge to spread good practices within the forensic community (Klumpp, 1990). Arguing for the value of this kind of studies, Herbeck (1990) notes that “the newcomers to the activity need more practical works which explain details ranging from case construction to refutation” (p. 2), a position that is also supported by Dean (1990).
This trend has resulted in a number of articles discussing best practices as regard various aspects of forensics.
One key issue that has been addressed by the practical studies is judge adaptation. Gill (1988), for example, recommends that debaters should attempt to analyse their target audience, that is, their judges before the round, either relying on rumours from other debaters, or referring to the judging philosophy files sometimes compiled for tournaments, which contain the judges’ expectations about debating. In another study, Eichenberg (1993) proposed a set of questions to include in judging philosophy files to aid debaters in addressing their audience in the most efficient manner.
Other studies published were aimed at sharing course materials within the forensic community. Hochel (1990), for example, shared a lesson plan on using court cases to teach about the freedom of speech and expression in tertiary classrooms;
Dean and Levasseur (1989) outlined the syllabus for a forensics-informed public speaking course aimed at providing additional educational opportunities to talented students. (For other research areas, see Croucher, 2006.)
The common concern among these articles is the spreading of best practices within the community; however, owing to the fact that they usually represent the idiosyncratic tips and tricks of individual debaters and coaches and lack scientific
rigour, they are difficult to qualify as research. A similar claim is made by Porter (1990), who emphatically called for more research, conducted with scientific rigour, in forensics; Croucher (2006), while seeing value to the dissemination of best practices, also argues that “the forensics community must recognize the need for more theoretical rigor in its research” (p. 1).
2.1.2 The results of empirical research
The second major trend in forensic publications has been empirical studies conducted on various aspects of forensics, ranging from descriptive research on the participants in different debate events, through the empirically grounded analysis of judging philosophies, to the advantages of participating in debating.
The majority of the empirical studies conducted on academic debate have focused on the question of diversity. Stepp (1997) examined five national Cross Examination Debate Association (CEDA) tournaments between 1991 and 1995 and concluded that women are minorities are underrepresented in intercollegiate debates, although Fugate (1997) found that the number of women and minorities participating in academic debate had increased significantly. More recently, Allen, Trejo, Bartanen, Schroeder, and Ulrich (2004) investigated a larger sample that included all of the most prominent organizations operating within the United States.
Because it is usually former intercollegiate debaters that act as judges (and coaches), a number of studies have tried to investigate the members of this “highly trained ‘expert’ audience” (McGee, 2002, p. 124). Cross and Matlon (1978) asked judges to respond to different judging situations, and based in these responses, they distinguish four judging philosophies: (a) the policymaking judge (i.e., a judge performing a cost-benefit analysis of the proposition’s vs. the opposition’s policy
system), (b) the hypothesis testing judge (i.e., a judge who focuses on the viability of the proposition case alone and thus ignoring off-case arguments), (c) the stock issues judge (i.e., a judge who requires that the proposition team should win each stock issue whereas the opposition should win only one), and (d) the “tabula rasa” judge (i.e., a judge who expects debaters to debate the theoretical framework of the debate within which the decision should be made, for criticism, see Bunch, 1994; in support of the idea, see McGee, 1998). Other analyses of judging philosophies have applied content analysis to judging philosophy booklets (e.g., Amsden, 2003; Gordley, 2003) or the way judges communicated their decisions at the end of the debate round (e.g., Birkholt & Diers, 2004).
Other studies have focused on the reasons behind participation in academic debates. For example, Rogers (2002) used a self-report questionnaire to compare debaters and non-debaters in five key areas (social responsibility, cultural understanding and tolerance, academic success, moral and ethical issues, and psychological adjustment), found positive outcomes in each key area in favour of debaters. In another study, Jones (1994) used interviews to investigate the reasons behind debater’s choice to participate and found a mixture of cognitive, competitive, heuristic, and social reasons for their involvement.
Among the largely descriptive studies in the literature, Gernant (2001) attempted to investigate if the type of the proposition (i.e., whether it focuses on questions of fact, value, or policy) has a significant effect on the win/loss ratios in a given form of academic debate. Conducting a content analysis of judge’s written feedback at the end of the round, she found no statistically significant effect of the type of proposition on the judge’s decision.
To summarize, although in volume secondary to practical, how-to articles (Croucher, 2006), research on academic debates has also produced some articles reporting on the results of empirical investigations. There is, however, a pronounced lack of research that deals with the argumentative aspects of academic debating, which prompted Herbeck (1990) to argue for “reforging the connection between argumentation and debate” (p. 8).
2.1.3 Theoretical research on argumentation and debate
As van Eemeren et al. (1996) point it out, academic debate and argumentation theory share a mutually beneficial relationship. On the one hand, academic debate has been providing argumentation theoreticians with problems in need of explanation, a recurrent phenomenon in the discipline of speech communication, which instead of the deductive approach of testing grand theories in practice has attempted to
“construct theories as needed to explain problems encountered in practice” (p. 196).
On the other hand, textbooks for debaters (i.e., practitioners) nowadays tend to include extensive instruction on argumentation theory, for example, on the application of Toulmin’s (1958) layout of arguments (e.g., Freeley & Steinberg, 2005; Meany &
Shuster, 2002; Wood & Goodnight, 1995). (For a review of textbooks on debating, see Tindell, 1999.) Apart from Toulmin’s layout of arguments, however, the classifications of motions and stasis theory (cf., stock issues) have also found their way into debating. For a comprehensive coverage of the first, see Tankó and Tamási (2008); the second topic will be covered in detail in the following.
Freeley and Steinberg (2005) define stock issues as the “issues common to most debates on given types of propositions. In value debate they refer to the definitive and the designative issues; in policy debate they include harm [ill],
inherency [blame], and solvency [cure]” (p. 53). These issues date back to antiquity, and Hultzén (1966) actually argues for a correspondence between “the system of status for forensic” (p. 104) in Cicero and the four frames for the deliberative, which he calls Malum or Ill (cf., harm/ill), Sanabilitas or Reformability (cf., blame/inherency), Remedium or Remedy (cf., solvency/cure), and Pretium or Cost (cf., cost/disadvantage), respectively. (For an overview of other taxonomies, including those by Aristotle, Cicero, Quintilian, and Hermogenes, see Nadeau, 1958.) Although it was mostly before the 1960s that there was a general agreement among theorists on them (Ulrich, 1992), stock issues still feature prominently in textbooks on argumentation (Freeley & Steinberg, 2005; J. R. Richards & Ricket, 1995; Rybacki &
Rybacki, 2000; Sproule, 1980; Wood & Goodnight, 1995; Ziegelmueller & Kay, 1997). The following description is based on Freeley and Steinberg (2005), Meany and Shuster (2002), Ulrich (1992), and Ziegelmueller and Kay (1997).
Firstly, the first proposition speaker has to create a logical link between the motion announced and the case he or she is going to argue for (i.e., the stock issue referred to as topicality). Secondly, he or she should define the key terms of the motion. When choosing from among the possibilities to define, the speaker may opt for (a) giving an example (e.g., “By a national program of public work, we mean a program similar to the WPA of the 1930s”, Freeley & Steinberg, 2005, pp. 54-55); (b) using the expressions in their common usage (e.g., “By labor organizations we mean the type of organization popularly referred to as unions”; p. 55); (c) referring to an external authority (dictionary, encyclopaedia, etc.); (d) presenting the definitions as they are realized within the plan (see below); (e) specifying what is not covered by the definition (e.g., “We do not mean the corner drugstore, we do not mean retail businesses, and we do not mean service businesses [italics added]; we mean steel,
autos, transportation, mining, oil, and gas”; p. 57); (f) referring to another concept, object, etc. as a means of comparison (e.g., “similar to the National Association of Broadcasters”; p. 58); or (g) referring to the etymology of the word (e.g., “prejudice
… [means] before judgment”; p. 58).
After the definitive issues, the speaker has to turn to what some scholars call the core of the case, namely, harm and inherency (Wood & Goodnight, 1995).
Through the harm, the speaker must demonstrate that either a qualitatively or quantitatively significant harm exists within the status quo, which “absent a significant change in policy action … is likely to continue” (Freeley & Steinberg, 2005, p. 66). This inherency of the harm may be attributed to a law or structure in place that prohibits change (viz., structural inherency), the attitude of the population that obstacles the solution of the problem (viz., attitudinal inherency), or the fact that although “the philosophical commitment of the present system admits that the ills [harms] could be solved, but claims that the ills [harms] should not be solved because greater ills [harms] would result” (Ziegelmueller & Kay, 1997, p. 175). If the inherency is identified correctly, it should be the major contributing cause to the isolated harm. For an overview of cause and effect argumentation, see Trapp and Luqman (2003).
Once the significant harm and its major contributing cause, the inherency, have been delineated, the speaker may proceed to introducing the plan. Ideally, the plan features four or five planks, one for agency (i.e., identifying the body or organization to undertake action, viz., the who), one for mandates (i.e., identifying the actual steps to be taken by particular the body or organization, viz., the what), one for enforcement (i.e., identifying how the execution of the plan will be imposed), one for
funding and staffing (i.e., identifying the material and personnel resources), and one optionally for addendum (i.e., miscellanea), respectively.
After the plan has been detailed, the speaker has to demonstrate that (a) the plan works (viz., workability), (b) the plan eradicates the harm identified before (viz., solvency), and (c) the steps specified within the mandates are warranted (viz., justification). Although logically first proposition speakers may stop at this point because they have, through the previous steps, constructed a prima facie case (Bickenbach & Davies, 1997; Freeley & Steinberg, 2005; Ulrich, 1992), they may also want to highlight additional independent advantages to the plan. That is, they may want to show that apart from solving the harm, the plan also has other desirable qualities.
After the first proposition speaker has established the case, the first opposition speaker has to respond to what has been said. While textbooks on argumentation are usually very precise about the systemic nature of constructing a case (e.g., Freeley &
Steinberg, 2005; Wood & Goodnight, 1995), with regard to the strategies opposition speakers may select from, they merely present a checklist of stock issues that may be attacked and ways in which the speakers may respond to the given stock issues.
Therefore, another research niche can be delineated in terms of comparing actual practice with the textbook descriptions.
2.1.4 Summary
Having reviewed the relevant literature, one may delineate an apparent research niche as (a) practical, know-how articles lack academic rigour; (b) the empirical studies that have been conducted, despite van Eemeren et al.’s (1996) claim otherwise, have mostly ignored the argumentation component of academic debates
and focused instead on other issues (Herbeck, 1990); and (c) the textbooks that have been published on academic debate lack empirical foundations.
2.2 Linguistic analyses of argumentation
Argumentation theory is “interested in the problems involved in the production, analysis, and evaluation of argumentative discourse … in the light of the actual circumstances in which it takes place” (van Eemeren, et al., 1996, p. 12). As such, it encompasses, among other subjects, philosophy (e.g., Aczél, 2005; Toulmin, 1958); rhetoric (e.g., Perelman & Olbrechts-Tyteca, 1969); argument interpretation (e.g., Németh T., 2000; van Eemeren, Grootendorst, & Henkemans, 2002; van Rees, 2001); and formal, informal, and propositional logic (e.g., Bickenbach & Davies, 1997; van Eemeren, et al., 1996).
In her article, van Rees (2007) draws attention to the fact that argumentation theory and discourse analysis can mutually benefit from a collaboration between the two disciplines by combining the “theoretically motivated normative model of argumentation … [and] the empirical insights about how argumentative discourse is organised, produced, and understood” (p. 1462). In line with this suggestion, discourse analysis (e.g., Hoey, 2000; Hosman, 2002; Lauerbach, 2007; van Dijk, 1980) and composition studies (e.g., Connor, 1986, 1990, 1991; Connor & Lauer, 1985; Connor & Takala, 1987) have studied argumentation from a discourse analytical viewpoint.
In the last decade, a number of studies published by Hungarian authors have also explored written argumentation. For example, Károly (2002) investigated the text-organizing function of lexical repetition in foreign language argumentative discourse, Tankó (2005) established the Argumentative Move Structure (AMS)
Model for the analysis of Anglo-American formal argumentative essay, and Árvay (2007) analyzed promotional leaflets, magazine and newspaper advertisements, and direct marketing brochures for traces of manipulation. More recently, Tankó and Tamási (2008) investigated the thesis statements that appear in the argumentative essays of EFL student writers.
While there is an abundance of research on written argumentation, there are much fewer studies dealing with oral argumentation, and almost exclusively, they deal with contemporary media discourse. Such examples include Emmertsen (2007), who examined British broadcast panel interviews; Hess-Lüttich (2007), who described the genre of political (show) television debates; Lauerbach (2007), who used an integrated approach to investigate celebrity talk show interviews; and Blum- Kulka, Blondheim, and Hacohen (2002), who traced back the controversial and fierce style displayed in Israeli talk shows to Talmudic academies.
Kinneavy (1971) in his book A Theory of Discourse gave possibly the most comprehensive account of persuasive discourse, that is, the “kind of discourse which is primarily focused on the decoder and attempts to elicit from this a specific action or emotion or conviction” (p. 211). Based on Aristotle, Cicero, and Quintilian, Kinneavy describes the structure of persuasive discourse in five functional constituents: (a) introduction; (b) narration (i.e., “the recital of the background facts necessary to understand the issue”; p. 267); (c) confutation (i.e., “arguments that destroy the opponent’s position”; p. 270; viz., refutation); (d) confirmation (i.e., “arguments that prove one’s own position”; p. 270); and (e) conclusion. Apart from the structural analysis, however, Kinneavy also examines topoi (i.e., “in contemporary terms stereotyped arguments”; p. 245) as well as the semantic and grammatical
characteristics of persuasive style, especially based on Roosevelt’s “First Inaugural Address”. (For the critique of Kinneavy's work, please refer to Swales, 1990.)
The most coherent theoretical model of dialectic argumentation, pragma- dialectics was developed by van Eemeren, et al. (1996). In the pragma-dialectical research program,
argumentation is a phenomenon of verbal communication which should be studied as a specific mode of discourse, characterized by the use of language for resolving a difference of opinion. The quality and possible flaws of argumentation are to be measured against criteria that are appropriate for the purpose of such discourse. Because the study of argumentation should encompass the descriptive as well as the normative dimensions of argumentative discourse, it is to be construed as a special branch of linguistic pragmatics in which both dimensions are methodically integrated. (p. 275)
Van Eemeren, et al. (1996) distinguish four analytical stages in the process of resolving a difference of opinion: (a) confrontation stage (i.e., “a difference of opinion presents itself through an opposition between a standpoint and nonacceptance of this standpoint”, p. 281); (b) opening stage (i.e., “the protagonist and the antagonist in the dispute … and their initial commitments [are identified]”, van Eemeren et al., 1996, p. 281); (c) argumentation stage (i.e., “the party that acts as the protagonist methodically defends the standpoint at issue against critical responses of the antagonist”, p. 282); and (d) concluding stage (i.e., “the protagonist of a standpoint and the antagonist determine whether the protagonist’s standpoint has been successfully defended against the critical responses of the antagonist”, p. 282).
Besides identifying the four stages, van Eemeren, et al. also enumerate the speech acts (assertives, commissives, directives, and usage declaratives) that characterize the individual stages. Because of its conception as discourse dialectic, pragma-dialectics is claimed to “overcome the sharp and infertile ideological division between rhetoric and dialectic” (van Eemeren & Houtlosser, 2000, p. 297).
An advantage of the pragma-dialectical approach is that it has been used in a number of empirical studies. In addition to describing the theory behind rhetorical analysis conducted within a pragma-dialectical framework, van Eemeren and Houtlosser (2000) also present a brief analysis of advertorial. Van Eemeren, Houtlosser, and Henkemans (2008) report on research focusing on the linguistic indicators of argumentative topoi2. Moreover, Árvay (2007) used the list of fallacies compiled within the pragma-dialectical framework in her research.
Considering the review of studies present in this section, it may be concluded that while there has been considerable amount of research conducted on written argumentation, the field of formal spoken argumentation (e.g., academic debates), despite its significance in education, has been somewhat neglected. Although media genres featuring debating (e.g., talk shows) have received some research attention and informal argumentation (both written and spoken) have been explored by proponents of the pragma-dialectical approach, a research niche presents itself, namely, that of exploring academic debates.
2.3 Research on genre
After the review of the linguistic analyses of argumentation, in the present chapter, in line with the recommendations of Bhatia (1993, 2004), an overview of the
2 Following Kneale and Kneale, van Eemeren et al. (1996) translate the Greek “topos” as “move”.
However, in order to distinguish the concept of a move as used in the Swalesian vs. the Aristotelian
relevant theories and methods of genre analysis will be provided. Firstly, based on Paltridge (1997), I am going to review some of the general predecessors to the concept of genre as it is used by the British English for Specific Purposes school (Bhatia, 1993, 2004; Swales, 1990, 2004) today; more precisely, some of the contributions by (a) literary genre analysis, (b) the ethnography of speaking, and (c) conversation analysis will be outlined. Although Paltridge (1997) enumerates further influences on the development of the notion of genre (e.g., linguistic anthropology or rhetoric), because these are more closely related to the systemic-functional, the new rhetoric, or the British ESP school, these will be discussed under the respective schools. Finally, I am going to review the empirical studies on spoken genre analysis, with reference to Bhatia’s (1993, 2004) classification of empirical research studies, originally developed for studies conducted in written genre analysis.
Firstly, literary genre analysis has had an important influence on the evolution of the concept of genre (Paltridge, 1997). Baldick (1990) defines literary genre as “a recognizable and established category of written work employing such common conventions as will prevent readers or audiences from mistaking it for another kind”
(p. 90). Corbett (2006) draws attention to the fact that while in the classical Greek literary theory lyric, epic, and dramatic poetry were distinguished based on the use of narrative voice, this distinction held gradually less true by the time of neoclassicism due to the mixing of genres, which subsequently led to the concept of literary genre falling out of favour and even achieving the status of a stigma. He points out that 20th century structuralist critics tend to view genre as a “set of expectations that a reader acquires from his or her reading” (p. 26), which then may be fulfilled or frustrated.
Fowler (as cited in Swales, 1990), on the other hand, shifts the focus from the reader
to the writer and points out that genres are also useful for the authors in the writing process.
Secondly, the ethnography of speaking (Hymes, 1974; Saville-Troike, 2003) has had a significant effect on the thinking about genre (Paltridge, 1997). As Philipsen and Carbaugh (1986) argue, the central motive of those working within the ethnography of speaking has been “to create a theory of linguistic communication which is grounded in the comparative analysis of many communities and their distinctive ways of speaking” (p. 387), a motive that can be traced back to the work of Malinowski (1923, as cited in Halliday & Hasan, 1989, p. 7). Hymes (1974) outlines an analytical tool created in order to assist the exploration of various ways of speaking in speech communities. The SPEAKING grid invites researchers to focus on the (a) setting and scene, (b) the participants, (c) the ends, (d) the (speech) act sequence, (e) the key, (f) the instrumentalities, and (g) the genre of the speech event.
For Hymes (1974), “by genres are meant categories such as poem, myth, tale, proverb, riddle, curse, prayer, oration, lecture, commercial, form letter, editorial, etc.”
(p. 61), and as such these often coincide with speech events, that is, “activities … directly governed by rules or norms for the use of speech” (p. 52). (For a bibliography of studies in this research tradition, please refer to Philipsen & Carbaugh, 1986.) Studies following the ethnography of speaking have been mostly characterized by an inductive methodology and used interviewing, observational methods, diaries, and so on in their investigations.
As opposed to the ethnography of speaking, conversation analysts reject ethnomethodology (e.g., interviewing and observational methods) as a research paradigm and advocate the primacy of the text in conducting analyses (Heritage, 1984, 1995). The emphasis for conversation analysis is on analytic induction, that is,
attempting to detect patterns about “the uses and organizational properties of particular conversational practices” (Heritage, 1995, p. 399) in a wide range of (linguistic) data. Conversation analysis, unlike, for example, the systemic-functional school described below, rejects the stand that “that determinate and obligatory rules of interactional sequencing provide for ‘well formed’ discourse” (p. 399) on grounds that is incompatible with notion of conversational implicature.
2.3.1 Schools of genre analysis
After the brief overview of some of the influences on the notion of genre, the present section aims to provide an overview of the three majors schools of genre analysis: (a) the Australian systemic-functional linguistics school, for whom “is a staged, goal-oriented, purposeful activity” (Martin, 1984, p. 25 as cited in Eggins, 2004, p. 55); (b) the American new rhetoric school, for whom genres are “typified rhetorical actions based in recurrent situations” (Miller, 1984, p. 159); and (c) the British English for Specific Purposes (ESP) school, for whom genre is a
“conventionalized communicative event” (Bhatia, 1996, p. 46). In each of the cases, the discussion will include (a) the aim of the school, (b) the most significant influences on the development of the school, (c) the way the school views genre, (d) the more recent developments that have borne out of the particular school, and (e) criticism targeted at the theory and/or methodology employed by proponents of the given research trend. In order to summarize and synthesize the section, some of the agreements and the disagreements among the three different traditions will be highlighted.
2.3.2 Australian systemic-functional linguistics: Genre as a staged, goal- oriented social process
One of the major influences on contemporary genre theory has been the work of those operating within the systemic-functional framework developed by the Sydney school of scholars. According to Hyon (1996), the school is unique both in terms of addressing the needs of a different target population and hence in terms of pursuing different goals. While the majority of the studies conducted by those working within the framework of the new rhetoric movement and the British English for Specific Purposes (ESP) school (both discussed below) concern the tertiary context, the research of the Australian school has been motivated by the need for the improvement of literacy skills and hence has focused on primary and secondary educational contexts as well as (immigrant) adult education. Because the ultimate aim in this case has been empowering students by providing them with the assistance necessary to succeed in their education, the Sydney group of scholars have advanced a number of instructional frameworks to implement genre-based pedagogy (for a review, see Hyon, 1996). As Eggins (2004) argues, “what is distinctive to systemic linguistics is that it seeks to develop both a theory about language as social process and an analytical methodology which permits the detailed and systematic description of language patterns” (p. 21, emphasis in the original).
The systemic-functional approach has its roots in British contextualism and the work of J. R. Firth (Eggins & Martin, 1997; Ventola, 1984). As Eggins and Martin (1997) argue, the work of the Australian school was influenced by the idea of contextual theory, which posited a “direct correlation between the functional organization of language and the organization of context” (p. 239), where context included both the “context of situation” and “the context of culture” (Malinowski,
1923, as cited in Halliday & Hasan, 1989, p. 7). Firth, influenced by Malinowski’s ideas, also argued for the importance of context in analysing texts. In his view, the following frames could be used in order to systematically describe the features that are relevant from the perspective of the text produced within a given context:
A. The relevant features of the participants: persons, personalities.
(i) The verbal action of the participants.
(ii) The non-verbal action of the participants.
B. The relevant objects.
C. The effect of the verbal action. (Firth, 1950/1957, p. 182)
At the same time, however, he warned against treating these frames as tangible entities and claimed that they are “best used as schematic constructs to be applied to language events and … they are merely a group of related categories at a different level from grammatical categories but of the same abstract nature” (Firth, 1957/1968, p. 154). (For an overview of Malinowski’s contribution to linguistics, please refer to Hasan, 1985.)
Firth’s idea that context defined meaning was further elaborated by Halliday and his colleagues (e.g., Halliday & Hasan, 1976, 1989; Halliday, McIntosh, &
Strevens, 1964/2007). For Halliday, language is considered in a social semiotic perspective. Central to his theory is the claim that “a systematic relationship [exists]
between the social environment on the one hand, and the functional organisation of language on the other” (Halliday & Hasan, 1989, p. 11). In order to better characterise this social environment, Halliday, et al. (1964/2007) established a conceptual framework that consists of three “abstract components of the context of the situation”
(Halliday & Hasan, 1989, p. 29): (a) the field of discourse, that is, “what is happening, the nature of the social action that is taking place” (p. 12), the features of which
activate experiential/ideational meaning; (b) the tenor3 of discourse, that is, “who is taking part, the nature of the participants, their statuses and roles” (p. 12), the features of which activate interpersonal meaning; and (c) the mode of discourse, that is, “what part language is playing, what it is that the participants are expecting the language to do for them in the situation” (p. 12), the features of which activate textual meaning.
As a corollary to the systematic relationship between context of situation and functions of language, it also follows that certain patterns will be more typical than others, and in other words, the “notion of the context of situation [may be extended]
to ‘issues of linguistic predictability’” (Paltridge, 1997, p. 24) an idea that, in Halliday’s model, is captured by the notion of register (Halliday & Hasan, 1976, 1989; Halliday, et al., 1964/2007). In his view, a register may be
defined as a configuration of meanings that are typically associated with a particular situational configuration of field, mode, and tenor. But since it is a configuration of meanings, a register must also … include the expressions, the lexico- grammatical and phonological features that typically accompany or REALISE these meanings. (Halliday & Hasan, 1989, pp. 38-39, emphasis in the original)
In line with this definition, the studies attempting to address the issue of register variation, irrespective of the language under consideration, may be characterised by a focus on (a) the description of actual discourse; (b) language varieties as opposed to, for example, idiolects; (c) formal linguistic features as they are realised at different linguistic levels; and (d) the description of situational features (Atkinson & Biber, 1994). However, because the procedure involved in establishing the register of the
3 Originally, Halliday, et al. (1964/2007) used the term style instead of tenor; however, in subsequent publications this was revised, following the terminology used by Spencer and Gregory (as cited in
text based on, primarily, realised lexico-grammatical features was relatively complicated and because discrepancies arose between feature-based and intuitive classifications, alternatively, text patterning was proposed as a criterion to assist decisions regarding the register membership of a given text (Ventola, 1984).
According to Corbett (2006), genre research in systemic-functional linguistics was the result of the fact that previous studies on register could not provide sufficient explanation for contextual variation, but the relationship between register and genre remains a contested issue in the systemic-functional school. For Halliday, genre is included under the mode of discourse and used synonymously with the rhetorical mode of the text (Halliday, 1978; Halliday & Hasan, 1976). This idea was further developed by Hasan (Halliday & Hasan, 1976), for whom genre is closely related to text structure and may be regarded as the structural aspect of register. In her view, given field, tenor, and mode configurations will enable predictions about “the OBLIGATORY … and the OPTIONAL … elements of a text’s structure as well as their SEQUENCE … vis-à-vis each other and the possibility of their ITERATION”
(p. 56, emphases in the original). Therefore, she claims, it is possible to describe each text that can be an appropriate instance of a genre by listing the above features, a concept that is covered by the term “generic structure potential” (Halliday & Hasan, 1989, p. 64)4.
While for Halliday and Hasan (Halliday, 1978; Halliday & Hasan, 1976, 1989) genre is, from a certain aspect at least, sub-ordinated to register, scholars in the Sydney school (e.g., Eggins & Martin, 1997; Martin, 2006) conceptualise genre as a super-ordinate term. For Martin, “a genre is a staged, goal-oriented, purposeful
4 According to Ventola (1984), there is some confusion as regard the terminology generic structure vs.
(generic) structure potential as used by Halliday and Hasan, respectively. Halliday’s generic structure
“seems to be the actual structure detected in texts” (p. 285), while Hasan’s “structure potential aims to capture the possibilities of variance in text structures” (p. 285).
activity in which speakers engage as members of our culture” (Martin, 1984, p. 25 as cited in Eggins, 2004, p. 55), and therefore both the context of culture and that of the particular situation is necessary to interpret the text (Martin, 1984, as cited in Paltridge, 2007). Genres invariably involve (a) a certain register configuration, (b) a given schematic structure, and (c) fixed realizational patterns (Eggins, 2004). In this sense, Eggins and Martin (1997) claim that context on its own and in itself is not capable of accounting for the recurring configurations of the Hallidayan variables, or their culture-specificity, and argue that an additional level of genre, “posited above and beyond the field, mode, and tenor register variables” (p. 243), is necessary.
Genres, registers, and lexico-grammatical features (cf., realizations of registers) form three concentric circles, where “genre is a pattern of register patterns, which are in turn patterns of discourse semantics patterns and so on” (Martin, 2006, p. 284).
While some scholars deliberately distinguish genre from style, others use them synonymously. Biber (1995), for example, has argued that the differences between the two terms that have been pointed out by previous researchers are rather minute, largely abstract, and devoid of practical implications as far as classifications are concerned; therefore, he uses the term register to cover “all aspects of variation in use” (p. 9).
While extensive research has been conducted on comparing registers along single situational parameters (Atkinson & Biber, 1994), more recently, the focus of research has shifted to register variation and multi-dimensional (MD) studies (e.g., Biber, 1995, 2003, 2006; Biber & Barbieri, 2007; Biber, Conrad, Reppen, Byrd, &
Helt, 2002). A defining feature of research conducted within the MD paradigm is that
“different co-occurrence patterns are analyzed as underlying dimensions of variation”
(Biber, 1995, p. 30, emphasis in the original). Biber (1995), for example, establishes
the following dimensions: (a) involved vs. informational production, (b) narrative vs.
non-narrative discourse, (c) situation-dependent vs. elaborated reference, (d) overt vs.
non-overt expression of argumentation, (e) non-abstract vs. abstract style, (f) on-line informational vs. edited/not informational elaboration marking stance, and (g) academic hedging; Biber (2003) uses (a) oral vs. literate discourse, (b) procedural vs.
content-focused discourse, (c) narrative orientation, and (d) academic stance as dimensions. After these dimensions have been established, the registers may be placed on a cline between the two extremes: classroom management as an extremely procedural kind of discourse, and textbooks as extremely content-focused discourse, for example (Biber, 2003).
The shift towards multi-dimensional studies has been motivated by the realization that (a) in order to analyse register variation, the full range of spoken and written registers must be simultaneously subjected to analysis and (b) focusing on one specific linguistic parameter fails to account for the comparisons and contrasts among different registers (Biber, 1995). Due to these assumptions, MD studies have worked within the mixed methods paradigm5 and as such have been characterised by the need for generalizability (cf., Biber, 2003) and, as a prerequisite, representativeness (cf., Biber, 1995), on the one hand, and by the need for an extensive survey of linguistic features (most prominently applying factor analysis as a method), on the other.
According to Biber (1995), representativeness in MD studies must be established at three separate levels: those of (a) registers, (b) texts, and (c) linguistic features (p. 27). First, a wide enough range of registers must be selected for analysis in order to enable the detection of cross-situational variation, and second, a representative sample of texts must be analysed (cf., the qualitative methodology
5 For Biber (1995), dimensions of variation possess both a linguistic and a functional component.
While it is necessary to investigate the former using quantitative methodologies, the latter, that is, the underlying functions of linguistic features, necessitate qualitative techniques.
employed in ethnographic studies, see below). Third, “a wide range of linguistic features should be analyzed in each text, representing multiple underlying parameters of variation” (p. 27). To fulfil these requirements, MD studies have typically been conducted on sizeable corpora, for example, the London/Oslo/Bergen Corpus (e.g., Biber, 1995), the TOEFL 2000 Spoken and Written Academic Language Corpus (e.g., Biber, 2003, 2006; Biber & Barbieri, 2007), and/or the Michigan Corpus of Academic Spoken English (e.g., Csomay, 2005, 2006).
Apart from employing corpora in their investigations, those working within the MD framework tend to focus on the “linguistic co-occurrence of linguistic features” (Biber, 1995, p. 30) and thus are concerned with register markers and register features. The former, register markers, Biber (1995) defines as “distinctive indicators of a register” (p. 28), while the latter, register features, for him are the “core lexical and grammatical characteristics found to some extent in almost all texts and registers” (p. 29). As an example, Biber (1995) cites inning and home run as examples of register markers that would typically occur in texts that deal with baseball games.
On the other hand, he claims, lexical units are rarely “distinctive indicators” (p. 28) of a given register: both inning and home run could occur in a variety of registers, ranging form a broadcast of a baseball match to a romance novel. Contrary to register markers, register features can be used as litmus tests in identifying the register membership of a given text by virtue of the fact that the distribution of tense and aspect markers, prepositional phrases, nominal forms, modals, etc. displays considerable variation across different registers (Biber, 1995).
Although register studies, largely owing to the more extensive use of corpora and information technology, have evolved considerably over the past fifty years, they
are nevertheless criticised on a number of grounds. For example, as Widdowson (1979) argued:
The fact that scientific English text exhibits a relatively high proportion of certain syntactic features and a relatively low proportion of others may be useful for identifying scientific English texts should we ever wish to do such a thing. In fact this approach has proved useful for establishing authorship; it can reveal, with the help of a computer, who wrote what. But it cannot reveal the communicative character of what was written.
(p. 56)
Swales (1990), on the other hand, has emphasised that using register labels may be a convenient albeit misleading shorthand because “they overprivilege a homogeneity of content at the expense of variation in communicative purpose, addresser—addressee relationships and genre conventions” (p. 3). Others have criticised the systemic school due to their marked orientation to production (cf., reception). As Poynton (1993) points out, there is a contradiction between claiming that meaning resides in the text and parallel to this focusing only on production as opposed to reception.
2.3.3 American new rhetoric: Genre as typified rhetorical actions based in recurrent situations
As opposed to the Sydney group of scholars, those working following the traditions of the American new rhetoric movement have focused primarily on university contexts and the professions, their ultimate aim being to help students succeed (Freedman & Medway, 1994a; Hyon, 1996; Miller, 1984) by virtue of the fact that “genres serve as keys to understanding how to participate in the actions of a
community” (Miller, 1984, p. 165). As Freedman and Medway (1994a) point it out, the rhetoric turn in composition studies meant that students were asked to consciously think about their audience as well as their purpose, which, in combination with the process approach to writing, has significantly transformed writing classrooms.
While there is a considerable amount of overlap between the Sydney and the North American schools of genre analysis (e.g., the importance attached to context;
see Freedman & Medway, 1994a), there are also a number of differences between the two approaches. If for the proponents of systemic-functional approach the aim was developing both a theory of language and a framework for analysing its language patterns (Eggins, 2004), for the American school of scholars the exploration of genres has been deemed important so far as they “can serve both as an index to cultural patterns and as tools for exploring the achievements of particular speakers and writers” (Miller, 1984, p. 165). In line with this idea, the proponents of the American school attribute relatively less significance to the examination of formal text characteristics and instead tend to focus on “exploring sociocontextual aspects of genres and the action a particular genre aims to accomplish, as well as how these aspects might change through time” (Paltridge, 1997, p. 16). This is possible only if genres are viewed as dynamic and open-ended rather than static and fixed (Bakhtin, 1986; Freedman & Medway, 1994a; Schryer, 1994).
The American approach to genre has its roots in the study of classical and contemporary rhetoric. As Paltridge (1997) argues, one important influence has been Aristotelian rhetoric, but others working within the field of rhetoric (e.g., Toulmin, 1958) have also played an instrumental role. Kinneavy (1971), for example, in his A Theory of Discourse provides an overview of the evolution of rhetoric from Aristotle to contemporary rhetoric. Kinneavy (1971) bases his taxonomy of genres on the
communication process, that is, the basis for classification becomes the emphatic element in the process. For example, if the text focuses on the sender; it is expressive, if it focuses on the receiver, it is discursive; if it focuses on the form of the text, it is claimed to be literary; and if it attempts to describe some reality in the world, it is then described as a referential genre. (For the critique of Kinneavy's work, please refer to Swales, 1990.)
As Freedman and Medway (1994a) argue, another important influence on the North American school of genre studies has been the work of Mikhail Bakhtin (most importantly, Bakhtin, 1986). Bakhtin (1986) views utterances as the basic units of linguistic analysis, which “reflect the specific conditions and goals … not only through their content (thematic) and linguistic style, that is, the selection of the lexical, phraseological, and grammatical resources of the language, but above all through their compositional structure” (p. 61). While admitting that all utterances are highly idiosyncratic and individual, Bakhtin (1986) also draws attention to the fact that “each sphere in which language is used develops its own relatively stable types of these utterances” (p. 61), which he refers to as speech genres. Speech genres, for Bakhtin (1986), are thus “certain relatively stable thematic, compositional, and stylistic types of utterances” (p. 65). Another influence by Bakhtin (1986) that is clearly traceable in the work of the North American school is the idea that the analysis of speech genres should aim to connect changes taking place within the society with those taking place in language. In line with this, the proponents of the New Rhetoric school tend to use ethnographic methods to provide thick descriptions, for example, in the case of the evolution of experimental articles in physics (Bazerman, 1984) or patents (Bazerman, 1994).
