Genre is a Verb:

Research on Academic Writing in Critical Perspective

Qualifying Paper #2, Draft #3

Matthew C. Bronson

December 5, 2001

I. Introduction and Summary

II. Textually Oriented Studies

III. Systemic Functional Linguistics

IV. Contrastive Analysis

V. Genre Analysis

VI. Socio-Cultural Approaches

VII. Criticalist Approaches

VIII. Re-envisioning EAP: Language Socialization

I. Introduction and Summary

Academic life is s a buzzing beehive of literate activity. A defining characteristic of the university is that is a place where people read, write, exchange and respond to a dizzying variety of texts in the context of disciplinary or interdisciplinary study. The genres of academic writing which members of the community--professors and students alike--encounter on a daily basis include a startlingly broad range, from research reports and e-mail communiqués to requests for reprints and term papers. Each of these, in turn, can be definitively characterized neither by the presence of particular rhetorical and linguistic forms nor even in terms of the conventionalized purposes which the host discourse community associates with the genre. Under a critical and criticalist gaze, the a priori integrity of a category like "research paper" dissolves into a congeries of institutional, disciplinary and situational contigencies (Prior, 1998). The nested complexity in which academic writing tasks are framed, interpreted and executed is both problematic and potentially empowering for the non-native writers with whom the field of English for Academic Purposes ("EAP") and this paper is especially concerned (Benesch, 2001a).

My aim here is to broadly sketch the trajectory of academic writing research in the past forty years and, especially, the fields of English for Academic ("EAP") and Special Purposes ("ESP") from the point of view of a criticalist and socio-culturally oriented researcher. EAP originates in an intention to identify and teach the language and literacy skills that students need to know to succeed in school, and especially in disciplinary contexts. EAP has its roots in formal comparisons of texts and evolves toward studies which encompass increasingly "narrower" and "deeper" treatments (Swales,, 1990: p.3) of the social and institutional contexts in which texts are actually authored, circulated and evaluated. The story of EAP and its disciplinary cousins can be told in miniature by considering how researchers have imagined language and learning and how they have variously defined and operationalized three key terms: genre, discourse community and task. My introductory task is to outline the structure of the paper in a summary of how various approaches to EAP have evolved over time. These approaches will be covered in the same order and in more detail in the following sections and will be subjected in turn, to a critique grounded in my own criticalist and socio-cultural orientation.

The English for Special Purposes, or ESP approach to academic writing research was begun in the 1960s and 1970s under the rubric of "needs analysis" and derives from a "conduit metaphor" (Reddy, 1974; Lakoff & Johnson, 1980) view of language in which examples of academic genres as "autonomous texts" (Olson, 1994) hold intrinsic meanings. More precisely, the field shifted over time from a construal of text as a "linguistic object" to a view of text as a "vehicle for meaning" (Johns & Davies, 1983: p.1, cited in Dudley-Evans & St. John, 1998: p. 26)--which is itself a step toward the social construction of meaning. These textually oriented studies are rooted in an asocial view of language inasmuch as meaning is understood to reside "in" the text rather than emerging from socially situated "transactions" between readers and texts (Rosenblatt, 1989). Work in this mode consists largely of a search for specific pre-defined formal features of an academic genre or register via comparison of exemplars within and across disciplines as constituted in professional publications. The goal of such efforts is to identify the forms of target discourses so that students can more easily acquire mastery of those forms and thereby advance their academic and professional careers within their chosen disciplines. Genre, task and discourse community are, in this approach, under-theorized from the perspective of subsequent research.

While other some approaches are clearly based in a "conduit" view of language, others are explicitly attentive to aspects of culture and social context. The "Australian School" of genre studies, emphasizes the relationship between the forms of texts and the conventionalized purposes or functions which they aim to realize. The Australian School is largely derived from "systemic functional linguistics" ("SFL") (Christie and Martin, 1997; Cope and Kalantzis, 1993) which, while still centered almost exclusively on techniques for textual analysis, aims to represent the social and institutional meanings which always accompany the "content" of a text.

Studies in the tradition of "contrastive rhetoric" typically compare, via rhetorical, rather than linguistic analyses, examples of the "same" academic genre as it is realized in different linguistic and cultural contexts. The goal in such studies is to uncover the cultural dimension of textual organization as revealed in the structure and form of the overall text as well as in the presence and sequencing of specific rhetorical "moves".

"Genre Analysis"(1) (Swales, 1990), the focal approach in the next section, was initiated in the 1980s and early 1990s. Findings from earlier textual analyses had been found to be largely devoid of practical implications, given the complexities of teaching and learning academic writing in vivo (Widdowson, 1979). Genre Analysis articulated an approach which attempted to harness theoretical and methodological developments in the Human Sciences--especially linguistics, anthropology and rhetoric-- to inform research and practice in academic writing. Genre here has been re-conceptualized in largely socio-cultural (equivalent here to "social constructionist", see Dudley-Evans & St. John, 1998: p.117) terms and is recognized as inextricably linked both to the discourse community in which it is embedded and to the conventionalized tasks which the community ascribes to its members' (or aspiring members') writing. The Genre Analysis model has proven very influential in EAP and genre studies generally. This is largely owing to its explicitness, completeness and capacity to investigate and "explain" complex relationships between text and context in educationally appropriate terms (for example in Swales', 1990, treatment of the research article). Genre Analysis, while emblematic of, and perhaps pivotal in the turn toward socio-cultural approaches, continues to hold positivist assumptions in its terms which must be explicitly bracketed in ethnographic or criticalist research.

Researchers have appropriated, challenged and adapted the definitions and methods of Genre Analysis as they have set out to realize various aspects of its ambitious program. A third section focuses on what can be loosely characterized as "socio-cultural" approaches which gained momentum in the early to mid-1990s. The "socio-cultural" or "socio-historical" (Berkenkotter & Huckin, 1988; Samraj, 2002a, 2002b; Prior, 1998) approach extends the methods of genre studies to include an increasingly "thick" (Watson-Gegeo, 1992) account of the actual conditions and contexts in which the forms of academic writing are produced and interpreted (in, for example, a particular undergraduate class, as in Benesch, 2001, or graduate seminar as in Prior, 1998, over the course of a semester).

Genre, discourse community and task can be profitably explored in an ethnographic or "grounded" mode as intersubjectively situated constructs which dynamically unfold in the joint understandings and practices of diversely positioned individuals (c.f., Prior, 1998). Rather than depending on intuitive or abstract conceptions of these key terms and their a priori import in any particular context, the socio-culturalist approach investigates them as they are revealed in all their complexity within situated practice. This contrasts with the positivist idea--implicit in the conduit metaphor-- that genre, discourse community and task (terms which have been explicitly linked by Swales, 1990) refer to stable classes of "things" out in the world rather than intersubjective and situated constructs. Socio-cultural inquiry in academic literacy embraces a turn toward the indeterminate nature of "reality" (Lyotard, 1984); it aspires not to an ultimate accounting of "truth" but to a "thick record" derived from empirical investigation of multiple participants' lived experiences (Watson-Gegeo, 1992, 2001). Central in this mode of research are ethnographic and qualitative treatments of the complex, ever-changing and often conflicting meanings ascribed to academic writing tasks and genres by differently situated (Lave and Wegner, 1993) subjects in real-world settings.

Increasing attention to the situated, social and institutional contexts of literacy has rendered the fields of EAP, and its close cousins, second language writing (Silva, 2001) and contrastive rhetoric (Connor, 1996), more permeable to a "criticalist" (Benesch, 2001; Freire, 1974, 1994; Bordieu, 1991; Fairclough, 1989, 1992; Chouliaraki & Fairclough, 1999; Carspecken, 1996) point of view in the pre- and post-millennial period. The criticalist position argues that there is an ethical necessity for educational research and practices that challenge "business as usual" in the academy. The criticalist research program (c.f., Welton, 1993) seeks to redress pressing social, political and economic inequities through inquiry that acknowledges, for example, the agency and concerns of "marginalized" or "novice" participants. "strategies of resistance" (Candela, 1999; Gillmore, 1999) employed by students and genres as "sites of contestation" (Prior, 1998; Bakhtin, 1981) are objects of study within the criticalist program. Such a stance is necessary as a counterpoint to the normalizing effects of standard needs analysis which takes "target" text types and the social status quo that support them as unchallenged givens which students have no choice but to accept (Pennycook, 1994; Benesch, 2001a, 2001b). Importantly, the criticalist approach foregrounds the ethical dimension of praxis, an acute necessity given the post-colonial heritage of ESP/EAP.

Finally, "language socialization" ("LS" henceforth; see for a full acccount, Watson-Gegeo & Nielsen, i.p; Watson-Gegeo, 2001) with roots in first language research of the 1980s is an approach which articulates a set of best practices and an alternative metaphor for second language and literacy "acquisition" research in both a socio-cultural and criticalist mode. Language socialization embodies in its methods a genuinely dialogic and constructivist image of language and socio-literate activity. EAP aims to help students master the genres of disciplinary academic writing and began in a comparison and analysis of target texts. The project of EAP can be more profitably and ethically realized when learning to write is viewed as a complex process of negotiation and dialogue between situated (Lave and Wegner, 1993) social actors, as a form of "literacy socialization."

II. Textually Oriented Studies

The categories by which this paper organizes studies and approaches in sections and sub-sections are not meant to be mutually exclusive; indeed, superior studies incorporate multiple methods and data sources (Carspecken, 1996; Watson-Gegeo, 1992). It is, however, my intention to foreground quite distinct ideas about the nature of language and and their associated research programs. Specifically, my goal is to discern the overall trajectory of English for Academic Purposes and related fields in the evolution of how researchers have variously construed language, genre, discourse community and task. All of the work reviewed in this document is "textually oriented" owing to the nature of the research. The citations in this section share a common root orientation to language as a a "conduit" (Reddy, 1974) and text as "autonomous" (Olson, 1994); i.e., the social is generally backgrounded or, at best, added in to an analysis post hoc as another "kind" of meaning. The associated image of learning is the "banking metaphor" (Freire, 1974, 1994) wherein, the ESP teacher's job is to make the deposit into the mind of the student. Genre is taken to be a "reality in the world" embodied in the form of published texts by recognized experts and amenable to linguistic and rhetorical analysis. The discourse community is conceived of here as an unproblematic and homogenous social space with "experts" at the center and aspiring "novices" at the periphery. The task of the learner within this conception is to emulate, or even directly imitate, the expert's skillful use of academic genres.

A conduit view, (i.e., an instrumental and relatively asocial image of language and learning) and an emphasis on textual analysis have been common to much of the work in ESP and Contrastive Rhetoric, Even so, as the following brief discussions will show, each field has continued to evolve so as to encompass more socially grounded and critically oriented inquiry.

English for Specific Purposes: Origin and Evolution(2)

A recent comprehensive review of the history of the field of ESP (Dudley & St. John, 1998: pp. 19-33) represents the expansion of English to become the preeminent language of world commerce and scholarship in the last half of the twentieth century as a "naturally occurring and inevitable occurrence" (Benesch, 2001: p. 25):

The original flowering of the ESP movement resulted from general developments in the world economy in the 1950s and 1960s: the growth of science and technology, the increased use of English as the international language of science, technology and business, the increased economic power of certain oil-rich countries and the increased numbers of international students studying in the UK, USA and Australia (Dudley & St. John, 1998: p. 19).

As Benesch (2001a) documents, the "flower" of ESP was no wildflower whose seed happened to be blowing in the wind; rather it was consciously and firmly planted in the soil of an emergent post-colonial Anglo-American economic order. Moreover the "flower" was carefully tended by a specific alliance of industry, political and academic groups who actively worked to implant English in the Middle East and throughout the developing world as the language of power and commerce (without regard to how such efforts might negatively impact local languages and cultures).

For example, as Benesch demonstrates, an Arab-American joint oil venture (ARAMCO) justified its massive expenditure on training local Saudis in English as part of its overall mission to help them "join the modern world", make them "company men" (p. 29) and thereby to coax (coerce?) them out of their dependence on "the family, the tribe, the religion, the government…" (Johnson, 1971: pp. 68, cited in Benesch, 2001a: p. 29). While the post-colonial legacy of ESP will be taken up in more detail below (under criticalist approaches), the point to keep in mind for now is that ESP originated in a kind of "vulgar pragmatism" (Allison, 1996; Benesch, 2001b), i.e., the most "applied" and unreflective end of the theoretical-applied linguistics spectrum. This has licensed, until very recently, a largely ahistorical and uncritical stance towards ESP practices and methods which has tended to obscure its (inevitable) ideological and political entanglements (Benesch, 2001a, 2001b; Pennycook, 1994).

The instrumental orientation of ESP is also consistent with the major method of this area of research, which entails the careful analysis of authentic target texts as a way of isolating the grammar and forms of discourse which should be taught (i.e., "transmitted") to students. With its roots in the 1960s, this type of inquiry continues, with some modifications, to the present day. The idea is that the ESP practitioner's first task in any setting is to perform a "needs analysis"; the logical starting place is a linguistic deconstruction of genuine texts produced by experts for real-world purposes. This mode of ESP research, associated with the tradition of "register analysis" (Dudley-Evans & St. John, 1998: p. 21), helps the practitioner identify and prioritize aspects of textual genres that students must master to be considered competent members of the discourse communities they aspire to join. For example, Barber's (1962) article on grammatical and lexical features of "modern scientific prose" is cited by Swales (1988) as a seminal text for ESP. The Barber (1962) article used approaches to descriptive linguistics (that were eventually documented in Halliday, Stevens and McIntosh, 1964) to provide a lexicostatistical profile of a very broadly construed "scientific register" based on a corpus drawn from several disciplines (Dudley & St. John, 1998: p.20). Key findings of the Barber (1962) study were that the continuous tenses were so rare in scientific prose that "they could virtually be discounted" (Swales. 1990: p.2) and that the passive voice figured much more prominently in scientific than in non-scientific discourse.

Following Barber's (1962) lead,

…much material was produced as a result of the practitioner engaging with the teaching situation, carrying out a limited text analysis and then writing a handout or series of handouts. The activity then may have been written up and published as an article, forming the basis of more extensive research. (Dudley-Evans & Huckins, 1998: p.20)

Thus, Huddleston, (1971) investigated the occurrence of verb forms in "scientific English" (as cited in Swales, 1990: p.2). These early studies were motivated by the reasonable goal of "…provid(ing) (within their limitations) a descriptively adequate account of distributional frequencies in the target language variety and thus offer(ing) a basis for prioritizing teaching items in specialized ESL materials" (ibid).

The research article and its sub-parts are among the more well-researched genres as Swales' (1990: pp.131-132) list of studies of this type exemplifies (the following citations were taken directly from the list). These studies can be characterized on three dimensions: linguistic or rhetorical feature(s) that were focal, the field (e.g. "engineering") and the corpus on which the analysis was performed. Typical in this regard are studies of tense in engineering texts (Lackstrom et al., 1972, 1973), and types of lexis (Inman, 1978) modals (Ewer, 1979) and topic sentences (Popken, 1987) in research articles across a range of disciplines. Linguistic, especially grammatical features are a predominant interest, e.g., "tense and aspect" (Ard, 1982), "NP-development" (Dubois, 1982), and "voice" (as in "active" or "passive"; Tarone at al., 1981). From the perspective of later research, there was an inadequate attention in these early studies to such variables as task and target audience as influences on text structure. On the practical level, "…investigations into sentence length, voice, and vocabulary" have had little to contribute to the process of teaching academic writing skills (Swales, 1990: p.3). This was because the final research reports culminated in "discrete item surface feature assemblies of data" which lacked sufficient context to "…be of much interest to those concerned with L1 (or for that matter L2) composition" (ibid.).

ESP-type analyses became both narrower and deeper in the 1980s (ibid). This shift in research mirrored a contemporaneous trend in the Human Sciences generally towards critique of many "common sense" distinctions between disciplines (see the Foucauldian perspective on this turn in Dreyfus and Rabinow, 1983; a socio-rhetorical approach can be found in Geisler, 1994a, 1994b; Prior, 1998; for a post-process writing perspective on disciplinarity, see Kent, 1998). In this case, broad registral labels like "scientific", "medical", 'legal" or even "newspaper" English were being shown to be "too wide", i.e., they assume that the content of the discourse must be the controlling variable of textual form. These early studies were thus not attentive to how textual forms responded to "variation in communicative purpose, addresser-addressee relationships and genre conventions" (Swales, 1990: p.3).

As Dudley-Evans and Huckins (1998: p.21) note, the first significant ESP (really EST, or English for Science and Technology) textbook by A.J. Herbert The Structure of Technical English (1965) exemplifies both the promise and the limitations of register analysis. While teachers admired the treatment of technical vocabulary, they found the book "difficult to use". Moreover, theoretical objections were raised: that

…the concentration on a restricted range of grammar and vocabulary was an insufficient basis for a textbook on EST and that this concentration on form needed to be replaced by a concentration on language use and communication. (Dudley-Evans & Huckin, 1998: p.22).

This complaint represents the pressure which helped to drive the next stage of research in ESP. The field came increasingly to draw from analyses of texts that extended to the rhetorical and discourse levels, (in contrast to earlier quantitative studies which focused only on lexico-grammatical elements at the sentence level or below).

Emblematic of the rhetorical turn in ESP is a contrastive study focusing on when and how writers use we+active verb as opposed to its passive alternate for rhetorical effect in two astrophysics papers (Tarone et al. 1981 cited in Swales, 1990: p.3). This study has since been updated and replicated with a cross-cultural comparative twist (Tarone et al., 1998). Tarone's (1998) research team found that

(1)We + active verb indicates the author's procedural choice whereas the passive indicates the standard procedure; (2) we is used describe the author's own work and the passive the work of others unless the other's work contrasts with that of the author in which case the active is used; (3) the passive is used for the author's proposed studies. And (4) the use of the active is determined by focus due to the length of an element or the need for emphasis" (p.113).

These results hold up for a comparable corpus from Russian astrophysics articles. Once such "secrets" of academic textual construction are revealed, they can inform discipline-based writing instruction in a way that context-free lexicostatistics or mere frequency counts for various grammatical features can not. Similar rhetorical or discourse-level analyses include those which focus on paragraph development (Lackstrom et al., 1972; Weissberg, 1984), authorial comment (Adams-Smith, 1984) or overall rhetorical structure (Hill et al., 1982)--see Swales, 1990: pp.131-132 for a more complete list of studies of this type.

Thus, as models of language in the 1980s came to emphasize "communicative competence" (Hymes, 1974), the overall effectiveness of the writer's message was foregrounded at the expense of more prescriptive formulations of style. Recent editions of the journal English for Specific Purposes show that this line of rhetorically inspired research continues to the present. A recent study of the Acknowledgments sections drawn from various disciplines (biology, economics, mathematics, medicine, linguistics, and sociology) is a case in point (Gianoni, 1997).

Though apparently only a minor part of research literature, such sections encode a wide range of communicative strategies, whose sociopragmatic constraints can be interpreted in terms of "generic meaning potential" (p.1).

Employing a method of rhetorical analysis similar to that originally developed by Swales (1981) and elaborated in Swales (1990)--see appendix A from Swales, (1990: pp.141,143) for an example--the Acknowledgement macrostructure is divided into three rhetorical units or "moves", consisting of several sub-steps. Highlighting the interpersonal nature of scientific discourse, the study concludes that " like linguistic politeness, Acknowledgements are affected by the relative status and social distance of participants, with senior academics earning more credits for the same type of support" (Gianoni, 1997:p.1).

Other similar research has focused on the use of pronouns in scientific journal articles. Viewing written text as interaction, Kuo (1998) investigates how the use of personal pronouns may reveal writers' perceptions of their own role in research and their relationship with expected readers as well as the scientific-academic community. Findings include:

…first-person plural pronouns are used far more frequently than other types of personal pronouns. A further analysis of first-person plural pronouns suggests that they can have a number of semantic references and perform multiple functions in the journal article. Examples from sampled texts show how writers use strategically exclusive we to refer to writers themselves or inclusive we to refer to either writers and readers or the discipline as a whole for different communicative purposes. The use of second-person, third-person and indefinite pronouns also reflects a writer's intention to secure cooperation from, and stress solidarity with, readers (p.121).

Such findings are based on a simultaneous consideration of linguistic features (pronouns in this case) and the underlying rhetorical rationale for the use of such features. This study and its findings are representative of the rhetorical/discourse analysis approach to ESP needs analysis which, having gained momentum in the mid-1980s is still very much alive in ESP.

As mentioned earlier, the focus of ESP narrowed during the 1980s to accounts of particular linguistic or discoursal features within specific disciplinary genres like the research article introduction (Swales, 1990) or the Acknowledgements section (Gianoni, 1997). This narrowing has been accompanied by a concomitant "deepening" (Swales, 1990:p.2) or perhaps more precisely, "thickening" (Geertz, 1973; Watson-Gegeo, 1992) of the research record to include consideration of authorial purpose and social context, thereby aspiring to a "multi-layered" rather than "one-dimensional" account of textual form (Swales, 1990:p.3). Thus, as in Tarone (1998), an increasing focus has been placed upon why authors decide to use particular words and grammar (c.f., Bhatia, 1997; Hopkins, 1985; Crookes, 1986; Swales and Najiar, 1987; Samraj, 2002a). A typical recent study (Holmes, 1995) in this mode analyzed the Discussion sections of thirty social science research Articles, ten each from the disciplines of history, political science and sociology. The articles were analyzed in terms of the sequence and structure of their rhetorical moves (see appendix B for an exemplar of this kind of analysis). "It was found that, although there were fundamental similarities to the natural sciences, social science Discussion sections also displayed some distinctive features. History texts were particularly distinctive, and of the three disciplines bore the least resemblance to those of the natural sciences" (p.321) Such studies point to the ways in which academic genres can vary significantly across disciplines, which has important implications for ESP needs analysis.

"Work in ESP was by the middle 80s, not merely interested in characterizing linguistic effects; it was also concerned to seek out the determinants of those effects" (Swales, 1990: p. 4). ESP, thus, came to be concerned with discourse as communication and interaction in a social context. The trajectory of ESP toward a socio-cultural orientation can be discerned in researchers' changing ideas about the objects of their study. For example, a text and its genre are viewed as "autonomous" (Olson, 1994), i.e. as "conduits" (Reddy, 1974) for pre-existing and autonomous meaning in earlier registral analyses. The discourse community in these studies is an unexamined "constant", a monolithic space of expert readers. Whether based on register or discourse/rhetorical analyses, studies of this type are based in the common sense idea of disciplines as distinct and disparate "fields" which are, in turn, embodied in canonically shaped texts. Similarly, the task of the learner is reflexively construed: she must internalize through direct instruction and practice the textual norms of the discourse community which she aspires to join.

Work in the mid to late 1980s began to focus more extensively on text structure as a realization of the writer' s communicative purpose and less on morpho-syntactical elements of the sentence level (see McKinlay, 1984; Belanger, 1982; Hopkins and Dudley-Evans on the "discussion section" of research articles). To this extent, the previously monolithic images of the text-as-conduit and the discipline-as-discourse community were being construed in increasingly functional, contextual and social terms. Researchers' notion of genre expanded from an unproblematic "type of text" to include conventional purposes and aspects of social context associated with a given textual structure. Contemporary work in ESP, as exemplified in recent issues of the ESP Journal includes an increasing number of ethnographically oriented studies which replace after-the-fact intuitions about intended meanings with empirical investigations of how real people in real settings produce and interpret textual genres in vivo. This trend will be taken up in more detail in the Socio-Cultural section below.

The next approach which I will touch on incorporates a much more social and context-oriented view of language and text from its inception. In some respects, Systemic Functional Linguistics begins about where textually-oriented studies in ESP leave off.

III. The "Australian School": Genre, Discourse Community and Task in Systemic Functional Linguistics

No discussion of academic genres would be complete without at least an outline of how the tradition of SFL, or Systemic Functional linguistics has responded to this area of study. SFL presents a theory in which form and function are always considered together; language-in-use and the grammar of genre itself are forms of social action. My plan here is to briefly describe the SFL framework, then to indicate its methods and approach to research with a few brief examples of how researchers have understood language, register, genre, discourse community and task along the way. I will then bring out some critiques of SFL that have been raised (for example that it is more "semiotic" than "social" (Fairclough, 1999) and assess the possible contribution of this line of research to critically oriented and situated studies of academic literacy. I'll close with a review of SFL in terms of the key themes that organize this paper.

Criticalist Roots & Key Distinctions in SFL

SFL, even beyond its substantial research base on academic writing genres (see Cope and Kalantzis, 1997 for a recent sampling), connects with the overall criticalist thrust of this paper. It has deep roots in movements for educational equity and social change, particularly in the context of providing access to the "genres of power" to disadvantaged immigrant and aboriginal populations in the Australian context (ibid.). Similarly, "the interactionist nature of systemic linguistics implies a notion of ideology quite consistent with that developed by Foucault and a notion of discourse consistent with that of Bakhtin" (Sullivan, 1995: p.1) Foucault's and Bakhtin's work is also routinely sampled by criticalists who seek an explicit means to explore the inter-articulation .of power in the social order and patterns in language. SFL is attractive to the criticalist project because it aims at making explicit the social ground of language, an intention which is not achievable within the positivist--and currently hegemonic--"conduit" or "banking" metaphor.

SFL research (even to the novice eye) approaches issues in language and text structure with a theoretical and methodological rigor that deserves much more attention than I can give it here (after four decades of work by hundreds of scholars in settings all over the world). It also merits special attention owing to its status as a full-blown theory of language and grammar rather than as a mere set of heuristics for text analysis. I'll start with an overview and then plunge into a more detailed consideration and critique of the place of "register" and "genre" within SFL.

Largely parallel to, and contemporaneous with the ESP approach, SFL (see Thompson, 1996; Christie & Martin, 1993; Cope and Kalantzis, 1997) has carried out a wide-ranging research program, mostly within Australia, on academic genres. SFL, which is notable for its substantial impact on K-12 curriculum, begins with an idea that was only late in coming to ESP (and Anglo-American linguistics generally, see "Linguistics Wars", Harris, 1996): that form and function of language must always be considered together in a theoretically complete formulation, especially one with pretensions to be applicable to real-world use and social interaction. In this and several other ways, the assumptions of SFL differ significantly from the Chomskyan-Generative tradition (see for an overview, Pinker, 1994, or for a more concise formulation, McGroarty, 1998). Generative approaches imagine a syntax autonomous from semantics or meaning in which the sentence (The S node in the generative "tree") is the prime and starting point of all analysis.

Social context, in the generativist formulation is not part of the formal, mathematical "deep" structure of any given sentence; rather a sense of context, along with lexical items is "added" to the structure of a sentence to yield the meaning intended in a given "performance". Sentences are, at root, mathematical expressions and "meaning", or the functional import of a given sentence results from values (e.g. lexical item) binding to variables (e.g., "N" or "V") of a deep structure, (say, the passive version of a sentence), in a particular context. Context necessarily remains outside the purview of linguistic theory, as a "performance" issue. ESP was apparently strongly influenced by this "sententiocentric", formalistic and asocial momentum within linguistics, particularly in its formative moment.

SFL, by contrast, finds its initial inspiration, not in isolated sentences made up by grammarians but in studies of how people actually use texts to do things in the world (Halliday, 1974; Halliday & Hassan, 1976) and thus takes context as a constitutive element of both grammar and meaning generally. SFL also has quite a different epistemological project from the generativists, inasmuch as it aims to describe not the "intra-organism" but the "inter-organism" (Halliday, 1974: p.10) dimension of language, i.e. "language-in-use" (p. 211). From the point of view of SFL, generativist linguistics consists of "elegant self-contained systems that are of only limited application to any real issues" (Halliday, 1978: p.3). Even a cursory treatment of the well-articulated SFL terminology and methodology is outside the scope of this paper (see Thompson, 1996 and Martin, 1997 for this). Nevertheless, a closer look at the SFL image of language and its associated notions of genre, discourse community and task highlights the possible contributions of the "Australian School" --as well as some of its possible limitations as an approach to EAP and genre studies generally.

"Functional linguistics is fundamentally concerned with showing how the organization of language is related to its use" (Martin, 1997: p.4). SFL's principal unit of analysis is an "entire text" (ibid) rather than a sentence. While also providing an account of sentential structure, SFL seeks to uncover how the various elements of coherence (i.e., " the emerging sense") and cohesion (i.e, "the way the text 'hangs together', e.g., pronouns") are distributed across sentences and throughout texts in ways that are characteristic of particular genres.

A quick example might clarify this point. An early SFL treatment focuses on an analysis of a brief, humorous parable by James Thurber about a couple of parrots disturbed by the amorous noises of two neighbor hippopotamuses (Halliday, 1978). The analyst shows how the writer makes use of grammatical resources such as transitivity, verb tense, pronouns and the generic structure of the parable to realize a genre of "narrative with dialogue" (p.148). Such explorations get at the ways that texts represent the specific linguistic choices writers are making to achieve their aims against the backdrop of all the other ways they might have expressed themselves, i.e. the "meaning potential". "We see the text as actualized potential; it is the actual seen against the background of the potential" (p.40)

In the SFL image of language and social context, they are taken to "metaredound", i.e. to realize each other : "language realizes social context, social context realizes language" (Martin, 1997: p.4).. Thus. "language construes, is construed by and (over time) reconstrues social context" (ibid.). Language is a "social semiotic" (Halliday, 1978) i.e. language is interpreted "within a sociocultural context in which the culture itself is interpreted in semiotic terms--as an information system, if that terminology is preferred" (p.2).

Language functions are sub-divided into three key metafunctions: "ideational", "interpersonal" and "textual". "Ideational linguistic resources are concerned with representation, interpersonal resources with personal interaction and textual resources with information flow" (ibid). In SFL, this functional organization is "projected" onto context, redounding, in turn with the context variables of "field" (what is being talked about), "tenor" (the social relationship between sender and receiver) and the "mode" or channel in which the communication is being realized (e.g., spoken vs. written). "Register" is used as a term to subsume the field, tenor and mode context variables at the level of meta-function. Thus, the "register" of a text can be defined by the fact that it refers to a particular field, say "medicine", or indexes a particular relationship, say "doctor-patient" or by the mode in which it is transmitted, say, a "written" prescription. One could properly speak of a "medical register", a "doctor-patient register" or a "written register" within SFL in identifying functional aspects of texts. As might be expected, the more one knows about the field, tenor and mode in a given situation, the stronger "predictions" one can make about the language that will be used to realize this particular configuration or "register" (p.32). "Register" accounts for "variation in use" versus "dialect" which accounts for "variation in the user." The predictive ability inherent in register goes both ways: if I presented you with the text: "Directions: shampoo, rinse, repeat", my guess is that you could tell me where and in what context this text would likely be found. Your ability to immediately make that kind of connection between a text and its context are exactly the kind of thing that SFL aims to explain formally with the idea of register.

Problematics with Genre and Register

I must insert a caveat at this point: SFL, like any broad-based intellectual movement, has evolved and changed considerably over time and its many partisans do not necessarily agree, particularly with respect to the use of the key terms "register" and "genre" and the exact relationship between them (Ventola, 1984 is the first to trace this in detail; Butler, 1989 follows the story through the 1980s). In the canonical formulation (Halliday, 1978), the idea of genre still seems fluid: for example, the "mode", or relationship of the channel of language to the situation is said to encompass Hymes' (more or less informal) idea of "genre" (p.63). Similarly, "The concept of genre…is an aspect of what we are calling the 'mode'" (p. 145). In another place, the author seems to accept the everyday or literary understanding of the term as a kind of text, with "generic structure (being) the form that a text has as a property of its genre" (p.133).

Elsewhere we find "the various genres of discourse, including literary genres are the specific semiotic functions of text that have social value in the culture" (p.45). Immediately following this passage he presents the contrasting (?) notion of register as "the semantic configuration (which) defines the variety…of which the particular text is an instance" (ibid). This early formulation of register and genre was confusing for me as the author seemed to be moving between "everyday" notions of genre as a "kind of text" (e.g., fable, humorous story, narrative with dialogue) and a more technical formulation which I had trouble sorting out from the statements about register. On a similar note, I struggled with the idea of genre as a mode, while at the same time trying to think of genre as a "superordinate" or controlling term to register (which supposedly encompasses mode). Apparently, I'm not the only one who is confused. Swales (1990) echoes Ventola's (1984) uncertainty: "The relationship between genre and the longer established concept of register (within SFL) is not always very clear" (p.40)

Swales (1990) holds out the work of Couture (1986) as an exception in the clarity of its formulation of register versus genre in SFL: "Registers impose constraints at the linguistic levels of vocabulary and syntax, whereas genre constraints operate at the level of discourse structure" (p.40). Couture feels that the two need to be carefully distinguished: "genres (research report, explanation, business report) are completable structured texts, while registers (language of scientific reporting, language of newspaper reporting, bureaucratic language) represent more stylistic choices" (ibid). There is a potential tension between genre and register, inasmuch as a writer might choose a genre that carries the expectation of high explicitness (say, a business report) while at the same time choosing a less explicit register (say, bureaucratic language). In such a situation the writer must decide "which criteria for explicitness he or she wishes to dictate linguistic choice" (Couture, 1986: p.87, cited in Swales, 1990: p.40).

SFL has identified genre as "a category that describes the relation of the social purpose of text to language structure" (Martin, 1993: p.2). Genre is brought forth as a term which emphasizes the conventional purposes which shape otherwise disparate "registers". Thus, "science textbooks", whether they be for Physics or Biology, share similar structural features because they serve similar communicative intentions. SFL conceives of genre as a "cultural script" which guides the unfolding stages of linguistic form, especially as these are elaborated in academic texts (Veel, 1997) and connected chunks of classroom discourse (which may span over an entire instructional unit-c.f., Christie, 1993, 1997).

"Genre" in J.R. Martin's (1993) formulation is designed "to account for relations among social processes in more holistic terms, with a special focus on the stages through which most texts unfold" (Martin, 1997: p.6). Research in the (largely Australian) tradition of Martin (1985) has used his definition of genre as "staged, goal-oriented purposeful activity (p.251) to describe the inherent meaning structure of a text and to uncover the generic structure potential" (Samraj, 1995: p.1) of particular genres. For example, Rose (1993) seeks to unpack the "structural potential" of an article on Fisheries Management by presenting text as "nested" within its register, i.e., within its context (field, tenor and mode). The register is in turn "situated" within its context of "culture" (the level of genre). The result of the analysis is an explicit description of the "specialized grammar" that writers in scientific genres employ, " a grammar that has evolved over hundreds of years, with science at its cutting edge, to construct the world in different ways than talking does" (Martin, 1993: p.134). This grammar includes strategies for chaining concepts, building arguments and marking key rhetorical and structural transitions which are expected of scientific and academic genres.

In keeping with the educational emphasis of SFL, Christie (1993, 1997) extends the idea of genre to include the curriculum and classroom lesson. One study she conducted reveals the distinct stages of instruction by which the lesson proceeds through discourse analysis of classroom talk in a science class. In the course of several lessons, the everyday proposition, "machines make work easier" is incrementally elevated to a technical concept, "mechanical advantage" (a grammatical metaphor in SFL terms). Her study shows how students gradually master the material and move from "consumers" to "owners" of the technical terms in the lesson, culminating in their oral research reports on their own experiments. Lemke (1990) offers an additional set of studies aiming to unpack in SFL terms the challenges of teaching scientific genres in the context of secondary science. Christie's and Lemke's work show how SFL provides a theory of grammar which can be a very useful tool in the analysis and teaching of genre, especially in educational settings.

Discourse Community and Task

While "discourse community" is not a term in SFL(3) , the linguistic resources with which bureaucracies and other institutions construct their authority are being actively investigated under its flag (c.f., Iedema, 1993, White, 1993; Veel, 1993; Coffin, 1993). The discourse community is also implicitly evoked when students are considered-as they are generally within SFL--as apprentices or "novices". Within SFL, the principal task of the student-as-apprentice is to internalize--gradually and over time-- the "domain knowledge" and "rhetorical forms" (Geisler, 1994a, 1994b) that mark the "expert's" mastery of academic genres.

Critiques of SFL

While there is clearly much of value in SFL and its possibilities for genre-oriented research are large, a balanced treatment will mention some of its limitations as well, particularly as these pertain to the criticalist orientation which I am attempting to articulate here. The challenges to SFL which I will consider briefly can be glossed under reification, thinness and criticalist


SFL's methods and formal statements point continually toward the social and situated nature of discourse (witness for example, the term "text-in-situation" Halliday, 1978:p.192, as an alternative formulation of text to express how it is always already interpreted in context). Its proponents are very careful in their formulations vis-à-vis language as a "shared meaning potential, at once both a part of experience and an intersubjective interpretation of experience" (p.1). Simlarly, "It is impossible to draw a line between what he (the writer) said and how he said it since this is based on a conception of language in isolation from context" (p.34). If one understands "what is said" as the "content" and "how he said it" as the "expression", this last statement is hard to reconcile with SFL's "fundamental separation of the content and expression planes" (Martin, 1993: p.5).

In an obvious and (it could be argued, inconsequential way) the word "content" evokes the conduit view of language, in which meanings are reified and "transmitted" "in" words. Despite its commitment to the social character of meaning-making, SFL is apparently not immune to a conceptual metaphor (Lakoff & Johnson, 1980) and associated "cultural models" (Quinn) which are not in alignment with that commitment. Thus, Halliday (1978) makes comments like the following (italics mine):

"The most everyday uses of language transmit to the child the essential qualities of the society and the nature of the social being" (p.9)

"Language is the medium through which a human being becomes a personality…"(p. 15).

"A common speech form transmits much more than words…" (p.26).

"Language is a principal channel of communication" (p.55).

"The code transmits or controls transmission of the underlying pattern of a culture or sub-culture" (p.111).

"The meanings are encoded in and through the semantic system…" (p.139).

As noted, SFL in many of its formal statements carefully avoids the simplistic terms of the conduit metaphor. Still, researchers' understandable "lapses" into "commonsense conduit-talk" must be taken as something more serious when it is pointed out that these "lapses" are sanctioned and codified in an emergent information-based semiotic paradigm with which SFL specifically aligns (Halliday, 1978). The larger problem, of which this point may be symptomatic, is that SFL seems to some to "privilege the semiotic at the expense of the social and the text at the expense of the system" (Toolan, 2001: p.15 in a review of Chouliaraki & Fairclough, 1999: Ch. 8); this sentiment is exactly echoed in Sullivan, 1995). This critique may perhaps be a reflection of the reification problem inherent in any formal system which is attempting to explain an interpretive process "after the fact" (c.f., Kent, 1998 for an exploration of this idea for "post-process" composition theory; for the deeper ontological and epistemological problems inherent in formal accounts of engaged activity, see also Heidegger, 1962 and Dreyfus & Rabinow, 1986)


Much SFL research appears to consist of (sophisticated) textual analysis supplemented by "common sense" extrapolations about motivations for particular structural features. Nevertheless, the social and intersubjective nature of meaning continues to intrude into the neat analyses of genre. In an extended technical analysis of the resources for APPRAISAL in English (of the novel Rubyfruit Jungle, for example, Martin confronts disparate accounts of the evocations, or instantiations of the resources in an actual text which depend on the different points of view of a teacher and a student: "…it (this disparity) means that analysts need to declare their reading position- since the reading one makes depends on the institutional position one is reading from" (Martin, 1997: p.25). It is notable that position has to be "added" as an adjunct to an otherwise "subjectless" analysis. My critique, from a socio-cultural perspective, is that SFL, while aspiring to describe an "intersubjective transaction" (in this case, a response to a controversial novel), ends up adding subjectivity after the fact to what is fundamentally an objectivist and conduit-derived model of meaning. A related problem lies in the fact that ethnographic studies show that the sentence-centered analysis of SFL and the assignment of a unique function for every sentence can not account for the multiple speech acts which can occur simultaneously in a single utterance. For example, studies of the discourse of the Emergency Room show that an attending physician's statement, "Let's just pop a couple of little stitches in" can accomplish multiple functions simultaneously, including a request for the patient to get into position, for the nurse to prepare for the procedure and conveying a calming, informal attitude of professional competence (Candlin, et al. 1976, cited in Swales, 1990: p.11 as part of a critique of SFL). It is difficult to imagine how all of these meanings can reside "in" the statement simultaneously.

The focus on texts qua texts also apparently limits the methodological scope of SFL to include only the point of view of the expert analyst about the situation in question. This leads to a lack of triangulation among multiple points of view, and an empirically "thin" record. If my cursory reading of several recent books in SFL is any indication, there has been a systematic exclusion of students' and others' points of view from the actual research record (even when their discourse is included as part of the research record as in Christie, 1993). This exclusion is licensed by the fact that SFL is a formal and technically elaborated system; thus, the statements of students and other non-initiates do not have the status of officially sanctioned knowledge within the SFL paradigm. Informal statements must be translated into "SFLese" before they count as knowledge. Moreover, "analysis of Halliday's methodological stance suggests that his work does not normally proceed by the explicit formulation, in print, of specific hypotheses that are then subjected to testing and rejected if crucial counterevidence can be found" (Butler, 1999: p.1).

The thinness of the record typically provided in SFL accounts (in contrast with, for example, ethnographic accounts) makes it necessary to "fill in" a great deal about the nature of context from sources outside the research itself. For example, Rose (1993) depends on an a priori theory of society (Bernstein, 1971) to explain observed variations in modality in science texts from the way modality is employed in everyday language. Inasmuch as SFL analyses depend only on the analyst's own intuition without consideration of parallel counter-narratives and multiple interpretations and possible points of view, without other "voices" and triangulated data sources, they fail the test of providing a genuinely "thick" record (Geertz, 1973; Watson-Gegeo, 1992). SFL, from this point of view is thereby limited in its ability to contribute to a "thick explanation" of social life. To say it another way, as Chouliaraki & Fairclough (1999), Sullivan (1995) and Hyon (1996) have documented, SFL is limited by its insufficient emphasis on the social in general vis-a-vis the semiotic, on the instance (the text) as opposed to the system, and by its lack of recognition of a system corresponding to the 'order of discourse' (Bordieu, 1991). For these reasons, as Chouliaraki & Fairclough, (1999) and Sullivan (1995) show, SFL on its own and unmodified does not allow for an adequate explanation of hybrid texts, (that mix discourses, genres, and registers nor of power relationships). It also systematically undertheorizes the positionality of analysts themselves to the detriment of the model's potency for criticalist work.


One well-articulated critique of SFL (Sullivan, 1995) shows how this sophisticated theory of text and grammar can be augmented and modified within a criticalist program to respond to many of the above criticisms. The author reconsiders in detail a very sophisticated SFL experiment (Hake & Williams, 1981) which involved having teachers evaluate student-written essays under various conditions. One of the twists in the research design was that some of the student papers were re-written with one significant change: statements of the type: (1) I prefer life in the city were re-written as. 2) My preference is for life in the city. In SFL terms, the second statement uses the resource of nominalization to express the idea in less concrete, more abstract terms. The authors found that the teachers did not rate the modified essays down for being less explicit as expected, and they explained this as a training problem: teachers were not putting into effect the standards of "clarity" for the essay as they had been trained to do. Sullivan's alternate explanation shows how SFL's commitment to a "hierarchy of functions" based in social constructivism (Jakobson, Firth) limits its analysis in this case. The authors assumed--based on the a priori categorization of the situation-- that the "ideational" function was dominant given that essays were being evaluated. They were leaving out of consideration that it was an "examination" seen from the students' point of view, and therefore that the use of nominalization concurrently realized the important interpersonal function of deference to the preferred style of school and teachers:

As we have seen, nominal and verbal styles also realize differences in the social relations presumed to be operating in the situation. The indirectness of the former is seen, generally, as marking asymmetrical social relations, whereas the directness of the latter is associated with more symmetrical relations. Thus, the preference for the nominal style represents as much a preference for deference as for abstraction. Such an explanation is consistent with readers observing, yet forgiving, substantive problems in nominal texts. These texts appropriately represented not only information but also attitudes toward authority--their examiners. These are the texts of students who show "promise" (p.18).

Critical theory allows for the bracketing of SFL's foundational commitment to a hierarchy of functions determined uniquely by the situation "type", a commitment which it inherits as a branch of "systemic linguistics" (Jakobson, 1960; Firth, 1957). Sullivan shows how, "through reuniting the theorizing power of critical theory and the methodological power of systemic linguistics, we can analyze the discursive tactics and strategies of those whose encounters constitute and remake the Contact Zone" (p.28). The Contact Zone is understood here as a space of hybridity, heteroglossia and non-converging difference in contact, an image which stands in stark distinction to the idealized and essentialized "communities" common to SFL and most linguistic theories (Faigley). Fairclough (1999) and Sullivan (1995) converge in their recognition of SFL as a powerful and useful tool within the criticalist project, but as insufficient in and of itself to achieve its professed ideological aims. Chouliaraki & Fairclough (1999) find in SFL an account of text which adds analytical depth and rigor to his own critical discourse analysis approach ("CDA"). CDA, in turn, provides a rich articulation of the social order which, from his point of view, is under-theorized within SFL. Sullivan (1995) sees useful connections to be made between SFL and critical theory with a similar symbiosis. Hyon (1996) proposes a combination of SFL and New Rhetoric to the mutual benefit of both fields. The time and trouble these researchers took to master SFL formalisms and terms attest to its perceived value as a theory of text and grammar within a criticalist program.

SFL and Educational Praxis

The genre-based approach to literacy involves identifying the key resources employed in target discourses and apprenticing students to the "generic potential" of those resources. The success and appeal of SFL may well lie in this fact: it provides a theoretically consistent, classroom-tested and, by now, well-documented set of tools by which researchers and educators can exploit the social and linguistic construction of texts for the benefit of students.

In SFL-inspired teaching,, form-focused instruction, including explicit attention to grammar as a resource for meaning-making,, helps to build students' meta-linguistic resources within the context of real-world communication.. SFL has, unlike its Generative "rival" in the space of theoretical linguistics (see McGroarty, 1998 for a summary and critique), contributed substantially to curricular reform in the area of literacy instruction. Some years after the "traditional" i.e., canon-based curriculum was discarded in Australia in the 1970s, it was clear the "progressivist" (process writing, whole language) curriculum that replaced it was also "not producing the goods" (p.1) and it was systematically displaced.

The progressivist "vacuum" has been largely filled and the genre-based SFL approach to literacy, now well established in K-12 instruction in Australia, is spreading to North America, Britain, Scandanavia and Israel (ibid). This further attests to the maturity and relevance of the SFL paradigm as a robust player in genre studies.

IV. Contrastive Rhetoric: The Cultural Dimensions of Textual Variation

Whereas SFL focuses on the interaction of linguistic form and context, Contrastive Rhetoric operationalizes "culture" as an influence on textual structure by comparison of texts produced within, for example, the "same" academic genre but in different languages and cultural contexts. The roots of Contrastive Rhetoric (Connor, 1996) lie in a comparative study of how students from various non-English L1 backgrounds organize written paragraphs in English differently from native writers (see the famous "squiggles" article which started it all: Kaplan, 1966). For example, the strings of coordinated clauses which typify L1 Romance students' sentence structure in English are considered good academic style by readers and writers in their host cultures but seem florid and obtuse when translated wholesale into the Anglo-American context. Many of the "aberrations" in L2 writing (per Kaplan's, 1966, argument) result from the "transfer" of rhetorical and cognitive conventions into L2 from students' L1.

The research program has evolved in the ensuing years from an emphasis on L2 Writing to encompass studies of how a rhetorical structure of genre can vary within a discipline across linguistic and cultural boundaries. For example, a comparison of abstracts in linguistics journals in Swedish and in English (Melander, et al., 1998) shows that the nature of the "academic market" in a given discourse community has a bearing on the rhetorical forms of its genres. Since linguistics in Sweden is not nearly the large-scale "industry" it is in North America, it possesses the kind of small and intimate feel possible in an academic community in a country of only seven million. As a result, writers for linguistics journals in Sweden make more assumptions about a common base of knowledge (e.g., via insider references and rhetorical leaps) than their U.S. and British counterparts. Similarly, Swedish linguistics texts were found to be less transparent or "readerly" in their discussion of methodology and reasoning than those in the Anglo-American sample. The authors build a convincing case for the observed generic variations based on a demographic and socio-rhetorical re-construction of the Swedish and Anglo-American contexts. In an edited volume, Duszak presents five similar examples, mostly contrasting Polish/English realizations of disciplinary genres (Duszak, 1997).

One study of abstracts which encompasses both cross-disciplinary and cross-cultural perspectives is that by Melander et al. (1997) which compared abstracts from Biology, Medicine and Linguistics (by native speakers from the U.S. and non-native speakers from Sweden). "While this study and others on disciplinary variation in research articles have given us some insights into generic variation within one cultural/linguistic community due to the influence of disciplinary values, our knowledge of the interaction of generic and disciplinary norms in text structure is still in its infancy"(Samraj, 2002c: p. 4).

Connor (1996) attempts to articulate a theory of contrastive rhetoric based on a review of three decades of research inspired by Kaplan's (1966) article. She traces how the field has been influenced by the trajectory in applied linguistics from "error analysis" to "transfer" to "interlanguage." Her review is notable for its inclusion of research on genres in a variety of cross-cultural writing contexts including Arabic. Chinese, Japanese, Korean, German, Finnish, Spanish and Czech. After a comprehensive review of over forty research reports, Connor concludes:

Since Kaplan's first study, a wealth of research has compared writing patterns and styles in many languages and cultures. Kaplan's study contrasted English essay writing of the E S L writing of international students with the largest concentrations in the United States in the 1960s, namely students from Asia, the Middle East, and Europe. Subsequent studies, such as the great many studies focusing on Chinese, Japanese, and Arabic, echoed similar contexts and goals. Yet significant changes in emphases of research are obvious. Although committed to the general goal of better understanding E S L writing, research has not been limited to analyzing L2 student writing. Instead, writing patterns and styles in a variety of first languages have been addressed both by E S L experts working in the United States and by EFL (English as a foreign language) experts in many other countries. The initiatives of contrastive rhetoricians working in many countries and languages have reduced the Anglo American bias in the earliest contrastive rhetoric (p.55)

The field has also changed in its orientation toward the "Whorf Hypothesis" (see Gumperz and Levinson, 1996 and Lee, 1994 for significant updates on this false and misleading term). The field has also evolved to "include a diversification of the genres of writing studied: newspaper writing: both news stories and editorials; academic writing; and professional writing of all sorts, particularly business correspondence and facsimiles" (Connor, 1996:p.55). A very recent study focuses on computer-mediated communication in contrastive rhetoric terms (Scoggins, 2001), highlighting the possibility for intercultural misunderstandings online owing to differing discoursal expectations. The expansion of the scope of the field of CR in both the United States and other countries reflects the realization that English is now the international language of science and Commerce (Pennycook, 1994). For millions of people who use English as a non-native language, there is a lot at stake in understanding the questions which are at the core of CR inquiry: How and why do texts vary in their form across culture and language groups? How is interpretation and evaluation of such texts problematic owing to differences between readers' and writers' unconscious--and largely culturally derived-- assumptions?

From a criticalist perspective, researchers in the tradition of contrastive rhetoric provide an important counterpoint to essentialist models of genre and discourse community which tend to "creep in the back door." Recent, more critically oriented work within this originally ethnocentric tradition has appropriated contrastive rhetoric to articulate the position that "alternate" formulations of genre should be "equal" to "expert" formulations of genre from the point of view of African American women (Comfort, 2001) "queers" (McBeth, 2001), women (Micceche, 2001) and other traditionally silenced others. Similarly, the Anglo-American hegemony over the standards and genres of discourse is being challenged in the movement toward recognition of international and World Englishes in business and elsewhere (Connor, 1996: pp.16-18; Woolever, 2001; Pennycook, 1994).

In some measure, research in the tradition of contrastive rhetoric responds to a critique I raised in discussing other text-analysis studies. Rather than depending on external models and common sense to unpack cultural variables, CR empirically explores variation in genres based on cultural and disciplinary factors. Still evident is the exclusively textual orientation of most studies. Contrastive rhetoric and its updates continue to be limited in their explanatory power by the "thinness" of their data, e.g., the lack of genuinely ethnographic accounts of the contexts in which contrasted texts are produced and interpreted. Still, the field articulates an important voice of conscience in the form of its founder, quoted here, some thirty-five years after his little article launched an influential research tradition:

The individual who does not participate in the monolingual, monocultural assumptions that dominate the composition classroom is faced with the task of answering five terrible questions:

(1) What may be discussed?

(2) Who has the authority to speak/write? Or: Who has the right to write to whom under what circumstances?

(3) What form(s) may the writing take?

(4) What is evidence?

(5) What arrangement of evidence is likely to appeal (be convincing) to readers? -(Kaplan, 2001: p.ix)

Kaplan's concern, raised originally in the context of composition studies, applies equally to content area or discipline-based instruction. More long-term ethnographic case studies are necessary that focus on how students actually learn to answer Kaplan's (2001) questions in differing contexts (c.f., Prior, 1998; Berkenkotter at al. 1988). As a "critical mass" of such studies accrues, the capacity of the field of contrastive rhetoric to improve content area writing instruction for diverse L2 writers will be more fully realized.

This section has traced the evolution of the concept of genre and related ideas in academic writing research as these have played out in several textually-oriented traditions. English for Special Purposes began in quantitative analyses of academic genres and became both "narrower"--in its increasing emphasis on both the communicative impact of specific forms-- and "deeper" in its growing attention to the overall structure of texts as a response to communicative purpose and context (Swales, 1990: p.2). Systemic Functional Linguistics presents an elaborate, self-contained and highly technical approach to these issues. While perhaps optimally combined with more critically potent approaches, SFL has proven its applicability and usefulness in K-12 educational settings where the goal is to design and execute a staged apprenticeship for students in the interpretation and production of academic genres. Finally, Contrastive Rhetoric has rendered the cultural dimension of academic genres amenable to empirical investigation, thereby opening the field of academic writing research to critical and relevant questions about power and authority in a globalizing, post-colonial world.

IV. Genre Analysis: Pulling It All Together?

This section considers one approach to the study of academic genres which is significant for its wide-ranging influence. In the context of the evolving narrative in this paper, "Genre Analysis" (Swales, 1990) can profitably be considered a "pivot point" between textually oriented and "genuinely" socio-cultural treatments. While the terms of Genre Analysis are sensitive to developments in, and the importance of, socio-culturally oriented approaches for genre research, aspects of this formulation have been challenged even as they have been adapted by subsequent researchers with a more critical or ethnographic focus. The theoretical and methodological approach articulated in Swales' volume--which I gloss here as "Genre Analysis"-- has been very influential among researchers and practitioners interested in tertiary and graduate-level academic writing. In the context of the present discussion, Genre Analysis represents an attempt to integrate multiple methodologies and data sources in the study of academic genres in English, especially from linguistics, rhetoric and anthropology. The overall aim is to frame systematic inquiry that can inform "best practices" in the design of English for Academic Purposes curricula.

My reading for this paper began with an ERIC search on all articles that came up under the key terms "English for Academic Purposes" and "Contrastive Rhetoric" since 1980. I decided to look for overall patterns in the theoretical frameworks and references in the corpus of over 250 abstracts that resulted as a way of uncovering canonical sources that might ground my own work. I also reviewed over a hundred abstracts from the past five years' volumes of Second Language Writing, a journal devoted to these concerns to get a sense of the scope and methods of this emergent field. Swales (1990) emerged as one of the most frequently cited works in this corpus. Genre Analysis considers genre to be a dynamic but stable, socio-rhetorically constructed text-type. Genres can be usefully conceived of as emergent from the evolving conventions and domain knowledge of a discourse community (Ramanathan and Kaplan, 2000).

Texts which are taken as exemplars of a given genre can for this reason not be distinguished only in terms of a set of necessary and sufficient conditions as required by classical set theory and the unmodified logical-positivist theory of categories (Lakoff 1987; Swales 1990). This stems not merely from the logical impossibility of constructing any formal set of categories which is both "consistent" and "complete" (per Gödel's theorem, see Hofstader, 1992). Rather it is a function of genre as a socially constructed discourse type; as such it is subject to judgements of acceptability based on the internalized norms of members of a discourse community. Genre should not, therefore, be interpreted only as a "kind of text" without reference to certain conventional forms and purposes targeting a particular audience in a particular setting.

Swales (1990) presents a comprehensive review of the term as it has been deployed in folklore, literary studies, linguistics, anthropology and other disciplines. His is a careful delineation of genre, as well as the necessarily linked terms of discourse community and task By examining his proposed criteria for each term in a critical light, my aim is to foreshadow how these terms were refined by subsequent researchers.

"Discourse Community"

I will now enumerate each of Swales' (1990: p.28 ff.) criteria for discourse community followed by my own critique. A subsequent and similar treatment of chapters on genre and task will complete this section.

While the term "discourse community" has been defined in many ways (see Harris, 1989; Swales, 1993, for a critique see Faigley), in the context of academic writing research, a comprehensive set of criteria provided by Swales (1990, p.28) provides a good starting point for further inquiry. First, it is necessary to recognize the distinction between a "speech community" and a "discourse community" as these are intended to be a contrast set in this framework. As Swales puts it, all normal-functioning people become members of at least one speech community through a developmental process of socialization. Discourse communities can only be joined on purpose; they depend on special training and experience with appropriate genres as part of a professional or other intentional association.

1. A discourse community has a broadly agreed-upon set of common public goals.

One of the problems with this criterion from a criticalist point of view is that L2 writers may not perceive the "common public goals" the same way that people from the dominant culture do. Academic cultures, for example, differ in how they understand plagiarism (Duszak, 1998; Li, 1996) and thus the goal of clearly delineating the author's voice from the recitation of canonical sources is not "agreed upon" a priori by all aspiring members of a community. The mismatch between the L2 writer's previous literacy socialization and what is expected in the target discourse community can even create career-threatening challenges to an academic career when, for example, a student naively cites a source without proper attribution (Li, 1996; Bloch and Chi, 1995). One especially salient source of difficulty is that the "common public goals" are so obvious to expert members of the community that they are largely unconscious and unaddressed in classes and in the published literature itself. Culturally different others, and, indeed, all students, can benefit from explicit and continuous discussion of the goals of the community and the corresponding expectations of community members vis-à-vis citation practices, rhetorical form, scholarly voice, etc. (Delpit, 1995; Gee, 1990; Kern, 2000).

(2) A discourse community has mechanisms for intercommunication among its members.

The question of how access to these channels of communication is allocated occurs to the critically minded researcher. The point to emphasize here is that people from different disciplines and cultures may possess quite distinct conventions, purposes and positions with respect to intended audiences. People will necessarily interpret the "same" genre, say what constitutes a valid research article, very differently depending on where they sit.

In this connection, let's consider the peer review process whereby scholarly articles are subject to scrutiny prior to publication. Even within the relatively homogenous cultural and disciplinary context of an editorial review board, the standards applied to determine whether a given submission is appropriate for publication are idiosyncratic, often contentious and, always, at least to some extent, ideologically loaded. Reviewers who agree with an author's overall theoretical orientation will tend to be more forgiving on questions of form and those with differing orientations, less forgiving. The often capricious nature of this "gate-keeping function" means that the submissions of L2 scholars are especially vulnerable for exclusion from English-language journals for rhetorical or linguistic lapses that are ancillary to their content (Duszak, 1998; Cecceto, 1998).

Regardless of whether the writing shows sufficient domain knowledge and reports important findings to the research community, as long as its form does not match the internalized and largely unconscious rhetorical conventions expected by the review board, the submission will simply never be considered "appropriate" and, thus, will never be published (Geisler, 1994).

(3) A discourse community uses its participatory mechanisms primarily to provide information and feedback .

"Thus, membership implies uptake of the informational opportunities. You are not really a member of the discourse community if you subscribe to a journal but never read it (Swales, 1990: p.26). In the context of the questions underlying this review, this statement brings to mind how L2 academic writers' lack of linguistic skill or genre knowledge can prevent their full participation in the exchange of information both as readers and writers. This exchange of information should not be envisaged in a social vacuum, however. The rhetoric and content of the professional literature, in addition to delivering the substance of its messages in an efficient format, is itself a resource for the socialization of new community members. The feedback function encompasses, for example, professors' marginal comments on students' written assignments and peer review comments on article submissions. Feedback lets community members know how well and in what ways their writings approximate what is expected of community members (Anson, 2000; Straub & Lunsford; Prior, 1998).

(4) A discourse community utilizes and hence possesses one or more genres in the communicative furtherance of its aims.

(5) In addition to owning genres, a discourse community has acquired some specific lexis.

One of the hallmarks of a disciplinary discourse community is the existence of a shared specialized vocabulary. For example, acronyms like EAP, ESP, ESL, SLA and ELL, can mark the user as a member of a certain discourse community within education and applied linguistics, Such vocabulary serves a referential function in the sense of telegraphing the technical concepts of the field. It also serves an indexical function, in the sense that understanding or using it appropriately (or inappropriately) marks social position vis-à-vis the discourse community. No one can deny that there is always a social function present in any human interaction. To naively construe discourse communities merely as places where information and feedback are exchanged misses that every discourse community is, in the first instance, a social formation for the "presentation of self in everyday life" (Goffman, 1968).

Technical vocabulary can prove embarrassing and difficult for L2 speakers and writers who want to be part of the "in" group. Besides being difficult to pronounce, the Greek and Latin roots often employed in such vocabulary are especially opaque for those with limited exposure to academic English. The sheer volume of new terms one is required to master in some fields can present a formidable challenge for any student, much less one whose native language is not English. A limited vocabulary can make the acquisition and demonstration of new knowledge in a field more difficult for the non-native on a variety of levels, from understanding oral language in class (Flowerdew, 1996) to reading (Fortes, 2001) and writing (Silva and Matsuda, 2001).

(6) A discourse community has a threshold number of members with suitable degrees of relevant content and discoursal expertise.

Since the point of this formulation is to provide a definition of "discourse community" that is appropriately constrained, this statement can be summarized as a requirement for a "critical mass" of fully socialized members. From a critical perspective, the highly interpretive nature of words like "suitable", "relevant" and "expertise" leaves them open to abuse by an élite who will tend to use them to enforce and normalize any existing hierarchy. Experts will generally manage the discourse community at least as much for the benefit of themselves as for the pursuit of some neutral "truth" (Geisler, 1994). A discourse community is typically "hegemonic" in the sense that its members tend to embody an internalized sense of the "rightness" of the existing power structure. Every discourse community is also, to some extent a "guild." (Geisler, 1994; Duszak, 1998) or a "community of practice" (Lave & Wegner, 1993). To this extent at least, the discourse of expert members is profitably viewed as the production of professionals who are attempting to create personal social capital. Publishing in journals, for example, fulfills the official purpose of disseminating knowledge. On a social level it allows authors and editors to build a career while making friends and influencing people. Everyone who belongs to a discourse community is more than a neutral agent of truth (Duszak, 1997). Indeed the whole idea of a harmonmious, homogenoeus "community" as Swales envisions is impossibly "utopian" (Faigley, 1992). The texts exchanged in any community will inevitably embody multiple purposes and goals beyond those explicitly mentioned in the editorial statements of peer-reviewed journals (Prior, 1995, 1998; Geisler, 1994; Samraj, 2002a, 2002b).

To provide a counter-weight to the "hegemonic" distortion of discourse, a more encompassing definition of "discourse community" may be in order. In addition to "experts" at the "center", a vital "discourse community" must also have a threshold number of novices, outsiders and other "members", who, for whatever reason are positioned at the "periphery", or at the "borderlands" (Anzaldúa, 1996; Faigley, 1992; Sullivan, 1995). My point is that there is no "inside" without an "outside"; the "experts" need "non-experts" for their position to make any sense. The more elaborated the discourse community, the more nuanced and articulated become the various positions available to aspiring "members."

Let me explicitly challenge the implications of the original statement about threshold membership: The vitality of a discourse community should not be measured solely in terms of how many at the "center" fully embody its internalized norms at a given moment in time. A discourse community needs its share of apprentices, wannabes, rivals, detractors and upstarts if it is also to have successors. Those who dwell at the borderlands are an important part of the story of a discourse community; they are not merely "deficient" or "incomplete" versions of the experts at the center. (Candela, 1998; Gillmore, 1998) They are not automatically destined to replicate the practices of their intellectual elders. Indeed, as the natural agents of change, peripheral or even resistant members of the discourse community will largely author the future chapters of the story.

Swales' (1990) formulation of "discourse community" as summarized in these six points is useful for distinguishing the kinds of social formations definitive of academic life from the innumerable possible constellations of socio-literate practice. The socio-culturally oriented researcher can find much of value here, for example, by being alert to the specialized vocabulary and usage which are hallmarks of particular discourse communities. By highlighting the "participatory mechanisms" by which the community conducts its business, Genre Analysis reminds us to look at the gate-keeping functions if we hope to understand, for example, how discourse is managed and regulated.

In critical anthropological terms, the Genre Analysis (Swales, 1990) view of discourse community (in so far as it occludes a priori the "agency" of peripheral members) is "hegemonic". Applied unreflectively, such exclusion of the periphery as a site of positive action will tend to normalize the status quo with its extant inequities. Critical research in and about academia will optimally inform a program of social change only if that research begins with a view of "discourse community" which is ample enough to encompass, not only the starring players but the full humanity of all the actors on the academic stage.

Swales' "Working Definition of Genre"

A genre comprises a class of communicative events the members of which share some set of communicative purposes. These purposes are recognized by the expert members of the parent discourse community, and thereby constitute the rationale for the genre. This rationale shapes the schematic structure of the discourse and influences and constrains choice of content and style. Communicative purpose is both a privileged criterion and one that operates to keep the scope of a genre as here conceived narrowly focused on comparable rhetorical action. In addition to purpose, exemplars of a genre exhibit various patterns of similarity in terms of structure, style, content and intended audience. If all high probability expectations are realized, the exemplar will be viewed as prototypical by the parent discourse community. The genre names inherited end produced by discourse communities and imported by others constitute a valuable ethnographic communication, but typically need further validation. (Swales, 1990; p.58).

For the purposes of the present review, this definition has much to offer. It encapsulates a principled yet open-ended concept with theoretical potency in the context of a socio-culturally oriented study of academic writing. Especially important is the articulation of the specific ways that genre is more than a cluster of textual features " in the world" as it had generally been conceived of in much prior work (see Swales, 1990: p.131-132 for a comprehensive list of such studies of the research article). Genre conceived here looks less like a class of objects and more like a nexus of socio-literate activity where the conventionalized purposes of a discourse community find expression in "appropriate" rhetorical and linguistic designs (for "prototypical" instances, at least).

There are also entailments which are not so helpful in the present context, particularly when one evaluates this formulation from a critical theoretical or feminist (Tisdell, 1993) viewpoint which questions the right of "experts" to, in some sense "own" the genre and to ultimately determine what the valid rationales are (Lakoff, 1984). Additionally, the socially contested nature of genres is minimized in the Genre Analysis version inasmuch as there is an inherent faith that the discourse community as a whole could actually agree on the protypicality of any given set of exemplars. There is a difference between "intersubjective" and "objective" ideas of truth which we must be careful to distinguish if we are to fully tap the value of this discussion. Intersubjective reality is located in social relationships and is therefore dynamic, subject to constant change and re-interpretation. It is a site of constant, dialogic negotiation between and among subjects (viz. Bakhtin's, 1981, "heteroglossia") rather than a necessary convergence of opinions about the "facts" of the situation.

The "native" or "emic" nomenclature of a discourse community for its genres constitutes a beginning point for empirical investigation of the target forms for academic writing instruction in the community. It also can serve to identify classes of texts that are likely to be problematic, both to understand and to produce, for academics from diverse backgrounds (and other outsiders). Scholars for whom English is a non-native language are often unable to ever fully master the genres of academic discourse as these are understood by the experts (Farell, 1997). They are especially vulnerable to exclusion from the "emic" understandings which underlie the common, largely implicit norms of the mainstream academic community (Kaplan, 2001).

Additionally, when we background the intersubjective nature of genre and imagine, as Genre Analysis appears to do here, that there is an a priori prototype for each genre shared by expert members of the community, we risk over-simplification (for example, equating a preponderance of "expert" opinion with unanimity). Similarly, we can miss the (often vast) disparity between what experts actually do and what experts think they do. We also normalize "what is," i.e., the existing state of affairs in academia, and foreclose "hope" (Freire, 1994) for "what ought to be" or "what might be." (Geisler, 1994).

In order to maximize the critical potential of research on academic writing, i.e., to facilitate maximum inclusion of traditional "others" in non- subordinate roles, researchers must acknowledge that yesterday's heresy is today's dogma, yesterday's transgression is today's canon. From a criticalist point of view, even peripheral members of the community like "novices" and "L2 writers" play an active role not just in reproducing but in (de-)constructing, expanding, and re-inventing the genres of academic discourse over time (Prior, 1998; Gillmore, 1999 ;Candela, 1999). Genre is easily reified, normalized, frozen in time as a concept. Perhaps this is because it is generally considered a noun which, grammatically speaking, names a "thing" in the world (Whorf, 1956). Ultimately, genre may be more effectively conceived of as a verb for the purposes of research. If it were a verb, genre would be a verb which the critical researcher would freely use with subjects beyond "expert members of the parent discourse community" and inflect in more than the present indicative tense.


All professors are also literacy instructors inasmuch as they design and execute activities which both develop and assess students' socialization into disciplinary literacy. Students must interpret what is required, often based on mere clues as to what the professor actually expects. The third and final key concept to be treated by Genre Analysis (Swales, 1990) is that of the language-learning tasks which are put forward as the "methodology" by which the instructional designer (as informed by genre theory) gives students access to developmentally appropriate writing experiences. (See fig. 1, from Swales: p.69). Moreover, as a recent major cross-disciplinary review concludes:

It has been shown that the category "task" as used by researchers generally, is widely applicable and has psychological reality. Much, if not most, of human activity, whether in employment or in the classroom can be seen as a series of tasks-some having a communicative aspect, others not. (Crookes, 1986: p.32, cited in Swales: p.73).

Notable in this connection is the failure of research to show convincingly that one language teaching method is better than another (Swaffar, Arens and Morgan, 1982, cited in Swales, 1990: p.73). The apparent reason, which has only become clear through ethnographic classroom studies (e.g., Cazden, 1988; Candela, 1999) is that there are so many shared classroom practices (for example, the Initiation-Response-Evaluation discourse routine) that cut across methods. Task has proven to be a more appropriate unit of analysis for research on educational practice than method. As will be demonstrated below, the Genre Analysis conception of task can be refined to be more explicitly "transactional" and "situated" and to keep research engaged with the multiple points of view at work in any given setting.

After a review of some literature on task and task-based learning (also called TBL in case you were hankering for another acronym). Swales settles on Candlin's (1987) definition:

(Task is) One of a set of differentiated, sequenceable problem-solving activities involving learners and teachers in some joint selection from a range of varied cognitive and communicative procedures applied to existing and new knowledge in the collective exploration and pursuance of foreseen or emergent goals within a social milieu (p.23).

To bring in a criticalist slant, task construal is thus fundamentally social, and as such is subject to multiple interpretations. For example, students and evaluators often have very different ideas of the tasks involved in responding to a given writing prompt. L2 writers will be especially subject to these kinds of misunderstandings, often with unfortunate consequences for their academic careers (Bruce & Schneider and Fujishima, 1995; Li, 1996).

By way of responding to this formulation of task It will be helpful to refine the idea in terms that will help us keep track of the different "moments" of interpretation and production in academic writing "transactions" and therefore where there are potential misunderstandings between students and teachers. Researchers, at least those who are responding to the perspectives of the process-writing movement (Flower, 1981, see Kent, 1998 for a critique), share a common understanding of the stages that students go through in preparing written assignments and the (potential) role of teacher input at each stage, with an emphasis on "formative" rather than "summative" feedback" to facilitate students' progress through multiple drafts. Researchers differ in the parts of the composing and responding process which they foreground and about which they gather data, owing to their distinct interests, theoretical and methodological orientations. ,

For the purpose at hand, a useful analogy can be made between tests and writing tasks. (Spolsky, 1994) has shown how reading tests involve multiple layers of interpretation and introduces multiple reader/writer roles. Writing tasks of all types in academic settings are similarly constrained and, owing to the requirement for interpretation by differently positioned teachers and students, subject to misunderstanding at each of the stages in the process (Sperling, & Freedman, 1987). By substituting "writing assignment" for "reading test" in Spolsky's (1994) original figure (as adapted in Kern, 2000, p.269), we can delineate the different "readers" and "writers" potentially relevant to the idea of task being developed here:

Stages of Production and Interpretation in Response (SPIR)

1. Writer 1       Reader 1
(author)       (student)
produces =>
<= interprets
original source-
reading passage
2. Writer 2        
(teacher) =>
<= Reader 1
(writing assignment)
  (student )
3. Writer 3 =>
<= Reader 2
(written assignment)
produces       interprets
4. Writer 4 =>
<= Reader 3*
(teacher response)
produces       interprets
5. Writer 5 =>
<= Reader 4**
(Second or nth draft)
produces       interprets
*= in single or final draft situations, the process stops here
**=this process can be iterated indefinitely until a final draft is accepted

Anyone who is researching academic writing instruction must necessarily have in focus some sub-set of the steps in this process with corresponding data. With some variations, this sequence of productive and interpretive "moments" also can account for autobiographical or expressive writing (substitute "personal experience and reflection" for TEXT 1), and situations where, for example, the creator of the assignment--say a professor-- and the evaluator--say a teaching assistant--are not the same person. It can also help us to keep in mind the default assumptions of research designs in which, for example, teachers are asked to respond to student work "as if" it were the work of their own students or students are asked to interpret commentary "as if" it were from their teachers.

This paper does not deal with the problem of how students actually interpret writing tasks (see Bruce & Murphy, 1988 for a book length consideration of the problematic genre called "writing prompt"). This paper is also not about the first step in the process, i.e. how readers interpret texts per se, and as such it will not attempt to summarize the vast research in reading and literary theory (See Kern, 2000 for a summary). I would like to retain for the current discussion, however, a key term from the reading and literary theoretical literature: the "transaction" (Rosenblatt, 1989); this term has been used by others attempting to frame inquiry into teacher response (Probst, 1989); "Transactional theory respects this fundamental fact: meaning resides in the person rather than in the dictionary" (Probst, 1995, p. 68). More precisely, the meaning emerges in the transaction between reader and text which allows interpretation in the first place, for "a novel or play remains merely inkspots on paper until a reader transforms them into a set of meaningful symbols." (Rosenblatt, 1989, p. 25). Instructors' feedback on student work similarly, is mere inkspots until the student reads them.

When the assignment of tasks, instructor feedback and, indeed, every instance of "interpretation" in Figure 3 is treated as a set of "transactions," used in Rosenblatt's (1989) sense, as a term to designate "relationships in which each element conditions and is conditioned by the other in a mutually constituted situation," then we can recognize a possible refinement to Candlin's (1974) definition of task quoted earlier. While Candlin includes "joint selection" as one of the criteria for identifying a task, he does not highlight the ways in which those individuals who are "jointly" selecting a task will necessarily diverge to some small or large extent in how they interpret it. The SPIR model reminds us that tasks must be separately construed by professors and students. This is not a trivial issue, especially in the context of L2 writers who may miss culturally nuanced contextual cues that are transparent to others.

Descriptions of academic writing tasks which do not specify the significant stages of production and interpretation for the response writing task per the SPIR figure above, require the researcher and/or the critical reader to "fill in" the missing pieces, based on some kind of default "schema" for those stages which are not under conscious consideration. Without such "filling in," it would not be possible for respondents to complete the task of responding, nor could the critical reader make connections from the research findings to the real world of educational practice. If, for example, a researcher in academic writing does not include the student's point of view in how a task was interpreted, the research record is incomplete and insufficiently thick to fully encompass the complexity of the transaction.

This transactional model of task also brings to mind another potentially problematic point: the articulation of tasks throughout a student's graduate studies. Ideally, a graduate curriculum should prepare students for dissertation writing through an increasingly rigorous sequence of courses and qualifying examinations involving substantive feedback and opportunity for revision in each transaction and at each stage of study. To the extent that writing asks are not explicitly articulated or developmentally appropriate and extensive feedback is not provided, the curriculum is not optimally supporting students' progress (Prior, 1995, 1998; Cooley and Lewkowitz, 1998). In such a context, students succeed despite, rather than because of, the tasks defined by their professors and institutions. To re-iterate with a different spin a point made earlier: task is method, i.e., the tasks students are asked to do (and the way they understand those tasks) is a more defensible and generative beginning point for situated and ethnographic studies of instruction than the formal "methods" supposedly in use in the setting.

Genre Analysis (Swales, 1990), despite problems in its formulations from a criticalist point of view, articulates a set of clearly defined criteria for discourse community, genre and task and a pluralistic methodology which make it a canonical point of reference in a very diverse area of research. The next section shows how research in an ethnographic mode of the real-life contexts and conditions in which academic writing occurs can complement purely textual methods, significantly extending the explanatory power of genre studies.

VI. English for Academic Purposes: a Socio-Cultural Approach

A salient point of agreement in recent socio-cultural treatments is the following: the genres of academic discourse must be understood as socio-rhetorical, historically situated constructions if they are to be appropriately de-mystified, both as objects of research and as standards for teaching or practice (Kern 2000; Swales, 1990; Geisler 1994; Prior, 1995, 1998; Olson, 1994). For example, Swales (1990) traces the evolution of the modern research report from its origins in the written accounts of experiments provided to members of the Royal Society in the Eighteenth Century who were unable to attend the live demonstrations performed during the Society's meetings. The canonical forms of the scientific report genre appear natural and atemporal are in fact the residue of particular social and historical processes Without such background knowledge and a critical stance, the ever-present hazard in EAP is that genres, once "discovered" or reified, enter the canon as nothing more than decontextualized formulas for generations of students to replicate ad nauseam, e.g., the much maligned yet immortal five-paragraph essay.

Too often the prescriptions that underlie a good deal of writing instruction are not actually based in empirical research on what proficient writers actually do. Popken's (1987) example of the topic sentence is emblematic of this point. It was found in this study that professional writers' practices in writing topic sentences differed significantly from what students were told to do in writing classes. In fact, the functions of a topic sentence were distributed across the paragraph or were simply not present in the professional writers' texts. There is a disconnect, evident in the research on topic sentences, between prescriptions of the classroom and the actual literacy practices of the non-school world. This disconnect is emblematic of the general mystification of academic language forms for students. By "mystification" I refer to the way that genres are often presented as givens, as grounded in "authoritative" rather than "internally persuasive" discourse (Bakhtin, 1981; Halasek, 1999). Without an historical and critical context, students do not automatically make the connection between the language of schooling and the language of the non-school world. There is a tendency to rely on academic genres as immutable expressions of good practice. Socio-culturally oriented scholars think that students should not be taught to emulate genres as though they were immutable and sacred.. Students need to approach genres as social structures that emerge form a particular set of embedded power relations as they cultivate "critical language awareness" (Fairclough, 1989). It is necessary to understand these forms as socially and historically constructed "available designs" (Kern, 2000; New London Group, 1999) if they are to master the norms of academic language without minimizing their capacity to challenge and re-invent those norms in their own terms.

EAP, inasmuch as it touches on written genres, has focused largely on "needs analysis" for instruction in disciplinary literacy and therefore concerns itself with formal characterizations of target texts written by experts in a given field (Swales, 1990). Comparative studies include those which focus on variation of a genre across disciplines (Samraj, 2002a). Studies have also been conducted which focus on the ways in which student writing in graduate seminars approximates the expectations of the target disciplinary community (Samraj, 2002c). EAP is traditionally associated with applied linguistics in terms of its theoretical orientation, goals and methods, therefore there is a tendency to pay more attention to the grammatical and lexical levels as opposed to the rhetorical level (Samraj, 1998; see Swales, 1990: p. 131-132, for a list of more textually oriented studies).

Such work aims for more than merely descriptive adequacy in its accounts of the textual features attested in exemplars of a given genre. It aims to explain those textual features in terms of how they express both the conventionalized purposes of the discourse community and the author's purposes with respect to the particular audience being addressed. To this extent, the text becomes more than an amalgam of autonomous textual features as envisaged in many studies; text is re-imagined as a nexus of socio-literate activity. Concepts of genre are grounded, in this more contemporary view, in a rigorous attempt to observe variations among textual features based on the associated purposes and contexts which they realize and respond to. Only ethnographic observation of the actual social conditions in which the text is produced and attention to the lived experiences of readers and writers can provide a record that is "thick", i.e., full of relevant contextual details gathered over time and triangulated from multiple points of view. A research program in academic writing needs a sufficiently thick description of context to legitimately claim to have "explained" the rationale behind any given aspect of a text or its organization (Watson-Gegeo, 1992). It is this kind of "explanatory adequacy" to which many contemporary researchers aspire.

Samraj (2002a) attempts such explanation by harnessing a multi-layered model of context.ample enough to account for the multiple and complex factors that condition students' responses to content area writing tasks. In so doing she helps us in our task in this section by putting Swales' (1990) discussion of "discourse community" into perspective. The definition of discourse community as advanced thus far in this paper is so highly constrained that it only encompasses one dimension of possible influence on the writer; it corresponds only to the "discipline" level of context. As such its influence is usually filtered through other levels and can be expected to account only for some aspects of the final text. The task looms large if student writing is the text in question. It is not the "discourse community" or the "discipline" in general which assigns the task which students must respond to in assignments. Science won't be giving anyone a passing grade; Professor Smith, (with some luck!) will.

The institutional context, largely in the person of the professor(s) in charge and the source texts assigned, is also a significant factor in how students write. Samraj's (2002c) study explores the types of student writing produced by students in a master's program in environmental science in a School of Natural Resources and the Environment at a mid-Western University in the U.S. Environmental science was chosen because it is an interdisciplinary field, a complex context and "growing in importance world-wide" (p. 4). She selected writing in three courses that captured the broad spectrum of courses and sub-disciplines included in the school. She based her methods on elements of naturalistic investigations of writing conducted by composition specialists like Walvoord and McCarthy (1990). "The researcher attended the courses for the semester in which the data were collected so that she could become familiar with subject matter and the types of meaning making valued in each class" (Samraj, 2002c: p.4). The study documented oral class discussions of the writing tasks and hence could provide a much richer understanding of the context of the writing than would otherwise have been possible.

Owing to the relatively thick record which she has assembled, Samraj (2002c) can provide convincing connections between aspects of context and previously studied textual features like tense, passive voice, hedges, impersonal constructions and first person reference in the abstracts of students' papers. In fact, Samraj's (1995, 2002a, 2002b, 2002c) work generally is exemplary of how rhetorical and linguistic analyses can be combined with ethnography to go beyond mere "description" toward actual "explanation" in research on academic research articles.

The "advocacy for urgent measures" for example, which seemed to be an obligatory "constituent move" in conservation wildlife abstracts, was missing in the wildlife behavior abstracts which were gathered from the comparison classes (Samraj, 1995). Samraj presents a convincing case that the absence of this move in the wildlife behavior domain reflects the fact that it is socially constructed as a "neutral" biological science Samraj (2002a, 2002b). Moreover, the obligatory "call to action" makes sense in the wildlife conservation abstracts since advocacy for social and ecological change is constitutive of the field. In fact, the field was specifically conceived as a response to the dire state of the world's wildlife populations.

Betty Samraj, a student of John Swales who also served as her dissertation chair at the University of Michigan, has continued in the tradition of her mentor and extended and refined his initial methodology significantly in her investigation of the variations in a genre in closely allied but distinct disciplines. She has also investigated students' attempts at approximating disciplinary genres through ethnographic studies of writing tasks in graduate seminars. To this extent she has demonstrated in concrete terms the necessity for a critical stance with respect to these categories, to avoid "essentializing" them into a homogenous, unproblematic set of features prior to empirical investigation.

To the extent that we essentialize a category like "abstract" and assume a priori that it represents a concept similar for all disciplines and for people in all times and places, we have limited our empirical investigation of how the term is differently interpreted by differently positioned subjects; we risk relegating genre to a category of objects "in the world" which is inherently assumed to exist independently of the social context in which it is employed. Such essentialization precludes both a "grounded" analysis ( Glaser & Strauss, 1967) which aims to describe qualitative data on its own terms and the phenomenological attitude of the anthropologist which gives primacy to the "native's" actual point of view rather than to the theorist's a priori expectations.

Samraj's work demonstrates that disciplinary writing in appropriate genres is always a matter of responding to aspects of context as these are "lived" or experienced by the author, regardless of whether they emerge from "macro-", i.e. institutional and structural or "micro-", i.e., situational or task variables. Her work also demonstrates how students learn to replicate aspects of genre without explicit instruction. For example, they learn to include an "advocacy for urgent action" in their abstracts and submitted papers for Wildlife Conservation without ever being told explicitly that this was expected (Samraj, 1995). The socialization of students into disciplinary genres through "cognitive apprenticeship" (Lave and Wenger, 1993) rather than explicit instruction is evident here.

Relevant to the context of this paper, there is a vast cultural and linguistic repertoire that the native apprentice can draw upon in elaborating academic genres which may not be accessible to the non-native apprentice. Samraj's studies can alert us to the general lack of explicit attention to disciplinary genres in content area instruction. This lack of attention to genre has important implications for L2 writers which deserve further research. L2 graduate writers, particularly, are largely thrust into disciplinary environments with minimal preparation for the task of figuring out what mainstream others seem to grasp "naturally" (remember Kaplan's questions in the Contrastive Rhetoric sub-section). These advanced L2 writers deserve to be included in many more ethnographic studies of disciplinary writing. From such studies (c.f., Prior, 1995, 1998; Schenider and Fujishima, 1995) we can discover the "lived" nature of their challenges as well as learn from, and celebrate, their creativity and persistence in so often succeeding, even against all odds.

This last mentioned line of research (Prior, 1995, 1998) in a socio-cultural or "socio-historic" mode further exemplifies the turn toward ever "thicker" accounts of the processes involved in academic writing. This research on writing and response in graduate seminars seriously challenges the unity of the disciplinary or discourse community (as presupposed in the textually oriented studies reviewed in section I). Even the common sense notion that students "enter" a disciplinary community is negated. These longitudinal, ethnographic studies reveal, not that students unproblematically join a homogenized disciplinary space, rather, they learn, when successful, to "participate in communities of practice" (Prior, 1998: p. xi, citing Lave and Wegner, 1993). Task and genre are similarly problematized in Prior's work:

Even in advanced seminars, academic writing tasks (typically glossed as "the assignment") appear not as a single shared task with definite criteria of assessment, but as a collage of task representations, in which hybrid actions and understandings weave together personal, artifactual, institutional, as well as disciplinary histories. In this view, graduate students are not entering the autonomous social and cognitive spaces of discourse communities, but engaging in active relations with dynamic, open interpenetrated communities of practice (to use Lave & Wenger's, 1993 terms) (p. xii).

"Prior perceives the apparent coherence of a completed text (taken as a given within textually oriented studies) as a mangle of practice and the lamination of experience" - terms borrowed from Andrew Pickering and Erving Goffman" (Bazerman, 1998: p. vii).

This formulation aligns with Swales' (1990) in its construal of "genre (as) an open family of utterances (written and spoken) with fuzzy and contested boundaries" (ibid). It follows that "the joint work of alignment, that is of making or avoiding resemblances, must be central to the process of genrification" (p. 70). Prior differs from the Genre Analysis approach in understanding the process of alignment. Genre Analysis argues that "family resemblance" is too unconstrained (anything can resemble anything), and grounds its categorization in a cognitive idea of a prototype with "communicative purpose" as the central axis of each distinct genre. Swales (1998) has attempted to update his notion of purpose to keep it intact as the criterion for genre membership in light of these kinds of critiques

Prior's (1998) "socio-historic" approach questions the idea that it is only the conscious intentions expressed in texts that define their generic status. It emerged in an attempt to account for the range of responses to a writing task for two American Studies seminars and an Applied Linguistics seminar (after attending every class, interviewing the professor and students and collecting all the student papers with the professor's final grades and comments). As a key term in "sociohistoric" research, Prior (1998) proposes the "semiotic genre". Borrowing from Bakhtin's notion of the "speech genre", the "semiotic genre", like its root term "semiosis" emphasizes that contextual and unconsciously assimilated messages about a text can influence its interpretation as much as the intended content of the text itself; "semiosis does not imply intentionality, whereas communication normally does" (p. 71).

A clever hypothetical scenario ( pp. 71-72) illustrates this point. He posits that he is a professor of constitutional law and that he has assigned his students "the task of drafting a national constitution for a country of their choice as a final project (worth 15% of the grade)." One of the students turns out to be from "a small Pacific Island on the verge of independence, and she has been designated to draft national constitution and present it to the island's constitutional convention." He imagines finding her draft constitution in his departmental mailbox box with 30 others. He points put that in this case he would contextualize it as an "assignment" rather than as a "draft constitution" and would evaluate it only in terms of variations in the quality of the text or its conformity to the task. If he were aware that the "assignment" was intended to circulate in a sphere of world and national politics, he would presumably respond with quite a different reading. The evaluation of the constitution as just another assignment in this case can not be construed as a "feature of its internal communicative design" (p. 71). The contextual and incidental dimensions of semiosis, e.g., the presence of the document in a pile of other student papers, are not treatable within the textually oriented and Genre Analysis approaches given a common privileging of conscious purpose. Nevertheless, these seemingly incidental aspects of context can prove definitive in a given participant's actual understanding of the task at hand. As Prior's (1998) findings demonstrate, a "semiotic" as opposed to a communicative approach to genre can explain both the deceptive unity of academic genres and student's multiple and conflicting responses to the "same" assigned task.

For example, in the case of the applied linguistics seminar, the professor augmented the skeletal description of the final paper which appeared in the syllabus with numerous spontaneous and conflicting comments during class about what he expected. On several occasions he suggested that students should choose something they are really interested in as a step toward a qualifying paper or dissertation proposal, on others he indicated that they should choose something "doable" and "not too ambitious" (p.73). He ended up evaluating the papers as though they were proposals for dissertation projects submitted by his own advisees, (which they were not) subjecting them to a rigorous scrutiny which disadvantaged the more speculative or imaginative studies. The fact that some students were responding to his invitation to take a stab at a difficult as opposed to immediately doable project was not problematic for the professor and he evaluated and graded each submission by the same rubric. This resulted in Bs for those who took the "difficult" path and As for the students who chose the "doable" path. A careful comparison of the papers revealed that the different kinds of responses were normalized into the single genre of "student research proposal" through a process of semiosis. Post-course interviews with students and professors confirmed multiple and dysjunct interpretations of the academic writing task and also showed that none of the participants were consciously troubled by the discontinuity which was apparent to the researcher. Students and professor alike had assembled internally coherent, but mutually inconsistent, understandings of the writing task and its place within the seminar based on their selective attention to elements in a set of conflicting or "heteroglossic" (Bakhtin, 1981; Halasek, 1999) messages. The context of the seminar and the default expectations of the professor provided a sense of unity that was lacking in the framing of the task itself.

The socio-cultural approach has only begun to penetrate the field of EAP and academic writing research generally. With some precedence in a handful of earlier investigations of writing tasks situated in particular university classes (Herrington, 1985; McCarthy, 1987), Prior's work, begun in 1989, and Samraj's begun in the early 1990s represent an advance in theoretical and methodological sophistication and in the completeness of the ethnographic record from which they draw (Vaj Rmanathan, personal communication). They go along way toward demonstrating how ethnographic methods can be combined with textual analyses to create a more satisfying, and explanatory account of the real-life complexities of learning and teaching academic genres. They also highlight the institutional and social character of literate activity in the academy. Socio-cultural approaches necessarily de-center pre-existing categories and bracket common sense unities in genre, discourse community and task.

In this way, a socio-culturalist program tends to interrupt the hegemony of existing common sense and power structures, rendering research in academic writing permeable to a "critical turn" which is evident in the Human Sciences generally (Fairclougth, 1994, 1999; Macedo, 1993; Dreyfus & Rabinow, 1983; Carspecken, 1996; McClaren, 1992). This turn toward a criticalist stance opens a space for inquiry where issues of authority and the ethical and political dimensions of EAP practice and knowledge construction can be explicitly foregrounded. Significantly, it seems that the more time academic writing researchers spend in actual educational institutions observing "socio-literate activity" (Prior, 1998:p.3) rather than merely analyzing "complete" texts, the more they become concerned with the inequity of existing practices and the rights of students.

VII Critical English for Academic Purposes

Following Freire (1974a, 1974b, 1994), "Criticalist EAP" (Benesch, 2001a, 2001b) encompasses "Rights analysis" which examines the entitlements of students as full members of the academic community. "Rights analysis" is a strategy of inquiry which complements and "problematizes" traditional "needs analysis" in EAP by focusing on the ethical dimension and actual consequences of this kind of research for real people. This is poignantly necessary because ESP began in the 1960s in the Middle East oil industry as part of an Anglo-American expansionist regime (Benesch, 2001b: p.25). The "hidden history" of ESP indicates that it was launched as a deliberate and vital step in the creation of a post-colonial economic hegemony (which inherently ignores the rights of non-native-English speakers as fully human subjects entitled to use their own cultures and languages for their own purposes). Nevertheless, in their thirty-year retrospective of ESP, Dudley-Evans and St. John (1998) portray the ascendance of English as a "natural" and unabashedly positive development (per Benesch, 2001). Research which focuses only on ESP-type "needs analysis" without concomitantly focusing on the rights of students will tend to replicate this post-colonial and self-justifying "regime" (Dreyfus & Rabinow, 1983). Teaching people academic English is treated in the absence of "rights analysis" as a technical rather than a human problem (Habermas).

While criticalist intentions ("making things better") have been voiced by virtually all of the researchers included above under the SFL, genre analysis and socio-cultural approaches, the criticalist project can only be achieved through a conscious "bracketing" or suspension of many of the "common sense" assumptions of traditional genre analysis. For example, criticalist researchers challenge the idea that the genres of academic discourse and their associated entitlements uniquely belong to the "experts" at the "center" or "pinnacle" of a homogenous and hierarchical disciplinary space. Within criticalist research, genre, discourse community, task-- and discipline itself-- are taken to be open-ended and indeterminate "sites of contestation" where the inherently "heteroglossic" (Prior, 1998; Bakhtin, 1981; Halasek, 1999; Anzaldua, 1987) nature of literate activity can be discerned in an unfolding social drama. For the criticalist, the "margin" has its own part to play in negotiating, (re)constructing and renovating academic genres as well as the institutions in which they are embedded.

As an example of how a criticalist "rights analysis" works in practice, Benesch (2001a) presents the case of an EAP seminar linked with a large lower-division anthropology class. The EAP class provided a forum for students to work on their content-related readings and written assignments and to jointly assemble comprehensive notes on lectures from the 450-student class. Under her guidance, the course becomes a "third space" (Anzaldúa, 1987; Candela, 1999) where students explored the intended meanings of the professor's lecture points or explanation of class assignments They went on to build a kind of "class solidarity" and a joint inquiry into the appropriateness of the anthropology professor's requirements. This meta-inquiry eventually empowered them to select a delegation, make an appointment and successfully re-negotiate an assignment with the professor. Another outcome of the "rights analysis" approach was that the anthropology professor began to re-think her curriculum in response to ongoing dialogue with the EAP instructor, making her expectations more explicit and more responsive to students' needs and interests. Such a collaboration in which the EAP instructor is not automatically subservient to the content area instructor is apparently an anomaly in the short history of such linked courses (Benesch, 2001: p.42) and represents another advance for criticalist EAP.

Criticalist EAP seeks to re-align educational praxis with a liberatory pedagogical program, to interrupt its connection with a genealogy of oppression and exclusion in a gesture of "hope" (Freire, 1994) for a more just and equitable world. Part of the program is to concretely name the power dynamics of the classroom and to support students' tactical and strategic assertion of their rights as full members of the academic community rather than as passive containers for officially sanctioned knowledge. Students are also encouraged within this approach to question their own relationship to the "genres of power" (Cope and Kalantzis, 1998) and the existing structures of the dominant culture. EAP easily becomes, in the absence of a criticalist sensibility, a process of reflexively socializing non-natives to prescribed English genres and their associated cultural norms. Students' valorization of their indigenous ways of knowing and being inevitably suffers, as do the vitality, authenticity and variety of academic writing. Criticalist EAP finds in the open-ended and negotiable nature of academic genres, tasks and discourse communities a possibility for re-writing not only the genres of academic writing but the relations of power in which those genres are inscribed. EAP in a post-colonial and criticalist mode focuses on ways that students--and professors--can negotiate agency in the face of institutional "limit situations" (Freire, 1974) to appropriate the genres of power for their own purposes and on their own terms.

VIII. Re-envisioning EAP for the New Millennium: Language Socialization in Advanced Disciplinary Literacy

So far this paper has explored the general trajectory of research in academic writing and EAP as this can be discerned in evolving and variegated understandings of language, genre, discourse community and task. Incorporating methods from linguistics, rhetoric, anthropology, literary and critical theory, researchers have, over time, construed the import of text in increasingly social, cultural and institutional terms. The "autonomous text" looks more and more like a nexus of socio-literate activity which is "heteroglossic" and indeterminate in its meaning except in a situated transaction with a particular human subject in a particular setting. Genre looks more and more like a verb and less and less like a noun as time goes on

Critically grounded (Carspecken, 1996; Fairclough, 1989; Chouliaraki & Fairclough, 1999) definitions of the key terms of "genre," "task," and community" are necessary starting points for research on academia which doesn't automatically replicate the inequities of the status quo. The definitions provided in a foundational text in genre studies (Swales, 1990) need some careful refinement to be appropriate for use in critical ethnographic studies of literacy socialization. The ESP/EAP (Dudley-Evans & St. John, 1998) and SFL approaches (Martin, 1993) help to clarify the target discourse expected of a competent member of a disciplinary community. The contrastive rhetoric approach (Connor, 1996; Duszak 1998, Panetta, 2001) can foreground the potential mismatch between students' prior language socialization and the largely unconscious norms of the target discourse community. .Samraj's (2002a, 2002b, 2002c) and Prior's (1998) combination of textual research and field observation contributes a model for the nuanced comparison of exemplars of a genre within a thickly described context. Criticalist studies within the most contemporary strands of contrastive rhetoric research demand a re-thinking of the status quo in educational practice and writing instruction by challenging existing power structures and norms (Panetta, 2001). An incipient tradition of case-study research of ESL and other graduate students is beginning to put the writing challenges they confront into the context of the lived complexities out of which those challenges emerge (Berkenkotter et al., 1988; Prior, 1998). Those who design and execute case studies of education in advanced disciplinary settings can benefit by using terms and concepts derived in genre theory. They can check some of the less helpful entailments of the key terms in genre theory by thinking of genre as a verb, of a discourse community as a community of practice rather than a place and of tasks as the most important "method" of instruction.

Where does EAP go from here? The "Language Socialization" ("LS") approach to second language acquisition will lend rigor to EAP's social and criticalist turns and, by providing a framework in which multiple points of view and multiple methods and data sources can be reconciled, will help to re-vitalize it as an academic discipline. LS articulates a set of "best practices" by which EAP can significantly advance its research program in an ethnographic and qualitative mode while subjecting to a radically empirical and criticalist inquiry the actual practices of the literate academy. It lays out a way of looking at human language based on the premise that "linguistic and cultural knowledge are constructed through each other," and that 'language-acquiring children and adults are active and selective agents in both processes" (Watson-Gegeo & Nielsen, i.p; Schieffelin & Ochs, 1986 cited in Watson-Gegeo, 2001).

The learning of language, cultural meanings, and social behavior is experienced by the learner as a single continuous (though not linear) process. Learners construct "a set of (linguistic and behavioral) practices that enable" them to communicate and live among others in a given cultural setting. A second premise is that all activities in which learners regularly interact are "not only by definition socially organized and emebedded in cultural meaning systems, but are inherently political" (Watson-Gegeo, 2001: p. 21).

This is because "all communicative contexts involve social, cultural and political dimensions that affect which linguistic forms are available or taught and how they are represented" (ibid). The restricted notion of context evident in the evolution of EAP is replaced in LS with a notion of context as "the whole set of relationships in which a person is situated.."

Language socialization began in studies of first language acquisition. Throughout the 80s and 90s researchers reached a consensus that there were important differences between the language practices of school and home (c.f., Bryce-Heath 1983, Ochs and Schiefflin, 1983, Boggs 1982; Cummins, 1981; Cazden, 1988; Watson-Gegeo, 1992; Watson-Gegeo and Nielsen, 2000; Watson-Gegeo, 2001; Cummins & Swain, 1986). More importantly for the task of this paper, there are significant differences between the literacy practices employed in higher-level disciplinary settings and the "general or universal literacy" which is the goal of the "general education" curriculum (Geisler, 1994,p. xi). These differences are often based on specialized vocabulary and rhetorically complex constructions (c.f., Lemke, 1990, 1995; Rose, 1997; Christie, 1993, 1997; Schleppegrell, 2000) that can prove especially problematic for people from diverse cultural and linguistic backgrounds. Language Socialization is still an as yet largely unrealized research program. Yet, owing to its inherently criticalist orientation, its emphasis on longitudinal research in naturalistic settings, its holistic conception of language and culture and its emphasis on multiple methods and points of view, it seems a natural next step in the trajectory of EAP research.. The fact that LS is consistent with the latest developments in neuroscience, cognitive anthropology and literacy studies (Watson-Gegeo, 2001) should provide additional incentive for the forward-thinking researcher interested in the teaching and learning of academic genres in real-life settings.

As the potential of the language socialization approach becomes manifest, the persistent remnants of the conduit metaphor and its unhelpful entailments can be fully dislodged from genre analysis and its associated methods. Under the rubric of language socialization, researchers of a new generation can manifest more fully the vast potential of socio-cultural and criticalist approaches to inform educational practice and reform. This emergent approach can also play a role in the re-invigoration of EAP, by articulating a "higher ground" of methodological rigor and transparency to which future researchers will aspire. Moreover, as it implements an LS research program, EAP can recast itself as a more socially responsible and ethically grounded field of inquiry.


Adams-Smith, D.E. (1984) Medical discourse: Aspects of author's comment, The ESP Journal, 3, pp. 25-36.

Allison, D. (1996) Pragmatist discourse and English for Academic Purposes, English for Specific Purposes , 15(2), pp. 85-103.

Anson, C. (2000) Response and the social constrcution of error, Assessing Writing, 7(2000), pp. 1-16.

Anzaldúa, G. (1987) Borderlands/La frontera: The new mestiza (San Francisco, Aunt Lute).

Ard, J. (1982) The semantics of tense and aspect in scientific discoursePaper given at Ohio State Conference on Semantics of Tense and Aspect, May 1982

Bakhtin, M. (1981) The dialogic imagination (Austin, TX, The University of Texas Press).

Barber, C.L. (1962) Some measurable characteristics of modern scientific proseContributions to English syntax and phonology (Stockholm, Almquist and Wiksell).

Bazerman, C. (1998) Editor's Introduction Writing/Disciplinarity: A socio-historic account of literate activity in the academy

Belanger, M. (1982) A preliminary analysis of the structure of the discussion sections in ten neuroscience journal articles (mimeo)., .

Benesch, S. (2001a) Critical English for Academic Purposes (Mahwah, NJ, Lawrence Erlbaum Associates).

Benesch, S. (2001b) Cultural pragmatism: A Politics of L2 composition, in: T. Silva & P.K. Matsuda (Eds)

Berkenkotter, C. .

Berkenkotter, C., Huckin, T.N. & Ackerman, J. (1988) Conventions, conversations and the writer: Case study of a student in a rhetoric Ph.D. program, Research in the Teaching of English, 22, pp. 9-44.

Bernstein, B. (1990) Class, codes and control: The structure of pedgaogic discourse (London, Routledge).

Bhatia, V.K. (1983) An applied discourse analysis of English legislative writing (Birmingham, UK, The University of Ashton Language Studies Unit).

Bloch, J. & Chi, L. (1995) A comparison of use of citations in Chinese and English academic discourse, in: D. Belcher & G. Braine (Eds) Academic writing in a second language (Norwood, NJ, Ablex).

Boggs, S.T., Watson-Gegeo, K.A. & McMillan, G. (1985) Speaking, relating and learning: A study of Hawaiian children at home and at school (Norwood, NJ, Ablex).

Bourdieu, P. (1990) The logic of practice (Stanford, CA, Stanford University Press).

Bourdieu, P. (1991) Language and symbolic power (Cambridge, MA, Harvard University Press).

Butler, C.S. (1989) Systemic models:Unity, diversity and change, Word, 40(1-2), pp. 1-35.

Candela, A. (1999) Student's power in classroom discourseAmerican Anthropological Association, 98th Annual Meeting (Chicago, IL,

Candlin, C.N. (1981) Discoursal patterning and the equalizing of interpretive opportunity, in: L.E. Smith (Ed) English for cross-cultural communication (London, Macmillan).

Candlin, C.N. (1987) Towards task-based language learning, in: C.N. Candlin & D.F. Murphy (Eds) Language learning tasks (Englewood Cliffs, NJ, Prentice-Hall International).

Carspecken, P.F. (1996) Critical ethnography in edicational research (New York, Routledge).

Cazden, C. (1988) Classroom discourse: The language of teaching and learning (Portsmouth, NH, Heinemann).

Ceccheto, V. & Stroinska, M. (1998) Systems of reference in intellectual discourse: A potential source of intercultural stereotypes, in: A. Duszak (Ed) Culture and styles of academic discourse (New york, Mouton de Gruyter).

Chomsky, N. (1991) Manufacturing consent (Cambridge, MA, M.I.T. Press).

Chouliaraki, L. & Fairclough, N. (1999) Discourse in late modernity: Rethinking critical discourse analysis (Edinburgh University Press: Edinburgh,

Christie, F. (1993) Curriculum genres: Planning for effective teaching, in: B. Cope & M. Kalantzis (Eds) The powers of literacy: A genre approach to teaching writing (Pittsburgh, PA, University of Pittsburgh Press).

Christie, F. (1997) Curriculum macrogenres as forms of initiation into a culture, in: F. Christie & J.R. Martin (Eds)

Christie, F. & Martin, J.R. (1997) Genre and institutions: Social processes in the workplace and school (London and New York, Continuum).

Coffin, C. (1997) Constructing and giving value to the past: An investigation into secondary school history, in: F. Christie & J.R. Martin (Eds)

Comfort, J.R. (2001) African-American women's rhetorics and the culture of Eurocentric scholarly discourse, in: C.G. Penetta (Ed)

Connor, U. (1996) Contrastive rhetoric (Mahwah, NJ, Larence Erlbaum Associates).

Cooley, L. & Lewkowicz, J. (1998) Developing awareness of the rhetorical and linguistic conventions of writing a thesis in English: Addressing the needs of EFL/ESL postgraduate students, in: A. Duszak (Ed) Culture and styles of academic discourse (New York, Mouton de Gruyter).

Couture, B. (Ed.) (1986) Functional approaches to writing: Research perspectives (Norwood, NJ, Ablex).

Crookes, G. (1986) Toward a validated analysis of scientific text structure, Applied Linguistics, 7(57-70).

Cummins, J. & Swain, M. (1986) Bilingualism in education: Aspects of theory, research and practice (New York, Longman).

Delpit, L. (1995) Other people's children: Cultural conflict in the classroom (New York, The New Press).

Dreyfus, H.L. & Rabinow, P. (1983) Michel Foucault: Beyond structuralism and hermeneutics (Chicago, University of Chicago Press).

Dubois, B.L. (1982) The construction of noun phrases in biomedical journal articles, in: J. Hoedt, et al. (Ed) Pragmatics and LSP (Language for Special Purposes) (Copenhagen, Copenhagen School of Economics).

Dudley-Evans, T. & St. Johns, M.J. (1998) Developments in English for specific purposes: A multidisciplinary approach (Cambridge, Cambridge University Press).

Duszak, A. (1997a) Cross-cultural academic communication:A discourse community view, in: A. Duszak (Ed) Culture and styles of academic discourse

Duszak, A. (Ed.) (1997b) Culture and styles of academic discourse (New York, Mouton de Gruyter).

Duszak, A. (1998) Analyzing digressiveness in Polish academic texts, in: A. Duszak (Ed) Culture and styles of acdemic discourse (New York, Mouton de Gruyter).

Ewer, J.R. (1979) The modals in scientific discourse: Function, meaning and use, Research Report, Department of English, Santiago, Chile, .

Faigley, L. (1992 ) Fragments of Rationality: Postmodernity and the Subject of Composition (Pittsburgh, University of Pittsburgh Press).

Fairclough, N. (1989) Discourse and social change (Cambridge, Polity Press).

Fairclough, N. (1992) Language and power (London, Longman).

Farrell, L. (1997) Doing well...doing badly: An analysis of the role of conflicting cultural values in judgements of relative "academic achievement", in: A. Duszak (Ed) Culture and styles of acdemic discourse (New York, Mouton de Gruyter).

Firth, J.R. (1957) Papers in linguistics 1934-1951 (London, Oxford University Press).

Flower, L.S. & Hayes, J.R. (1981) A cognitive process theory of writing, College Composition and Communication , 32, pp. 365-387.

Flowerdew, J. (1994) Academic listening: Research perspectives (Cambridge, UK, Cambridge University Press).

Fortes, M. (2001) A Case study of emergent ESL readers, an unpublished dissertationDivision of Education (Davis, CA, University of California Davis).

Frederickson (2001) The queen's English: A queery into contrastive rhetoric, in: C.G. Panetta (Ed) Contrastive rhetoric revisited and redefined (Mahwah, NJ, Lawrence Erlbaum Associates).

Freire, P. (1974a) Education for critical consciousness (London, Sheed and Ward).

Freire, P. (1974b) Pedagogy of the oppressed (New York, Continuum).

Freire, P. (1994) Pedagogy of hope: Re-living pedagogy of the oppressed (New York, Continuum).

Gee, J. (1996) Social linguistics and literacies: Ideologies in discourses (Philadelphia, The Falmer Press).

Geertz, C. (1973) The interpretation of cultures (New York, Basic Books).

Geisler, C. (1994a) Literacy and expertise in the academy, Language and Learning across the Disciplines, 1, pp. 35-57.

Geisler, E. (1994b) Academic writing and the nature of expertise (New York, Harper Colins).

Gianoni, D.S. (1997) Credit where it's due: a sociopragmatic interpretation of Acknowledgements sections in English academic articles, TESL-EJ, 6(3), pp. 221-242.

Giddens, A. (1979) Central problems in social theory (Berkeley, CA, University of California Press).

Gillmore, P. (1999) Seizing academic power: indigenous subaltern voices and counter narratives in education American Anthropological Association, 98th Annual Meeting (Chicago, IL,

Glaser, B. & Strauss, A. (1967) The discovery of grounded theory: Strategies

for qualitative research (Chicago, Aldine).

Goffman, I. (1972) The presentation of self in everyday life (Berkeley, CA, University of California Press).

Group, N.L. (1996) A pedagogy of multi-literacies: Designing social futures, Harvard Educational Review, 66(1), pp. 60-92.

Gumperz, J.J. & Levinson, S.C. (1996) Rethinking linguistic relativity (Cambridge, Cambridge University Press).

Halasek, K. (1999) A pedagogy of possibility: Bakhtinian perspectives on composition studies (Carbondale, IL, Southern Ilinois University Press).

Halliday, M.A.K. (1978) Language as a social semiotic (Baltimore, MD, University Park Press).

Halliday, M.A.K. & Hassan, R. (1976) Cohesion in English (London, Longman).

Halliday, M.A.K. & Macintosh, A. (1964) The linguistic sciences and language teaching (London, Longman).

Halliday, M.A.K., McIntosh, A. & Stevens, P.D. (1964) The linguistic sciences and language teaching (London, Longman).

Harrington, S.M., Rickly, R. & Day, M. (Eds.) (2000) The Online writing classroom (Cresskill, NJ, Hampton Press).

Harris, J. (1989) The idea of community in the study of writing, College Composition and Communication, 40(1), pp. 11-22.

Harris, R.A. (1993) The linguistisc wars (Oxford, Oxford University Press).

Heidegger, M. (1962) Being and Time (New York, Harper & Row).

Herrington, A. (1985) Writing in academic settings: A study of the conetxts for writing in two chemical engineering courses, Research in the Teaching of English, 19, pp. 331- 359.

Heslot, J. (1982) Tense and other indexcical markers in the typology of scientific texts in English, in: J. Hoedt & e. al. (Eds) Pragmatics and LSP (Language for Special Purposes) (Copenhagen, Copenhagen School of Economics).

Hill, S.S., Soppelsa, B.F. & West, G.K. (1982) Teaching ESL students to read and write experimental research papers, TESOl Quarterly, 16, pp. 333-347.

Hofstadter, E. (1979) Gödel, Escher, Bach: An eternal golden braid (New York, Basic Books).

Holmes, R. (1996) Genre Analysis and the Social Sciences: A study of the introductions, background sections and discussion sections of research articles in history, political science and sociology, ESP, 16(4), pp. 321-337.

Holmes, R. (1997) Genre analysis in the social sciences: An investigation of the structure of research article discussions in three disciplines, English for Specific Purposes, 16(4), pp. 321-337.

Hopkins, A. (1988) A genre based investigation of the discussion sections in articles and disserttaions, English for Specific Purposes, 7, pp. 113-122.

Huddleston, R.D. (1971) The sentence in written English (Cambridge, Cambridge University Press).

Hyon, S. (1996) Genre in Three Traditions: Implications for ESL, TESOL, 30(4), pp. 693-722.

Iedema, R. (1997) The language of administration: Organizing human acticity in formal institutions, in: F. Christie & J.R. Martin (Eds) Genre and institutions: Social processes in the workplace and school (London and New York, Continuum).

Inman, M. (1978) Lexical analysis of scientific and technical proseEnglish for Special Purposes: Science and technology (Corvallis, OR: Oregon State University,

Jakobson, R. (1960) Closing statement: Linguistics and poetics, in: T.A. Sebeok (Ed) Stle in language (Cambridge, MA and New York, MIT Press and Wiley).

Kaplan, R. (2001) Foreword, in: C.G. Panetta (Ed) Contrastive rhetoric revisited and redefined (Mahwah, NJ, Lawrence Erlbaum and Asociates).

Kaplan, R.B. (1966) Cultural thought patterns in iintercultural communication, Language learning, 16, pp. 1-20.

Kern, R. (2000) Literacy and Language Learning (Oxford, Oxford University Press).

Kretzenbacher, H.L. (1998) Mind your metaphors! Historical and theretical notes twoard a constructivist theory of metaphor in scientific communication, in: A. Duszak (Ed) Culture and styles of academic discourse (New York, Mouton de Gruyter).

Kreutz, H. & Harres, A. (1998) Some observations on the distribution and function of hedging in German and English academic writing, in: A. Duszak (Ed) Culture and styles of academic discourse (New York, Mouton de Gruyter).

Kuo, C.-H. (1998) The use of personal pronouns: Role relationships in scientific journal articles, English for Specific Purposes, 18(2).

Lackstrom, j., Selinker, L. & Trimble, L. (1972) Grammar and technical English, English Teaching Forum, X(5), pp. 127-136.

Lakoff, G. (1984) Women, fire and dangerous things: What categories reveal about the human mind (Chicago, University of Chicago Press).

Lakoff, G. & Johnson, M. (1980) Metaphors we live by (Chicago, University of Chicago Press).

Lakoff, R.T. (1988) Talking Power (Berkeley, CA, University of California Press).

Lave, J. & Wenger, E. (1993) Situated learning: Legitimate peripheral participation (Cambridge, Cambridge Univesrity Press).

Lee, P. (1996) The Whorf theory complex: A critical reconstruction (Amsterdam and Philadelphia, John Benjamins).

Lemke, J.L. (1990) Talking science (Norwood, NJ, Ablex).

Lemke, J.L. (1995) Textual Politics: Discourse and social dynamics (London, Longman).

Lyotard, J.-F. (1984) The Post-modern condition: A report on knowledge (Minneapolis, MN, University of Minneapolis Press).

Macedo, D. (1994) Literacies of power: What Americans are not allowed to know (Boulder, CO, Westview Press).

Martin, J.R. (1985) Process and text: Two aspects of human semiosis, in: J.D. Benson & W.S. Greaves (Eds) Systemic perspectives on discourse (Norwood, NJ, Ablex).

Martin, J.R. (1993) A contextual theory of language, in: B. Cope & M. Kalantzis (Eds) The powers of literacy: A genre approach to teaching writing

Martin, J.R. (1997) Analysing genre: Functional parameters, in: F. Christie & J.R. Martin (Eds) Genre and institutions:Social processes in the workplace and school (New York, Continuum).

McCarthy, L.P. (1987) A stranger in strange lands: A college student writing across the curriculum, Research in the Teaching of English, 21, pp. 233-265.

McClaren, P. (1992) Collisions with otherness: "Traveling" theory, post-colonial criticism and the politics of ethnographic practice-The mission of the wounded ethnographer, International Journal of of Qualitative Studies in Education, 5(1), pp. 77-92.

McGroarty, M. (1998) Constructive and constructivist challenges for applied linguistics, Language Learning , 48(4), pp. 591-622.

McKinlay, K. (1984) An analysis of dicussion sections in medical journal articlesApplied Linguistics, unpublished Master's thesis (Birmongham, UK, University of Birmingham).

Melander, B., Swales, J.M. & Fredrickson, K.M. (1998) Journal abstracts from three academic fields in the United States and Sweden: national or disciplinary proclivities?, in: A. Duszak (Ed) Culture and styles of acdemic discourse (New York, Mouton de Gruyter).

Miccicche, L.R. (2001) Contrastive rhetoric and the possibility of feminism

Miller, C. (1984) Genre as social action, Quarterly Journal of Speech, 70, pp. 151-167.

Murphy, H.A. & Hildebrandt, H.W. (1984) Effective Business Communications (4th Edition) (New York, McGraw Hill).

Olson, D.R. (1994) The world on paper: The conceptrual and cognitive implications of writing and reading (Cambridge, Cambridge University Press).

Pally, M. & Diallo, A. (1995) The Man who Mistook "Wet Paint" for a Verb: A Chronicle for Thinking about Language, Culture and Writing, .

Panetta, C.G. (2001) Contrastive rhetoric revisited and redefined (Mahwah, NJ, Lawrence Erlbaum Associates).

Pennycook, A. (1994) The cultural politics of English as an international language

Pinker, S. (1994) The language instinct (New York, William Morrow).

Popken, R.L. (1987) A study of topic sentence use in academic writing, Written Communication, 4, pp. 209-228.

Preston, D.R. (1986) The fifty some-odd categories of language variation, International Journal of the Sociology of Language, 57, pp. 9-47.

Prior (1998) Writing disciplinarity: A socio-historic account of literate activity in the academy (Mahwah, NJ, Lawrence Erlbaum Associates).

Probst, R.E. (1989) Transactional theory and response to student writing, in: C.E. Anson (Ed) Writing and response: theory, practice and research (Urbana, IL, National Council of Teachers of English).

Ramanathan, V. & Kaplan, R. (2000) Genres, authors, discourse communities and application for (L1 and L2) Instructors, Journal of Second Language Writing, 9(2), pp. 171-191.

Reddy, M. (1974) The conduit metaphor: A case of frame conflict in English, in: A. Ortony (Ed) Metaphor (Cambridge, UK, University of Cambridge Press).

Rosch, E. (1975) Cognitive representations of semantic categories, Journal of Experimental Psychology (General) , 104, pp. 192-233.

Rosch, E. (1978) Principles of categorization, in: R. E. & B. Lloyd (Eds) Cognition and categorization (Hilldsdale, NJ, Lawrence Erlbaum and Associates).

Rose, D. (1997) Science, technology and technical literacies, in: F. Christie & J.R. Martin (Eds)

Rosenblatt, L. (1989) Writing and reading: The transactional theory, in: J. Mason (Ed) Reading and writing connections (Needham Heights, MA, Allyn & Bacon).

Samraj, B.T. (1995) The nature of academic writing in an interdisciplinary fieldLinguistics (Ann Arbor, University of Michigan).

Samraj, B.T. (2002a) Introductions in research articles: Variations across disciplines, English for Specific Purposes, 13(2), pp. 28-42.

Samraj, B.T. (2002b) Texts and contextual layers (unpublished manuscript to be published in English for Special Purposes) English for Special Purposes

Samraj, B.T. (2002 (ms. to be published)) Disciplinary variation in abstracts: the case of wildlife behavior and conservation biology, in: J. Flowerdew (Ed) Academic discourse (Norwood, NJ, Pearson Publication).

Santos, T. (2001) The place of politics in second language writing, in: T. Silva & P.K. Matsuda (Eds)

Schleppegrell, M. (1999) The language of schooling (unpublished ms.)

Schneider, M. & Fujishima, N.K. (1995) When practice doesn't make perfect: The case of an ESL graduate student, in: D. Belcher & G. Braine (Eds) Acdemic writing in a second language (Norwood, NJ, Ablex).

Scoggins, D. (2001) Contrastive rhetoric theory in an electronic medium: Teaching ESL writers to become bricoleurs in in a computer-assisted classrooom, in: C. Panetta, G. (Ed) Contrastive rhetoric revisited and redefined (Mahwah, NJ, Lawrence Erlbaum and Associates).

Silva, T. & Matsuda, P.K. (2001) On second language writing (Mahwah, NJ, Lawrence Erlbaum Associates).

Sperling, M. & Freedman, S. (1987) A good girl writes like a good girl, Written Communication, 4, pp. 343-369.

Spolsky, B. (1994) Comprehension testing, or can understanding be measured?, in: G. Brown, M. Malmakjaet, P. A. & J. Williams (Eds) Language and understanding (oxford, Oxford University Press).

Straub, R. & Lunsford, R.F. (1995) Twelve Readers Reading: Responding to college student writing (Cresskill, NJ, Hampton Press).

Sullivan, F.J. (1995) Critical theory and systemic linguistics: textualizing the contact zone, JAC, 15(3), pp. 1-28 (electronic version accessed).

Swafar, J.K., Arens, K. & Morgan, M. (1982) Teaching classroom practices: Refining method as task hierarchy, Modern Language Journal, 66(24-33).

Swales, J. (1981) Aspects of article introductions(Birmingham, UK, The University of Aston, Language Studies Unit).

Swales, J. (1990) Genre analysis: English in academic and research settings (Cambridge, Cambridge University Press).

Swales, J. & Najiar, H. (1987) The writing of research article introductions, Written Communication , 4, pp. 175-192.

Tarone, E., Dwyer, S., Gillete, S. & Icke, V. (1998) On the use of the passive and active voice in astrophysics journals: With extensions to other languages and other fields, English for Specific Purposes, 17(1), pp. 113-132.

Tarone, E.S., Dwyer, S., Gilette, S. & Icke , V. (1981) On the use of the passive in two astrophysics journal papers, The ESP Journal, 1, pp. 123-140.

Thompson, G. (1996) Introducing functional grammar (New York, St. Martin's Press).

Tisdell, E. (1993) Feminism and adult learning: Power, pedagogy and praxis

Toolan, M. (2001) Review of Chouliaraki and Fairclough (1999), Journal of Sociolinguistics, 5(1), pp. 112-117.

Veel, R. (1993) Learning how to mean-scientifically speaking: Apprenticeship into scientific discourse in secondary school, .

Ventola, E. (1984) Orientation to social semiotics in foreign language teaching, Applied Linguistics, 5, pp. 275-286.

Walvoord, B.E. & McCarthy, L.P. (1990) Thinking and writing in college: A naturalistic study of students in four disciplines (Urbana, IL, National Council of Teachers of English).

Watson-Gegeo & Nielsen, S. (2000a) Language socialization in second language acquisition research, in: M. Long (Ed) Handbook of second language research (London, Blackwell).

Watson-Gegeo, K.A. (1992) Thick explanation in the ethnographic study of child socialization and development: A longitudinal study of the problem of schooling for Kwara'e (Solomon Islands) children, New Directions in Child Development, 58(Interpretive methodologies for the study of childhood socialization), pp. 51-66.

Watson-Gegeo, K.A. (2001) Mind, language and epistemology: Toward a language socialization paradigm for SLA. Pacific Second Language Research Forum-Invited Plenary Speech (Honolulu,

Weissberg, R. (1984) Given and new: Paragraph development models for scientific English, TESOl Quarterly , 18, pp. 485-500.

Welton, M.R. (1993) The contribution of critical theory to our understanding of adult learning An update on adult learning theory (San Francisco, Jossey-Bass).

White, P. (1997) Deat, disruption and the moral order: The narrative impulse in "hard news" reporting, in: F. Christie & J.R. Martin (Eds)

Whorf, B.L. (1956) Language, thought and reality (New Haven, CT, Yale University Press).

Widdowson, H.G. (1979) Explorations in applied linguistics (Oxford, Oxford University Press).

Woolever, K.R. (2001) Doing global business in the information age: Rhetorical contrasts in the business and technical professions, in: C.G. Panetta (Ed) Contrastive rhetoric revisited and redefined (Mahwah, NJ, Lawrence Erlbaum Associates).


1. "Genre Analysis" will be used in this paper to refer specifically to the approach articulated in Swales (1990) whereas "genre analysis" will refer to the activity without reference to any particular set of procedures or theory. (return to text)

2. ESP is technically a superordinate term which encompasses both English for Academic Purposes ("EAP") and English for Occupational Purposes ("EOP"). EAP encompasses EST ("English for Science and Technology and EOP, in turn, encompasses such sub-fields as EBP ("English for Business Purposes") which is one of the fastest growing sub-branches of ESP. (Dudley-Evans, 1998: p.) (return to text)

3. As evidence of this, consider the fact that a review of the indices of three books in the SFL tradition (Christie & Martin, 1993; Cope & Kalantzis, 1993; Thompson, 1996) found not one reference to "discourse community" or even "audience". There are, however, numerous implicit invocations of the idea of a discourse community in the form of, for example, "science," "technology" and "secondary school" as conditioning institutional environments ("registers") and as the sources of target forms for the curriculum. (return to text)