Multimodal Genre Systems in EAP | Writing Pedagogy: Reflecting on a Needs Analysis ~~

Using ethnographicmethods, an investigation of academic genres in several classrooms inthree academic disciplines civil and environmental engineering, archi-tecture, and music found thre

Multimodal Genre Systems in EAP

Writing Pedagogy: Reflecting on a

University of Illinois at Urbana-Champaign

Urbana, Illinois, United States

This article reports on a genre-based needs analysis for a graduatecourse in English for academic purposes (EAP) at a large public U.S.university In particular, it describes the theoretical reconceptualiza-tions of genre analysis that the data provoked Using ethnographicmethods, an investigation of academic genres in several classrooms inthree academic disciplines (civil and environmental engineering, archi-tecture, and music) found three complexities that challenged the origi-nal premises of the needs analysis: (a) that academic genres existed ingenre sets and systems that involved process and pedagogical genres aswell as genres of disciplinary or academic presentation; (b) that genreswere routinely multimodal in process and form; and (c) that the dis-cursive character of particular texts was routinely quite hybrid Thisarticle discusses and illustrates each of these findings and argues forunderstanding them as dimensions of multimodal genre systems

As Canagarajah (2006) points out, scholars in the field of TESOL havestopped looking for conclusive answers and have accepted that, both

as researchers and as practitioners, we need to abandon “the comfort of

solutions” (emphasis in the original, p 30) and embrace the messy

prac-tice of “continued questioning and searching” (p 13) As we strive todeepen our understanding of the challenges we face, such questioningand searching is our only feasible research goal This article emerges out

of that mangle of practice, as Pickering (1995) describes such processes: It

began as a needs analysis aimed at identifying different genres for anEnglish for academic purposes (EAP) course and ended by destabilizingour initial conceptualization of genres and how they operate

In the teaching of writing to speakers of other languages in general,

and in EAP in particular, the efforts of researchers and instructors to findanswers to pressing questions about the needs of students has oftenturned to genre analysis (Swales, 1990) EAP researchers have pursueddetailed analyses of the types of academic texts that students produce indiverse disciplines (e.g., Bhatia, 1993; Braine, 1995; Coe, 2002; Hyland,2004; Swales, 1990, 2004), the skills that such texts require of students(e.g., Belcher, 1995; Johns, 1997; Leki & Carson, 1997), and the ways inwhich genres function in specific contexts (e.g., Casanave, 2002; Prior,

1995, 1998) Genre analysis has considerably increased our ing of the complexity of genres and their linguistic and social realization

understand-It has also offered valuable perspectives on possible approaches to ing genre to diverse populations of students and shed light on some ofthe most urgent needs of students in particular contexts At the same

teach-time, EAP research has illustrated that the term needs can be

conceptu-alized very differently: ranging from form–function correlations (Bhatia,1993) to becoming aware of one’s rights as a student (Benesch, 2001).EAP research has suggested, furthermore, that no needs analysis can befinal or exhaustive, and that “needs assessment itself [is] in need ofcontinual reassessment” (Belcher, 2006, p 135) Needs analysis has ideo-logical roots, and EAP approaches to it are grounded in (often strong)expectations about the kinds of generic discourse, values, and practicesthat the students will encounter

This article discusses the conceptualization and execution of a based needs analysis for an EAP course for international graduate stu-dents at a large Midwestern U.S university More specifically, the articlereports on the ways in which some results of the needs analysis forced us

genre-to consider genres from more deeply processual, semiotic, and dialogicperspectives The narrative of our journey is organized as follows Wefirst discuss the initial goals of the research and describe in general termsthe nature of the theoretical reconceptualizations that the data pro-voked Second, we briefly describe the setting and methods of the needsanalysis Third, we examine each of the three intriguing findings (oranomalies) that destabilized the initial theoretical framework We con-clude by discussing possible pedagogical implications

INITIAL GOALS AND SUBSEQUENT

RECONCEPTUALIZATION OF THE RESEARCH

Many scholars and practitioners involved with academic writing haveconcluded that genre-oriented instruction is one key to preparing stu-dents for the writing that they may be expected to produce during theiracademic careers (e.g., Bazerman, 2004; Berkenkotter & Huckin, 1995;

Bhatia, 1993; Dias, Freedman, Medway, & Paré, 1999; Hyland, 2004;Johns, 2002; Paltridge, 2001; Russell, 1997; Swales, 1990, 2004; Swales &Feak, 2004).1Such a focus on genre, however, presents EAP instructorsand researchers with three immediate challenges: Which genres should

be taught in the EAP classroom? What kinds of features will best acterize these genres? And how are genres best taught or learned? Un-derneath these challenges lies a more fundamental set of questionsabout what genres are; what work they do; and how they are learned,used, and transformed

char-EAP researchers have generally attempted to address at least the firsttwo questions by conducting some form of needs analysis It was in view

of this broad consensus and in response to local concerns about therelevance and theoretical grounding of an EAP writing curriculum at amajor U.S university that one of us (Molle) decided to pursue a needsanalysis in her role as an instructor of an EAP course for internationalgraduate students at this university and asked the second author (Prior)

to direct this study for a master’s thesis The research sought to gate students’ needs through a contextualized analysis of the types ofgenres that students were asked to produce during two semesters in asample of disciplines The research design used ethnographic methods:interviewing students and instructors, collecting course documents andstudent texts, and observing some classes We anticipated that the endproduct of the study would be a list of proposed changes (e.g., differentgenres, different ways of handling process) to the existing genre-basedcurriculum of the EAP writing course that would reflect the study’s find-ings as well as recent developments in EAP research

investi-As the analysis proceeded, however, it became increasingly apparentthat the theoretical approach that had prompted the design did notmatch up well with the data collected The analysis was pointing towarddifferent understandings from what typical genre theory and needsanalysis would anticipate in three respects First, needs analysis has typi-cally aimed to uncover clearly identifiable genres (or tasks) We soonconcluded, however, that analysis of genre as an isolated phenomenon

1 Considerable diversity exists in genre theories A common classification (Hyland, 2004; Hyon, 1996) identifies three schools, systemic functional linguistics (SFL), English for specific purposes, and new rhetoric, though others (e.g., Johns, 2003) suggest researchers and practitioners routinely blend insights from these schools Nevertheless, key differences

are noticeable, for example, in their definitions of genre: new rhetorical approaches take

up an open set of whole genres in specific contexts (e.g., a lab report in a biology room), whereas SFL defines genres as a finite set of linguistic processes (e.g., recount, narrative, account) that typically combine in whole texts (with recount appearing in lab reports, letters, novels, news reports, etc.) The approach we take, influenced by the intersection of phenomenological sociology with Bakhtinian views of genre as forms of situated practice (Hanks, 1996; Bazerman & Prior, 2005), positions us closest to new rhetoric and farthest from SFL.

class-would be incomplete and misleading and that it could not account forthe ways we saw genres functioning in these contexts What we found was

a complex web of relationships among different genres, within whichsome texts had ambiguous status as genres and some seemed primarilyoriented to goals other than communication Second, genre-based needsanalysis has typically aimed at identifying a set of textual features thatcould describe particular genres in particular contexts, and genre analy-sis, from Swales (1990) to Hyland (2004), has primarily focused on thelanguage of texts Yet, we found that multiple modes or media coexistednot only at every level of the texts (from sections to phrases) but alsothroughout the writing process, and these modes of representation co-constructed the meaning of the text Third, although genres (such asscientific reports) are seen as constituted by sections with different pat-terns of language use (e.g., the way the textual–rhetorical patterns of anintroduction differ from those of a methods section), genre analysistypically assumes a basic unity in the discourse, voice, or style of eachgenre In analyzing student work, however, we began to see a moreheterogeneous blend of discourses appearing (and being accepted) inthose texts

These anomalies, as we have called them, about the nature of genre in

the disciplines certainly challenged the main premises of the particularEAP curriculum that the research had aimed at revising However, theyalso challenged the initial formulation of the research itself, raising in-triguing questions about how to conceive of genres, needs analysis, andgenre-based teaching in light of these observations

SETTING AND METHODOLOGY

As already mentioned, the needs analysis was undertaken to ine the basis for an EAP writing course for international graduate stu-dents at a major public university in the United States The course in-volves two 2-hour class sessions per week for a semester (15 weeks); itsmultiple sections are taught primarily by masters-level graduate teachingassistants Almost all sections of the course enroll students from acrossthe campus (i.e., not from specific degree programs) We planned tocollect data from students and instructors in four disciplines The num-ber of disciplines selected for the study could not be larger for practicalreasons (balancing resources to the goal of achieving some depth ofanalysis) We believed, however, that a sampling of several different dis-ciplines should assist us in finding some common academic ground (ifsuch ground existed) and also allow us to partially test the fit of theexisting genres taught in the EAP course to those in some locally impor-

reexam-tant disciplines Four departments at the university were selected: tecture, Civil and Environmental Engineering (CEE), Music, and Psy-chology These disciplines were chosen on the basis of two criteria: (a)the number of students in the discipline who attended the EAP course inquestion during the five semesters between spring 2002 and fall 2004 and(b) the range of disciplines selected for the study.2Unfortunately, par-ticipation by students and faculty in psychology proved too limited tosupport in-depth analysis, though the small amount of data collected wasconsistent with findings from the other fields.

Archi-The study spanned two semesters and investigated the writing ventions in three disciplines through genre analysis of student texts;in-depth interviews with faculty and several international students; analy-sis of course materials such as syllabi, handouts, and evaluation sheets;and class observations Course documents and student texts were col-lected from the student participants as well as the instructor participants

con-in the study All the writcon-ing assignments that the primary student ticipants completed for the EAP course and for their disciplinary courses

par-in the period of data collection (two semesters) were obtapar-ined, alongwith course documents from the students’ disciplinary courses In thosecases in which no student participants were enrolled in classes to which

we had access through instructors, we obtained course documents and(with student consent) student texts from the instructors We sought andobtained permission to analyze the texts from their authors The studentauthors (the great majority of whom were international students) of thetexts we collected from instructors were not interviewed for this studyand are thus not designated as primary student participants

The assignments collected varied across disciplines and across ent courses in the same discipline; they included reaction papers, bookreports, posters, and term papers In most cases, we were able to obtainstudent texts written in response to these assignments with instructorcomments When we could not obtain students’ written texts with in-structor comments, the instructor’s assessment of the student texts wasdiscussed during the in-depth interview with the primary student partici-pants

differ-With one exception, all primary student participants had taken or

2 CEE, architecture, and music were chosen not only because they were among the six disciplines that contributed for five semesters the highest number of students to the EAP course studied but also because they contributed students each semester We were also looking for departments in different colleges as well as for disciplines that represented different types of fields (professional vs more conventionally academic) We were hoping that this would expose us to a greater variability of genres and conventions This first consideration eliminated disciplines like electrical engineering (which was in the same college as CEE) The second consideration led to the choice of psychology as the fourth discipline to be included in the study: unlike the other three, it is a more academic than professional discipline.

were taking the EAP writing course when data collection began Thesestudents were international graduate students from around the world,most of whom were enrolled in doctoral programs at the university Allprimary student participants were interviewed The interviews focused

on the students’ experience of writing in the disciplines and on theirown assessment of their needs in relation to writing

The primary instructors who participated in the study were members

of the faculty in the students’ disciplines and all identified as nativeEnglish speakers from the United States All primary instructor partici-pants were interviewed The interviews focused on the assignments theinstructors gave their students, the rationale behind the assignments,and the instructors’ criteria for a successful text Course observation wassuccessfully negotiated with some of the instructors As Table 1 indicates,four courses were observed, two in architecture, one in CEE, and one inmusic The CEE course was observed only once because, according to theinstructor, this was the only time the writing assignment for the course (aterm paper) was described The courses in architecture and music wereobserved every time the students gave presentations related to their writ-ten texts or the class discussed writing assignments, and the number ofobservations varied considerably (from two sessions to all class meetings

TABLE 1 Overall Description of the Data Collected for the Study per Department

Department

Primary instructor

participants

Student participants primary/total

Course assignments

Student texts

Courses studied*/ observed Architecture 3 2/3 5 15 3/2

analysis was grounded in theories of genre as situated practices (e.g.,Bakhtin, 1986; Bazerman, 1988; Berkenkotter & Huckin, 1995; Devitt,2004; Prior, 1998) Data were selected for presentation in this article toinclude a range of courses and participants but also to display well thetypes of phenomena that led us to make our central arguments.

As Table 1 indicates, the needs analysis was modest in scope: 3 plines, 7 primary instructors, 6 primary students (and 14 other studentparticipants), 21 assignments, 37 student texts, and 4 courses observed(with some data on 13 other courses) Nevertheless, the depth and range

disci-of the data, combined with the local relevance disci-of these departments,permitted a critical analysis of how well the academic tasks these gradu-ate students were assigned in their courses fit with the existing EAPcurriculum and ultimately also of how well the initial theoretical frame-work fit with the data In short, needs analysis cannot present a totalpicture of “what students need,” but these data did test the adequacy ofcurricular approaches and theoretical frameworks Although the re-search began with questions about the relevance of the current genres tothe curriculum covered, it also began with the assumption that a carefulexamination of genres would point toward a revised set of genres andperhaps some additional writing practices As we have already noted, thefindings quickly led us to more basic questions about that way of imple-menting genre theory We turn next to the results of the research as wepresent and analyze the three dimensions of genres that the data analysispushed us to consider: the relationships among genres, the multimodal-ity of genres, and the hybrid discourse of particular genres

GENRE SYSTEMS ACROSS THE DISCIPLINES:

RESULTS OF THE NEEDS ANALYSIS

From Genres to Genre Systems

As we analyzed genres in courses across these three disciplines, wesoon came to see that texts that represented particular genres werelocated in chains of texts that included other genres and that some textswere generically ambiguous or complex in interesting ways Over the last

15 years, in different terms and with somewhat different emphases, butwith increasing clarity, genre analysts have been moving from a focus ongenres as isolated phenomena to a recognition of how specific types oftexts are constituted by systems of genres Genres have been described in

terms of chains (Fairclough, 2004; Räisänen, 1999; Swales, 2004), colonies (Bhatia, 2002), repertoires (Devitt, 2004; Orlikowski & Yates, 1994), sets (Devitt, 1991), systems (Bazerman, 1994, 2004), and ecologies (Spinuzzi,

2004) Theorists have also begun to highlight ways that genre theory hasprivileged public texts and taken up a perspective that treats the primaryfunctions of genres as communicative Swales (1996, 2004), for example,

identified the category of occluded genres, or genres that are not typically

publicly shared In addition, Spinuzzi has argued that genres should beunderstood as mediational means or as collective tools that control prac-tices “from the outside” (p 114) Situated genre analyses in specific sites(e.g., Bazerman, 1999; Berkenkotter, 2001; Prior, 1998) have also de-picted ways that literate activity involves multimodal chains of genres(e.g., from planning talk to a written text that may then be reviewedorally and in writing), many of which are relatively occluded and fore-ground mediation of activity (e.g., a checklist that functions for the user

as a mnemonic device and prompts a sequence of actions)

The ways in which the genres analyzed in the current study interacteddiffered among and within disciplines, but evidence of such interactionwas clearly visible in most of the content courses included in the study

In a class on gender and race in architecture, for example, students’book reviews were shaped by the instructions, samples, and evaluationsheets handed out by the instructor The written book reviews were alsoaccompanied by 10-minute oral presentations In a seminar on music,the research papers that students composed were based on their earlieroral presentations and were shaped by the instructor’s guidelines for thepaper and the musical analyses the instructor had modeled throughoutthe semester In a mechanics course in CEE, the students’ term papersgrew out of their project proposals and progress reports The analysis ofgenres in these disciplines revealed strong intertextual relationshipsamong representatives of the same type of genre (e.g., samples of bookreviews composed during past semesters and assigned book reviews),among texts that represented different genres but shared a medium(e.g., assignment descriptions, project proposals, and research papers),and among texts in different media (e.g., oral presentations and researchpapers) It also identified texts that seemed designed to support learningand other activities and that were harder to categorize in usual genreterms We turn now to an example of how genre systems appeared in thework of a CEE course

The assignment, or series of assignments, that we have chosen toillustrate the intertextual nature of genre in the disciplines comes from

a graduate course in CEE related to advanced topics in the science andtechnology of materials used in civil engineering construction Weshould note that this assignment may not be representative; the instruc-tor indicated in an interview that it was the first time he had used it.Nevertheless, it demonstrates a pattern of relating genres that we sawacross disciplines in content courses in which the instructors were genu-inely concerned about the quality of their students’ writing and wanted

to help them improve it through a structured and linked sequence ofactivities.

The professor of the advanced materials course in CEE taught tworelated courses Instead of assigning the same type of paper in bothcourses, he decided that he would use one of the courses to give studentsthe kind of guidance in writing up research that he felt he had neededbut had not received when he was a graduate student Consequently, he

developed what he called a research exercise for that course His main

purpose was to teach his students how to approach research critically and

so be better able to meet the expectations of experts in CEE In lar, he wanted to show students how each new piece of research is con-nected to research done in the past and points to research that may bedone in the future The tasks were carefully staged and structured, sothat they built on each other and gradually became more and morecomplex The research exercise consisted of four different tasks Thefirst task had an open format, the second one was a critical review, thethird was argumentative, and the fourth was argumentative and con-tained a summary Each task was approximately 1,000 words in length At

particu-one level, this exercise fits with various observations of school genres (see,

e.g., Berkenkotter & Huckin, 1995; Dias et al., 1999), that is, of genresdesigned for learning and/or evaluation However, what we want toemphasize is the staged design of the tasks and how complexly relatedthey are to one another and to the disciplinary–professional practicesthey indexed

Task 1 asked students to (a) choose a paper on a particular topic andexplain why they had chosen that topic and (b) select an article related

to that topic and describe how they located it The task provided a wayfor the instructor to ensure that students could make use of the libraryresources The instructor also put special emphasis on the citation forthe paper The structure of the first part of Task 1 resembles a part of the

Introduction section in an empirical research paper (establishing a search territory, to use the terms in Swales & Feak, 2004) As we have

re-already mentioned, this resemblance was the result of a deliberate choice

on the instructor’s part to prepare students for the critical thinking theyneed to do when they write research papers The second part of theassignment consisted of a list of the steps that students took in finding ajournal article As a whole, this task rehearsed both pedagogical func-tions (student learning) and the backgrounded processes of academicwork (critical thinking, library searching)

Task 2 asked students to write a critical review of the article they hadselected They had specific instructions on what to include in their cri-tique: After they summarized the content of each section of the paper—

background, experimental tests (if any), modeling (if any), findings and clusions—they had to identify at least one strength and one weakness of

con-the article and support con-their opinion Then con-they were asked to “identifyone significant study, test series, model, etc that is not currently in thepaper but that would make the paper stronger or more convincing than

it currently is.” The following paragraph illustrates how Ji-Hyun,3 agraduate student in CEE from Korea, described the strengths of thearticle she was reviewing:

The use of C3S eliminated any effects of ettringite formation on sion Hence, results could be analyzed focusing on the role of gypsumformation Since two kinds of C3S—monoclinic crystal with impurities(MgO, Al2O3, Fe2O3 and alkalis) and triclinic crystals without impuri-ties—were used and both specimens showed same trend for expansion insulfate solution, it reduced the possibility that expansion is caused byother chemicals (minor substances in cement such as MgO, Al2O3, Fe2O3and alkalis) The use of two kinds of sulfate solution—Na2SO4 and(NH4)2SO4—can be understood in same manner As a result, the experi-ment was composed well to generalize the results The literature reviewpart was helpful to overview the contradicting and supporting theoriesabout gypsum formation Due to this review session, experiment resultscan be judged objectively

expan-Task 3 pushed students one step further in their analysis of the journalarticle and showed them two fundamental features of research work: that

it builds on past studies in a particular area and that it offers a uniquecontribution In this task, students were first asked to provide full cita-tions for “three previous papers [ .] which report findings that lead(either sequentially or in parallel) to the work reported in the paper.”Then they were asked to “document the procedure used to locate andidentify the selected background papers,” “identify or map out links intechnical ideas/developments between the background papers and thecurrent paper,” and then “briefly describe how [the paper] providestechnical contributions/innovations over the selected background pa-

pers.” The first three parts of the task (provide citations, document dures, identify or map out topical links) involve backgrounded research processes, and the last one (describe contributions/innovations) is a typical

proce-feature of literature reviews

Task 4 asked students to engage with a partner’s work In a researchinterview, the instructor indicated that his intentions were twofold: toencourage students to do their best on the tasks by telling them that aclassmate was also going to look at their writing and to give students anopportunity to engage in another research area besides their own thatwas related to the content of the class Task 4 asked students to summa-rize their partner’s thoughts and conclusions in Tasks 1–3 and then, on

3 All student names are pseudonyms.

the basis of the summary, to “envision and describe the ‘next step’ in thatspecific research area which would provide a contribution beyond thatalready established.” In the following excerpt, a CEE graduate studentfrom China, Ke, describes two possible “next steps” in the research dis-cussed by his peer:

As my partner noticed, this model may also be applicable to simulate thecarbonation process, another mechanism causing corrosion of concreterebar It is therefore suggested to extend their model to take carbonationinto account Again, in their extended model, they can consider only thecarbonation process and can compare the results with many publisheddata on carbonation If their model can work well in carbonation simu-lation, then the model would be valuable as it provides a tool for studyingcorrosion of rebar, a big problem in civil engineering field

A more challenging “next step” may be extending the model to 3-Dversion, as marine structures have different exposures: submerged zone,tidal zone, and atmospheric zone The transport processes, or corrosionprocesses are different in these different zones If the extended 3-D modelcan analyze this complicated case (3-D boundary value problem), it would

be useful in efficient concrete design: different parts of the concretestructure with different durability suiting for their different exposure

None of these tasks, with the possible exception of Task 2, can bedescribed as belonging to a named genre Nevertheless, all of themcontain elements that, in their content, form, and/or function, resemble

certain genres Task 1 has features (establishing a territory) that could be

incorporated into the Introduction section of a research article Hyun’s summary for Task 2 could be part of any report of the literature.Task 3 also includes elements that could be part of literature reviews orcritical reviews Task 4 points toward several generic possibilities: theconclusions/implications section of a research article or a critical review(where next steps for the research may be projected); the introductorysections of a research report or proposal (where the researcher identifies

Ji-a gJi-ap in previous reseJi-arch Ji-and explJi-ains how his/her reseJi-arch hJi-as Ji-dressed or will address that gap); and certain peer review genres (writtenresponses to grant proposals and articles) Although Tasks 3 and 4 arereminiscent of critical reviews, on the whole these tasks seem less ori-ented to formal genres of scientific and technical communication than

ad-to occluded genres of technical work and pedagogy All tasks also clude a number of features of backgrounded or situated activity thathighlight mediation (e.g., the descriptions of library searching in Task 1and Task 3, the instruction to identify an additional study, model, etc., inTask 2 and “map out links” in Task 3, and the discussion of the “partner”

in-in Task 4) The CEE assignment illustrates that genres do not existindependently but are links in a long chain of texts with varied purposes

In teaching materials for EAP such as Swales and Feak (2000, 2004),

some genres (e.g., summary, abstract, or literature review) are deemedimportant to consider because they serve as components of larger genres(e.g., research papers and dissertations) Swales and Feak highlight thediverse genres (occluded and foregrounded) that come together in aca-demic contexts and describe them as forming a genre network Theyusefully note chains of genres in background activity like the search foracademic positions Nevertheless, they go on to treat genres such as theconference abstract and the literature review in separate chapters andonly lightly link together forms of academic communication (e.g., re-quests and reminders) Such approaches, rich as they are in many re-spects, may reinforce the notion that genres can be studied on their own.Representing a loose collection of genres in a network does not signalthe complex web of relationships among genres that we found in ouranalysis.

A theoretical framework that seems to account better for our ence of genres in the disciplines is one that links the notions of activitysystems to genre sets and genre systems (Bazerman, 1994, 2004; Devitt,1991).4 In this framework, genre sets are “collections of types of textssomeone in a particular role is likely to produce,” while genre systemsconsist of “the several genre sets of people working together in an orga-nized way, plus the patterned relations in the production, flow, and theuse of these documents” (Bazerman, 2004, p 318) In giving examples ofthese notions in relation to classrooms, Bazerman clearly includes infor-mal and occluded as well as formal genres and multiple media as well aswritten texts (e.g., students’ notes on lectures, rough drafts, oral ques-tions in class, e-mail questions to the instructor; maps and clay models;instructor’s notes for lectures, replies to student questions and com-ments in class or online, syllabi and assignment sheets, class lectures, andhallway discussions) Genre systems are in turn shaped by, and give shape

experi-to, activity systems—durable cultural-historical configurations of tools,people, and sociomaterial environments aimed at achieving certain ob-ject(ive)s (Bazerman & Prior, 2005; Berkenkotter, 2001; Russell, 1997)

A theoretical framework that views individual genres as constituents of asystem should be of pedagogical significance As Bazerman (2004) points

4Devitt (1991) defined genre sets in terms of the genres of a particular category of actor in

an institutional context Analyzing tax accountants, she noted that “together with oral genres and tax returns, these kinds of texts form the accountant’s genre system, a set of genres interacting to accomplish the work of the tax department” (p 340) Devitt (2004)

does not differentiate between genre sets and genre systems (and indicates that she prefers sets

to systems because she is concerned that the latter term might suggest a fixity and order beyond what is typical) She also takes up genre repertoire to point to the somewhat broader

range of possible genres in a sphere of activity Bazerman (1994), on the other hand, proposed understanding genre systems more broadly as the whole range of genres that all participants in a context might produce (perhaps in particular sequences) In our discus-

sion, we are roughly following Bazerman’s use of set and system.

out, such a framework may help instructors to “identify all the forms ofwriting a student must engage in to study, to communicate with theteacher and classmates, and to submit for dialogue and evaluation” andthus help “define the competences, challenges, and opportunities forlearning” present in a particular content course (p 318).

The value of this theoretical framework seems evident in the context

of the advanced materials course in CEE The nature of the assignmentand the relationships among its parts were shaped by the system ofactivity in which the professor and the students were engaged In thiscase, the system of activity was driven by a focus on teaching and learning

as well as on the backgrounded professional practices of engineers; it was

a system of activity whose primary motive was to help the students learn

to think and act like engineers The public genres of engineering mayhave played only a secondary role in these four tasks, but the activitysystems of engineering influenced all the aspects of the texts that thestudents were asked to produce: their form, content, process, and com-municative purpose It seems crucial, then, to prepare students for thepossibility that the different genres they may be asked to produce in theircontent courses can be connected to each other and that both thisconnection and the genres themselves can be influenced by diverse pur-poses

Multimodality of Genre in Text and Process

Scholars in the field of EAP have long recognized that EAP instructorscan only have limited understanding of the writing that students com-pose for their content courses and have thus advocated different forms

of collaboration between EAP instructors and instructors in the plines (e.g., Bhatia, 1993; Dudley-Evans, 2001; Flowerdew & Peacock,2001; Johns, 1997; Starfield, 2001) Discussion of the strengths and weak-nesses of EAP instructors’ interpretation of writing in the disciplineshave tended to be general, however The specific features of texts in thedisciplines, other than specialized knowledge and lexis, that limit theaccess of nonexperts are rarely detailed Analysis of student writing inthis study suggests that one key reason for the inaccessibility of disciplin-ary texts may be their particular forms of multimodality

disci-The papers we studied could not be neatly divided into nonverbalsections (such as a series of equations, plots, or sketches) and verbalsections (i.e., sections consisting exclusively of linguistic elements) The

visual elements in the student texts we collected occurred within the

verbal text and were integral to the texts, even down to the level ofphrases in sentences The text in Figure 1 is a case in point, It comesfrom a paper composed by Peter, a master’s student in music from