Вы находитесь на странице: 1из 53

Critical Discourse Analysis in Education: A Review of the Literature Author(s): Rebecca Rogers, Elizabeth Malancharuvil-Berkes, Melissa Mosley, Diane

Hui and Glynis O'Garro Joseph Reviewed work(s): Source: Review of Educational Research, Vol. 75, No. 3 (Autumn, 2005), pp. 365-416 Published by: American Educational Research Association Stable URL: http://www.jstor.org/stable/3515986 . Accessed: 24/10/2012 08:02
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

American Educational Research Association is collaborating with JSTOR to digitize, preserve and extend access to Review of Educational Research.


Reviewof EducationalResearch Fall 2005, Vol. 75, No. 3, pp. 365416

Critical Discourse Analysis in Education: A Review of the Literature

Rebecca Rogers, Elizabeth Malancharuvil-Berkes, Melissa Mosley, Diane Hui, and Glynis O'Garro Joseph Washington University in St. Louis to have turned researchers Duringthepast decadeeducational increasingly CriticalDiscourseAnalysis(CDA)as a set of approachesto answerques-

tions about the relationships between language and society. In this article the authors review thefindings of their literature review of CDA in educational research. Thefindings proceed in thefollowing manner: the multiple ways in which CDA has been defined, the theories of language included in CDA frameworks, the relationship of CDA and context, the question of methods,

and issues of reflexivity. The findings illustrate that as educational researchers bring CDAframeworksinto educational contexts, they are the reshaping boundaries CDA. of

KEYWORDS: critical discourse analysis, CriticalDiscourse Analysis, critical discourse studies,educationaldiscourse. This year marksthe 25th anniversary the publicationof two seminal books: of Kress,andTony Languageand Control,by RogerFowler,RobertHodge, Gunther Trew, andLanguageAs Ideology, by RobertHodge andGuntherKress.These two books have influencedthe way in which scholarsapproachquestionsof language and society andhave become cornerstonesin what we know as CriticalDiscourse Analysis (CDA). Of course, the history of the critical study of discourse can be traced back much fartherto language philosophers and social theorists such as Bakhtin (1981), DuBois (1903/1990), Pecheux (1975), Volosinov (1930/1973), and Wittgenstein (1953), among others. We might also think of the history of critical discourse studies in terms of the emergence or the evolution of the term Critical Discourse Analysis, which has been attributedto the publication of Fairclough'sLanguageand Power in 1989. The emergenceof the interestin relatingthe studyof discourseto social events did not takeplace in isolation.The 1970s were characterized the transformation by of linguistictheoriesandmethodsin the social sciences, fromtraditional linguistics to interactional linguistics,to criticallinguistics.Indeed,duringthatdecade,linguists became aware that traditionallinguistics needed to consider questions relatedto society. MichaelHalliday's(1975, 1978) theoryof systemic functionallinguistics, which informedcriticallinguisticsandthenCDA, emphasizedlanguageas a meaning-makingprocess,completewith options;Halliday'stheorywas synergisticwith the criticalstudyof language.At the same time, therewas dissentandrevolutionin 365

Rogers et al.

society at large. We can look to the Vietnam War and the peace movement, the women's movement,the disabilitymovement,andthe civil rightsmovementin the UnitedStates,to namejust a few examples.All of thiswas accompanied a broader by linguisticturnin the social sciences, a movementaway from methodologicalindiof and vidualism,andthe proliferation post-structural post-moder theories. Theintellectual workof combining socialtheories withlinguisticworkwas, atfirst, conducted a disparate However, by groupof scholars,each at theirown universities. in theearly 1990sa groupof scholars(Fairclough, Kress,vanDijk,vanLeeuwen,and Wodak)spenttwo days at a symposiumin Amsterdam discussingtheoriesandmethods specific to CDA. These scholarscame from somewhatdiverseacademicback(vanDijk, 2001). grounds,andCDA reflectstheirinterdisciplinary approach Education researchers turnedto discourseanalysisas a way to make sense of the contexts.Earlyexamplesof linways in whichpeople makemeaningin educational guistic analysis in education research grew out of the work of sociolinguistics 1976),linguisticanthropology (Gumperz,1982;Labov, 1972;Sinclair& Coulthard, of (Silverstein& Urban, 1996), and the ethnography communication (Gumperz& Sinclairand Coulthard, example, introducedan for Hymes, 1964; Hymes, 1972). elaborate framework coding teachers'and students'discourseacts in classroom for talk.Theirintentionwas to providean extensive structural model of discourseorganizationin classroominteractions. classic workof Cazden(1988/2001)grewout The of such descriptiveanalysesof classroomtalk. Aroundthe same time that scholars were describingthe micro-interactions occurredin classrooms,scholarsfrom that fields such as sociology and culturalstudies were also looking to classroomsand schools to theorizeaboutthe ways in which social structures reproduced are through educational institutions (Bourdieu,1979/1984;Bowles & Gintis,1976;Oakes,1986; Willis, 1977).1 Drawingon criticalsocial theory,these studiessoughtto examinethe of rituals,andtraditions ways in whichmacro-structures out in the interactions, play the classroom.Culturaltheorists,however, do not often turnto a close analysis of discoursestructures Bernstein,1971, for an exception).On the otherhand,lin(see and guistic anthropologists conversationanalystsoften do not turnto social theory or attemptto connecttheirmicro-levelanalyseswith broadersocial forces. Critical Discourse Analysis was an attemptto bring social theory and discourse analysis and togetherto describe,interpret, explainthe ways in which discourseconstructs, becomes constructed represents, becomes represented the social world. and by, by Duringthe past decade, educationresearchers increasinglyhave turnedto Critical Discourse Analysis as an approachto answeringquestionsaboutthe relationships between language and society. This proliferation in scholarship, as we demonstratein this review, poses a series of focused questions for education researchers interestedin CDA. Indeed,discourseanalysis of all types comes from fields outsideeducation,andmuch of it is tied to linguisticsin one way or another. As such work crosses into the boundaryof education,interestingand substantive concerns arise about how it is applied to educationalissues, how it affects other research and approaches in education, and how it might be reviewed in the non-educationresearchtraditionsfrom which it came. This articleprovidesa critical,integrativereview of CDA across five databases in the social sciences. We presenta review of the literature we interrogate the and methods, and implicationsof the literaturereviewed. We intend that this theory, review of CDA in the field of educationbe viewed in the context of the original 366

CriticalDiscourse Analysis in Education

CDA founders.The following questions frame our review: What happens when Critical Discourse Analysis crosses the boundariesinto education research?In what ways do educationresearchersuse CDA? How can the use of CDA in educationalcontexts informus aboutmethodand theory? Critical Discourse Analysis: Key Concepts Critical discourse studies stem from three overlappingintellectual traditions, each emphasizingthe linguisticturnin the social sciences. These traditions disare course studies (e.g., Benveniste, 1958/1971; Derrida,1974; Foucault, 1969/1972; Pecheux, 1975), feministpost-structuralism (e.g., Butler, 1990;Davies, 1993), and criticallinguistics (e.g., Fowler, Hodge, Kress, & Trew, 1979; Halliday& Hasan, 1989; Hodge & Kress, 1979/1993; Pecheux, 1975; Pennycook, 2001; Willig, 1999). Critical Discourse Analysis focuses on how language as a cultural tool mediates relationshipsof power and privilege in social interactions,institutions, and bodies of knowledge (see, for example, Bourdieu, 1977; Davies & Harre, 1990; Foucault, 1969/1972; Gee, 1999; Luke, 1995/1996). Gee (2004) makes the distinctionbetween the capitalized term "CriticalDiscourse Analysis"(which the abbreviation CDA represents)and "criticaldiscourse analysis"in lowercase letters,a distinctionthatis quite relevantto this review. He argues that CDA refers to the brandof analysis that has been informedby Fairclough, Hodge, Kress, Wodak,van Dijk, van Leeuwen, and followers. Lowercase "criticaldiscourseanalysis"includesa "widerarrayof approaches" 20)-Gee's (p. own form of analysis (1992, 1994, 1996, 1999), that of Gumperz(1982), Hymes (1972), Michaels (1981), andScollon, & Scollon (1981), andthe workof otherdiscourse analystsin the United States and elsewhere. These scholarsare conducting criticallyorientedformsof discourseanalysisbut do not specificallycall theirwork CDA. Gee (2004) points out that critical approachesto discourse analysis "treat social practicesin termsof theirimplicationsfor things like status,solidarity,distributionof social goods, andpower"(p. 33). Because languageis a social practice and because not all social practicesare createdand treatedequally, all analyses of languageare inherentlycritical. In the next section we discuss some foundationalprinciplesthatare relevantin around any discussionof CriticalDiscourse Analysis. The discussionis structured the key constructs:"critical," and "analysis" "discourse,"
What Is "Critical" About CDA?

The Frankfurt school, the groupof scholarsconnectedto the Instituteof Social Research at the University of Frankfurt, focused their attentionon the changing natureof capitalismandits relationto Marxisttheoriesof economic determinism. Adorno,Marcuse,andHorkheimer-the scholarsmost commonly connectedwith the Frankfurt School-initiated a conversationwith the Germantraditionof philosophical and social thoughtof Marx,Kant,Hegel, andWeber.While rejectingthe strict economic determinism(the view that economic factors determineall other aspects of human existence) associated with Marxism, they continued the view that injustice and oppression shape the social world. The Frankfurtschool and scholars from across disciplines engaged with critical theory and attemptedto locate the multiple ways in which power and dominationare achieved (Kinchloe & McLaren,2003). 367

Rogers et al.

Thus the Frankfurt school and other neo-Marxistscholars of society and language (e.g., the BakhtinCircle)openedthe debateaboutwhetherlanguagebelongs to the economic base or the culturalsuperstructure, whetherit is determined and materialconditions or, in fact, determinesthese conditions (Ives, 2004). It is by school was rising importantto rememberthat at the same time that the Frankfurt in academic popularity,the works of W. E. B. DuBois (1903/1990) and Carter Woodson (1933/1990) also mounted serious challenges to the dominant EuroAmerican scholarlyparadigm.However, Horkheimer,Adoro, and Marcuse are commonly associatedwith criticaltheory, whereasDuBois and Woodson remain invisible in the scholarlycanon in criticaltheory (Ladson-Billings,2003). This is importantbecause critical theory, a set of theories that attempt to locate and confront issues of power, privilege, and hegemony, has also been critiquedfor power knowledge relationsand constructingits own regime of truth. reproducing Or, as Yancy (1998) puts it, critical theory is often "the words of white men engaged in conversationswith themselves"(p. 3). Evidence of this can be seen in the strikingabsence of issues of race in much of criticaltheory. Criticaltheoryis not a unifiedset of perspectives.Rather, includescriticalrace it neo-colonialstudies,queertheory,and theory,post-structuralism, post-modernism, so on. Criticaltheoriesaregenerallyconcernedwith issues of powerandjustice and the ways thatthe economy, race, class, gender,religion,education,and sexual orior entationconstruct, social systems.Althoughtherearemany reproduce, transform different "moments"when research might be considered critical, the various to Criticaltheorists,for examapproaches criticalresearchsharesome assumptions. Facts constituted ple, believe thatthoughtis mediated historically powerrelations. by areneverneutraland are always embeddedin contexts.Some groupsin society are privileged over others, and this privilege leads to differentialaccess to services, goods, and outcomes.Anothersharedassumptionis thatone of the most powerful forms of oppressionis internalizedhegemony, which includes both coercion and consent (Gramsci,1973; Ives, 2004). Criticalresearchers intenton discovering are the specificsof domination However,powertakesmanyforms:idethrough power. Criticaltheoristsgenological,physical,linguistic,material, psychological,cultural. in of and erallyagreethatlanguageis central theformation subjectivities subjugation. the Post-structuralism, intellectual movement with which Michel Foucault is often associated,was a rejectionof the structuralist movementof the earlier20th assumedthatrelacenturyandis intimatelyrelatedto criticaltheory.Structuralism in tionshipsexisted between structures systems and thatexaminingthose relationships could help us to understand the entirety of a system. The theory of structuralismpermeatedacross disciplines and could be seen in studies of the economy (Marx),language (Saussure),psychology (Freud),and anthropologyspecifically,cultureandkinshiprelations(Levi-Strauss).Foucault,once himself a brokefrom structuralism arguedthatwe cannotknow something and structuralist, based on a system of binariesand static relationships.Post-structuralism pointed out the inevitable slipperiness of social constructs and the language that constructedand representedsuch constructs (Peters & Burbules, 2004). Foucault's (1969/1972) concept of discourse and power has been importantin the development of CDA, as discussed in the next section. Scholarswho situatethemselves within the CDA traditionoften separatetheir work from otherforms of "non-critical" discoursesanalyses by arguingthattheir 368

CriticalDiscourse Analysis in Education

of analysesmove beyond descriptionand interpretation the role of languagein the social world, towardexplainingwhy andhow languagedoes the work thatit does. Criticaldiscourseanalystsbegin with an interestin understanding, uncovering,and conditions of inequality. The startingpoint for the analysis differs transforming depending on where the critical analyst locates and defines power. Critical discourse analystslocate power in the arenaof language as a social practice.Power, however, can take on both liberatingand oppressiveforms. WhatIs Discourse in CDA? Recentdevelopments Critical in DiscourseAnalysisarerootedin muchlongerhistoriesof languagephilosophy(Austin,1962;Gramsci,1973;Searle, 1969;Wittgenstein, 1953); ethnomethodology (Garfinkel,1967; Cicourel, 1974), the functional in tradition theUnitedStates(Gumperz, & 1982;Silverstein Urban,1996), linguistics and SystemicFunctional Linguisticsin England,Canada,andAustralia (Halliday& Therearemanysubsections discourseanalysiswithinthe socialtraof Hasan,1976).2 dition, includingspeech act theory(Goffman, 1959, 1971), genre theory(Bakhtin, 1981; Martin,1985;Hasan& Fries, 1995), intertextuality (Bakhtin,1981; Kristeva, 1980, 1986, 1989;Lemke, 1992), discursiveformations (e.g., Foucault,1972, 1979, 1981; Lemke, 1992), conversationanalysis(Collins, 1986; Gumperz,1982; Sacks, Schegloff,& Jefferson,1974;Schegloff,Ochs,& Thompson,1996),narrative analysis (Gee, 1992, 1994;Labov, 1972;Michaels,1981;Propp,1968;Scollon& Scollon, 1981; Wortham, 2001), discursivepsychology (Davies & Harre,1990; Edwards& of Potter,1992), ethnography communication analysis (Hymes, 1972), multi-modal (Gee, 2003; Hodge& Kress,1988;Kress& vanLeeuween,1996;Scollon& Scollon, 2003), andcriticaldiscourseanalysis. Theword"discourse" comes fromtheLatindiscursus,whichmeans"torunto and fro."The word "current" comes from the same Latinroot.Withina CDA tradition, discourse beendefinedas languageuse as socialpractice. has Thatis, discourse moves backandforthbetweenreflecting constructing socialworld.Seen in thisway, and the languagecannotbe consideredneutral,because it is caughtup in political, social, formations. CDA is whatFairclough racial,economic,religious,andcultural (1992) has referred as a textuallyoriented to formof discourseanalysis(TODA).To develop this textualanalysis,Faircloughbroughttogetherthe linguistictheoryof Systemic Functional Linguistics(Halliday& Hasan,1976;Halliday,1985) withthe social theory of discourseas it evolved in the workof Foucault(1969/1972, 1979, 1981). Systemicfunctionallinguistics(SFL)explainslanguageuse in termsof the form andfunctionof interactions. can SFL theoristsposit thatevery interaction be understood at threelevels: textually,interpersonally, situatedin a widersocietalconand text.Furthermore, languageuserswe choose fromthe meaning-making as potentials thatare availableto us to representand constructdialogue.Thus languageuse is a creativepractice.Young and Harrison(2004) point out that SFL and CDA share severalcharacteristics. First,both view languageas a social construction.Second, bothview languagedialectically,whichmeansthatlanguageinfluencesthe contexts in which it occurs and the contexts influencelanguageproduction.And third,both emphasizethe culturalandhistoricalacts of meaningmaking. Foucault' theoriesof discoursehave had a tremendous s impacton the social sciences. Foucault ultimately rejected the tenets of structuralism(that there exist binarydistinctionsbetween constructsand that we could remove ourselves from 369

Rogerset al. the structureof language) and began the intellectual movement known as poststructuralism. Foucaulttheorized that the traditionaldistinction between speech and language (parole and langue) did not provide explanatorypower. Rather, Foucault sought to understandthe history and evolution of constructsthat were considerednatural(normality,justice, intellect, and so forth) and how such constructsarea productof power/knowledgerelationships.Ordersof discourse,a key constructin Foucault'sunderstanding social practices,are the discursivepracof tices in a society or institutionand the relationshipsamong them. Faircloughdistinguishesbetween Foucault's analysis of discourseand his own approach,which he refersto as a textuallyorientedapproachto discourseanalysis (TODA). for Gee's (1996, 1999) theoryof discoursehas been particularly important eduin cation researchers the United States. Gee's theoryis inherently"critical" the in sense of assertingthatall discoursesare social and thus ideological, andthatsome discoursesarevaluedmorethanothers.Gee distinguishedbetween ("littled") discourse and ("big D") Discourse. "Big D" Discourse refers both to language bits and to the culturalmodels thatare associatedwith Discourses. For instance,there is a universityDiscourse thatincludes certainlanguagebits thatmay be particular to academia,and there are also associated ways of thinking,believing, and valu"Littled" in ing thatareconnectedwith membership the Discourseof the university. discourserefers to the linguistic elements-the languagebits-that connect with such Discourses.Of course,the languagebits (little d, discourse)andthe social and culturalmodels (big D, Discourse) areconstitutiveand worktogetherto construct, maintain,and transforminteractions.The importantthing to keep in mind about Discourse (both big and little d) is that they are social and political and have histories of participation are saturated power relations. that by CDA brings togethersocial theory and textual analysis. To provide a succinct overview of the sharedassumptionsaboutdiscourseheld by manywithinthe CDA tradition,we turnto Faircloughand Wodak (1997), who outlinedcommon tenets here: of discourseundera criticalumbrella,paraphrased * Discourse does ideological work. * Discourse constitutessociety and culture. * Discourse is situatedand historical. * Power relationsare partiallydiscursive. * Mediationof power relationsnecessitates a socio-cognitive approach. * CDA is a socially committedscientific paradigmthat addressessocial problems. * Discourse analysis is interpretive,descriptive, and explanatoryand uses a "systematicmethodology." * The role of the analystis to study the relationshipsbetween texts and social practices. WhatIs the "Analysis"in CDA? There are many approaches to CDA, including French discourse analysis (Foucault, 1969/1972; Pecheux, 1975), social semiotics (Hodge & Kress, 1988; Kress, 2003), sociocognitive studies (van Dijk, 1993), the discourse historical method(Wodak,1996;Wodak,Meyer,Titscher,& Vetter,2000), andmulti-modal methods(Hodge & Kress, 1988; Kress& van Leeuween, 1996). CDA departsfrom 370

CriticalDiscourse Analysis in Education

discourseanalysisandsociolinguisticanalysesin its movementfromdescription and to constructs versionsof interpretation explanationof how discoursesystematically the socialworld.Furthermore, of criticalanalysespositionsubjectsin relations power and thananalyzinglanguageas (bothliberatory oppressiveaspectsof power)rather a way of explainingthe psychologicalintentions,motivations,skills, andcompetencies of individuals(Luke, 1995/1996).Eachof these perspectiveson CDA has been appliedto relevantsocial problemsin a wide rangeof disciplines,includingpolicy, social work,linguistics,andeducation.Eachperspectivehas developedits own set of analytictools thatmightbe broughtto bearon a set of problemsor questions. Fairclough(1989, 1992, 1995) outlined a three-tieredframeworkthat is very common among critical discourse analysts (see Fairclough, 1989, for a visual heuristic of this framework).The frameworkincludes analysis of texts, interactions, and social practices at the local, institutional,and societal levels. The first goal of the analystis to describethe relationships amongcertaintexts, interactions, and social practices(this is accomplishedby describingthe grammatical resources that constitutesuch relations,an issue to which we will return).A second goal is to interpretthe configurationof discourse practices. A third goal is to use the to descriptionand interpretation offer an explanationof why and how social practices are constituted,changed,and transformed the ways thatthey are. in Fairclough'sanalyticframeworkis constitutedby three levels of analysis:the text, the discursive practice, and the socioculturalpractice.In other words, each discursiveevent has threedimensions:It is a spokenor writtentext, it is an instance of discoursepracticeinvolving the productionand interpretation texts, and it is of a partof social practice.The analysis of the text involves the studyof the language structuresproducedin a discursive event. An analysis of the discursive practice involves examining the production,consumption,and reproductionof the texts. The analysis of socioculturalpracticeincludes an explorationof what is happensocioculturalframework. ing in a particular Analysis at the textual level involves use of Halliday's systemic functional and linguistics andthe threedomainsof ideational,interpersonal, textualanalysis. The ideationalfunctions include meta-narratives circulatein society. Analythat sis at this level includes transitivity,which involves the different processes, or types of verbs, involved in the interaction.The interpersonalfunctions are the in meaningsof the social relationsestablishedbetweenparticipants the interaction. Analysis of this domainincludes an analysis of the mood (whethera sentenceis a statement,question, or declaration)and modality (the degree of assertivenessin the exchange). The textualdomaininvolves the thematicstructure the text. of Fairclough's second dimension, discursive practice, involves analysis of the distribution,and consumption.This dimenprocess of production,interpretation, sion is concernedwith how people interpretand reproduceor transform texts. The third dimension, sociocultural practice, is concerned with issues of power-power being a construct that is realized through interdiscursivityand hegemony. Analysis of this dimension includes explorationof the ways in which discoursesoperatein variousdomainsof society. Proliferationof CDA in Educationresearch Critical discourse analysts tend to work on applied topics in a wide range of domains, including political discourse, ideology, racism, economic discourse, 371

et Rogers al. advertisement with promotionalculture,medialanguage,gender,institutionaldiscourse, education,and literacy (Blommaert& Bulcaen, 2000). This is seen in the proliferatingnumber of journals, conferences, and special editions of journals devotedto CDA. ThesejournalsincludeLanguageand Politics; CriticalInquiries in Language Studies:An InternationalJournal; CriticalDiscourse Studies; Text; Linguistics and Education; Language and Society; Discourse & Society; Discourse Studies; and Discourse. In addition, there are many online resources for critical discourse studies, including Critics-Land Language in the New Capitalmaintained ism, to namejust two. The Linguist List (http://www.linguistlist.org/) at Wayne State University, is a very accessible website with many resourcesfor discourse studies, including book reviews, majorconferences,journals, a list of 20 linguists,andlanguageresources.A studygroupof approximately people meets at majorliteracyconferences.Thereareuniversityprogramsestablished regularly for the study of critical discourse at the University of Lancasterand a minor at Alfred University. Two inauguralCriticalDiscourse Analysis conferences were held in 2004. The firstInternational Conferencein CriticalDiscourseAnalysis was held in 2004 (http://www.uv.es/cdaval/) Valencia, Spain.The School of Educain tion at IndianaUniversity held the first U.S. conference devoted to CDA in June of 2004. In December of 2004, the National Reading Conference (NRC) held a series of workshopsfocused on methodology, and CDA was the focus of one of the sessions (Burs & Morrell,in press). In the same year, the NationalCouncil of Teachersof English held a pre-conferenceworkshopdevoted to CDA. CDA has not gone withoutcritique,andthe critiquesarepartof the overall context in which we intend this review to be read. The three most common critiques are (a) thatpolitical and social ideologies arereadinto the data;(b) thatthereis an imbalance between social theory, on the one hand, and linguistic theory and method, on the other; and (c) that CDA is often divorced from social contexts (Flowerdew, 1999; Price, 1999; Schegloff, 1999; Widdowson, 1998). How does CDA conductedin educationalcontexts hold up to these critiques?To answerthis databaseof educationresearchusing CDA. question,we reviewedthe proliferating Methodology Review of Databases We reviewedfive databasesin the social sciences with the searchterm"critical discourseanalysis"from the years 1980 through2003. The databaseswere Web of Science, MLA, PsycINFO, ERIC, and ArticleFirst.We also used bibliographic from otherresearchers. reviewed 1991-2003 abstracts We branchingand referrals of articlesfromLinguistics Education(Vols. 3-14), the tablesof contentsof Disand in course & Societyfrom 1993 through2003, andthe abstracts Languagein Society from 1998 through2003. We reviewed only researchthat was publishedin peerreviewed journals.We requiredthat the authorsuse the terms "criticaldiscourse abstracts. analysis"somewherein the article.We did not review dissertation We integrated important books throughout the review where appropriate, because emerging theories and researchoften appearin books first, and later in articles. Examples of such books are Critical LanguageAwareness (Fairclough, 1992); An Introductionto Discourse Analysis (Gee, 1999);AnalyzingDiscourse: Textual Analysis for Social Research (Fairclough, 2003); Discourse in Late 372

Critical Discourse in Analysis Education Modernity (Chouliaraki & Fairclough, 1999); Classroom Discourse Analysis (Christie, 2002); An Introductionto Critical Discourse Analysis in Education (Rogers,2004); Literacyand Literacy:Texts,Power, and Identity(Collins & Blot, 2003); SystemicFunctional Linguisticsand Critical Discourse Analysis: Studies in Social Change(Young & Harrison,2004); andA CriticalDiscourse Analysis of Family LiteracyPractices (Rogers, 2003). We reviewed only studies that were conducted in or that pertainedto formal education (in classrooms) or informal education (e.g., after-school programs, museums, family literacy programs) or that pertained to an educational issue (policy documents).Ourrationalewas thatwe wantedto see the rangeof perspectives, approaches,and theoriesin the pool of researchthat specifically referredto CDA. The searchterm"Critical Discourse Analysis"resultedin a total of 803 references. Many of these articlesinclude criticalperspectives,criticalthinking,and discourse analysis. We read all of the abstractsto determinewhetherthe authors were using CDA as a theoryor methodand not simply providingcriticalperspectives on discourse analysis or critical thinking and discourse analysis. This next level of analysisfound 284 works thatused CriticalDiscourse Analysis. Of these, 56 were situatedin the discipline of education.Of those 56 articles, 16 were overlappingreferencesacross the databases.Therefore,the original searchresultedin a total of 40 articles that used CDA in the context of education.We collected an additional6 referencesthroughbibliographicbranching. AnalyticProcedures We developeda codebookto standardize reviews(see AppendixA). We used our ourresearch to questionsandeachstudy'sfeatures developa codingscheme.We also includedaspectsof CDA thatwere relevantto researchin education(theoryof disfor we course,implications education). retrieval, used samplestudDuringliterature ies to refinethe coding scheme.Afterreviewingand coding a subsetof the studies, we selected10 studiesto determine We interrater partsof reliability. eachhighlighted the articlethatdealt with the issues in the codebook.Each of the articleswas read twice-once by the lead researcher once by a research assistant. The codebooks and were comparedfor reliability.All disagreements were discussedand resolved.Our That and analysiswas ongoing,informedby the literature, constant-comparative. is, as we reviewedstudieswe soughtout similarities differencesacrossthe studies and andmadenote of themes.Once all of the articleswere reviewedand the codebooks filledout,we beganto summarize each of the articles(see AppendixB). This level of analysishelped to clarify trendsin the data.Fromthere,we pulled out four major themes(whichwe reporton below) thatranacrossall of the articles.In addition,we askedtwo scholarswho workin CDA to reviewthe summary chart(Appendix and B) tryto suggestotherwritingsthatwe mightincludein the review. Limitations We do not claim to have includedevery articleon CriticalDiscourse Analysis and education, particularlyresearch published after 2003. We have taken on a review of researcharticlesin educationthat explicitly define themselves as CDA andareset in an educationalcontext.Because CDA is a relativelynew "discipline" (whetherit might be or should be considereda discipline is open for debate), we sought to bring togetherdiverse lines of educationresearchto take stock of what 373

Rogers et al.

had been done so far. In doing so, we have inevitably made the field seem more synthesized than it really is. However, we maintainthat the presentis a suitable point in the historyof CDA in educationresearchfor such a synthesis. In limiting our review to studies that have explicitly called themselves CDA or Critical Discourse Analyses, we have inevitably left out importantlines of scholarship-lines that include discourse analyses conducted from critical perspectives and those that assume that all language is ideological and thus critical. Many articles have multiple perspectives and draw on social semiotics, hermeneutics, intertextuality, post-structuralism,popular culture, and media studies thatbringtogethervariouscriticaltheories and modes of discourse analysis. Authorswho write within these traditionshave shapedthe types of analyses that have been conducted. We also did not review studies in intertextuality, though we realize that important work has been done in this subset of CDA (Beach & Anson, 1992; Fairclough, 1992; Hartman,1992; Kamberelis& Scott, 1992; Lemke, 1992; Short, 1992). We recognize that what we have offered in this review is a modest synthesis of currenteducation researchthat is informed by and informs CDA. Organizationof the Review In the following sections, we presentthe findings of the review. The first section is a summaryof the findings across all of the reviews. Next, we presentfive of the themes thatran across all of the studies.Table 1 is a summativetable of the findings from the review. We provide a descriptionof the findings that emerged from each theme and subset of the theme and identify studies that illustrateeach theme. To be as succinctas possible, we describeonly those studiesthat particular best representthe findings.In some cases, studiesillustrated more thanone theme; thereforewe describe the study underthe theme it mostly illustrates.Finally, we summarize all of the results in a discussion section, with particularattention focused on implicationsand futureresearchwith CDA in education. TABLE1
Findings by theme

Theme Articles reviewed


Findings N= 46
N= 39

Theoretical articles N= 7 of Modeof language empirical in articles 66%(26/39)Interactional (analysis spoken of Theory language

of 33%(13/39)Analysis written language of 28%(11/39)No theory language

language,or spoken and writtenlanguage)


85% (22/26) Took place in middle school, high school, or highereducation 15%(4/26) Took place in elementaryschool or with childrenunderthe age of 10 20% (8/39) Empiricalarticlesdid not comment

on theiranalytic procedures


Findings AppendixB is a descriptivechartof the findingsfrom the 46 articlesreviewed, organizedby the main sections of each of the articles(definitionof CDA, research focus, context,datasources,anddataanalysis).Thereis an interdisciplinary group of scholarsusing CDA to analyze and theorizeabouteducationalissues. We have designatedthe geographiclocation of the authorsnext to theirnames in this chart. The type of articleis abbreviated either"E"(empirical)or "T"(theoretical). as The chartdemonstrates thatwhile all of the articleswe reviewed were situated within an educationalcontext or pertainedto educationalissues, therewas a great deal of diversityin the focus of the articles.The researchfocus of these articlesvaried from exploringthe relationshipbetween personhoodand literacy,to how history standardsare presented to the public, to how knowledge is constructedin chemistryclassrooms. In what follows we reporton the majorthemes across the articles.We begin by exploringthe multipleways in which CDA has been definedin educationresearch. Next, we explore whetherand how educationresearchersusing CDA have overcome the written language bias that historically has characterizedCDA. From there, we explore the context in which CDA work is situatedand the relationship of CDA to context.We thenturnto the questionof methodsandthe ways in which education researchersusing CDA have taken up the methodological aspects of CDA. Reflexivity is an importantaspect of any criticalwork, and in the next section we illustratethe ways in which educationresearchershave dealt with issues of reflexivity. Finally, in the discussion, we turn to the findings of the articles reviewed to answerthe question,Whatdo we know as a resultof CDA work done in educationresearch? TheMultipleMeanings of CDA As Fairclough and Wodak (1997) pointed out, there are many different approachesto CDA, including Frenchdiscourse analysis (Foucault, 1969/1972; Pecheux, 1975), social semiotics (Hodge & Kress, 1988; Kress, 2003), sociocognitive studies(van Dijk, 1993), andthe discoursehistoricalmethod(Wodak, 1996; Wodak,Meyer, Titscher& Vetter,2000). Each of these perspectiveson CDA has been appliedto relevantsocial problemsin a wide range of disciplines including policy, social work, linguistics, and education. Despite the many different perspectives of CDA, most of the researchwe reviewed drew mainly on Fairclough to that (1989, 1991, 1993, 1995). We were surprised despiteWodak'scontribution the development of CDA as a theory, method, and research program(Wodak, 1996; Wodak,Meyer, Titscher,& Vetter,2000, Wodak & Reisigl, 2001) and her work as the directorof the WittgensteinResearchCenteron Discourse, Politics, andIdentity,therewere very few referencesto her (see Corson,2000, andRogers, 2003, for exceptions). The articlesreviewed here definedCDA in four ways. First,they definedCDA in relationto post-structuralism. is clear that CDA work in educationresearch It continues to draw on the relationshipbetween CDA and post-structuralism, parfeminism and Foucault.While CDA drawsheavily on ticularlypost-structuralist post-structural theory, Fairclough (1995) made a distinction between CDA and Foucault's theory of language. He aimed for CDA to be a textually oriented 375

et Rogers al. discourse analysis (TODA), whereas post-structural analyses were often lacking in close textualanalysis. Second, the articlesdefinedCDA in termsof its goals, aims, or functions.The articlesthat defined CDA in such terms assertedthat aims of CDA are to disrupt discourses, challenge restrictivepedagogies, challenge passive acceptanceof the status quo, and reveal how texts operate in the constructionof social practices. More researchtendedto define anduse CDA as a tool of critiquethanas a tool for re-imaginingthe social world. A thirdgroupof authorsdefinedCDA on the basis of its associationwith SystemicFunctionalLinguistics,criticallinguistics,or interactional sociolinguistics. A fourthset of authorsdefined CDA througha description of the analyticframeworkthatthey employed. Each of the authorsreferredto the CDA framework as a three-tiered framework and made reference to Fairclough's work. Some authors merged Fairclough's description with other frameworks(Chouliaraki,1998; Collins, 2001). All of the authorsseem to agree thatthe framework bringstogethera microandmacroanalysisandoffers a descripand explanation of social events. Three articles mentioned tion, interpretation, CDA but did not define it. Mode of LanguageAnalyzedand Theoriesof Language: OvercomingWritten LanguageBias CDA sets out to describe,interpret,and explain the relationshipsbetween language, social practices, and the social world. Languageindexes social relations, expresses social relations,constitutessocial relations,and challenges social relations. Language,in this framework, dialogic, intertextual, historicallybased. and is CDA has been seriously critiquedfor failing to addressinteractionalor dialogic texts andfocusing insteadprimarilyon writtentexts (newspapers,lists, policy documents, health care documents).Teo (2000) wrote, "CDA typically concentrates on data like news reporting,political interviews, counseling, andjob interviews that describe unequal encounters, or embody manipulativestrategies that seem neutralor naturalto most people" (p. 12). Similarly,Rampton(2001) pointed out thatinteractionand dialogism are rarelybroughtout in CriticalDiscourse Analysis. We wondered, as we began this review, if this critique would hold up with analyses conductedin educationalcontexts. It did not. It appearsas if education researchersusing CDA are beginning to overturnthis critiqueas more and more studies are using CDA with interactional data. Of the 39 empirical articles dataor interdata(eitherjust interactional reviewed,26 (or 66%)used interactional actional data and writtendata). See Appendix B for a descriptionof articles that includedeitherwrittenor interactional data. data While an impressivenumberof studiesfocused on analysesof interactional (ratherthan on writtentexts), the analystsdid not frametheir analyses within the historyof discourseanalysis and socio-linguistic analysis.CDA has also been critiqued for not paying attentionto socio-linguistic predecessors(Schegloff, 1993; Sawyer, 2002). A few studies mentioned the relationshipbetween different discourse analytictraditions.Peace (2003), for example, discussed the pros and cons and of "top-down"(critical discourse analysis that draws on post-structuralism) and conversationanalysis) theories of language "bottom-up" (ethnomethodology and assertedthat "bothapproachescan be problematic" 164). While it is true (p. thatthe two approaches discourseanalysishave some incompatibletenets, most to 376

Critical Discourse in Analysis Education criticalanalyses drawon elements of earlierdiscourseanalyses but do not explicitly mentionCDA's connectionwith otherforms of discourseanalysis (for exceptions, see Collins, 2001; Heller, 2001; Moje, 1997; Rampton,2001). We then wondered about the relationshipsbetween the type of text analyzed (written,interactionalor a combinationof writtenand interactional)and the theory of language broughtto bearon the analysis. We learnedthat emphasisplaced on theoriesof languagevariedwidely acrossthe studiesfrom a carefuldescription of post-structural theoriesof discourseand SFL to a descriptionof post-structural discoursetheory or a descriptionof SFL, to no descriptionof language at all. We found this surprisingbecause CDA is a discursively based framework,and we expected there to be more careful attention to and description of theories of language. A numberof studies, particularlythose conductedin the United States and in literacy studies, collapsed Gee's theories of discourse under that of critical discourse analysis (Brown & Kelly, 2001; Egan-Robertson, 1998; Hinchman & Young, 2001; Rogers, Tyson, & Marshall,2000; Johnson,2001; Orellana, 1996; Young, 2000). While Gee's discourse theory and analysis assumes language is political and social and thus "critical,"he does not refer to his brandof discourse analysis as CDA, a point that he made specifically in Gee (2004). Nevertheless, Rogers, Tyson, and Marshall (who, in a 2000 study of three children, their families, teachers,and principalsacross two schools, explore the interplayof discourses, or living dialogues, in their neighborhood)classify Fairclough, (1989, 1992), Gee (1996), and Lemke (1995) togetherundercritical discourse theories. Johnson(2001), in a studyof pre-serviceteachers' visual narratives Furthermore, of a student teaching experience, draws on Gee's theory of discourse in his definition of CDA. Egan-Robertson(1998), in a study of how personhood is communicated through writing in a community writing program, cites Gee's theory of discourse. Of the studies reviewed, 28% (1 lof 39) do not addresslanguage theory at all. Bartu (2001), Bergvall and Remlinger (1996), Comber (1997), Collins (2001), Kumaravadivelu(1999), Thomas (2002), Stevens (2003), Johnson and Avery (1999), and Fox and Fox (2002) all lack a discussion of language. One area that criticaldiscourseanalystsneed to be more conscious of is thatthe theoriesof lanbased on Europeanlanguages.This is guage thatarebeing used arepredominantly in because, as we will demonstrate the next section, CDA is often used important in workwith historicallymarginalized groupsof people, and such groupsarelikely to have linguistic variation at the syntactic and morphological level as well as discourse patternsthat may not be accountedfor in a European-language-based discourseframework.We returnto this point in the discussion. CriticalDiscourse Analysis in Context An ongoing discussion in the journalDiscourse & Society has focused on the relationshipbetween conversation analysis and CDA (Billig, 1999; Schegloff, 1999). The big questionis how much of the context-beyond the here and now of the interaction-is important, necessary,to understanding interaction.Critor the ical discourse analysts pay attentionto the macro context-the societal and the institutionalas well as the local level of a text and the grammatical resourcesthat make up the text. Conversationanalysts, on the other hand, believe all that is 377

Rogers et al.

relevantis the "hereand now" of the interaction,not what came before or afterit. This groupof scholarsarguesthatCDA does not attendclosely enough to the linguistic resourcesthatconstituteinteractionsbut insteadfocus on how macrorelations are mapped onto micro interactions (Billig, 1999; Widdowson, 1998). Contextalso has been important because CDA has often been critiquedas "outof context," meaning that bits of texts and talk are analyzed outside the context of theirproduction,consumption,distribution, reproduction. and CDA has also been critiquedby anothergroupof scholars(primarilylinguistic anthropologists)for not paying enough attentionto ethnographiccontexts-the criticism being that the analyses are often based on decontextualized texts (speeches, policy documents,excerpts of talk) ratherthan on grounded,interactional datathatoccurwithin a largerframeof interactions(see Critiqueof Anthropology, volume 21, issues 1-2 for an in-depthdiscussion of this issue). It appears that CDA conductedin educationalcontexts may offer a way out of this theoretical and methodologicalquagmire.As we describe in the following sections, education researchersare bringing CDA frameworksinto a variety of educational settingsand askingquestionsthatdemandattentionboth to the linguistic details of the interactionand to the larger social, historical, and culturalcontexts in which the interactionsemerge. As was mentionedearlier,33% (13 of 39) of the studies reviewed for this article were analyses of writtentexts where the context was the text itself, for example, policy documents, newspaper articles, textbooks, and transcriptsof videos 2001; Collins, 2001; Hays, 2000; Luke, 1997; (Ailwood & Lingard,2001; Barnard, Pitt, 2002; Stevens, 2003). However, all of the studieswere located in educational contexts (meeting,classroom,interviews,writingclub). Of the interactionalstudies, 85% took place in middle school, high school, or higher education settings. Only 15% (4 of 26) took place in elementaryschools (Gebhard,2002; Orellana, 1996; Rogers, Tyson, & Marshall, 2000; Young, 2000). Of all of the empirical studies (39), 15% (6 of 39) of the studies were set in a higher educationcontext (Corson, 2000; Fairclough, 1993; Fox & Fox, 2002) or in university classrooms (Bartu,2001; Bergvall & Remlinger, 1996; Heberle, 2000). We found thatthe studies covered a wide range of contexts, including science classes (Moje, 1997; Myers, 1996), a social studies class (Brown & Kelly, 2001); literature classes (Hinchman & Young, 2001), after-school programs (EganRobertson, 1998; Rogers, 2002c), home schooling experiences (e.g., Young, 2000), interviews(e.g., Collins, 2001; Nichols, 2002), special educationmeetings (Rogers, 2002b), administrativeschool meetings (e.g., Corson, 2000; Orellana, 1996), or written documents (e.g., Ailwood & Lingard,2001; Anderson, 2001; Davis, 1997). or All of the empiricalstudies(100%)used some formof anthropological ethnorecordedin fieldnotes,documentcollecgraphicmethod(participant-observation tion, anddebriefing)(Chouliaraki,1998; Comber,1997; Hughes, 2001; Hinchman & Young, 2001; Egan-Robertson, 1998;Rogers,Tyson, & Marshall,2000; Rogers, 2002a; Young, 2000), interviewsor focus groups(Brown & Kelly, 2001; Collins, 2001; Nichols, 2002; Peace, 2003; Johnson & Avery, 1999; Young, 2000). The studies varied in the detail and descriptionprovidedabout fieldwork(length and texts, interviews),andresearch duration),datasources (writtentexts, interactional participants (ethnicity,how they were selected). Some studiesprovideda clear and 378

CriticalDiscourse Analysis in Education

detaileddescriptionof their data sources (Comber, 1997; Egan-Robertson,1998; Hughes, 2001; Hinchman& Young, 2001; Moje, 1997; Rogers, 2002a; Rogers, Tyson, & Marshall,2000; Young, 2000). Otherslacked such descriptions. Some authors had innovative ways of including context in their analyses. Nichols (2002), in a study thatexploredthe genderednatureof parents'accounts of theirchildren,builtthreecontextsinto the interviewprotocol(memoriesof their own literacyexperiences,descriptionsof home literacypractices,andobservations of their children's literacy related behaviors). While Hays (2000) situated her analysis primarily on newspaper texts covering educational conditions in Botswana in SouthernAfrica, she did make reference to the ethnographicfieldwork that she had conductedthere in her explanationof the newspapertexts (references of this kind are rare in the analysis of writtentexts). Similarly, Stevens (2003), in a study of how the federal governmentdefined reading,combined her observationsof the ReadingLeadershipAcademy in 2002 with a textual analysis of the documents from that conference. Other studies (Bergvall & Remlinger, 1996; Chouliaraki,1998; Fox & Fox, 2002; Peace, 2003) recordedinteractional datain classroomsthroughparticipant observationor conductedinterviewsbut did that not specify the amountor the duration fieldwork.Peace (2003) also reported of someone else had collected the interviewsthathe analyzed. in The diversityof the researchparticipants represented these studieswas quite broad.As mentionedearlier,the vast majorityof articles focused on participants were of middle school age or older. Furthermore, most of the researchparticipants students(with the exception of Comber's 1997 analysis of a teacher).The ethnicity of the participantsvaried as widely as the contexts in which the studies across the occurred.See AppendixB for the diversityof the researchparticipants articles.Overall,the researchers definedcontext in termsof the field of study and and participants did not theorizethe role of context in conductingCDA. Whatwe learnis that CDA is being "putto work"in context, but the multiplemeaningsof context have not yet been theorized. The Questionof Methods Van Dijk (2004) has proposedchangingthe name CriticalDiscourse Analysis to CriticalDiscourseStudiesbecause the term"analysis"suggests thatresearchers are interestedmainly in analysis, without much theory-when, in fact, CDA is a combinationof theoryand method(van Dijk, 2004). As reviewed in the introducCDA in the social sciences, from semition, there are many ways of approaching otic, to historical,to multi-modalanalyses.The problemor object of studymay be shared,but the authorsare eclectic in theirmethods;thatis, they use methodsthat they thinkwill help themlearnmoreaboutthe problemunderstudy.Analystsbring a rangeof theoreticalandmethodologicaltools to bearon theirresearchproblems and perspectives. Researchersand scholarsof CDA vary on the questionof whetherthe analytic across researchor whetherstanproceduresof CDA shouldbe more standardized dardization runscounterto the epistemologicaland ontological tenets of a critical paradigm.Verschueren(2001) and Martin(2000), for example, arguedthat CDA shouldbe appliedmore systematicallyand more rigorously.Those who arguefor more systematicanalyticprocedures tryingto countercriticswho say thatCDA are researchers searchtheirdatafor examples of what they aretryingto prove, instead 379

et Rogers al. of lettingthe data"speak." These criticsrecommendthatcriticaldiscourseanalysts examine actuallanguagepatternswith some degree of explicitness and reconnect these patterns with the social and political themes that inform their work. In responseto the issue of a more systematicCDA, Bucholtz (2001) wrote: to criticaldiscourse analyAny attempt foolproofguidelinesin anacceptable sis will be defeatedby its own universalistic urge.... It is difficultto imagine whatmight constituteadequateformalanalysisin advanceof actually carryingit out: must all analyses attendto phoneticdetail?To syntactic structure? 176) (p. Others (Bucholtz, 2001; Gee, 1999) argue that there needs to be a diversity of approachesand that such diversity strengthensthe frameworkand the method. Ourreview of the literature of indicatedthatthe actualanalyticprocedures CDA were carriedout and reportedon (or not reportedon) in a vast rangeof ways. The authors used Fairclough'sthree-tiered discourseframeframework, post-structural works, or discourseanalysis (not CDA, despite calling theirproceduresCDA), or did not specify their analytic procedures.See Appendix B for a summaryof the analysis carriedout in each of the articles. Although all of the articles claimed to conduct a CriticalDiscourse Analysis, some presenteda discourse analysis, not particularly CriticalDiscourse Analya sis. Brown and Kelly (2001), in a studyof the narratives AfricanCanadianhigh of school studentswho discussed the relevancy of the social studies curriculumin theirclassrooms,arguedthat"theirgoal is to highlightandexaminediscourseproductionandinterpretation it intersectswith the 'life-worlds' of a particular subas ject grouping, i.e., the high-school student of African descent" (p. 503). The authorsprovidedconversationsbetween studentsbut did not include a discourse analysis of the conversations.Rather,they presentedthemes or social narratives ratherthanan analysisof the discursiveconstruction texts (eitherwrittenor spoof at ken). Discourse analysis in this sense seems to be interpreted the social rather than the textual level and does not attemptto move beyond descriptionto interpretationand explanation. One clusterof studiespresentedbroadthemesfromtheiranalysesandthenused the examplesof discourseto support themes(Bergvall& Remlinger,1996;Rogers, & Marshall,2000; Nichols, 2002; Peace, 2003; Tunstall, 2001). Rogers, Tyson, Tyson, and Marshalldescribedtheir analysis as categorizingutterancesinto three broadthemes. There was no mention of what constitutesan utteranceor how the themeswere determined. Nichols (2002) analyzedthe parents'interviewusing the broad themes and the research literatureto demonstratethe themes. Similarly, Peace (2003), in a studythatexploredthe ways in which women socially construct masculinitiesthroughcross-genderundergraduate studentgroupdiscussions,used a groundedapproach discourseanalysis.He wrote:"[T]theinterviewtranscripts to were readrepeatedly;as broadcategoriesbegan to emerge they were increasingly related to the literatureand analyzed in terms of what they may achieve" (pp. 165-166). He does not include for what specific linguistic propertieshe analyzed the texts. The question that these analyses raise is, Why did the authorchoose certain partsof the text to analyze and not others?It appearsfrom the analyticsections of these articles that the authorsassumed that, if they had a critical orientationand 380

CriticalDiscourse Analysis in Education

attendedto some aspects of language in their analysis, then they would be contradition,drawson ductinga criticaldiscourseanalysis. CDA, in a Faircloughian SFL thatassumes thatlinguistic form is relatedto linguistic functionand thatcertain categories of linguistic functions do particularsocial "work."However, the authorsare not clear on how an analysis of transitivityrelates to the ideological commitmentof a text. They are also not clear abouthow over-lexicalization(the of availabilityof manywordsfor one concept)relatesto the representation the hisin tory standards one way but not in another.Several studies combine social theories with the CDA framework(Chouliaraki, 1998; Collins, 2001; Woodside-Jiron, 2004). Collins mergedthe NaturalHistoriesof Discourse framework(referential, and framework interactional, metadiscursive levels) with Fairclough'sthree-tiered (textual,discursive,and society-wide). In general,the authorsused aspectsof Fairclough'sthree-tiered framework but failed to specify what linguisticresourcesaccompanywhich set of functions.This may be, in part,becauseFaircloughdid not specify in his earlierwork(1992, 1995) whatgrammatical resourcescorrespond each level of analysis.However,Chouto liaraki and Fairclough (1999) and Fairclough (2003) moved closer to the SFL frameworkand described the specific linguistic resources that may be used for CDA levels. Overall,there was lack of conanalysis at each of the corresponding nectionbetween linguisticpractices,social practices,and wider social formations. Twenty percent (8 of 39) of the empiricalarticles did not describe their analytic proceduresat all (Ailwood & Lingard,2001; Barnard,2001; Bartu,2001; Brown & Kelly, 2001; Fox & Fox, 2002; Hughes, 2001; Pitt, 2002; Thomas, 2002). We returnto discuss this in the concludingsection.
Reflexivity and Role of the Researcher

Chouliaraki Fairclough(1999) cited reflexivityas an important and agendafor CDA research.Similarly,Bucholtz (2001) called for a heightenedself-awareness in discourse analysis. She called for a reflexivity where, "the analyst's choices at every step in the researchprocess are visible as a partof the discourse investigation, and critique does not stop with social processes, whether macro-level or micro-level,butratherextendsto the analysisitself' (p. 166). Reflexivityincludes at least three aspects: participatory constructionof the researchdesign, reciprocity, and turningthe analyticframeback on the researcher. Reflexive intentionsvary from buildingrigorin the researchto questioningthe authenticityof the researcher(Alvesson & Skoldberg, 2000). The intention of reflexivity depends on whether researchersview their aim as strengtheningthe rigorof social science researchor questioningthe epistemologicalandontological foundationsof the knowledgeclaims thatcan be made.Forexample,Myerhoffand communicativeproductsso thatthe Ruby (1982) define reflexivityas "structuring audienceassumes the producer,process and productare a coherentwhole ... scientistshave also been engagedin reflexiveactivities... scientistscontinuouslytest their own assumptions and procedures"(pp. 6-9). This statement implies that being reflexive is synonymous with being scientific. While Bourdieu and Wacone's own quant(1992) call into questionthe ideological natureof "monitoring" the thoughtsand actions, their reflexive intentionis to "strengthen epistemological moorings"of the research(p. 46). This intentionmight be viewed in much the same way as are traditionalclaims to validity, which often safeguardresearchers 381

Rogers et al.

from a self-reflexive research paradigm. That is, if we triangulate our data, member-checkwith participants, engage in peer review, establish and maintaina papertrailof ourtheorizingandanalyticmoves, we can claim thatourCriticalDiscourse Analysis is valid-or an accuraterepresentation "reality."Such a view of is problematic,especially in a CriticalDiscourse Analysis frameworkthat rejects the view of an objective and neutralscience. Reflexivity within a CDA frameworkarisesfrom a concernaboutthe stabilization of knowledge claims and the slipperinessof language.Thatis, the fundamental natureof languagehindersempiricalresearchthat is aimed at establishingthe and "truth." Indeed,Chouliaraki Fairclough(1999) assertthatreflexivityis caught up in social struggleand thatreflexivity assumes a discursive element that posits thatresearchersare partof the languagepracticesthey study. The intentionof the reflexive stance dependson the claims to knowledge and reality of the researcher on andthe extentto which the researchers these frameworks themselves,either turn methodologicallyor theoretically. Reflexivity is crucial in researchagendasinvolving CriticalDiscourse Analysis in educationresearch.Educationresearchersare often researchersof familiar educationalsettings.As membersandex-membersof the school communitiesthat we study, we bring with us (often successful) histories of participationin those institutions as students, teachers, and parents. Thus we have embodied what Fairclough(1992) refersto as "members'resources,"or what Gee (1999) refersto as "culturalmodels" around our participationin school that includes beliefs, assumptionsand values within these contexts. Thus the classic tension between in distanceandcloseness in the researchsettingis often blurred educationresearch. mostof the analyTo turnbackto thearticlesreviewedandtheissue of reflexivity, ses thatdealtprimarily with writtentexts did not includea high degreeof researcher 2001; Bloome & reflexivity(Ailwood & Lingard,2001; Anderson,2001; Barnard, Carter,2001; Fairclough,1993; Johnson& Avery, 1999; Luke, 1997; Pitt, 2002; often positionedthemselves as if Thomas,2002). In these studies, the researchers were outsidethe texts. Of course, we know thatthis is not true-and that any they discourseanalysisis a processof constructing meaning.Hays (2000) was a notable exception to this rule (see below). Although some studies involved interactional still analyses,the researchers did not locate themselvesin the research(Bergvall& 1998;Fox & Fox, 2002; Johnson,2001; Moje, 1997). Remlinger,1996;Chouliaraki, In a numberof studies, the researcherspositioned themselves mainly as text (Anderson, analysts,even thoughthey were clearlythe datacollection instruments 2001; Baxter, 2002; Corson, 2000; Hinchman & Young, 2001; Peace, 2003; Nichols, 2003; Hughes, 2001; Stevens, 2003). In her 2003 study, Stevens does not addressher role in the researchotherthannamingherself as the statereadingspecialist;however, in anotherpublicationshe does deal closely with mattersof CDA and reflexivity (Stevens, 2004). In other studies (Brown & Kelly, 2001; Collins, 2001; Young, 2000; EganRobertson, 2000; Rogers, 2002a, 2002b, 2002c), the authorsdo position themselves in the researchand comment on issues of reflexivity. Collins (2001), in a study of how teacherstake up (or resist) the discourses of educationalstandards and the ways in which the standardsecho larger socio-political educational reforms,presentshimself as a text analystor researcherand also as a memberof the educationalcommunitywithin the districtthathe is writingaboutand a parent 382

CriticalDiscourse Analysis in Education

of a child in the same district.Young (2000), in a study that explores how critical the literacyactivitiesin a home schooling settingsustainor transform participants' awarenessof genderedidentitiesand inequitiesin texts, writes the following: litAs a middleandhighschoolliteracyteacher, explored I manyalternative eracypracticesandinstructional optionsin an effortto findways to encourage studentsto become readers,writers,and learners.As a mother,I have fromtraoftenlongedformy sons' schoolliteracy to experiences be different ditionaltextbookmethods.(p. 312) Young (2000) squarelypositions herself in her researchas a motherand teacher, one aspect of reflexivity. She does not, however, turnthe criticaldiscourseanalysis frameworkback on herself to analyze how her participationin the research contributed the reproduction disruption powerrelations.Overall,we found to or of few examples of this type of reflexivity in the studies that we reviewed-an very issue to which we returnin the conclusion. At presentwe move to summarizeour findingsand point to implicationsand futureresearch. Discussion and Conclusion The Whatdo we know as a resultof CDA conductedin educationresearch? studies reviewedprovideeducationresearchers with a closer look at the ways in which educational issues are constructedand representedat micro and macro levels throughpublic documents,speeches, interactionsin classrooms,informalsites of data learning,andacrossthe lifespan.The emphasison interactional gives us insight into the ways in which the micro and macro contexts are linked togetherand the ways in which competingdiscoursescome into play. Indeed,in this corpusof studdisies we have seen how discoursesof educationdrawon hybridand intertextual courses, such as business and management(Anderson,2001; Comber,1997). A strongthreadrunningthroughmany of the findingswas the identificationof unintendedconsequencesof educationaldecisions, policies, and social practices. That is, educatorsoften intendedto open up liberatoryspaces in meetings, policies, teachingdecisions, and classroomlessons; but a closer analysisrevealedthat theiractionshad unintended consequencesthatresultedin furtheroppression(Ailwood & Lingard,2001; Chouliaraki,1998; Corson, 2000; Comber, 1997; Fox & Fox, 2002). Along the same line, the analysesthatwe reviewedprovideda detailed investigationof the subtletiesof power and privilege, the ways in which power is linked to historiesof participation variouscontexts, and how power is internalin ized rather thanreinforcedfromabove.Withthatsaid, most of the analysesfocused on the ways in which power is reproducedratheron how it is changed, resisted, towardliberatory ends. Luke (2004) arguesthat,historically,most andtransformed critical analyses have focused on uncoveringthe discursiveplaces where oppression and dominationoccur ratherthanon places of liberation.Luke (2004) stated: We need more researchand scholarshipthat documentsand analytically and explicatesanalysesthatfocus on affirmative, emancipating redressive texts anddiscoursepractices-turningour attention instanceswheredisto course appearsto lead systematicallyto the redistribution wealth and of power.(p. xi) We concurwith this assertion. 383

Rogers et al.

Overall, this review has outlined the major areas of emphasis, as well as the strengthsandweaknesses,of CDA in educationresearch.We can also returnto the common critiquesof CDA (political and social ideologies are read into the data; there is unequalbalance between social theory and linguistic theory;and CDA is often divorcedfrom social contexts) and ask, How do educationresearchers using CDA fare with these critiques? It appearsthatCDA thatis conductedin educationalsettings is moving toward overcoming written language bias. Indeed, 66% of the empirical articles analyzed interactionallanguage. Much of the researchthathas been conductedwith CDA outside the field of education has analyzed written texts (e.g., speeches, policy documents, letters, textbooks). In the context of education research, we have seen a shift from the analysis of writtentexts to the analysis of spoken texts. This shift could potentially reshape each of the levels of the CDA framework ("critical," "discourse," and "analysis"). As a result of bringing CDA into dynamiclearningsettings, researcherschange, modify, and adjustthe framework to suit the needs of their research designs and particularquestions. We might reflect on how researchers are shaped to think in certain ways because of the frameworks that exist and how the research that we are conducting is, in fact, reshaping the framework itself. This analytic move keeps CDA as a usable, reflexive framework,open to adjustmentsand adaptations,given the demandsof the research questions, the contexts, and the theoretical frameworks that are broughtinto line with it. While 66 % of the articles focused on interactionaldata, many of the articles did not providea clear descriptionof theirlinguistic framework-an oddity given thatCDA is a discourse-basedframework.Such unbalancedattentionto language theoryin CDA in educationresearchmay be due, in part,to the lack of trainingthat educationresearchersreceive in language studies. A real problem for education researchers who are interestedin CriticalDiscourse Analysis is theirrelativelack of experiencein dealing with the micro-structure texts. This is compoundedby of the relativelack of attentionto SFL in the Americancontext. methods All of the studiesthatfocused on interactional dataused ethnographic of participant observationrecordedin fieldnotes,interviews,documentcollection, and debriefingwith participants. Some studies includeddataacross time and contexts (Moje, 1997; Rogers,2002a). All of the studiesattendedto bothethnographic and linguistic contexts, althoughthe weight placed on one or the othervaries. The attention paid to local, institutional, or societal contexts varied as well. The researchin this review did not theorizethe role of contextbeyond the field of study and the participants the study. More theorizationof the role of context in critiin cal discoursestudies would be an important next step. Although most of the studies focused on what Luke (2004) calls the "deconstructive" rather thanthe "reconstructive" aspectsof power,the focus on classroom discourse and interactionaldata opens up possibilities for investigatingthe ways in which people resist and transformsocial relations towardemancipatoryends. Interactional datatendto be morehybrid(or less stable)thanwrittentexts andthus open to the possibility of change. More analyses of the intricacies of classroom and talk, within a democraticframework,could offer descriptions,interpretations, of how agency, productiveliterateidentities,and a sense of commuexplanations nity are formedand sustained. 384

CriticalDiscourse Analysis in Education

The focus on interactionsin classroomsin the studiesreviewed also resultedin a discussionof the role of criticaldiscoursestudies in learning.Severalof the articles discussed viewing learning in terms of changing discourse practices across time (Rogers, 2002b). Furthermore, Introductionto CriticalDiscourse AnalyAn sis in Education (Rogers, 2004), provides a collection of empiricalchaptersthat illustratethe ways in which CDA can illuminatelearningby studyingshifts in discourse practicesacross time and contexts. More researchis needed to investigate how shifts in discourse patternscan provide educators insight into the ways in which people of variousages learn. An overwhelming85% of the studies involved participants who were of middle school age or older. Only 15%of the studiesincludedparticipants who were in elementaryschool and under 10 years old. Ideologies are reproducedand transformed at very young ages. Therefore,descriptionsand explanationsof how this occurs and, more important,how the acquisitionof counterproductive ideologies is interrupted,are necessary. This suggests the importanceof extending CDA inquiriesto primarygrades.It also raisesthe questionwhetherthe critiquesof integratingcritical literacy into primarygrades extend to the usefulness of CDA as a theoreticaland methodologicalframeworkin primarygradeclassrooms. It was not surprisingto see in this review thatCDA was mostly used with participantswho have historicallybeen oppressed(e.g., women, AfricanAmericans, the poor and workingclasses). As Wodak & Reisigl (2001) pointedout: Languageis not powerfulon its own-it gains powerby the use powerful peoplemakeof it. This explainswhy CriticalLinguisticsoften chooses the of perspective those who suffer,andcriticallyanalyzesthe languageuse of thosein power,who areresponsible the existenceof inequalities who for and also havethe meansandopportunities improveconditions. 10) to (p. What was refreshing is that researchersin education also looked closely at the language of those who suffer (students,parentsin meetings, teachers)and found places of agency, creativity,and resistance.We need to proceed cautiously with conductingresearchon groupsof people who have been oppressedhistorically,as opposed to conductingresearchwith these people (an issue discussed earlier).The Luke (1995/1996) majorityof the studiesreviewedhere took the formerapproach. writes, "[W]hatis needed is a systematicattemptto build on minoritydiscourses in schools, classrooms,andotherpublic institutions" 39). We mightextend our (p. analyses beyond verbaldata to the nonlinguisticand emotional aspects of suffering, oppression,hope, and liberation. In the corpus of studies we reviewed, there were more analyses of gender (Bergvall & Remlinger, 1996; Pitt, 2002; Young, 2000) than of race (Brown & Kelly, 2001). The difference seems to be related to the ways in which race is silenced in educationresearch(Greene& Abt-Perkins,2003; Tate, 2003). Critical discourseanalyses should more consciously drawon the historyof scholarshipin Critical Race Theory (Bell, 2004; Crenshaw, 1988; Delgado, 1995; LadsonBillings & Tate, 1995; Tate, 1997), especially when engaging issues of race, because CDA frameworkstraditionally racism,and anti-racism.This is important drawon Euro-American bothin theoreticalandanalytic epistemologicaltraditions, frameworks.Such frameworkshave continuedto silence and oppresshistorically marginalizedgroupsof people. 385

Rogers et al.

The studies that we reviewed included multiple analytic methods. However, they surprisinglyuniformin terms of the frameworkthat the analysts drew upon in theiranalysis.None of the studiesthatwe reviewed drewon multi-modalanalyses. The use of CDA as a methodology is rapidlygrowing in educationresearch. s Manyof the studiesdrawon Fairclough' approach-ratherthanon the approaches of van Dijk, Wodak, Kress and Van Leeuwen, and so forth.This homogeneity in approachis a trend that van Dijk (2001) warned against because of the multidisciplinary nature of CDA. Future studies should pull from a hybrid set of approachesthatcan help to bring fresh insights to educationalquestions. The weakestlink in all of these studies seems to be the connectionbetween linguisticresourcesandsocial practices.Thatis, althoughsome of the authorsfocused on the linguistic details of interactionsand made social claims, they failed to representthe relationship betweenthe grammatical resourcesandthe social practices. Not even the studies that provided an analysis of the micro-linguisticaspects of texts gave a rationalefor why those aspects were includedor explainedhow they are connected to social practices. On the other hand, researcherswere equally inclined to point out social practicesthroughbroadthemes or discourses without indicatinghow such discourses were constructedor constrainedby grammatical resources.Clearly,establishinga link betweenthe two levels is necessary.Indeed, Meyer (2001) arguedthat often a range of linguistic indicatorsand variablesare used to analyze texts with no theoreticalcoherenceor grammar theory supporting the analysis. It seems importantto be clear about what grammaticalresources are being inquiredinto (pronounsor modality)andwhy.Thatis, SFL arguesthatevery utterance performsthree simultaneousfunctions:It presentsideas, it positions people in certainways, and it performsa textualfunction of organizingthe coherence of to talkingand/orwriting.It is important clarify which aspectsof languageperform which functionsto avoid being criticizedfor readingideologies into the data.Anaand lysts can takeresponsibilityfor adoptinga moregroundedapproach lettingthe networksof practices-and be read from the data.Educaideologies appear-as tion researchers should spendmore time incorporating SFL theoryandmethod(or compatiblelinguistic models) into theiranalyses. Therewas alarminglylittle reflexivityin the articlesthatwe reviewed. Some of the articlesdid include a researcher role section-a rhetoricalstrategythatis comresearch.However,many of the authorsdid not monplacein publishingqualitative move from reflection to reflexivity. This is a problem, especially in education withinthe whereresearchers oftenhave successfulhistoriesof participation research, educationcontextswherethey are conductingresearch.Therewere some surprises. Hays (2000), for example,includeda momentof reflexivityin her analysisof newspaperarticles.This is the only reflexive section in a writtenlanguageanalysisthat we found. Despite not using reflexivityto its full potential,some authorsoutlined theiranalyticdecisions very carefully,thusallowingthe readerto assess them. We reportedthatvery few of the articlesreviewed here moved towardemancipatory action with the results of their analyses. There were exceptions, such as Rogers, Tyson, and Marshall(2000), Young (2000), and Rogers (2002a, 2002c), where each of the researchersalso worked as a literacy tutorand plannedcritical interventions with the people with whom he or she was working.The lack of action in the rest of the studiesis surprising,given thatmany of the authorsdefinedCDA 386

CriticalDiscourse Analysis in Education

in termsof its liberatory goals andaims (as was discussedearlier).Bucholtz(2001) assertedthat it is not possible for scholars who do critical work to separatetheir researchfrom theirpolitical positions. Similarly,van Dijk (2001) referredto CDA as "discourseanalysis with an attitude"(p. 96). What is importantfor continued workin CDA is a methodologythatallows politicalpositionsto arisefromthe data ratherthanbeing readinto them. The twin goals of a rigorousanalysis and a social justice agendaneed not be incompatible. Bucholtz (2001) points out that CDA should not strive to enforce stricter methodological guidelines, because more rigorous and scientific methodologies would inevitablymove researchers of away fromrecognizingthe construction their discourse analysis. Instead,researchersshould closely attendto the specific conditions that shape peoples' lives and bringthe researcher'srole more clearly into vision. Based on the review of CDA in education,we would agree with Bucholtz (2001) thata formalizedset of methodologicalcriteriafor CDA will not silence the critiquesof the theoryand method.Indeed,CDA, by design, is a hybridset of theories and methodologies. The continued work within and across frameworks allows CDA to adaptand respondto ever-changingconditions in a late capitalist society. We departfrom her assertion,however, afterthis review of CDA in education. As we have demonstrated this review, many studieshave not reportedan in analyticproceduressection. Many studies included a linguistic analysis but were not clear aboutwhy certainaspectsof texts were chosen. Still othersmade sweeping explanatorystatements-of the type thatareoften easy to stateeven before the analysishas been conducted-without attentionto the links betweenthe microand the macro.If CDA as a theoryandmethodis to move beyond the presentcritiques, researchers might attendto the following: (a) the links between the micro and the macro;(b) explainingwhy certainlinguisticresourcesareanalyzedandnot others, and (c) clear analyticproceduresoutliningthe decision makingof the researcher. Directionsfor FutureResearch Despite this robust collection of literature,there are areas where theories of or learningareunderdeveloped not attachedat all to "critical"discourse theoryand social transformation. CDA offers a synergisticframeworkwith social constructivist and communityof practicemodels of learning.Indeed,CDA can be used to trace changes in discourse patternsover time and across contexts-changes that we might refer to as learning. Future research might focus on bringing sociocognitive models to bear on CDA. Clearly, more researchis needed in primary grade classrooms, particularlywith interactionaltexts. There is also a need for researchanalyzing multi-modaltexts (Web surfing,hypertexts,channel surfing, network communication).We might also study genre mixing within and across researchsites and projects. An issue not exploredin this article,but which may be of considerableinterest to educationresearchersinterestedin CDA, is the representation the analysis of and the findings. Clearly, given the space constraintsof publishing in academic journals,it is not possible to representall aspectsof multi-vocaland multi-layered analyses. Authors make choices, and it is hoped that they are clear about their choices. This issue speaks to the need to consider the limitations of print-based journalsas the primaryoutlet for work in CDA. Researchersmight considerother multi-modal outlets for their work, such as electronic journals and books. 387

Rogers et al.

We might also look more closely at how studies that defined CDA as having liband eratingaims were relatedto the researchparticipants the role of the researcher. Future studies might offer descriptions of the nuances of learning described than rather constructiveCDA approaches) througha CDA framework(productive, a criticalframework. This last recommendation researchoften seems to for simply be placedlast on the agendaof scholarsin educationwho areusing CDA. We hope thatmoreactionwill be takenas a resultof the CDA studies.Perhapsmultiplestudies conductedwith CDA can be used to help shape constructiveinterventionsin policy and practicein educationalcontexts. Notes seminar Thisreviewof CDA in education research started a projectin a doctoral as in discourse in 2001.Earlier drafts thatthefirstauthor (Rebecca analysis Rogers)taught of the articlewere presented the CDA Conference Bloomington,Indiana(June at in 2004); the first annualCDA Conferencein Valencia, Spain (May 2004); and the of of 2004). We wouldlike to thankmembers the audiUniversity Albany(November ence at these workshops theirhelpfulcomments, for questions,andfeedback.We are also gratefulto the CDA study groupthatmeets regularlyat the NationalReading for an and abouttheoretical methodologConference, providing ongoingconversation ical issues involvedwith CDA. And, finally,we thankCynthiaLewis andthe anonymousreviewersfor helpfulfeedbackon earlierdraftsof this article. 1 See Seigel and Fernandez,2002, for an overview of critical approachesin education. 2 An in-depth of treatment the historyof discourseanalysisand its sociolinguistic rootsis beyondthe scope of this article.However,manyarticlesandbookshavebeen writtenthat focus on that history (e.g., Coupland& Jaworski,1997; Jaworski& Stef Slembrouck's websiteanswersthe question"What 1999).In addition, Coupland,
is meant by discourse analysis?" (http://bank.rug.ac.be/da/da.htm). The site includes

a detailedhistoryof discourseanalysis,withbibliographic references. References B. Ailwood,J., & Lingard, (2001). Theendgamefor national girls' schoolingpolicies
in Australia?Australian Journal of Education, 45(1), 9-22. Alvesson, M., & Skoldberg, K. (2000). Reflexive methodology: New vistas for qualitative research. London: Sage. International Journal of Leadership in Education, 4(3), 199-216.

G. Anderson, (2001). Discipliningleaders:A criticaldiscourseanalysisof the ISLLC administration. NationalExamination Performance and in Standards educational Press. Austin,J. (1962).How to do thingswithwords.Oxford,UK:OxfordUniversity
Bakhtin,M. (1981). The dialogic imagination:Four essays byM. M. Bakhtin(C. Emerson

& M. Holoquist, Austin: Press. TexasUniversity Trans.). C. The Barnard, (2001). Isolatingknowledgeof the unpleasant: rapeof Nankingin
Japanese high-school textbooks. British Journal of Sociology of Education, 22(4),

519-530. H. Bartu, (2001).Can'tI readwithout 13(2), Readingin a ForeignLanguage, thinking? 593-614. disBaxter,J. (2002). Competingdiscoursesin the classroom:A Post-structuralist courseanalysisof girls' andboys' speechin publiccontexts.Discourse& Society, 13(6), 827-842. 388

CriticalDiscourse Analysis in Education

conHow discourses discoursetheoryandreader Beach,R. (1997). Critical response: stitutereader stancesandsocial contexts.Reader,37, 1-26. in Beach, R., & Anson, C. (1992). Stanceandintertextuality writtendiscourse.Linguistics and Education, 4(3), 335-358. Bell, D. (2004). Silent covenants: Brown vs. Board of Education and the unfulfilled hopes for racial reform. Oxford, UK: Oxford University Press.

in Benveniste,E. (1958/1971).Subjectivity language.In E. Meek (Ed.),Problemsin of generallinguistics(pp.223-230). CoralGables,FL:University MiamiPress. in and K. resistance, gender educational V., Bergvall, & Remlinger, (1996).Reproduction, Discourse& Society,7(4),453-479. discourse: roleof critical The discourse analysis.
Bernstein, B. (1971). Class, codes, and control. London: Paladin. ety, 10(4), 572-576.

Billig, M. (1999). Conversation analysisandthe claimsof naivety.Discourse& SociBlommaert, & Bulcaen,C. (2000). CriticalDiscourseAnalysis.AnnualReviewof J., Bloome, D., & Carter,S. P. (2001). Lists in readingeducationreform.Theoryinto UniUK: P. Bourdieu, (1977). Outline a theoryof practice.Cambridge, Cambridge of versityPress.
Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste (R. Nice, Bourdieu, P., & Wacquant, L. (1992). An invitation to reflexive sociology. Chicago: Bowles, S., & Gintis, H. (1976). Schooling in capitalist society: Educational reform and the contradictions of economic life. New York: Basic Books. spaces. British Journal of Sociology of Education, 22(4), 501-518. Practice, 40(3), 150-157. Anthropology, 29, 447-466.

in workpublished 1979) London: Trans.). Routledge. (Original of University ChicagoPress.

and Brown, D., & Kelly, J. (2001). Curriculum the classroom:Privateand public and in ofAnthroBucholtz,M. (2001).Reflexivity critique discourse analysis.Critique Burs, L., & Morrell,E. (in press). Criticaldiscourseanalysisin literacyresearch.
National Reading Conference Yearbook,56. Mahwah, NJ: Lawrence Erlbaum. Butler, J. (1990). Gender trouble: Feminism and the subversion of identity. London: Cazden, C. (2001). Classroom discourse: The language of teaching and learning (2nd pology, 21(2), 165-183.


workpublished1988) Press.(Original MA:Harvard ed.). Cambridge, University L. in Chouliaraki, (1998). Regulation progressivist pedagogicdiscourse:Individualized Teacher-pupil talk. Discourse & Society, 9(1), 5-32. Chouliaraki,L., & Fairclough, N. (1999) Discourse in late modernity:Rethinking Critical Discourse Analysis. Scotland: Edinburgh University Press. Christie, F. (2002). Classroom discourse analysis: Afunctional perspective. New York:

Continuum Press.

Cicourel, A. (1974). Cognitive sociology: Language and meaning in social interaction.

UK: Harmondsworth, PenguinEducation. In in instruction reading Collins,J. (1986).Differential (Ed.), groups. J. Cook-Gumperz UK: Cambridge The Social Construction Literacy(pp. 117-137). Cambridge, of Press. University discourseand social standards, Collins, J. (2001). Selling the market:Educational
inequality. Critique ofAnthropology, 21(2), 143-163. Collins, J., & Blot, R. (2003). Literacy and literacies: Texts,power, and identity. Camstudents' work in literacy lessons. Discourse: Studies in the Cultural Politics of Education, 18(3), 389-407.

Press. bridge,UK:Cambridge University the B. discourses: Comber, (1997).Managerial Tracking local effectson teachers'and


Rogers et al. Corson, D. (2000). Emancipatory leadership. International Journal of Leadership in Education, 3(2), 93-120.

Coupland,N., & Jaworski,A. (1997). Sociolinguistics:A reader. New York: St. Martin's Press. K. and Transformation legitimation in Crenshaw, (1988). Racereform,retrenchment:
anti-discrimination law. Harvard Law Review, 101(7), 1331-1387. Davies, B. (1993). Shards of Glass: Children Reading and WritingBeyond Gendered nalfor the Theory of Social Behavior, 20(1), 43-63.

Identities. Press. Cresskill,NJ:Hampton R. of The Davies,B., & Harre, (1990).Positioning: discursive production selves.Jourof for reform thebabel(babble) culture: and Davis,J. (1997).Educational Prospects the
standardsof foreign language learning. Moder Language Journal, 81(2), 151-163.

and for Delgado,R. (1995). Legal storytelling: Storytelling oppositionists others:A

Press. 60-70). Philadelphia: TempleUniversity J. JohnsHopkinsUniversity Press. Baltimore: Derrida, (1974). Ofgrammatology. DuBois,W. E. B. (1990). Thesouls of blackfolks.New York:VintageBooks. (Original workpublished1903) D., Edwards, & Potter,J. (1992).Discursive psychology.London:Sage Publications. A. Egan-Robertson, (1998). Learningaboutculture,language,and power:Understandingrelationships amongpersonhood,literacypractices,and intertextuality.
Fairclough, N. (1991). Discourse processes and social change. Cambridge, UK: Polity

plea for narrative. In R. Delgado (Ed.), Critical race theory: The cutting edge (pp.

N. and Fairclough, (1989). Language power.London: Longman.

Journal of Literacy Research, 30(4), 449-487.

Press. N. UK: Fairclough, (1992).Discourseandsocial change.Cambridge, PolityPress. N. of discourseanalysisandthe marketization publicdisFairclough, (1993). Critical course:Theuniversities. Discourse& Society,4(2), 138-168.
Fairclough, N. (1995). Critical discourse analysis: The critical study of language. New Fairclough, N. (2003). Analyzing discourse: Textual analysis for social research. New Discourse as social interaction (pp. 258-284). London: Sage.


York:Routledge. N., Fairclough, & Wodak,R. (1997). Criticaldiscourseanalysis.In T. van Dijk (Ed.), D. a of discourse Towards new perspective EFL Figueriredo, (2000).Critical analysis: in Flowerdew,J. (1999). Descriptionand interpretation criticaldiscourseanalysis.
Journal of Pragmatics, 31(8), 1089-1099. Foucault, M. (1972). The archeology of knowledge and the discourse on language. reading. Illha do Desterro: A Journal of Language and Literature, 38, 139-154.

Foucault, M. (1979). Discipline and punish: The birth of the prison. Harmondsworth,

New York:Pantheon Books. (Original workpublished1969)

UK:PenguinBooks. UK: Foucault,M. (1981). History of sexuality(Vol. 1). Harmondsworth, Penguin Books. Fowler,R., Hodge,B., Kress,G., & Trew,T. (1979). Languageand control.London: Routledge. in Fox, R., & Fox, J. (2002). Thepower-discourse higheredurelationship a Croatian H. Hall. New Garfinkel, (1967). Studiesin ethnomethodology. York:Prentice M. and schoolreform, secondlanguage Gebhard, (2002).Fastcapitalism, pracliteracy
tices. Canadian Modern Language Review, 59(1), 15-52. Gee, J. P. (1992). The social mind: Language, ideology, and social practice. New York: cation setting. Education Policy Analysis Archives, 10(5), 1-16.

Bergin& Garvey.


CriticalDiscourse Analysis in Education

and as Gee,J. P. (1994).Firstlanguage acquisition a guidefortheoriesof learning pedagogy. Linguistics and Education, 6, 331- 354. Gee, J. P. (1996). Social linguistics and literacies: Ideology in discourses (2nd ed.). Gee, J. P. (1999). An introduction to discourse analysis. New York: Routledge. Gee, J. P. (2003). Whatvideo games have to teach us about learning and literacy. New

London: Taylor& Francis.

York:Palgrave Macmillan. In Gee, J. P. (2004).Whatis criticalaboutcriticaldiscourse analysis? R. Rogers(Ed.),

An introduction to Critical Discourse Analysis (pp. 19-50). Mahwah, NJ: Lawrence

Erlbaum. Anchor.

Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday Goffman, E. (1971). Relations in public: Microstudies of the public order. New York: Gramsci, A. (1973). Lettersfrom the prison notebooks (L. Lawner, Trans.). New York: Greene, S., & Abt-Perkins, D. (2003). Making race visible: Literacy research for cul-

& Harper Rowe.

& Harper Rowe.

New turalunderstanding. York:Teachers CollegePress. UK: Cambridge University Gumperz,J. (1982). Discourse strategies. Cambridge, Press. Ameriof J. Gumperz, & Hymes,D. (Eds.)(1964). The ethnography communication.
Halliday, M.A. K. (1978). Language as social semiotic: The social interpretation of language and meaning. London: Edward Arnold Press. Halliday, M. A. K. (1985). Spoken and written language. Geelong, UK: Deakin Uni-

ArnoldPress. Edward howto mean.London: Halliday,M. A. K. (1975). Learning

can Anthropologist, 66(6), 1-34.

versityPress. Longman. Halliday,M. A. K., & Hasan,R. (1976). Cohesionin English.London:

Halliday, M. A. K., & Hasan, R. (1989). Language, context and text: Aspects of language as a social-semiotic perspective. Oxford, UK: Oxford University Press. Hasan, R., & Fries, P. (Eds.). (1995). On subject and theme: A discourse functional

D. and Hartman, (1992). Intertextuality reading.Linguisticsand Education,4(3/4), 295-311. Amsterdam: Press. perspective. Benjamins A laments: criticaldiscourseanalysisof education, repreHays,J. (2000). Anguished and of Journalof theAfricanActivist sentation, development the Sanof Botswana. of Heberle,V. (2000). Critical principles criticaldiscourseanalyIntegrating reading: 115-138. Heller, M. (2001). Critiqueand sociolinguisticanalysis of discourse. Critiqueof Two adolescents but Hinchman, & Young,J. P. (2001). Speaking notbeingheard: K., Hodge, R., & Kress, G. (1979/1993). Language as ideology (2nd ed.). London: Routledge. UK: Hodge,R., & Kress,G. (1988). Social semiotics.Cambridge, PolityPress. withincurscientistidentities of the Hughes,G. (2001).Exploring availability student science. GenAn to riculumdiscourse: anti-essentialist approach gender-inclusive of Hymes,D. (1972).Modelsof theinteraction languageandsociallife. InJ. Gumperz & D. Hymes (Eds.),Directionsin sociolinguistics(pp. 35-71). New York:Blackwell Press. 391
der and Education, 13(3), 275-290. negotiate classroom talk about text. Journal of Literacy Research, 33(2), 243-268. Anthropology, 21(2), 117-141. sis and gender studies. Illha do Desterro: A Journal of Language and Literature, 38, Association, 28(3), 23-41.

Rogers et al. Ives, P. (2004). Gramsci's politics of language: Engaging the Bakhtin circle and the visual narrative.Narrative Inquiry, 11(2), 451-478.

school.Toronto, of Canada: Ontario, Frankfurt University TorontoPress. of tale: A dialogic re-reading a studentteacher's Johnson,G. (2001). A cautionary Johnson, & Avery,P. G. (1999). The powerof the press:A contentanddiscourse T., in as analysisof the UnitedStateshistorystandards presented selectednewspapers. of Kamberelis, & Scott,K. (1992). Other G., people'svoices:The coarticulation texts research. P. critical Kinchloe, & McLaren, (2003).Rethinking J., theoryandqualitative
In N. Denzin & Y. Lincoln (Eds.), The landscape of qualitative research: Theories Kress, G. (2003). Literacy in the New Media Age. London: Routledge. Kress, G., & van Leeuwen, T. (1996). Reading images: The grammar of visual design. Kristeva, J. (1980). Desire in language: A semiotic approach to literature and art. and subjectivities. Linguistics and Education, 4(3), 359-404. Theory and Research in Social Education, 27(4), 447-471.

and issues (2nded., pp. 433-488). Thousand Oaks:Sage. London: Routledge.

Oxford,UK:BlackwellPublishing. reader J. and Kristeva, (1986).Word,dialogue, thenovel.InT. Moi (Ed.),TheKristeva Press. (pp.35-61). New York:Columbia University
Kristeva, J. (1989). Language-The unknown: An initiation into linguistics (A. M. 33(3), 453-484.

New York:Columbia Press. Menke,Trans.). University B. Kumaravadivelu, (1999). Criticalclassroomdiscourseanalysis.TESOL Quarterly, of in Labov,W. (1972). The transformation experience narrative syntax.In W. Labov of Press. 354-396). University Philadelphia G. Ladson-Billings, (2003). Racializeddiscoursesand ethnicepistemologies.In N. issues (2nded., pp. 398-432). Thousand Oaks:Sage. G., Ladson-Billings, & Tate,W. (1995). Towardsa criticalracetheoryof education. and and research. Lemke,J. (1992). Intertextuality educational Linguistics Education, 4(3), 257-268.
Lemke, J. (1995). Textualpolitics: Discourse and social dynamics. London: Taylor & course analysis. Review of Research in Education, 21, 3-48. Teachers College Record, 97(1), 47-68. Denzin & Y. Lincoln (Eds.), The landscape of qualitative research: Theories and (Ed.), Language in the inner city: Studies in the Black English Vernacular (pp.

Francis. to An in Luke,A. (1995-1996).Textanddiscourse education: introduction criticaldisand effects of the words:Apologies,"stolenchildren" Luke,A. (1997). The material 343-368. Luke,A. (2004). Notes on the futureof criticaldiscoursestudies.CriticalDiscourse
Studies, 1(1), 149-152. Martin, J. (1985). Factual writing: Exploring and challenging social reality. Geelong, schools and communities. Washington, DC: Cassell. public discourse. Discourse: Studies in the cultural politics of education, 18(3),

UK: DeakinUniversity Press. Martin,J. (2000). Close reading.In L. Unsworth(Ed.), Researchinglanguage in of and Meyer,M. (2001).Betweentheory,method, politics:Positioning theapproaches (pp. 14-31). London: Sage. access time": Children's narratives Michaels,S. (1981)."Sharing stylesanddifferential Moje, E. B. (1997). Exploringdiscourse,subjectivity,and knowledgein chemistry
class. Journal of Classroom Interaction, 32(2), 35-44. to literacy. Language and Society, 10(4), 423-442. to CDA. In R. Wodak & M. Meyer (Eds.), Methods of critical discourse analysis


CriticalDiscourse Analysis in Education Myerhoff, B., & Ruby, J. (1982). A crack in the mirror: Reflexive perspectives in technology studies. WrittenCommunication, 13(1), 5-43.

Press. anthropology. Philadelphia: Universityof Pennsylvania and Myers,G. (1996). Outof the laboratory downto the bay:Writingin science and literature subconstruction theirchildrenas gendered, of Nichols, S. (2002). Parents'
jects: A critical discourse analysis. Journal of Early Childhood Literacy, 2(2),


Oakes, J. (1986). Keeping track: How schools structure inequality. Connecticut: Yale guistics and Education, 8(4), 335-365.

Press. University M. in Orellana, (1996).Negotiating meetings.Linpowerthrough language classroom of Peace,P. (2003). Balancingpower:The discursivemaintenance genderinequality
by wo/men at university. Feminism and Psychology, 13(2), 159-180. Pecheux, M. (1975). Language, semantics and ideology. New York: St. Martin's Press. Pennycook, A. (2001). Critical applied linguistics: A critical introduction. Mahwah,

NJ:Lawrence Erlbaum.

Peters, M., & Burbules, N. (2004). Poststructuralism and educational research. New

York:Rowman& Littlefield Publishers. Discourse& Society,13(2), 251-267. mother. Pitt,K. (2002). Being a new capitalist and discourse Discourseacquisition discourse Price,S. (1999). Critical pracanalysis: of Propp,V. (1968). Morphology offolk tale. Austin:University TexasPress. B. in 21(1), 83-107. Rampton, (2001). Critique interaction. Critique Anthropology, of discursive A Rogers,R. (2002a).Betweencontexts: criticalanalysisof familyliteracy, and 37(3), 248-277. practices, literatesubjectivities. ReadingResearchQuarterly Rogers,R. (2002b). A CriticalDiscourseAnalysisof the specialeducationprocess. Rogers,R. (2002c). "That'swhatyou'reherefor, you'resupposedto know":Teaching and learning critical literacy. Journal of Adolescent and Adult Literacy 45(8), Anthropology and Education Quarterly 33(2), 213-237. tices. TESOLQuarterly, 33(3), 581-595.


Rogers, R. (2003). A critical discourse analysis offamily literacy practices: Power in Rogers, R. (2004). An introduction to critical discourse analysis in education. Mah-

and out ofprint.Mahwah, Lawrence NJ: Erlbaum.

Erlbaum. wah,NJ:Lawrence E. Rogers,T., Tyson,C., & Marshall, (2000). Livingdialoguesin one neighborhood: of acrossdiscourse practices literacyandschooland Movingtoward understanding for G. Sacks,H., Schegloff,E., & Jefferson, (1974). A simplestsystematics the organizationof turn-taking conversation. for Language,50, 696-735. An Sawyer,K. (2002). A discourseon discourse: archeological historyof an intellecin Schegloff, E. (1993). Reflectionson quantification the study of conversation. A Schegloff,E. (1999). "Schegloffs texts"as "Billig'sdata": criticalreply.Discourse and S. Schegloff,E., Ochs,E., & Thompson, (1996).Interaction grammar. Cambridge, UK:Cambridge Press. University
Scollon, R., & Scollon, S. (1981). Narrative, literacy, andface in interethnic commu& Society, 10(4), 558-572. Research on Language and Social Interaction, 26(1), 99-128. tual concept. Cultural Studies, 16(3), 433-456. ing. Journal of Literacy Research, 32(1), 1-24.


Scollon, R., & Scollon, S. (2003). Discourses in place: Language in the material world. Searle, J. (1969). Speech acts: An essay in thephilosophy of language. Cambridge,UK:

London: Routledge.

Press. Cambridge University


Rogers et al.

classroomlearning withincollaborative Short,K. (1992). Researching intertextuality In S. Siegel,M., & Fernandez, L. (2000).Critical approaches. M. Kamil,P. Mosenthal, P. D. Pearson,& R. Barr (Eds.), Handbookof reading research (Vol. 3, pp. NJ: Erlbaum. 141-152). Mahwah, Lawrence Silverstein, & Urban.G. (Eds.).(1996).NaturalHistoriesof Discourse.Chicago: M., of University ChicagoPress.
Sinclair, J., & Coulthard, M. (1975). Toward an analysis of discourse. Oxford, UK: environments. Linguistics and Education, 4(3), 313-334.

OxfordUniversity Press. in of Stamou,A.G. (2002). The representation non-protesters a studentand teacher A in Comprotest: criticaldiscourse analysisof mewsreporting a Greeknewspaper. Stevens, L. (2003). Readingfirst:A criticalpolicy analysis.The ReadingTeacher, Stevens,L. (2004). Locatingthe role of the criticaldiscourseanalyst.In R. Rogers
56(7), 662-668. (Ed.), An introduction to critical discourse analysis in education (pp. 207-224). munication Abstracts, 25(2), 143-181.

NJ: Erlbaum. Mahwah, Lawrence to Who is my neighbor? Qualitative Tate,R. (2003). The "race" theorizeeducation: Tate,W. (1997). Criticalracetheoryandeducation: History,theory,andimplication. Teo, P. (2000). Racismin the news:A CriticalDiscourseAnalysisof news reporting education Thomas,S. (2002). Contesting policy in the publicsphere:Mediadebates over policies for the Queensland school curriculum. Journalof EducationPolicy, 17(2), 187-198. P. of Tunstall, (2001). Assessmentdiscourseandconstructions social realityin infant vanDijk,T. (1993).Principles critical of discourse analysis.Discourse&Society,4(2), 249-283.
van Dijk, T. (Ed.). (1997). Discourse as social interaction. London: Sage classrooms. Journal of Education Policy, 16(3), 215-231. in two Australian newspapers. Discourse & Society, 11(1), 7-49. Review of Research in Education, 22, 191-243. Studies in Education, 16(1), 121-126.

van Dijk,T. (2001). Multidisciplinary CDA:A plea for diversity.In R. Wodak& M. First AnnualInternational Conferenceof CriticalDiscourseAnalysis, Valencia, Spain. J. of Verschueren, (2001). Predicaments criticism.Critiqueof Anthropology, 21(1), 59-81.
Volosinov, V. (1973). Marxism and the philosophy of language. Berkeley: University Meyer (Eds.), Methods of critical discourse analysis (pp. 95-120). London: Sage. van Dijk, T. (2004, May). Critical Discourse Analysis and context. Presentation at the

Applied Linguistics, 19(1), 136-151. Willig, C. (Ed.). (1999). Applied discourse analysis: Social and psychological interWillis, P. (1977). Learning to Labor: How working class kids get working class jobs.

of California Press.(Original workpublished 1930) in Widdowson,H. G. (1998). The theory and practiceof critical discourseanalysis. ventions.Buckingham, OpenUniversity UK: Press.

New York:Columbia Press. University L. (G. Wittgenstein, (1953). Philosophicalinvestigations E. M. Anscombe,Trans.). New York:MacMillan. Wodak,R., Meyer,M., Titscher, & Vetter,E. (2000).Methods textanddiscourse S., of
analysis. London: Sage. Wodak, R., & Reisigl, M. (2001). Discourse and discrimination: Rhetoric of racism and anti-Semitism. London: Routledge. Wodak, R. (1996). Disorders of discourse. London: Longman.


CriticalDiscourse Analysis in Education

H. Woodside-Jiron, (2004). Language,power, and participation: Using criticaldiscourseanalysisto makesense of publicpolicy. In R. Rogers(Ed.),An introduction Erlbaum. C. NJ: Woodson, (1990).Themis-education theNegro.Trenton, AfricaWorldPress. of workpublished1933) (Original S. in Wortham, (2001).Narratives action.New York:Teachers CollegePress.
Yancy, G. (Ed.). (1998). African American philosophers: Seventeen conversations. to critical discourse analysis in education (pp. 173-206). Mahwah, NJ: Lawrence

New York:Routledge. ReadingResearch Young,J. P. (2000). Boy talk:Criticalliteracyandmasculinities.

Quarterly, 35 (3), 312-337. Young, L., & Harrison, C. (Eds.). (2004). Systemic functional linguistics and critical discourse analysis: Studies in social change. New York: Continuum Press.

Authors REBECCA in ROGERSis an AssistantProfessorof LiteracyEducation the Departmentof Education Washington at Universityin St. Louis,CampusBox 1183, One Her research BrookingsAvenue, St. Louis, MO 63130; rogers3948@aol.com. focuses on the discursiveconstruction literatesubjectivities acrossthe lifespan, of with a particular interestin teacherandstudentlearningwithinanti-racist learning contexts. ELIZABETH student theDepartment in MALANCHARUVIL-BERKES doctoral is a of Education Washington at Box 1183,OneBrookin University St. Louis,Campus She ings Drive, St. Louis, MO 63130; eberkes@wustl.edu. uses a graduate-level both and and background in education in molecular cellulargeneticsto examinescientificdiscoursepracticesin classrooms to fromkindergarten college andin informal science settings, with a particularinterest in understanding how specific affectpersistence scientificcommunities. in pedagogical practices at MELISSAMOSLEYis a doctoralstudentin the Department Education Washof in Box 1183, St. Louis, Drive,Campus ingtonUniversity St. Louis,OneBrookings MO 63130;mrmosley@wustl.edu. uses CDA andotherqualitative methodsto She discoursein the fieldof literacyandurban education. analyzeteacherandstudent UniDIANEHUIis a doctoral in student the Department Education Washington of at Box 1183,St. Louis,MO63130; Drive,Campus versityin St Louis,OneBrookings diane.hui@wustl.edu. current Her the and research examines possibilities constraints of online communitiesin supporting teacherlearningfor both experiencedand noviceteachers. at GLYNISO'GARRO JOSEPH a doctoral is in of student theDepartment Education Washington Universityin St. Louis, One BrookingsDrive, CampusBox 1183, St. In researchshe focuses on the Louis, MO 63130; gogarro@wustl.edu. her current socialandacademic of school;she experiences Blackgirlsin a suburban elementary of research to employsethnographic procedures examinethe interconnection race andgender.


APPENDIX A Code Bookfor CDA Publicationdate for study: Authors'names and institutionalaffiliation/location: Whattype of articleis this? (e.g., empiricalstudy, theoreticalpaper,review of literature, position paper) How is CDA defined?Use the author'swords to define CDA. What theorists/researchers cited in referenceto CDA? are What mode of languageis studied?(spoken, written,interactional) Whattheoreticalframeworksdoes the researcher in the paper?List the theoretical use frameworksand cite all theoristsreferenced. Whattheoriesof languageare used? Use the author'swords to describethe theoryof languageand cite linguists and discourseanalyststhe authorreferences. Who are the researchpopulation/participants? if Whatis the ethnicityof the researchparticipants applicable? if What is the gradelevel of the participants applicable? What is the geographiclocation of the study? What is the context of the study?(communityagency, newspaper,school) Whatis the researchquestion? How is learningaddressed(intertextuality references)? How is the analysis conducted?(e.g., what aspects of CDA are used) [Specifically describethe methodof conductingcriticaldiscourseanalysis.] What are the datasources? What is the role of the researcher(e.g., text analyst,participant observer)? Is therea theoryof learningin the research? What are the noted limitationsof the work? [the author'swords] Whatare critiquesof the work? How does this articlerelateto otherarticles?


APPENDIX B Summary studies reviewed of Publication Ailwood & Lingard (2001), Australia [E] Definitionof CDA Researchfocus/question Context of study

Anderson,G. (2001), USA [E]

"Theanalysis providedhere aims "The analysis providedhere Policy aims to disruptthis broad document to disruptthis broaddiscourse discourse of gender equity as of genderequity ... considering it is evident in the Gender the ways in which it is Equitytext, consideringthe historicallyand contextually situatedin relationto broader ways in which it is social and political historicallyand contextually situatedin relationto broader discourses"(p. 11). social and political discourses" (p. 1). There is a set of unexamined Written "Faircloughsuggests a threetheories in use that are standards dimensionalapproachto and test discourseanalysis. The most embeddedin the ambiguities of the standardsfor school immediatelevel is textual document leaders, which become visible analysis ... [next is the level of] discursivepractices... [and] throughan examinationof various linguistic strategies. finally, he suggests analysis of the text as social practice" (p. 202).


APPENDIX B (Continued) Publication Barnard,C. (2001), Japan [E] Definitionof CDA Researchfocus/question Contextof study

Bartu,H. (2001), Turkey [T]

"As with many applicationsof "To what extent, then, can we elucidate the relationship systemic functionalgrammar, one of the aims is to use this between texts themselves, the to model of grammar relate the meanings createdby the text being analyzedto its wider language of such texts, and social and culturalcontext. One the texts as they exist embeddedin their social and way this can be achieved is to culturalcontexts? In this investigate the range of meaning-making potential paper,I shall show that we can seek to answer this possessed by language and, by seeking to identify the specific question by adoptinga choices made in any particular linguistically grounded communicativesituation, analysis of the languageof textbooks"(p. 520). question why such choices have been made, and suggest what otherchoices could have been made and what different meanings would have been producedby these alternative choices" (p. 519). This is a position paperon the "Accordingto Fairclough... role of CDA in a critical every instanceof language use has three dimensions:it is a readingcourse. spoken or writtenlanguage text, it is an interaction... and it is also a social action. CDA itself also has three dimensions:the

Study of textbooks

College studentsin a universi classroom in Turkey

Baxter,J. (2002), UK [E]

Beach, R. (1997), USA [T]

description... interpretation... and the explanationof how the texts and the interaction processes relate to the social action"(pp. 595-596). CDA is definedin comparisonto People have arguedthat girls are often silenced by boys/men as PDA. Both CDA and PDA are interestedin the workings of speakersin public contexts. If this is really the case, would power throughdiscourse, girls find it more difficult than thoughconceptualizedrather differently.CDA assumes boys to meet the new GCSE assessmentrequirements? discourseto work dialectically in so far as the discursive event is shapedby and thereby continuouslyreconstructs events. CDA emphasizes social theory on behalf of dominated and oppressedgroups. None stated. "CDA provides a useful theoreticalperspectivefor how these understanding competing discourses constitute the meaningof social practices in discoursecontexts. These theoristsexamine the 'common sense' presuppositionsor ideological assumptions operatingwithin a particular discoursecontext"(p. 3).

Secondary students (mixed-se class of 14- and 15-yearolds) in a high Briti school.


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question Whatis the role of talkthathelps talk continueinstructional and talkthatdivergesfromit in or reproducing challenging genderroles in the classroom? Contextof study

Bergvall,V., & "CDAaimsto revealhow texts K. operate the construction in of Remlinger, socialpractice examining (1996), by the choicesthatdiscourse USA offers"(Kress,1991,p. 454). [E]

College class of 18-35 students, 40%fema American a universi in Michiga class discussion

Bloome, D., & "CDAexaminespower relations and ideologies embeddedin Carter,S. P. texts throughcarefuland (2001), USA [E] systematicanalysis" (p. 151-152).

"Questionshave not been raised aboutthe consequencesof using a 'list' for framing educationalpolicies and practices,regardlessof the contentof the list or type of item on the list. The purpose of this articleis to raise such questions"(p. 151). "Thegoal is to highlight and examine discourse production as and interpretation it intersectswith the

University colleges of education studying texts abou education reform.

Brown, D., & Kelly, J. (2001), Canada[E]

"CDA providesan example of... recenttheorization.... This type of study can be differentiated from other

12th-grade studentsof African origin

analysis of curriculumcontent given its tendenciesto merge orientationswith post-structural the criticaland interpretive" (p. 502). Chouliaraki,L. Draws connectionsbetween Bernstein'stheoretical (1998), UK [E] frameworkand Fairclough's methodology (p. 10).

living in Alberta, Canada,in a social studies class. How does the genre of Secondary individualizedtalk organize school in the UK, its linguistic options and in which ways does it privilege class discussion certaindiscourses over others?

"life-worlds"of a particular subjectgrouping,i.e., the high-school studentof African descent"(pp. 503).

Collins, J. (2001), USA [E]

"How can discourse analysis, in "[Fairclough proposes]three critical and analytic levels, one of which particular derives from traditionsof anthropologicalframeworks, contributeto our linguistic analysis, the other two from contemporary social of understanding the natureof the the appeal of standards, analysis. Althoughthe terms have shifted over the years, diffusion of influential Fairclough'sschema calls for argumentsand the resonance of standards rhetoricwith analysis of the textual, the discursiveand the society-wide" broadsocio-political (p. 144). developments?"(p. 144).

The call for national and statelevel education standards



APPENDIX B (Continued) Publication Comber,B. (1997), Australia [E] Definitionof CDA Foucauldian analysisandCDA areusedinterchangeably. "CDAcan be productive and positivework,whichcan in contribute, Foucault'sterms, to people seeingthatthey are 'muchfreerthanthey feel' (Foucault,1988,p. 10)" (p. 390). Researchfocus/question "Howdo managerialist discoursesconstitute'the child' as the disadvantaged subjectfor school literacy and programs whatlimitsand possibilitiesdo such discursive andinstitutional practices createfor students their and teachers?" 390). (p. Context of study

Case studyof an Italian teacherin a school in South Australia with a social justice agenda.

Corson,D. (2000), Canada[E]

"CDA goes beyond other forms of discourse analysis by focusing directlyon macro and micro power factorsthat operatein a given discursivecontext" (p. 98).

"TheCDA asks about the distortinginfluence that ideology has on the proceedingsin a formal school meeting. And it asks how that distortionshows up in the

A monthly meeting of a boardof trusteesin a seconda school in

discourse itself" (p. 98).

New Zealand.

Davis, J. (1997), UK [E]

No definition.

(1989) developed Egan"Fairclough Robertson,A. a... rubric... thatinvolves threecomponents: textual (1998), USA [E] of analysis,interpretation the interactional processesinvolved in text production and consumption,andexplanation of 'how interaction process relatesto social action' (p. 11)" (p. 454). "Textualanalysisincludes ideational,or content,analysis of social identities;

"Whydo we as language educatorsand as membersof other social groupsreact to reformistagendas in the ways that we do? Whatqualities of the socioculturalcontext into which these proposalsare introducedcondition our reactions?"(p. 152). "Thegoal of the analysiswas to explorethe relationships amongpersonhoodand literacypractices,using the constructof intertextuality.... I attemptto make visible the ways in which discursive in practicesconstructed the writingclub providedan for opportunity studentsto exploreissues of personhood based on constructing an analysisof intertextual

B< of oard trustees meeting,


Zealand secondary school.

3 adolescent girls (8th grade),of PuertoRican and African American descent, in a communit writing program. The schoo

4 z

APPENDIX B (Continued) Publication Definitionof CDA interactional analysiscombines textualanalysiswith interpretive analysis,and explanatory analysisbrings togetherthese two types of analyseswith analysisof socio-cultural practiceat the and situational, institutional, societallevels" (p. 454). Researchfocus/question links between issues of culture,language,andpower andtheirpersonalexperiences" (p. 455). Context of study

is a worki class New England urban school.

N. Fairclough, (1993), UK [E]

A social-theoretically informed mode of discourseanalysis. CDA views discourseas social practice-both shapedand shapingsocial relations.CDA drawsfromthe linguistic resourcesof SFL.

Whatis happening the to of authority academic institutions, academics,andto relationsbetween authority academicsand students, academicinstitutions the and public?Whatis happeningto the professionalidentitiesof academicsandto the collective identitiesof institutions?

UK, Higher Education

Figueriredo,D. EquatesCDA with critical (2000), linguistics. Offers a description of its basis in SFL and then USA [E] points out its aims: "to make available,throughthe analysis of language as social practice, a critiqueof discoursethat might lead to consciousness raising, emancipation,and empowerment" 141). (p. Cites Fairclough. Fox, R., & Fox, J. (2002), Croatia [E]

Review of literatureand plan of readingclass

A group of EFL studentsa the college level.

Gebhard,M. (2002), USA [E]

Not defined.

"[Sy]stematicallyexplore the 'opaquerelationshipsof causality and determination' between (a) the council membersdiscursive practices and (b) wider social and culturalstructures,relations and processes in the Croatian higher education system and Croatiaitself" (p. 3). "How do second language learnersassume, negotiate, and resist the role assigned to them by the discoursesof school reformat Web Magnet?"(p. 18)

Croatia, higher education council meetings.

3rd-4th-grad ESL students (Hispanic and teachersin an urban magnet school.


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question "My goal is to illustratethe ways that social hierarchiesare and perpetuated contested througha multitudeof mundaneand seemingly texts, and to straight-forward explore the relevance of discourse analysis and ethnographyto understanding and addressingproblems the surrounding development of formal education"(p. 27). Contextof study NA

Hays, J. (2000), "(Critical)discourseanalysis USA [E] becomes a locus of hegemonic struggle.By exposing hidden layers of meaning,then, we can pose a challenge to the existing hegemonic order"(p. 26).

Heberle,V. (2000), Brazil [T]

Authorlinks CDA to critical linguistics in general. "Two of the main purposesof these studies are to make people awareof how language is used to dominateor reinforcesocial inequalities, such as those between people of differentethnic, economic, social or intellectualgroups, and to analyze changes takingplace in social (p. organizations" 117).

The purpose ... is to offer a conceptualizationof how readingcan be looked at from the perspectiveof CDA, focusing also on issues of language and gender.

University studentsin Brazil taking reading courses in English.

Why are therecompetingvisions French at this momentof what it Canadian leadersin means to speak and be French? the politic What kinds of people sphere; participatein redefiningthis basic social category,and what one kinds of people find themselves policy document marginalized,voluntarilyor otherwise?In otherwords, who newspape clippings, gets to decide what speaking recorded and being Frenchmean in these contexts? narrative interview "Whatis the natureof Colin's Colin (10th Hinchman,K., "CDA investigatesthe in and Desuna's participation grade, Peyton relationshipsamong particular White discourseevents, discourse classroomtalk aboutthe text Young, J. over the course of one school male), norms,and social and historical (2001), Desuna USA [E] contexts, and how thatlanguage, year?How, and to what end, were situational,institutional, in turn,shapesthese (8th grade Black and societal contexts relationshipsand contexts. As constitutedin their female)Faircloughdescribesit, CDA both verba makes visible tracesof (p. participation?" 248). students discoursesoperatingand the chosen asymmetriesin power relations from a thatare constitutedin their groupof (p. operation" 246). 20 studen at various sites. Literature classes in middle an high schools. Heller,M. (2001), Canada [E] CDA is definedin relationto NorthAmericanand European the approaches, former attendingto linguistic and anthropology the latter attendingto isolated sections of texts. The authorarguesfor attentionto languageas well as to social theory.

APPENDIX B (Continued)

Publication Hughes, G. (2001), UK [E]

Definitionof CDA


Context of study

Johnson,G. (2001), Australia [E]

"Thisarticlepresentsdiscourse Middle and "Fairclough(1992) developed a [CDA] which combines the analysis of interviews with 3 secondary mixed genderpairs of students schools as micro-analysisof text from the study to illustrate well as a productionand interpretation with the macro-publichow differentsubjectivities college. can interactwithin the knowledge and social practice that constructentities such as constraintsand contradictions science' or gender of these competing discourses 'curriculum which framethe micro level of science to producedifferent discourse"(p. 279). outcomes"(p. 276). "[CDA]is a 'top down' Re-readingone studentteacher's University in text to extend and critiqueher pre-servic approach the sense that Discoursesor ways of being in commonsenseinterpretation teacher's the world-as opposedto from a social interaction and reflective discourseas forms of politicalperspective(p. 451). assignmen language-[...] thatare known to by the researcher operatein the worldoutsideof the picture book, areimposedon categorizations membership foundto operateinside the data" (p. 456). "How was the history standards debatepresentedto the public in selected US newspapers?" (p. 448).

"The aim of CDA is to uncover Johnson,T., & Avery, how languageworks to P. G. (1999), constructmeaningsthat signify USA [E] people, objects, and events in

US newspape

the world in specific ways" (p. 452). Kumaravadivelu, B. (1999), USA [T] CDA is definedas an educational The purposeis to conceptualize applicationof post-structuralism. a frameworkfor conducting critical classroom discourse "Ideologyand power that constitutedominantdiscourses analysis. are hiddenfrom ordinary people; criticallinguists seek to make these discourses visible by engaging in a type of CDA that is 'more issue orientedthan theoryoriented"'(p. 466). Defined in contrastto traditional What counts as discourse linguistics. "CDA offers an analysis in educational alternativeapproachto the research?(p. 10) analysis of educational disenfranchisement, enabling us to trackthe governmental, institutionaland professional constructionof deficit, disadvantageand deviance (e.g., Comber, 1996)" (p. 347). NA

Luke, A. (1997), Australia [T]

Variety of public speeches.

0 o1

"CDA is a political act itself, an Luke, A. interventionin the apparently (1995/1996), Australia naturalflow of talk and text in institutionallife that attemptsto [T] "interrupt" everydaycommon

"Withthe official recognitionof the educationalclaims of culturalminoritiesand indigenouspeoples and of girls and women,... the

Theoretical paper


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question Context of study

Moje, E. B. (1997), USA [E]

classical questions of sense.... Such an analysis destabilize[s]authoritative sociology and psychology of educationhave more relevance discourses and foreground[s] thanever: Who succeeds and relationsof inequality, who fails in schools? How and domination,and subordination. In its constructivemoment, why?" (p. 7). CDA sets out to generate agency (pp. 12-13). Whatcounts as knowledgein Whitehigh to "According Luke (1995), CDA school science withinthe oral and makesvisible how teachers' writtendiscoursesof one students and students'spokenand andtheir writtentexts shapeand high school chemistry classroom? teacherin constructpolicies and rules, chemistry knowledge,and indeed, class. 'versionsof successful and failing students'(pp. 11)" (p. 36).

Myers, G. (1996), UK [E]

This articles draws on CDA to "Analysis of ideology in texts, within a theory of discourse show how we might read noun phrases,clause structure, practicesand social discourse representation, and practices.... CDA startswith discoursepracticesin terms of the social categories of discourse a combined CDA and actor and genre and applies them to of networktheory approach. an understanding clauses" (p. 27).

Lancaster, newspape articles on the Heysham nuclear power stations.

Nichols, S. (2002), Australia [E]

The authorexplicitlystatesthat the interviewswere analyzed by using CDA. But thereis no realanalyticsectionor formal definitionof CDA.

How do interactions between discoursesof genderand discoursesof childhoodin parents'accountscome to constructthe male andfemale child as differentkindsof genderedandliteratesubjects?

Middle-clas Australia parents talking aboutthe literacy developm of their children.

Orellana,M. (1996), USA [E]

"An analysis of language and power. Power may be manifestedin discourse,by

powerful participants ... [I]t

Inquiresinto power relationsin primarygrade meetings.

may also be manifestedbehind discourse. [C]riticalanalysts examine implicit assumptions embeddedin discourse, and considerhow coherence is achieved"(p. 337).

Latino/a primarygrade students (Grades 1 2, 3); problemposing meetings in an urba


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question Context of study

bilingual elementar school classroom Peace, P. (2003), UK [E] "The 'top-down'criticaldiscourse ... analytictradition drawson and post-structuralism of emphasizesthe structuring of speech andconstitution by subjectivities culturally availablediscourses"(p. 164). "How do women socially constructmasculinitiesand play a partthroughtheir discursiveactivities,in bolsteringgenderrelations thatultimatelyoppressthem? How do they resistthese discursivelyconstructed masculinities?" 160). (p.

Second-year undergraduate psycholog studentsat Yorkshire University England.

Pitt, K. (2002), Not defined. UK [E]

The goal of this articleis to analyzethe pedagogic discourseof family literacy.

Interviews with 14 parents from England andWales

Price, S. (1999), Australia [T] Rampton,B. (2001), UK [T]

No formal definitionof CDA is given. The authorwrites that CDA is concernedwith a reconstruction the discourse of at the meta-discursivelevel. "CDA focuses on the naturalizationof inequalityin everyday common sense, on the way in which established ideologies, institutionalizedin the workingsof the lexicorecruitpeople to a grammar, view of the world particular withouttheirreally realizing it" (p. 97). "Critical discoursestudiesfocus on how languageas a cultural tool mediatesrelationshipsof power and privilegein social interactions, institutions,and bodies of knowledge"(p. 251).

Aims to point out the inadequacy NA of Fairclough'sCDA and Widdowson's discourse analysis. None stated.

Rogers, (2002a), USA [E]

"Whathappenswhen personal literaciescome in conflict with institutionalliteracies?What makes it possible for June (and her daughter)to experience school literacy failurewhile simultaneously considerable demonstrating literacycompetencein other contexts?"(p. 248).

Adolescent studentsin London; twenty 14-year-o in two multiethnic schools, one suburban one urban An African American family in a city in New York


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question Context of study

"CDA ... offers a theory of How was it that Vicky consented CSE to her placementin special language as a system (building meetings, on systemic functional educationwhen, as her 6th-grade African linguistics [SFL]), combined opening quote suggested, she with a social theory of language had been determinedto come American use (discourse).... Ordersof out of special education? girl, CSE discourse have a roughly (p. 222). team, parallelstatusto the parent. grammaticalaspects embedded in SFL" (p. 220). also "Wearguethatunderstanding 3 African Rogers,T., "Fairclough arguedthat who we are andhow we are American Tyson, C., languageis a domainof & Marshall, a children connectedor disconnected can ideology andtherefore site of E. (2000), best be understood (in 4th strugglesfor power.Discourse through USA [E] is shapedby powerrelationships criticaldiscoursetheories" grade), and social institutions society in acrosstwo (p. 2). as a whole, anddiscourseboth schools; affects social structures is and low-SES affectedby them,contributing Midwester to social continuityandsocial communit observatio change"(p. 4). Rogers, (2002b), USA [E]

Stamou,A.G. (2002), Greece [E]

Stevens,L. (2003), Australia [T]

CDA sees language-all semiosis -as discourse,that is, a form of social practice.... In order to uncover the way discourse operatesin society, CDA proceeds to a systematic textual analysis, since texts constitutethe medium through which discourseis enacted" (p. 657). "Thiscriticalpolicy analysis, drawingmethodsand techniquesfromthe work of Gee (1996), Fairclough(1989, 1992), Luke (1997), recognizes and worksfromthe situated meaningsof texts, and it documentsthe time-space of hybridizations local, and institutional, societal discourses"(p. 663). "CDA seeks an understanding of how 'discourseis implicatedin relationsof power'.... It provides a useful analyticaland political tool talkingback to discourse"(p. 189).

This articleexamines the in depiction of non-protesters the in the AthensNews, an English-languageGreek newspaper,in the light of the significantimpact that most of those depictions could have on their readersand thus on theirreceptionof the protest. "Withsuch high-stakes discoursesand ideologies at work,how was the federal governmentdefining reading?... How would this information both supportand constrainthem in shaping literacypolicy and practicein theirstates?"(p. 662).

in homes and communit of childre Nonprotesters at a studen -teacher protesttha took place in Greece

Reading Leadershi Academy in 2002, sponsored by U.S. Departme of Education

Thomas, S. (2002), Australia [E]


Investigatethe discursive Newspaper constructionsof curriculum articles about duringone policy initiative. Analysis focuses on newspaper physical debates over the inclusion of education Health and Physical Education in


APPENDIX B (Continued) Publication Definitionof CDA Researchfocus/question in the secondaryschool curriculum(p. 187). Tunstall,P. (2001), UK [E] "CDA was developed to try to put to work practicallyin the field of assessmentsome ideas critical/poststructuralist about languageand practice and theirrelationshipwith social reality"(p. 216). "CDAconsiderslanguageas a social practiceandassumes asymmetrical power withinand among distributions threedifferentsocial contextsan immediatelocal context,a widerinstitutional context,and context.It seeks the institutional to uncoverandunderstand these unequalpowerrelations" (p. 319). "Itexamines the various constructionsof personaland social reality and their associatedpower relations within infant classrooms and discusses the results of the analysis in terms of policy contestationand effect" (p. 216). How do criticalliteracy activitiesin a home-schooling settingsustainor transform the participants' awareness of genderedidentitiesand inequitiesin texts? Contextof study

Queenslan secondary schools. Childrenwh were 6-7 years of age in six schools in London.

Young,J. P. (2000), USA [E]

Fourmale, middleclass, White students (ages 10, 11, and 13 years)


participat in home schooling.

Note: E = empiricalarticles,T = theoreticalarticles,W = writtentexts, I = interactional texts, I/W = comb NA = not applicable.