Talk in the Second and Foreign Language Classroom:A Review of the Literature

Teachers College, Columbia University, USATeachers College, Columbia University, USATeachers College, Columbia University, USA

Classroom interaction has long been a rich site for scholarly research, as attested by the sizable body of literature surrounding classroom discourse. This paper reviews three frameworks currently informing studies of L2 and FL classroom talk, with a focus on qualitative analyses of turn-by-turn talk between teachers and students, and amongst students themselves. Findings from studies working within (1) language socialization and sociocultural theory, (2) critical discourse analysis, and (3) conversation analysis are explored, along with the ways in which these frameworks complement and complicate one another. Finally, we briefly consider possible areas of further research and implications for teacher training.

Keywords: classroom discourse, teacher-student interaction, qualitative research, language socialization, critical discourse analysis, conversation analysis

Correspondence should be addressed to Catherine Box, 525 West 120th Street Box 66, New York, NY 10027, USA. Email: cmb2150@tc.columbia.edu

INTRODUCTION

Of the many activities that occur in the classroom,talkhappens most (Adger, 2001). While a teacher may be well-prepared to carry out the daily agenda, and supervisors may have thoughtfully constructed the teaching philosophy undergirding curricula, it is through moment-by-moment talk that teaching plans unfold. As noted by Walsh (2006), the interactive decisions that teachers make are at least as important as their preparations, since student participation and successful task completion may be helped or hindered by what teachers actually say to students (see also Hall, 1998; Nystrand, 1997; Walsh, 2002). Hall and Verplaetse (2000) pointed out that nearly all classroom activities are carried out through talk, either between teacher and student or amongst students themselves. Unsurprisingly, then, a cursory glance at research during the past fifty years reveals that classroom talk has by and large remained a fertile ground for scholarly work.

In recent years, research on classroom discourse in second (L2) and foreign language (FL) classrooms has become increasingly popular. Interaction in these spaces takes on an especially significant role, given that it is “both the object of pedagogical attention, and the medium through which learning is accomplished” (Hall & Walsh, 2002, p. 187; see also Mori, 2002). Therefore, through examining teacher-student and student-student communication, scholars hope ultimately to gain insight into the potential link between classroom talk and second language development. Indeed, L2 researchers working within the sociocultural paradigm have maintained that language learning is not an internal assimilation of structural components of a language system, but rather a development that begins through social interaction as a learner engages in intellectual and practical activities (Hall & Walsh, 2002). Thus, a finer-grained understanding of second language development might be gained through analyzing how classroom talk engages students, and how students engage themselves in classroom talk.

In this article, we critically review pertinent research on interaction in the L2 and FL classroom, focusing on some of the different lenses through which scholarly studies have performed classroom discourse analysis. Reviews that take into account the strengths and limitations of qualitative research on classroom discourse have been lacking (Zuengler & Mori, 2002). Thus, in an attempt to contribute to this nascent conversation, we consider findings based on transcripts of naturally-occurring, unfolding classroom talk in the L2 and FL classroom. We focus on three frameworks that have been particularly influential in generating these findings: language socialization and sociocultural theory, critical discourse analysis, and conversation analysis, with all three emphasizing the social and interactional nature of classroom talk. We begin by introducing some of the foundational works in classroom discourse that may be considered wellsprings of later developments.

SEMINAL CLASSROOM DISCOURSE RESEARCH

Despite the similarities in the structure of everyday conversation and classroom exchanges, findings from research have led scholars to conclude that classroom interaction differs significantly from all other talk. Nassaji and Wells (2000), for example, differentiated classroom interaction from casual conversation, reasoning that student-teacher talk typically contains a pedagogical purpose, which is not the case in conventional person-to-person interaction. Scholars have also suggested that, as in many other institutional settings, there is often an unequal distribution of conversational power in the classroom, with the teacher controlling who speaks when (Drew & Heritage, 1992; Thornborrow, 2002; Markee & Kasper, 2004; McHoul, 1978). Consequently, unlike in most conversations, not everyone has equal participation rights. Also, McHoul (1990) found that teachers often “jump in” to correct students when they do speak, a move that most interlocutors do not readily engage in during typical conversations. Mehan (1979b) explored the content of teacher talk, particularly the teacher’s use of questions in the classroom. He demonstrated that, in stark contrast to ordinary questions in everyday discussions between two people, teachers most often ask questions of students to which they already know the answer. Therefore, the teacher, unlike a usual participant in a conversation, functions as both the questioner and the primary holder of information. This creates an asymmetrical relationship, in which the teacher is both a questioner and evaluator, determining the “correctness” of the student’s answer (see also Lemke, 1990).

Although early work on interaction in educational settings did not focus specifically on the L2 or FL classroom, these first forays into uncovering patterns of discourse in the classroom sparked research that laid the foundation for decades of scholarship, including analysis of discourse in language learning classrooms. Seminal work on coding teacher—and sometimes student—talk (e.g., Bellack, Kliebard, Hyman, & Smith, 1966; Fanselow, 1977; Flanders, 1970; Sinclair & Coulthard, 1975), revealed that classroom talk was systematic, and hence, could be categorized, quantified, and analyzed. According to Sinclair and Coulthard, there exists at least a tacit understanding amongst students and teachers concerning who speaks and when the talk occurs. They specify that this highly structured exchange consists of a teacherinitiationin the form of a question, resulting in a studentresponse, which is in turn followed by teacherfeedbackto the given response. The uncovering of the initiation-response-feedback (IRF) sequence has undergirded much of the research on classroom discourse that has since ensued, and has certainly influenced those working within the frameworks examined here. We will refer to this three-part dialogue asIRF, although it has been termed otherwise (e.g., Lemke, 1990; McHoul, 1978, 1990; Mehan, 1979). While Sinclair and Coulthard’s (1975) findings concerning the IRF sequence have informed subsequent research on classroom talk, the descriptive nature of their work assumed a detached, neutral tone towards classroom research. They neither condoned nor critiqued the pervasive IRF in educational settings, nor did they consider the effect of such talk on student participation or its effectiveness with regards to learning.

Subsequent researchers have expanded on this work, often taking a stance concerning the advantages and drawbacks of the IRF in the classroom. Some go further, challenging the usefulness of this type of descriptive coding when examining classroom discourse and eschewing the imposition of pre-set categories on talk that might be highly contingent on the activities at hand. Instead, they examine talk as it unfolds in moment-by-moment interactions, considering how such talk encourages or constrains subsequent dialogue. It is with this point that we begin our review. The work examined here stems from three distinct theoretical views concerning how talk operates in the classroom and how teachers and learners co-construct meaning through verbal exchanges within the classroom walls. Indeed, these frameworks do not espouse identical views on the role of context in interactions. Nevertheless, they all do work from the understanding that classroom language use goes beyond the static, unilateral descriptive labeling of verbal exchanges. These approaches have served to inform much recent qualitative work on classroom discourse in the L2 and FL classrooms.

LANGUAGE SOCIALIZATION (LS) AND SOCIOCULTURAL THEORY (SCT): UNDERSTANING DEVELOPMENT

Language socialization (LS) and sociocultural theory (SCT) stem from separate intellectual roots—LS championed by the linguistic anthropologists Ochs and Schieffelin (1984; also Shieffelin & Ochs, 1986) and SCT grounded in the Vygotskyan tradition of Soviet psychology. Nevertheless, the two have much in common in their stance towards the learning process (Duff, 2007). Both emphasize the importance of social interaction, especially interaction over time between novices and those who are capable peers, caregivers or experts. In the first work on language socialization, Ochs and Schieffelin (1984) traced adult-child interactions across three cultures. Through a close examination of observed verbal exchanges between mothers and their young children, they noted that the ways in which mothers spoke to their pre-verbal infant children differed cross-culturally. These linguistic differences are reflective of contrasting cultural beliefs concerning the appropriate ways to address interlocutors, and the children’s ensuing language development seemed to signal both an acquiring of language and an understanding of these cultural views on speaking. The findings in their case studies (Schieffelin & Ochs, 1986) led to the argument that language learning is an interactive, sociocultural process between more and less competent community members, in which two important and distinct process occur: language is taught (children are socialized touselanguage), and culture is taught (children are socializedthroughlanguage). In sociocultural theory, the centrality of such interaction between more or less competent members is epitomized in classic concepts such as scaffolding (Wood, Burner, & Ross, 1976), mediation (Lantolf & Thorne, 2006), and the zone of proximal development (ZPD) (Vygotsky, 1978).

Although neither LS nor SCT was initially concerned with L2 and FL studies, their influence on the latter is aptly captured in Patricia Duff’s (2007) plenary address, where she articulated the compatibility of the two approaches and their implications for what she calls Second Language Socialization (SLS) research. Scholars influenced by LS and SCT argue that L2 and FL learners, like their L1 counterparts, learn through social interaction with those who are more proficient in the target language. Echoing Shieffelin and Ochs (1986), these researchers argue that language learners are members of communities, and that their participation is crucial for language development. Through classic ethnographic methods such as participant observation, interviews, and—most importantly for the purpose of this paper—discourse analysis of naturally-occurring talk, they attempt to better understand how language learners are socialized into becoming competent members of such communities (cf. Kulik & Schieffelin, 2004, concerning ethnographic method).

Some research has focused on the socializing force of teacher talk. Poole (1992), for example, recorded verbal interactions as a participant observer in beginning-level adult ESL classes in the United States. The teachers, themselves middle-class European-Americans, tended toward speaking to the students in ways akin to what Ochs and Schiefflin (1984) uncovered in their observations of white, middle-class American caregivers and their children. For instance, like mother with their young, teachers often used IRF sequences to guide novice L2 speakers through language production. As the discourse analysis demonstrated, all interactions in the classroom were carefully steered by the teacher. Ohta (1999) tracked one student’s progress in a beginning adult Japanese class. Like Poole (1992), the teacher utilized IRF to lead students through the language production process. In this setting, the IRF format practiced by the teacher proved to be quite constraining for the student, whose contributions were restricted to the “R” portion of the dialogue. As a result, the participant was often not able to practice ways of giving feedback during interactions. This “F” slot, used to express alignment or affect, is a crucial slot in Japanese dialogue. Therefore, the teacher served as a living—albeit perhaps inhibiting—model, making expressions salient in the teacher-fronted discourse. As the semester progressed, the student was able to increasingly incorporate expressions uttered by the teacher into peer group work. Kanagy (1999) analyzed teacher modeling in a Kindergarten Japanese immersion classroom in the United States. Like the teacher in Ohta’s study, the instructor in this classroom carefully demonstrated proper greetings through slow, deliberate enunciations, with exaggerated accompanying gestures. At first, the students imitated the teacher’s moves and language, and produced partially correct chunks of words. Gradually, the teacher removed the support structures, until the students could independently participate in the greeting. For both Ohta and Kanagy, the results lent evidence to the importance of expert-novice interactions for developing FL proficiency. Such expert-novice interaction is implemented through what Kahn (2012) calls an open-task framework characterized by emerging foci in an adult ESL classroom, where the teacher and the students jointly worked through exploratory problems that were not worked into the lesson plan, but rather emerged during class activities. According to Kahn, such unplanned moments allowed for the fostering of new understandings within the learners’ zone of proximal development, an argument that has yet to be corroborated.

Others have sought to document the socialization of cultural values or behaviors in the second or foreign language. In her study on Chinese heritage language schools, He (2000) suggested that the interactional and grammatical organizations of teacher’s directives offered rich resources for socializing the children into Chinese cultural values. Duff’s (1995) work, conducted in an EFL history classroom in a Hungarian dual-language high school, the students were well socialized into a classroom format known asrecitation, in which a teacher-selected student must answer a series a display questions posed by the teacher. By contrast, they were novices in the student-centered, cooperative classrooms typical of Hungarian dual-language programs. When initially given the task of leading lectures, the students were dependent on formulaic expressions provided by the teacher to frame or conclude their talks. As the year progressed, students exhibited their adeptness at using language more creatively through asking for clarifications from the teacher, and providing their peers with feedback on improving student lectures. Duff concluded that the students were being socialized into becoming future EFL history teachers, performing as increasingly competent apprentices of this particular community.

Beyond the emphasis on expert-novice interaction, much analytical attention has been devoted to the socializing process or mediating benefits stemming from collaborative peer interaction. Looking at a dual language classroom, Angelova, Gunawardena, and Volk (2008) argue that peers as well as teachers can serve as experts. Their study shows the ways in which a group of students switch between expert and novice roles, depending upon whether classroom work is conducted in Spanish or English. That is, the more competent Spanish speakers take on the role of expert (and peer teacher) in Spanish-language activities, while the native English speakers do the same for work that takes place in English. Antón and DiCamilla (1998) showed the ways in which the first language was used as a mediational tool as students, working in pairs, performed word searches, attempted to translate, and organized their ideas during an essay-writing task in a Spanish FL classroom. Also situated in the Spanish FL classroom, Gutiérrez (2008) reported instances of microgenesis, or moments that trace the origins of the learning process, during collaborative activities amongst peers. Ohta (2000) closely examined the talk of two college-level L2 speakers in a Japanese FL classroom as they co-constructed a zone of promiximal development that scaffolded their performance as they carried out a translation task. Based on data from a French immersion and an adult ESL classroom, Swain (2000) argues that collaborative dialog between peers is a key form of mediated learning, which in turn plays a crucial role in the language development.

As some research has revealed, the language socialization process does not always progress neatly (Morita, 2004), and two studies from Hall (1995; 1998) demonstrated just that. Hall (1995) found that in a first-year Spanish class, the teacher attempted to engage students in verbal interactions. However, the talk had no overarching topic to lend it coherence. Instead, it was a string of talk in the IRF format, with the same information repeated and recycled. The highly constrained nature of the interactions precluded students from becoming active participants through elaboration or inquiry. In another study with the same Spanish instructor, Hall (1998) discovered that students did not have equal access to opportunities for linguistic interactions. In fact, it appeared that two status groups had formed, the “primary” group receiving considerably more interactional opportunities from the teacher than the “secondary” group. The primary group was able to successfully initiate turns, overlap other students? talk, and engage the teacher in discussing content. On the other hand, the teacher often either ignored initiations by secondary group members or critiqued their contributions based on linguistic form, blocking them from elaborating on content. Hall (1998) concluded that this differential treatment resulted in varied opportunities for participation in the community of learners, and as a consequence, different understandings amongst the students concerning their roles and rights as Spanish language learners. Morita (2004) tracked six adult English language learners during their year-long sojourns in a Canadian university. She found that socialization is indeed a dynamic process, involving constant and evolving negotiating of identities as students socially construct their roles as members of varying classroom communities. At times, the students struggled to find or use their voice, silenced by peers or instructors, while at other moments, their presence and opinions were valued.

Willet’s (1995) year-long research in an American elementary-school classroom resonated with Hall’s (1998) conclusion that students experience the language socialization process differently, and hence, follow different trajectories with regard to L2 development. She focused on a group of L2 students as they completed phonics seat work, and she noticed a growing disparity in language development. Indeed, three of the four L2 learners had banded together, helping one another with tasks and experimenting with increasingly larger language chunks while conversing. Their active participation and linguistic independence increased during the year. By contrast, the fourth child’s participation decreased, with his sparse linguistic contributions signaling an ‘outsider’ status. In a study of L2 learners in kindergarten with regard to student participation, Toohey (1996) concluded that participation in the classroom community is not tied to language proficiency. In fact, not all students who exhibited a strong grasp on the target language became active participants in the classroom. Some students preferred to associate themselves with those who rejected classroom activities. In order to retain a high social status in one community, it was necessary to withdraw from another one (the classroom community). On this basis, Toohey effectively posited the argument that more than one community functions in classrooms, and that the competing goals of these communities can facilitate or hinder the socialization process. She also asserted that in forcing themselves to retreat to the margins of a community, students were activelyresistingsocialization. These arguments were corroborated in Duff’s (2002) study, which discovered that some mainstreamed ESL high school learners in Canada refused to provide more than terse one-word answers to questions, even if they were capable of doing so.

The finding that some students either cannot or will not engage in active participation in a community interacting in the target language raises some challenges for a framework that assumes complete socialization as an endpoint. Even though researchers working within language socialization recognize that the process can be iterative, they have not traditionally considered factors that may temporarily or permanently disrupt the process. At the same time, however, scholars working within critical discourse analysis consider this kind of question(s) to be a launching pad for research in the L2 and FL classroom.

CRITICAL DISCOURSE ANALYSIS (CDA): UNDERSTANDING POWER

While all three frameworks examined here operate with the argument that language is socially co-constructed through interaction, Critical Discourse Analysis (CDA) alone insists that language is inextricably intertwined with ideologies and notions of power embedded in the dominant cultures that use the language (Kumaravadivelu, 1999). CDA further claims that interactions are shaped by external sociopolitical conditions. Thus, language use does not begin with spoken or written text, but with social and political issues (Fairclough, 2001). A critical analyst approaches the data with the intent to uncover larger sociopolitical issues operating within the culture. Fairclough, Mulderrig, and Wodak (2011) argue that CDA does not adhere to one particular research method. Instead, these researchers begin with a topic; methodology is the analytical mode by which the topic is refined, and the research questions emerge, which depends upon the subject under investigation. Earlier work discussed the use ofcriticalethnography(Simon & Dippo, 1986), which includes discourse analysis, participant observation, and interviews that one would expect of a classic ethnography. Yet, critical ethnographers share the expressed goal of discovering and examining the ideological discourse that is assumed to exist in the language.

The arguments outlined by CDA scholars have challenged many L2 and FL classroom researchers to consider ways in which power relations and cultural ideologies function in educational institutions. Popular target languages such as French, English, and Spanish have been linked to societies that have histories of colonization, racial aggressions, as well as socioeconomic and gender inequalities. Language learners are often members of formerly colonized societies and/or identify as members of populations that have been historically minoritized, whichaprioricreates a power dynamic within the classroom walls. Not only does the teacher have the authority to ask questions and evaluate student performance, but the instructor also may represent the cultural values and political stances of the communities that speak the target language as an L1. Therefore, these scholars have found it important to ask research questions concerning the ways in which cultural, racial, and gender ideologies embedded in such languages affect language development in L2 and FL learners. They are also interested in the relationship between language learning and social change (Norton & Toohey, 2004, p.1).

Like some of the researchers working in language socialization frameworks, scholars in critical frameworks have addressed issues of learner exclusion or unequal learner access to the socialization process. Critical researchers go one step further, however, as to link notions of inequality and exclusion to larger sociopolitical forces. Gutiérrez and Larson (1994), for example, discussed the IRF structure in terms ofhegemony, or the power exerted by a dominant group over marginalized groups (Gramsci, 1971). In their study of a 10th grade ESL class, the two researchers argue that the teacher’s strict adherence to the IRF confined students to giving short answers only. When students attempted to elaborate on their answers, the teacher would interrupt or latch onto the response; at times, she simply ignored them. This lack of interactional space for students to develop their own text or inject their own content knowledge into the lesson resulted in students having little opportunities to develop a range of language skills and sociocultural knowledge necessary for entry into the academic community (cf. Hall, 1995). Based on such findings, Gutiérrez and Larson then attempted to connect the IRF structure to larger notions of power in the classroom. They contend that the teacher’s adherence to the IRF pattern allowed her to assume the position of the sole possessor of knowledge. As such, student voices were marginalized in the presence of the teacher-dominated discourse. These findings are repeated and extended in Gutiérrez, Larson, and Kreuter (1995), which also revealed how the use of IRF imposed interactional constraints on students. When one student continually resisted these rules through speaking out of turn or ignoring directions, the teacher viewed the student as not only disruptive, but also academically deficient. Although the student’s verbal expressions demonstrated her growing literacy skills, the teacher only focused on her disregard for classroom procedure, and labeled her “remedial.”

While the teacher’s role as a member of the dominant culture was emphasized in both Gutiérrez and Larson’s (1994) and Gutiérrez, Larson, and Kreuter’s (1995) analyses of the discourse of classrooms, Canagarajah (1993) found that student resistance in an FL setting may produce differential results. More specifically, it was observed that an EFL classroom in Sri Lanka in which the entire class resisted the teacher-imposed interactional patterns. The teacher gave language instruction in the benefits of a student-centered, discussion-based classroom; unfortunately, he failed to engage his students, who were used to teacher-centered, grammar-based classes. Because of the local cultural view that teachers must impart their knowledge to students, the students would not accept an atmosphere in which knowledge was shared and students interacted with one another. As a result, many students stopped attending the class and enrolled in private tutoring classes instead in order to develop the L2 and FL skills necessary for the next course in the sequence. This way, they effectively marginalized the teacher.

On a different note, some researchers have discovered opportunities within the classroom walls through which students can develop as language learners outside of the sanctioned student-teacher or student-student exchanges. Canagarajah (2004) analyzed moments in which L2 students were able to communicate, usually surreptitiously, outside the strict IRF format imposed by the teacher. This kind of exchange, called asafehouseby Canagarajah, usually occurred in a secondary ESL classroom as off-task or between-task talk, or during whole-class interactions. In these moments, students experimented with vocabulary and syntax, and would at times code-switch in order to further clarify a point. Through a close analysis of the language used in these safe houses, Canagarajah convincingly identified a rich, multilingual site where students are exposed to important communication strategies such as style-switching, which involves having the interlocutor switch between formal and more informal registers. On this basis, he argued that more research needs to be conducted in such spaces.

Another prominent trend for researchers informed by CDA has been to closely analyze classroom discourse for the purpose of ascertaining how identities are negotiated. While identity negotiation has long been a point of interest for researchers from paradigms including the second and foreign language classrooms, it is somewhat surprising how few studies offer the close analysis of talk demonstrating how identity is actually constructed in the moment-by-moment of classroom interactions. Hruska (2004), in her year-long study of kindergarteners, examined ways in which linguistic realizations of gendered identities facilitated or hindered L2 learners’ classroom participation. In line with findings from language socialization research, she noticed that L2 proficiency played less of a role in determining access to interactions than identity constructions in unfolding talk (see also Hruska, 2007; Norton, 2001). For instance, boys tended to engage in what Hruska referred to ascompetitivediscourse, debating who had “won” or who was “bigger.” In contrast, girls often opted forromanticdiscourse, discussing issues such as “which classmates would marry each other.” The demarcations along gender lines constrained the types of interactions available for the class. Furthermore, when the teacher commenced a class-wide discussion on the youth soccer league—an activity that the students had verbally proclaimed to be “masculine,” girls retracted from speaking, despite the teacher’s prodding. Indeed, gender seemed to govern who could speak, and what kind of talk could ensue. As stressed by Hruska, these notions of gender did not direct the discourse in the form of overarching stereotypes, but rather, occurred at certain moments; they were also constantly in flux as children negotiated them in specific interactions. She concluded that moments in which gender ideologies emerged in the discourse evoked served to block certain students from their linguistic expression, and barred certain L2 learners from participating in the talk.

Theorists have noted that terms such as “husband/wife,” “married/divorced” are reflective of heterosexuality being the dominant discourse on sexual identity. Nelson (1999), among other scholars, has argued that L2 researchers need to be more sensitive to alternate sexual identities, and that it is important to provide opportunities for language learners to use expressive language to talk about it. In Nelson’s study of a community-college ESL grammar course, the teacher presented a graphic of two women walking arm-in-arm during a lesson on modal verbs. This was followed by a content-based discussion of cultural assumptions about homosexuality. Though it remained unclear as to how this verbal exchange of sexual identity played a role in hindering or facilitating L2 development and learner participation in linguistic interactions, the importance of having L2 learners discuss cultural assumptions surrounding sexual identity was affirmed.

All of the researchers reviewed in this section analyzed discourse with the expressed purpose of connecting the use of language to larger sociopolitical structures that operate in the target language culture. They located interactional moments when the language serves to dominate, resist, or negotiate the cultural ideologies in the community. The last framework to be reviewed below, conversation analysis (CA), converges in the sense that it also engages in close analysis of verbal interaction. Nevertheless, this approach favors working directly from the data, rather than from macrostructures such as race or gender, to form the research questions.

CONVERSATION ANALYSIS (CA): UNDERSTANDING ACTION

Founded by American sociologists in the 1960s and 1970s (Sacks, Schegloff, & Jefferson, 1974), CA as both a theory and a methodology comprises a set of assumptions that prioritize analytic induction and participant orientations: (1) social interaction is orderly at all points, such that no detail can be dismissedapriori; (2) participants orient to that order themselves, (i.e., order is not a result of the analyst’s conceptions or any pre-formulated theoretical categories); and (3) such order can be discovered and described by examining the details of interaction. The goal of CA is to uncover the methods and procedures participants utilize to understand each other and to be understood. Analysts work with audio or video recordings along with the transcripts of these recordings, which capture a full range of interactional details such as volume, pitch, pace, intonation, overlap, breaths, smiley voice, the length of silence as well as nonverbal conduct. It is in these minute details that evidence is located for how social actions such as requesting or criticizing are accomplished. The basic unit of interaction, for example, is an adjacency pair (APs) (Schegloff, 2007), where the first pair calls for the production of a second pair part. Through adjacency pairs, a wide range of actions may be accomplished: question-answer, invitation-acceptance, accusation-denial, complaint-apology, etc. When a question-answer adjacency pair is followed by a sequence-closing third such as “okay,” “good” or “thanks,” it begins to look, at least ostensibly, very much like the IRF structure in classroom discourse. The highly detailed nature of CA analyses has appealed to L2 and FL classroom researchers who wish to better understand the immediate interactional accomplishments or consequences of talk.

Like the other frameworks examined in this paper, researchers working in the CA tradition have taken interest in the IRF structure. A specific focus has been placed on the local effects of the third turn. Seedhouse (2004) demonstrated that IRF cycles “...perform different interactional and pedagogical work according to the context in which they are operating” (p. 63). He supported his argument by juxtaposing two L2 classroom extracts: one text contained strings in which the ‘F’ slot displayed a teacher’s preoccupation with evaluating the grammar of a student; another showed a teacher handling grammar correction incidentally, while attending to the content of the student response, which triggered further teacher-student interaction. Seedhouse then concluded that the content of the third part can either narrow or widen the interactional space allotted to the learner.

Waring (2008) uncovered one specific way in which the third turn can close down interaction. Her analysis of homework checks conducted in an adult ESL classroom showed that phrases such as “very good” in the third turn, termedexplicitpositiveassessments(EPA), are treated as a signal that the sequence had shut down; when students did have questions related to the answer that received the EPA, they did not pose the questions until long after the exchange related to the homework item had passed. Thus, positive assessments can actually hinder student participation through blocking a response from further elaboration or questioning by other students (see also Seedhouse, 1999).

Rather than declaring the third turn a blanket “evaluation” or “feedback” move, Lee (2007) argues that attention must be paid to its local contingencies. In his analysis of college-level ESL classrooms, Lee uncovers five different functions for the third turn, each highly dependent upon the immediate context created by the second turn. For instance, a teacher may use the third turn to steer students into a particular interactional trajectory, leading them to evaluate the grammatical (in)correctness of a peer response; or a teacher may attend to classroom management issues, such as soliciting answers from the class after a second turn silence, or asking a student to repeat an answer that went unheard. Also focusing on the third turn, Fagan (2012a) demonstrates the multifunctionality of “okay” as a teacher resource for giving feedback.

Aside from the F in IRF, conversation analytic interests have also been directed toward teacher questions. During what Markee (2004) callszonesofinteractionaltransition(ZIT), when a student summons a teacher to answer a question that a group of students has posed during peer work, the teacher often responds by inserting a counter-question, thereby regaining the initiation slot, and repositioning the student as the one whose answer is a candidate for evaluation or feedback. Teacher questions can also carry complex ramifications. Yes-no questions can convey a critical stance that does not encourage open and critical thinking (Waring, 2012a), and understanding-check questions launched at activity boundaries are often treated as a pre-closing rather than an invitation for questions (Waring, 2012b).

Some recent conversation analytic work has looked into aspects of teacher talk beyond questions and feedback. Fagan (2012b) examines how a novice teacher deals with unexpected learner contributions by glossing over learner contributions and assuming the role of information provider. In a follow-up larger scale study, Fagan (2013) details how an expert teacher manages learner contributions by consistently prioritizing learner competence and positioning learners as partners in a joint inquiry that underscores exploration. Jacknick (2013) specifies the teacher’s use of laugher in managing moments where her epistemic authority becomes questionable. Waring (2013b) documents how the teachers in the adult ESL classroom manage competing voices through eitherselectiveattending(directing attention to one particular student voice that has in some way risen above the chorus of voices) orsequentialattending(directing attention to multiple voices in a serial manner).

Aside from the focus on teacher talk, conversation analytic work has generated important insights into the active roles learners take in steering classroom interaction, such as moving out of the IRF (Waring, 2009), contesting activity shifts initiated by the teacher (Jacknick, 2011), challenging the teacher’s epistemic authority (Jacknick, 2013), and doing being playful (Waring, 2013). Waring (2011), for example, produces a typology of learner initiatives both within and outside the IRF and discusses the learning opportunities such initiatives may engender.

Work on learner talk has also described how gestures are used to aid repair and to complete an action in small group project work in the ESL classroom (Olsher, 2005, 2008) and howinteractionalshiftsare accomplished as students begin and end tasks (Hellermann, 2007, 2008; Hellermann & Cole, 2008). Markee (2005) untangled an exchange during a university-level ESL class, in which the social practice of inviting a friend to a party co-existed with the teacher-assigned task of discussing Günter Grass. Between the teacher giving directions and the start of the task, an interactional space unrelated to the task opened up for talk. At that moment, two students began what Markee called the “skillful schizophrenia” (p. 210), assuming multiple identities of inviter/invitee and students. Although the off-task exchange was subject to frequent interruption, the students used eye gaze and hand motions to signal alignment in finishing the business of the invitational act. One of the participants was able to answer a teacher question before turning back to the invitation sequence, deftly attending to his dual role as a student and an inviter. By locating moments in which students balance several identities and manage two unfolding sequences, Markee argues that students display their growing L2 competence both during and at the boundaries of tasks. As a matter of fact, these two moments did occur simultaneously in the study. Overall, work on learner interaction has found that L2 learners of all proficiency levels engage in meaningful talk when given the opportunity to work with one another. In some cases, such interactional space allows students to aid one another in comprehending vocabulary words, or assist one another in repairing errors (Markee, 2000). However, as noted by Markee (2000) and Mori (2002), the teacher’s idea about whatshouldbe going on is not always what is going on. Students may be attempting to grasp the meaning of a lexical item in context, for instance, or students may be struggling with managing an interaction. On the other hand, students may achieve a naturally flowing conversation in their FL during the planning phase, when the teacher might not be expecting it. Finally, without any specific focus on either teacher or learner talk, Waring (2013c) shows how learners develop their competence in mastering the sociopragmatic aspect (i.e. the social expectations of how speakers in a community should communicate) and the pragmalinguistic aspect (i.e. the resources available to the speaker in the target language to convey a socially appropriate message) of certain interactional routines in English over a nine-week span. Taken together, the CA framework aids researchers in finding out exactly when and how these fruitful talks occur, thereby shedding light on ways to engage students in meaningful exchanges.

CONCLUSION

This brief review of frameworks informing classroom discourse analysis demonstrates that talk in the classroom has proven to be highly structured, and as such, can be described and explored in various ways. For a scholar working in language socialization or social cultural theory, for instance, the IRF format may afford teachers an opportunity to guide students through the process of using the target language in increasingly sophisticated ways. For a critical discourse analyst, the rigidity of the IRF may contribute to issues of teacher dominance in the classroom, resulting in the teacher’s control over the issue of who holds the floor. For a conversation analyst, each turn of the IRF sequence can be examined to shed light on how a turn-at-talk might play its role in constructing the next spate of interaction. All of these approaches allow researchers to understand the intricacies of how talk functions within the classroom walls. The recent resistance to unilaterally code an utterance as “initiation,” for example, or “feedback,” has opened the possibility to consider the multi-faceted nature of classroom talk. A teacher question, for example, can serve to gather information—or as a managerial strategy or a method to spark class debate. Thus, studies that privilege the contingency of both teacher and student talk may result in a more nuanced understanding of classroom talk, and may provide a more detailed analysis of both the advantages and drawbacks of certain kinds of classroom discourse.

Research stemming from the three frameworks examined has also looked into the kinds of interactions that are not part of canonical classroom talk. Marginalized student-student exchanges, including off-task talk and unsanctioned interactions, have been shown to create spaces for students to experiment freely with the target language. These studies, while still small in number, contribute to a more complete understanding of how language is used in the classroom. Studies that seek to collect data of this kind of talk would help researchers and teachers alike to better understand what students grapple with or comment on when they believe the teacher is not listening. In addition, spontaneous teacher-student talk revealing emerging foci, which are not a part of a preset lesson plan, merits further research, as such studies could shed light on the kinds of language points that spark student questions, struggles, or debates. Through a continued, close exploration of both teacher-student and perhaps student-student interactions in particular, researchers can gain greater insight into the exact ways in which classroom discourse encourages or hinders student participation and language development in the L2 and FL classroom.

Future research can also harvest great gains from some cross-fertilization between various approaches, The strength of CA in illuminating interactional practices, for example, when combined with SCT’s powerful lens for explaining teaching and learning, can yield findings that are both empirically defensible and theoretically illuminating. In addition, researchers and practitioners alike are frequently concerned about the learning “outcome” of specific teacher practices, and such outcomes are typically not obtainable from the analysis of discourse alone. Questions such as whether a particular teacher practice is more effective than some other practice are readily answerable with experimental designs. As such, collaboration with colleagues outside discourse analysis, though easier said than done, can be immensely productive.

With greater insights into the nature of classroom discourse comes the possibility of linking theory to practice. Indeed, educating teachers about classroom language use might allow them to become more aware as to how students and teachers interact, and how such exchanges promote or hinder the possibility of the meaningful participation that purportedly leads to student learning. In particular, research on how experienced teachers handle the moment-by-moment contingencies of classroom talk would be of great use to teacher-educators and novice teachers alike. We envision teacher education programs that include analyzing actual classroom talk, in order for teachers-in-training to identify and maximize opportunities for their students to engage in the activities that are believed to lead to language development.

REFERENCES

Adger, C. T. (2001). Discourse in educational settings. In D. Schiffrin, D. Tannen, & H. Hamilton (Eds.),Thehandbookofdiscourseanalysis(pp. 503-517). Malden: Blackwell.

Angelova, M., Gunawardena, D., & Volk, D. (2008). Peer teaching and learning: Co-constructing language in a dual-language first grade.LanguageandEducation, 20, 173-190.

Antón. M. & DiCamila, F. (1998). Socio-cognitive functions of L1 collaborative interaction in the L2 classroom.CanadianModernLanguageReview, 54, 314-342.

Bellack, A.A., Kliebard, H.M., Hyman, R.T., & Smith, F.L., Jr. (1966).Thelanguageoftheclassroom. New York: Teachers College Press.

Canagarajah, A.S. (1993). Critical ethnography of a sri lankan classroom: Ambiguities in student opposition to reproduction through ESOL.TESOLQuarterly, 27, 601-626.

Canagarajah, S. (2004). Subversive identities, pedagogical safehouses, and critical learning. In B. Norton & K. Toohey (Eds.),Criticalpedagogiesandlanguagelearning(pp. 116-137). New York: Cambridge University Press.

Drew, P. & Heritage, J. (Eds.) (1992).Talkatwork:Interactionininstitutionalsettings. Cambridge: Cambridge University Press.

Duff, P. A. (1995). An ethnography of communication in immersion classrooms in Hungary.TESOLQuarterly, 29, 505-537.

Duff, P. A. (2002). The discursive construction of knowledge, identity, and difference: An ethnography of communication in the high school mainstream.AppliedLinguistics, 23, 289-322.

Duff, P. A. (2007). Second language socialization as sociocultural theory: Insights and issues.LanguageTeaching, 40, 309-219.

Fagan, D.S. (2012a). ‘Okay’ as a multifunctional resource for giving feedback in classrooms.Language&InformationSociety, 16, 9-41.

Fagan, D.S. (2012b). ‘Dealing with’ unexpected learner contributions in whole-group activities: An examination of novice language teacher discursive practices.ClassroomDiscourse, 3, 107-128.

Fagan, D.S. (2013).ManaginglearnercontributionsintheadultESLclassroom:Aconversationanalyticandethnographicexaminationofteacherpracticesandcognition. Unpublished doctoral dissertation, Teachers College, Columbia University, New York.

Fairclough, N. (2001). The discourse of new labor: Critical discourse analysis. In M. Wetherall, S. Taylor, & S. J. Yates (Eds.),Discourseasdata:Aguideforanalysis(pp. 229-266). London: Sage.

Fairclough, N., Mulderrig, J., & Wodak, R. (2011). Critical discourse analysis. In T.van Dijk (Ed.),Discourseanalysis:Amultidisciplinaryapproach(pp. 357-375). London: Sage.

Fanselow, J. F. (1977). Beyond Roshomon: Conceptualizing and describing the teaching act.TESOLQuarterly, 11, 17-39.

Flanders, N. A. (1970).Analyzingteachingbehavior. Reading: Addison-Wesley.

Gramsci, A. (1971).Selectionsfromtheprisonnotebooks,editedandtranslatedbyQuintonHoareandGeoffreyNowellSmith. New York: International.

Gutiérrez, A.G. (2008). Microgenesis, method and object: A study of collaborative activity in a Spanish as a foreign language classroom.AppliedLinguistics, 29,120-148.

Gutiérrez, K.D. & Larson, J. (1994). Language borders: Recitation as hegemonic discourse.InternationalJournalofEducationalReform, 3, 22-36.

Gutiérrez, K.D., Larson, J., & Kreuter, B. (1995). Cultural tensions in the scripted classroom: The value of the subjugated perspective.UrbanEducation, 29, 410-442.

Gutiérrez, K. D., Rymes, B., & Larson, J. (1995). Script, counterscript, and underlife in the classroom: James Brown versus Brown vs. Board of Education.HarvardEducationalReview, 65, 445-471.

Hall, J. K. (1995). “Aw, man, where you goin??” Classroom interaction and the development of L2 interactional competence.IssuesinAppliedLinguistics, 6, 37-62.

Hall, J. K. (1998). Differential teacher attention to student utterances: The construction of different opportunities for learning in the IRF.LinguisticsandEducation, 9, 287-311.

Hall, J. K. & Walsh, M. (2002). Teacher-student interaction and learning.AnnualReviewofAppliedLinguistics, 22, 186-203.

He, A. W. (2000). The grammatical and interactional organization of teacher’s directives: Implications for socialization of Chinese American children.LinguisticsandEducation, 11, 119-140.

Hellermann, J. (2007). The development of practices for action in classroom dyadic interaction: Focus on task openings.TheModernLanguageJournal, 91, 83-96.

Hellermann, J. (2008).Socialactionsforclassroomlanguagelearning. Clevedon: Multilingual Matters.

Hellermann, J. & Cole, E. (2008). Practices for social interaction in the language-learning classroom: Disengagements from dyadic task interaction.AppliedLinguistics, 30, 186-215.

Hruska, B. L. (2004). Constructing gender in an English dominant kindergarten: Implications for Second Language learners.TESOLQuarterly, 38, 459-485.

Hruska, B. L. (2007). “She My Friend”: Implications of friend ideologies, identities, and relationships for bilingual kindergarteners.MulticulturalPerspectives, 9, 3-12.

Jacknick, C. (2011). Breaking in is hard to do: Negotiating classroom activity shifts.ClassroomDiscourse, 2, 20-38.

Jacknick, C. (2013). “’Cause the textbook says...”: Laughter and student challenges in the ESL classroom. In P. Glenn & L. Holt (Eds.),Onlaughing:Studiesoflaughterininteraction. London: Continuum.

Kahn, G. (2012). Open-ended tasks and the qualitative investigation of second language classroom discourse.JournalofEthnographicandQualitativeResearch, 6, 90-107.

Kanagy, R. (1999). Interactional routines as a mechanism for L2 acquisition and socialization in an immersion context.JournalofPragmatics, 31, 1467-1492.

Kulik, D. & Shieffelin, B. B. (2004). Language socialization. In A. Duranti (Ed.),Acompaniontolinguisticanthropology. (pp. 349-368). Malden, MA: Blackwell.

Kumaravadivelu, B. (1999). Critical classroom discourse analysis.TESOLQuarterly, 33, 453-484.

Lantolf, J. P. & Thorne, S. L. (2006).Socioculturaltheoryandthegenesisofsecondlanguagedevelopment. Oxford: Oxford University Press.

Lave, J. & Wenger, E. (1991).Situatedlearning:Legitimateperipheralparticipation. Cambridge: Cambridge University Press.

Lee, Y. (2007). Third turn position in teacher talk: Contingency and the work of teaching.JournalofPragmatics, 39, 180-206.

Lemke, J. L. (1990).Talkingscience:Language,learning,andvalues. New York: Ablex.

Markee, N. (2000).Conversationanalysis. New York: Routledge.

Markee, N. (2004). Zones of interactional transition in ESL classes.TheModernLanguageJournal, 88, 583-596.

Markee, N. (2005). The organization of off-task talk in second language classrooms. In K. Richards & P. Seedhouse (Eds.),Applyingconversationanalysis(pp. 197-213). New York: Palgrave.

Markee, N. & Kasper, G. (2004). Classroom talks: An introduction.ModernLanguageJournal, 88, 491-500.

McHoul, A. W. (1978). The organization of turns at formal talk in the classroom.LanguageinSociety, 7, 183-213.

McHoul, A. W. (1990). The organization of repair in classroom talk.LanguageinSociety, 19, 349-377.

Mehan, H. (1979a).Leaninglessons:Socialorganizationintheclassroom. Cambridge: Harvard University Press.

Mehan, H. (1979b). “What time is it, Denise?”: Asking known information questions in classroom discourse.TheoryintoPractice, 28, 285-294.

Mori, J. (2002). Task design: Plan, and development of talk-in-interaction: An analysis of a small group activity in a Japanese language classroom.AppliedLinguistics, 23, 323-347.

Morita, N. (2004). Negotiating participation and identity in second language academic communities.TESOLQuarterly, 38, 573-603.

Nassaji, H. & Wells, G. (2000). What’s the use of the “triadic dialogue?”: An investigation of teacher-student interaction.AppliedLinguistics, 21, 376-406.

Nelson, C. (1999). Sexual Identities in ESL: Queer theory and classroom inquiry.TESOLQuarterly, 33, 371-391.

Norton, B. & Toohey, K. (Eds.). (2004).Criticalpedagogiesandlanguagelearning. New York: Cambridge University Press.

Norton, B. (2001). Non-participation, imagined communities and the language classrooms. In M. P. Breen (Ed.),Learnercontributionstolanguagelearning(pp. 159-171). London: Longman.

Nystrand, M. (1997).Openingdialogue:UnderstandingthedynamicsoflanguageandlearningintheEnglishclassroom. New York: Teachers College Press.

Ochs, E. & Schieffelin, B. B. (1984). Language acquisition and socialization: Three developmental stories and their implications. In R. Schweder & R. LeVine (Eds.),Culturetheory:Essaysonmind,selfandemotion(pp. 276-320). New York: Cambridge University Press.

Ohta, A.S. (1999). Interactional routines and the socialization of interactional style in adult learners of Japanese.JournalofPragmatics, 31, 1493-1512.

Ohta, A. S. (2000). Rethinking interaction in SLA: Developmentally appropriate assistance in the zone of proximal development and the acquisition of L2 grammar. In James P. Lantolf (Ed.),Socioculturaltheoryandsecondlanguagelearning(pp. 51-78). New York: Oxford University Press.

Olsher, D. (2005). Talk and gesture: The embodied completions of sequential actions in spoken interaction. In R. Gardner & J. Wagner (Eds.),Secondlanguageconversations(pp. 221-245). London: Continuum.

Olsher, D. (2008). Gesturally-enhanced repeats in the repair turn: Communication strategy or cognitive language-learning tool? In S. G. McCafferty & G. Stam (Eds.),Gesture:Secondlanguageacquisitionandclassroomresearch(pp. 109-130). New York & London: Routledge.

Poole, D. (1992). Language socialization in the second language classroom.LanguageLearning, 42, 593-616.

Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation.Language, 50, 696-735.

Schegloff, E. A. (2007).Sequenceorganizationininteraction:Aprimerinconversationanalysis,Volume1. Cambridge: Cambridge University Press.

Schieffelin, B. B. & Ochs, E. (1986). Language socialization.AnnualReviewofAnthropology, 15, 163-191.

Seedhouse, P. (1999). Task-based interaction.ELTJournal, 53, 149-156.

Seedhouse, P. (2004).Theinteractionalarchitectureofthelanguageclassroom:Aconversationanalysisperspective. Malden: Blackwell.

Simon, R. I. & Dippo, D. (1986). On critical ethnography work.AnthropologyandEducationQuarterly, 17, 195-202.

Sinclair, J. M. & Coulthard, R. M. (1975).Towardsananalysisofdiscourse:TheEnglishusedbyteachersandpupils. London: Oxford University Press.

Swain, M. (2000). The output hypothesis and beyond: Mediating acquisition through collaborative dialogue. In James P. Lantolf (Ed.),Socioculturaltheoryandsecondlanguageearning(pp. 97-114). New York: Oxford University Press.

Thornborrow, J. (2002).Powertalk:Languageandinteractionininstitutionaldiscourse. London: Virago.

Toohey, K. (1996). Learning English as a second language in kindergarten: A community of practice perspective.TheCanadianModernLanguageReview, 52, 549-576.

Vygotsky, L. S. (1978).Mindinsociety:Thedevelopmentofhigherpsychologicalprocesses. Cambridge: Harvard University Press.

Walsh, S. (2002). Construction or obstruction: Teacher talk and learner involvement in the EFL classroom.LanguageTeachingResearch, 6, 3-23.

Walsh, S. (2006).Investigatingclassroomdiscourse. New York: Routledge.

Waring, H. Z. (2008). Using explicit positive assessment in the language classroom: IRF, feedback, and learning opportunities.TheModernLanguageJournal, 92, 577-594.

Waring, H. Z. (2009). Moving out of IRF (initiation-response-feedback): A single case analysis.LanguageLearning, 59, 796-824.

Waring, H. Z. (2011). Learner initiatives and learning opportunities.ClassroomDiscourse, 2, 201-218.

Waring, H. Z. (2012a). Yes-no questions that convey a critical stance in the language classroom.LanguageandEducation, 26, 451-469.

Waring, H. Z. (2012b). “Any questions?:” Investigating understanding-checks in the language classroom.TESOLQuarterly, 46, 722-752.

Waring, H. Z. (2013a). Doing being playful in the language classroom.AppliedLinguistics, 34, 191-210.

Waring, H. Z. (2013b). Managing the competing voices in the language classroom.DiscourseProcesses, 50, 316-338.

Waring, H. Z. (2013c). “How was your weekend?”: Developing the interactional competence in managing routine inquiries.LanguageAwareness, 22, 1-16.

Wood, D., Bruner, J., & Ross, G. (1976). The role of tutoring in problem-solving.JournalofChildPsychologyandChildPsychiatry, 17, 89-100.

Willet, J. (1995). Becoming first graders in an L2: An ethnographic study of L2 socialization.TESOLQuarterly, 29, 473-503.

Zuengler, J. & Mori, J. (2002). Microanalyses of classroom discourse: A critical consideration of methods.AppliedLinguistics, 23, 283-288.