Ta

ST

a

AA

KCRCEL

I

amS2a

lialioi

r

0

Linguistics and Education 29 (2015) 1–14

Contents lists available at ScienceDirect

Linguistics and Education

j ourna l h omepa ge: www.elsev ier .com/ locate / l inged

eacher repair in a second language class for low-literatedults

eo Hyun Park ∗

he Ohio State University, 611 Cuyahoga Ct., Columbus, OH 43210, USA

r t i c l e i n f o

rticle history:vailable online 26 November 2014

eywords:onversation analysisepairlassroom discoursenglish as a second languageow-literate adults

a b s t r a c t

This paper complicates the practice of teacher repair in the second language (L2) classroomby specifically focusing on the subset of adult English learners who have a background of lowliteracy. Using Conversation Analysis, this study explores the interactional means locallyavailable for the low-literate learners to deal with teacher repair in the activity of vocabularyintroduction. The analysis shows that the organization of teacher repair is oriented to thelearners’ state of literacy, which for the teacher holds priority over what the literature hasfound as repair strategies sui generis either inside or outside the L2 classroom. The teacherprofitably tailors her (para)linguistic input for the learners to perceive and react to herrepair, and this rule-governed turn construction helps them identify the pedagogical intentof each turn. These findings will enrich the discussion of whether our existing knowledgeof L2 repair can be extended to all learners.

© 2014 Elsevier Inc. All rights reserved.

ntroduction and background

Considering that language is both the means and the object in second language (L2) classroom interaction, conversationnalysts have found that repair in L2 classrooms is reframed by the institutional object and thus distinct from repair inundane discourse (Jung, 1999; Kasper, 1985; Lehti-Eklund, 2013; Nakamura, 2008; Okada, 2010; Rolin-Ianziti, 2010;

eedhouse, 1997, 2004; van Lier, 1988). Based on and along with research on repair in generic classroom talk (Macbeth,004; McHoul, 1990), those studies of L2 classroom repair have shed light on how repair is exploited instructionally as wells interactionally towards the goal of language acquisition.

While following this strand of research, this paper focuses specifically on teacher repair towards low-literate adultearners’ troubles in L2 communication. Repair understood by this particular learner population is an intriguing topic forts own right because it is a dialogic site where the learners potentially negotiate meanings and keep engaged in learningctivities. This topic also merits attention to extend our knowledge of repair by and for widely defined types of languageearners. Liebscher and Dailey-O’Cain’s (2003) focused interest in advanced L2 learners in their repair study, for example,mplies that different L2 learner groups may develop and benefit from different repair trajectories. At the other extreme

f language proficiency, this paper explores the context-sensitive mechanisms that low-literate adults find appropriate fordentifying and resolving breakdowns in their English as a second language (ESL) classroom talk.

Low-literate ESL learners in this paper refer to those who cannot read or write in their first language (L1) but who currentlyead and write Roman alphabets and are in the process of practicing how to combine them to read and write English words.

∗ Tel.: +1 614 804 0633.E-mail address: park.1050@buckeyemail.osu.edu

http://dx.doi.org/10.1016/j.linged.2014.10.007898-5898/© 2014 Elsevier Inc. All rights reserved.

2 S.H. Park / Linguistics and Education 29 (2015) 1–14

These learners are to be distinguished from literate learners at a beginner level of ESL, who are familiar with basic skillsof reading and writing language as a semiotic system (Bigelow, 2010; Moore, 2007). ESL learners with low print literacyalso depart from native speakers (NS) of English with low literacy (Eme, 2011) in that they need to practice speaking andlistening to English as a new language as well as reading and writing. In this respect, the pre- or low-literate students in ESLclasses are primarily immigrants who recently arrived in the Anglosphere and whose background of L1 illiteracy may resultfrom a number of reasons: one’s schooling might be severely disrupted before entering the host country, or one’s ethnicgroup might traditionally prefer oral language performances such as stories or poems to literacy, compared to the prevailing“chirographic culture” in most developed countries (Ong, 1988, p. 2).

L2 researchers have called for more studies on this particular learner group in response to an exploding migrant popu-lation, the majority of whom is coming to the first-world countries where English language skills are essential for everydaylife (e.g. Dooley & Thangaperumal, 2011; Strube, 2010; Tarone, 2010; Wrigley, 2007; see also proceedings of Low EducatedSecond Language and Literacy Acquisition for Adults symposia, available at: http://www.leslla.org). Investigating adult L2learners with limited print literacy or formal schooling is an important and thoughtful move because they have been largelyunderstudied compared to school-age or higher-educated immigrants. There is evidence that pre-literate and low-literateadults acquire the oral processing of an L2 differently than literate learners, with little awareness of linguistic units likewords and phonemes (Tarone, Bigelow, & Hansen, 2009). These authors’ finding indicates that our current understanding ofhow “the human mind acquires L2s” may be in fact merely “based on data from some humans (the literate ones)” (Tarone &Bigelow, 2011, p. 6, emphases in original). The field of second language acquisition (SLA) must include learners’ alphabeticliteracy level and interrupted education in its research agenda for a more comprehensive and ecological theory-building.

It is of particular value to examine the practice of repair that occurs in low-literate adults’ language classes and howit is distinct from repair in classes for other learner groups or outside the classroom. In their ESL classroom contexts, theasymmetry of language competence is greatest between an NS teacher of English and her students. What is consideredrepairable in this situation will predominantly be language problems made by learners, who are positioned at an extremeof “not-yet-competent” interactants in the literate and English-speaking social world (Schegloff, Jefferson, & Sacks, 1977, p.381). Therefore, L2 classes for low-literate adults are likely to have a particular repair trajectory, where the teacher helps thelearners carry out smooth interactions, by both initiating and completing repairs toward their silence, cut-off or try-markingutterances, as presented in the following example:

Teacher: What do you see in this picture?Student: Dino-

→ Teacher: A dinosaur.Exploring how low-literate L2 learners navigate teacher repair in the classroom environment is also meaningful because

their access to repair as a resource determines how they modify L2 input and output, which may contribute to languagedevelopment (Hardy & Moore, 2004). By constantly adjusting what is said (trouble source) and what is heard (repair), thelearners notice the linguistic gap between the two, which is central to the acquisition of a language (Doughty & Williams,1998). Prior to identifying such a connection between repair and language learning, this paper investigates how trouble isrepaired in L2 classes for learners with low literacy. To examine the detailed fabric of talk surrounding repair using theirown perspective, this study utilizes conversation analysis (CA) as a methodological framework, which will be outlined inthe following subsections.

Repair in the L2 classroom

Transparent and smooth communication is built on a conversational mechanism that breaks through what is obstructedand clarifies what is cloudy. Defined as “practices for dealing with problems or troubles in speaking, hearing, and [. . .]understanding what someone has just said” (Schegloff, 2000, p. 207), repair is one of the most ubiquitous events in humaninteraction. Repair in the L2 classroom environment, among others, has been comprehensively investigated under the cate-gory of institutional talk (Drew & Heritage, 1992). The consensus among those studies is that the “institutional or professionalidentities” assigned to each participant constrain the organization of repair and hence result in certain features of L2 class-room repair sui generis (Drew & Heritage, 1992, p. 4). The effect of such contextualized repair is the transmission of L2knowledge, as Iles (1996) states that “language is demonstrated, experienced, and worked on by both teacher and learnerin repair trajectories” in L2 classroom discourse (p. 25).

Kasper (1985) represents one of the early efforts to explore the interaction between repair and L2 classroom contexts.Explaining the repair organization in language-centered and content-centered phases of foreign language lessons, the authorshows that the dominant type of repair and the teacher’s goal correspond to which phase the class belongs to: more other-repairs appear in language-centered phases and more self-repairs in content-centered phases. Seedhouse (1997, 2004)moves a step beyond Kasper by focusing on various pedagogical contexts of L2 classes that define what is repairable: form-and-accuracy, meaning-and-fluency, and task-oriented contexts. The pedagogical focus of each context is related reflexivelyto the organization of repair. Similarly, van Lier (1988) examines pedagogy as a key factor that differentiates the way repair is

dealt with in the L2 classroom from that in non-pedagogic settings. And this is why other-repair prevails in the L2 classroom,whereas self-repair dominates in daily conversation.

Other L2 repair studies have revisited Kasper, Seedhouse, and/or van Lier’s works by examining interaction betweenNS–NNS (Hellermann, 2009; Hosoda, 2006; Wong, 2000), NNS–NNS (Kasanga, 1996) or teacher-led classroom interaction

S.H. Park / Linguistics and Education 29 (2015) 1–14 3

Table 1The cross-cutting organization of IRE and repair.

Turn Speaker IRF Repair

1 Teacher Initiation

(oLawr

ltwhi

C

asTTT

riEr

tsed1(ios

ricti

ore(op

rib

2 Student Response Trouble source3 Teacher Feedback Repair initiation4 Student Repair completion

in addition to the works cited in the first sentence of this paper, Cho, 2008; Iles, 1996; Juvonen, 1989; Seong, 2004). Fewf these, however, narrow down and pay attention to a specific learner group’s characteristics, except for aforementionediebscher & Dailey-O’Cain’s (2003) research on repair by advanced L2 learners. Although Cho (2008) and Seong (2004)ccount for types and distributions of repair strategies depending on ESL grade levels, their results are rather taxonomic,hich therefore might not fully explicate the way the participants display to one another their orientation to the L2 classroom

epair.The primary concern of this paper is to explore deviations from the existing literature on L2 classroom repair when the

earner group is limited into the low-literate adults. As Tarone (2010) laments, they are one of the most understudied inhe field of SLA. They are not ready to speak out much yet and when they do, their use of L2 is telegraphic with one or twoords at best. Thus, it is the teacher who apparently shapes the interaction in this setting, and this teacher-centerednessas been broadly explored along with the conventional sequence of teacher initiation-student response-teacher feedback

n classroom discourse research.

A for reframing IRF

As stated earlier, the classroom environment finds more other-repairs than ordinary conversation, and this feature isttributed to pedagogy as an occasion that the participants are oriented to in this particular context. Indeed, McHoul (1990)hows that 55 percent of the repair sequences in his corpus consist of three turn constructional units as follows (p. 352):urn 1: Trouble source (Student)urn 2: Repair initiation (Teacher)urn 3: Correction (Student)

According to McHoul, the teacher elicits student self-correction in Turn 3 by delaying strategies such as cluing or questioneformulations in Turn 2.1 That is, to maximize the opportunity for the student self-repair, the teacher’s repair initiations placed in the third slot of Initiation-Response-Feedback (IRF) convention (Sinclair & Coulthard, 1975; Initiation-Reply-valuation in Mehan, 1979). Informed by Macbeth’s (2004) reformulation of McHoul’s (1990) idea, Table 1 exhibits theelationship between IRF and repair sequences.

Putting the two schemata together, Macbeth (2004) demonstrates that repair is relevant with an instructional goalowards which the teacher and the students orient themselves in classroom discourse. The repair sequence and the IRFequence are concurrent but distinctive, although the former is “relevant throughout” the latter (Macbeth, 2004, p. 719,mphasis in original). The three-part IRF sequence has been one of the paradigmatic exchanges of turns observed in classroomiscourse (Cazden, 2001; Hall, 1998; Hellermann, 2003; Nassaji & Wells, 2000; Nystrand, 1997; Seedhouse, 1996; Wells,993). This formulaic pattern as a lens to look into classroom talk, however, was not without criticism; Drew and Heritage1992), for instance, point out that the IRF model fails to specify “how participants show their orientations to the particularnstitutional context” due to its simplistic formulation. Waring (2009) argues that more learning opportunities can existutside the normative IRF sequences, which are “tightly chained to one another within the same exercise, offering fewpaces in between” for student agency (p. 807).

What Macbeth (2004), Drew and Heritage (1992), and Waring (2009) share in common methodologically is the use orecommendation of CA, a fine-grained tool to examine the speakers’ own perspectives and changing identities manifestedn talk-in-interaction. The CA framework complicates the IRF model by focusing on the sequential development of speakers’ollaborative effort in managing interactional problems. CA-oriented research also sheds wider light on learning opportuni-ies and learner agency in and around the IRF exchange by looking at “members’ methods for repair,” which are “indigenous,nteractionally-defined structural units of language” (Hellermann, 2009, p. 128).

This paper joins this line of research, noting that CA is of benefit in particular to highlight the complexity and dynamicityf the low-literate learners’ L2 interaction. Their conversation in language-learning environments may not be lexicallyich, because of the extreme lack of L2 knowledge; every conversation is socially rich, however, with the participants’ fullngagement in the logical sequencing of verbal and bodily actions and reactions to others. Here the social means, as Erickson1982) defines citing Weber (1978), “action taken in account of the actions of others” (p. 155). Thus, what this article focusesn through CA is not “a lexicalized straightjacket” (Huth, 2011, p. 300), but the interactional means locally available to thearticipants that fulfill the practice of repair in the context of L2 learning.

Analyzing five instances where a teacher and low-literate English learners recognize and resolve trouble sources sur-ounding a vocabulary introduction activity, this article reports two unique structural traits: (1) the recursive pattern ofnitiation-response-feedback-repetition of feedback (henceforth IRFF′) and (2) the distant but corresponding relationshipetween the first and the fourth slots of IRFF′. In addition, this article explores four ways of teacher repair particularly

4 S.H. Park / Linguistics and Education 29 (2015) 1–14

oriented to the learners’ low literacy: (1) the teacher’s straightforward repair without allowing any student to make anattempt to repair, (2) one-word elicitation of the pedagogical focus, (3) multiple rounds of repair with multifaceted lin-guistic approach and the moment of teacher’s laissez-faire decision, and (4) the teacher’s shifting treatment of her studentsbetween a cohort and individual speakers of trouble sources.

Methodology

The naturally-occurring conversations were video-recorded in an ESL class offered by a 501(c)(3) nonprofit organizationin an American Midwestern city, once a week for 130 min for eight weeks. Funded by the city’s foundation and a contractwith the county, several NGOs in the city support newly-arrived adult refugees admitted to the United States from overseasrefugee camps. The ESL program is one of the free services they offer. An American teacher had taught the observed class as avolunteer for eight months at the point of data collection, with the English for Speakers of Other Languages (ESOL) certificateobtained after her retirement as a librarian. All the students in her class were refugees over the age 25 whose level of alphabetliteracy was low. Their literacy level had been diagnosed by a placement interview. According to the program’s placementcriteria, students at the low level know fewer than 50 English words and how to pronounce and write the Roman alphabets.They may not know, however, how to combine alphabets to make a particular sound and pronounce unfamiliar vocabulary.The class had four Ethiopians, two Somalis, and one Burundi. The period for which the students had been in the UnitedStates varied from several months to two years. Appendix 1 displays how each participant’s pseudonym is abbreviated inthe transcripts of this paper.

I unobtrusively sat in the left front corner of the classroom, taking field notes and video-recording the conversation with acamera on a tripod from five-minute before the class starts to five-minute after the class is over. Watching the videos carefullyand repeatedly, I extracted random samples of repair from each lesson and activity for initial analysis. Of seven activitiesobserved in eight sessions in total (i.e. greetings, guided dialogs, vocabulary, writing, games, tests, small talk), I decidedto focus on the activity of vocabulary introduction and the sequential development of repair in it.2 Re-watching all thescenes of vocabulary introduction in the movie, I found 54 repair episodes and transcribed them following CA transcriptionconventions (Jefferson, 2004) (see Appendix 2). Through this method, as Hutchby and Wooffitt (2008) note, the transcript isconsidered as a “representation . . . of a determinate social event”, which is, in this paper, the practice of repair (p. 70). Beyondthis, I interviewed the teacher of the observed class after completing the data collection to member-check my understandingof classroom activity procedures and foci. Another follow-up interview was conducted with an Oromo-English interpreter,watching the selected video clips together, as an effort to comprehend what some students meant by their non-Englishutterances during the lessons. Both interviews were audio-recorded and transcribed for analysis.

Analysis

This section analyzes the phases of introducing new English words to learners with low literacy. I selected this particularactivity for two reasons: first, the activity of introducing new words transpired for a significant amount of time, from 20 to30 min (up to 25% of the entire session), in every class that I attended. This observation is not surprising, considering thatvocabulary is a prerequisite resource for L2 learners to acquire before they can perceive or produce more complicated turnsin the form of phrases or sentences (Dehghani, Motamadi, & Mahbudi, 2011). Second, repair was observed the most oftenwhile the participants were involved in the activity of vocabulary introduction. Researchers have dealt with vocabularyintroduction as one of the primary classroom activities for both beginning L1 (Levy & Lysynchuk, 1997) and L2 readers(Sagarra & Alba, 2006). Once the learners acquire the ability to treat unfamiliar words by analogy to known words, theteacher can reduce emphasis on vocabulary during the in-class lesson. For those at beginner levels, however, the teacherinvests a great amount of time in vocabulary and presents each unfamiliar word metalinguistically as a unique form toapprehend.

Such metalinguistic instruction of L2 vocabulary in my data largely comes in the form of the mechanical question (eitherdirect or indirect) and answer (Q–A) sequence, e.g. “what is x?” (Teacher)–“(It is) y” (Student(s)). Of course, this base pairtransforms and expands in numerous ways, as will be analyzed throughout this section. By maintaining the above uncom-plicated structure of classroom talk for teaching vocabulary, the teacher orients to what she believes to be the specialinstructional needs of low-literate English language learners. Below is an extract from an interview with the teacher:

Most of them are really illiterate in their own language. So they have very little experience of school. Some of the mostbasic concepts of language is- at least I found this with my students- they didn’t have the clear idea as one might thinkabout what a letter was, what a word was, and certainly what a sentence was. You know, one thing I try to be verycareful about when I write on the board, for example, is to put a sufficient space between each word to emphasizethe fact that they’re individual words. I try to emphasize that over and over again . . . That sounds like a very simplequestion but it isn’t.

The teacher’s understanding of the learners’ linguistic history as well as their current status, as above, also results in therecursive pattern of both repair and classroom activity across the lessons; 48 out of 54 repair episodes occurred within theidentical buildup of the following three phases of vocabulary introduction3: In Phase 1, new words are preannounced inprint or image. In Phase 2, the structured, even ritualistic, Q–A exchange aforesaid occurs. The teacher asks the students how

S.H. Park / Linguistics and Education 29 (2015) 1–14 5

ttblt

P

tsiE

1234567891111111

tatt

Fig. 1. A scanned textbook page on the screen.

o read a word and the students try to answer, in a response that is usually repairable. In Phase 3, the Q–A pair expands ashe repair trajectory becomes longer as occasion arises. These phases of the particular classroom activity will be a situationalackground against which this article explains the sequential characteristics of the repair practice in the L2 class for low-

iterate learners. Since this paper illustrates one such progression during the teaching of new vocabulary across the lessons,he five excerpts to be analyzed in this section are from the data collected on the same day.

hase 1: Preparation and transition

The gist of what happens in the first phase is that the students are exposed to the printed words or pictures prior to theeacher’s explicit guidance on what to do with them. For example, Fig. 1 is a scanned textbook page projected on the frontcreen, and Excerpt 1 details the classroom interaction while and after the teacher sets up a laptop and beam projector tontroduce to her students the names of rooms in American houses.xcerpt 1

T: (Looking at her laptop monitor and the front screen by turns several times))◦Okay um-

(4.0) T: ◦Oka:y. There it is. Amazing. Um-◦= Ha ((Looking at the screen)) =◦Home. ((Coughing)) T: ((Scrolling the scanned page down on the screen)) Okay. What we’re going to be looking at (0.5) are (1.0) places inda- (.) in the home. Here’s the word, HO:me. Let’s all say that.0 (1.5)1 T: Home.2 Ss: Home. =3 T: = Home. >Yeah.< In our houses.4 (0.5)5 Ma: Hou[ses6 Sa: [House.

In the process during which technology becomes ready for the lesson, the teacher’s three “Okays” (lines 2, 4, 7) marks

he transition of activity. According to Beach (1993), a speaker’s transitional employment of “Okays” displays not only anttention to prior turn but simultaneously “state of readiness for movements to next-positioned matters” (p. 329). Theeacher in Excerpt 1 gradually establishes “Okays” along with her nonverbal actions, such as looking at the computer andhe screen by turns or scrolling down the screen, in the transition from one activity to another. Although the teacher does

6 S.H. Park / Linguistics and Education 29 (2015) 1–14

not verbally clarify any pedagogical focus until line 8, her turn, “There it is,” signifies that the pedagogical focus was alreadyexposed to the students at that moment. Another evidence for the introduction of a pedagogical focus is Haji’s turn “Home”(line 5). Thus, the learners are provided a cue to perceive the transition of activity (or the start of a lesson) and what to focuson pedagogically not only by the teacher’s verbal production of “Okays” but also by the visual provision of a target form.

In the following turns in lines 7–8, the teacher creates a context for the new activity, providing the learners with an overallpicture of what they are going to do. The subject part in her turn “what we’re going to be looking at” notifies the learners inadvance that its complement will be the pedagogical focus that they need to pay attention to. This notification is reinforcedwith her two pauses and a stress on the word “home.” By additionally combining “here’s the word” with “HO:me” as a formmore emphasized phonetically than the previous one, “home,” the teacher introduces the lexical item in focus. Subsequently,the teacher requests her students to say the word, which is the first pair part (FPP) of an adjacency pair (Schegloff & Sacks,1973). Now the context makes the next turn relevant; the students are expected to accept such an instructional request byreciting the word “home” immediately after the teacher’s turn. The students, however, fail to produce the second pair part(SPP) of the paired sequence. The pause in line 10 displays that they have a problem of hearing or understanding. By repeatingthe lexical item (line 11), the teacher diagnoses the previous pause as the students’ trouble with hearing or understanding.In this sense, the teacher’s repetition plays a role of repair initiation. And the students complete the repair, displaying theirunderstanding of what has been requested by the teacher’s reciting the targeted word (line 12). The teacher confirms andevaluates such completion (line 13). As Pomerantz (1984) observes that preferred responses are characteristically performedwithout delay, the teacher’s evaluation latches the previous turn with an ensuing positive marker, “yeah.”

The episodes of repair that share the sequential feature with Excerpt 1 (i.e. setup – transition mark – learning new words)appeared reasonably often during my observation. On all such occasions, the use of visual aids preceded the teacher’s verbalintroduction to new words. This structure plays two roles. First, as researchers have proved (Hammerly, 1995; Pouwels,1992), the combination of words and pictorial aids facilitates L2 learning. Second, the visual aids in this researched classallow the teacher to spare her breath to navigate the teacher-led classroom activities. Put otherwise, visual cues often speaklouder than words, especially for illiterate or low-literate learners. Moore (2007) points out that to ESL beginners with lowliteracy the teacher’s “language of the instructions” is often more challenging than “the language actually needed to performthe task” (p. 7). Note that the teacher’s relatively lengthy explanation on the upcoming activity in lines 7–8 in Excerpt 1 doesnot work well.

Rather, the trouble in Excerpt 1 is resolved by the teacher’s structurally even simpler turn, “home” (line 11). That is, thestudents successfully produce the pedagogical focus without the teacher’s verbose instruction. And this one-word elicitationof the learners’ target word production in unison occurs over and over throughout the activity. Hence, the teacher’s repairinitiation in line 11 in Excerpt 1 not only functions as her effort to resolve the trouble source at hand but lays a foundationfor the linguistic form through which the learners can access repair as a lexical resource throughout the activity. I will comeback to this issue in the discussion of Phase 2.

Phase 2: The question–answer sequence for vocabulary introduction

In Phase 1, the teacher embarks on the new word activity in the form of the one-word elicitation. The teacher simplifiesthe lexical input in a word to ease access for her students. In Phase 2, the participants’ tacit agreement to such organizationof repair for vocabulary learning is maintained in an ongoing activity. Overall, the work of repair in Phase 2 reflects a typicalIRF triad reviewed in the previous section. In other words, the repair trajectory displays a specific institutional goal, “theinstruction of novices” (Macbeth, 2004, p. 721) in Phase 2.Excerpt 2

1 T: Yah. Okay. (.) Um number three:: i::s-2 Ha: Be:d?3 T: BEdroom.4 S1: Bed[room.5 S2: [Bedroom.6 T: Bedroom.=7 Ma: =Bederoo:m.8 T: Number FOu:r i::s-9 Ma: Living?10 Ha: Living-11 T: LIving room.12 Ss: Living room.13 T: Uh-hum, (.) ((Looking at Mayike)) oh I see Mayike. ((Pointing at a piece of paper14 on Mayike’s table)) >You’ve got a paper. Hehe.< ((Turning her eyes back to the15 screen)) Okay number FI:ve (.) i:s-

In the first line of Excerpt 2, the teacher invites her students to co-complete her turn, by cutting it off after the verb “is”

and lengthening the vowel sound in the verb. The same is found in line 8. Such a cutting-off and phonetic emphasis guidesthe students directly to the pedagogically-focused lexical item. However, Haji and Mayike have trouble to producing the fulltarget words (lines 2, 9, 10). The data do not tell us whether such failure results from the learners’ trouble with pronunciation,or with retrieval of the lexical item or of the knowledge about the lexical item itself. What Haji’s and Mayike’s turns tell us is

S.H. Park / Linguistics and Education 29 (2015) 1–14 7

Table 2The cross-cutting organization of repair in Excerpt 2.

Line Speaker IRF Repair

8 T Initiation9 Ma

Response Troublesource10 Ha

11 T Feedback Repair12 Ss Repetition of feedback (F′)

tc

aalrct

tbWitilNoas

ko2osuEwr

atar

arl

hat their construction of the turns with an accompanying upward intonation accomplishes a specific goal. By exhibiting aontoured guess, Haji and Mayike create “the accommodating environment for other correction” (Seedhouse, 1997, p. 562).

The social action that Haji and Mayike aim to perform through these “try-markers” (Sacks & Schegloff, 1979) is understoods such by the teacher, who immediately provides repairs (lines 3, 11). At this point, the teacher’s corrections not only displayn orientation to prior turns, but also create a pedagogical environment where the other students can recite the correctedexical items, as shown in lines 4, 5, 7 and 12. Intriguingly, lines 8 through 12 above identically repeat the pattern of other-epair in lines 1–7. These patterns support Macbeth’s (2004) argument that classroom correction and the work of repair areross-cut in the lesson problems. Table 2 schematizes this cross-cutting organization of repair in lines 8 through 12, adoptinghe form of Table 1 in the present paper.

A marked distinction here from Table 1 is that Table 2 does not distinguish repair initiation from repair completion; theeacher does not withhold repair completion until any student makes an attempt to repair. This decision of the teacher maye also linked to the students’ language proficiency, but this assumption cannot be verified by the transcript data alone.hat we see for sure is that the teacher’s turn in Excerpt 2 straightforwardly offers the correction as a feedback follow-

ng the students’ try-markers. What is added in this sequence (and in Table 2), instead of the students’ self-correction inhe fourth turn, is the multiple students’ (Ss) reproduction of the teacher feedback, or the teacher’s other-repair. Unliken Excerpt 1 where the teacher explicitly gives the students a direction by saying “let’s all say that,” the students inine 12 in Excerpt 2 are not explicitly required to repeat the lexical item corrected by the teacher in the prior turn.otwithstanding, they produce the pedagogical focus without instruction. This phenomenon, the students’ response inne accord to the teacher’s one-word elicitation in the discussion of Excerpt 1, occurred in the gamut of classroom inter-ction that I observed. Additionally, this trend was member-checked during the follow-up interview with the teacher, whoaid:

Obviously I’m the one who talks the most in the class, but I’m always conscious of the kind of word or expressionI choose to say. Sometimes I just say a word over and over, until they get familiar with it. It’s almost impossible toexplain them what the word means in English. They won’t get it. So a lot of utterances that I make during the classconsist of one or two words, especially when the word is what I want them to learn.

The teacher talk is tailored to the learners’ ability not only to comprehend L2 input but also to make L2 output. The L2nowledge becomes available, through the form of single words, for the students to perceive and recite. This achievementf linguistic accessibility by the teacher’s one-word elicitation can be explained in the conversation data as well. In Excerpt, the students’ repetition with one voice is clearly oriented toward the teacher’s original request for the co-completionf the targeted word. For instance, the students’ chorus in line 12 is attributed to the teacher’s cut-off turn in line 8. Thetudents in lines 4, 5, and 7 also return to the teacher’s invitation to co-completion in line 1. This orientation implies a sharednderstanding among participants concerning the teacher’s management of the class as a cohort (Payne & Hustler, 1980).xperienced teachers, standing in front of dozens of students, establish a relationship with the class as a whole rather thanith each individual separately. Payne and Hustler remark on how the assembled parties in the classroom accomplish this

elationship:

The teacher does not announce that the lesson is beginning, nor does he explicitly describe the pupils as a collectivity;rather it is through the organization of his talk that these actions are being made available to the parties in the occasionand to us as hearers/readers/observers of the interaction (Payne & Hustler, 1980, p. 53).

Likewise, the teacher in Excerpt 2 does not announce exactly whom she wants to complete her cut-off turns in lines 1nd 8. The teacher’s feedback (i.e. the third slot of IRFF′ in Table 2) is also oriented toward the entire class, in spite of the facthat, with regard to the practice of repair, she merely corrects one or two students’ try-marked guesses. And the students,ware of this mechanism, respond to the teacher’s incipient request as a cohort with one voice. To display the correspondingelationship between the first and the fourth turns in the IRFF′ quadriad, Table 2 can be updated as below.

In the first pair part, the teacher heralds a new pedagogical focus by asking the students to guess and orally produce it. After “trial and error” interval, all the students duly reach the second pair part. This paired relationship between the teacher’sequest/question and the students’ repetition of the teacher’s feedback is maintained even when the teacher feedback is

onger than a word, as found in Excerpt 3.

8 S.H. Park / Linguistics and Education 29 (2015) 1–14

Table 3The relationship between L2 classroom repair and adjacency pair.

Speaker IRF Adjacency pair Repair

T Initiation First pair partS Response Trouble sourceT Feedback Repair

′

Ss Repetition of feedback (F ) Second pair part

Excerpt 3

1 T: Uh-hmm, (1.0) ((Moving the mouse pointer from one picture to another on the2 screen)) what do you think this is.3 (3.0)4 Ha: House?5 T: It’s also a- (.) a house in the sense that people live. But this is an aPA:rtment6 building.7 Ha: Apart[ment building.8 Hu: [Apartment building.9 Sa: [Apartment.=10 Ma: =Building.

After a three-second pause, Haji tries to answer the teacher’s question. Unlike what we have seen in Excerpts 1 and 2, herethe ensuing feedback consists of more than one word (lines 5–6). Nonetheless, the students selectively recognize what thepedagogical focus is there and produce it from lines 7 through 10, as the SPP toward the FPP, the teacher’s original questionin line 2. To apply this string to Table 3, Excerpt 3 proceeds as Initiation (line 2) – Response (line 4) – Feedback (lines 5–6) –Repetition of Feedback (lines 7–10).

Of interest in this teacher feedback is the moment-by-moment course of teacher-repair. Above all, the teacher’s initiationof repair in line 5 concurs with Seedhouse’s (1997) assessment of a teacher’s tendency to avoid overt negative evaluationson students’ linguistic errors. The teacher first partially agrees that Haji’s attempt is not wrong in a sense. However, thefollowing conjunction “But” is a cue which indicates that his answer of “house” is not what the teacher expects to hear. Priorto this, the students are likely to perceive that Haji’s turn in line 4 has a problem already from the moment when the teacherutters the word “also” in line 5. Beyond demonstrably denoting the existence of trouble, the teacher’s use of the word “also”projects the upcoming conjunction “But.”

Hutchby and Wooffitt (2008) claim that the projectability in turn-construction units enables the participants to foresee“what sort of unit it is and at what point it is likely to end” (p. 50). The data can only collect spoken data and not how thoseturns are understood by the interlocutors. Peering into the structure of the teacher feedback in Excerpt 3, however, we caninfer that the students are already aware that the teacher’s entire turn in lines 5–6 will be a long explanation, even beforeher turn comes to an end. On hearing her word “also,” the students take a collaborative action to refrain from interruptingher in mid-utterance, waiting for a decisive clue of other-repair in her turn. At last, the teacher provides a correction: sheconstructs the word “apartment” with an emphatic stress and a lengthened vowel. The students display their understandingof the teacher correction and pedagogical focus at the moment by reciting the lexical item in partially overlapped (lines 7–9)and co-completed ways (lines 9–10).

Formulating her repair initiation in this circumlocutory way, the teacher achieves the effect of softening her disagreementwith the student’s try-marked answer. Although mitigated, structural features of the design of the teacher’s turn are ademonstrable cue for the students to draw a conventionalized inference about the kind of action the turn is performing. Asnoted earlier, such structural features have been investigated in CA as the notion of (dis)preference (Pomerantz & Heritage,2013; Sacks, 1987). In case of Excerpt 3, “also” and “but” are dispreference markers which are characteristic of the teacher-correction turn. Furthermore, the participants in the classroom display their knowledge of what kind of specific actions getsaccomplished by this institutionalized way of speaking.

To summarize Phase 2, the teacher either directly asks a question or indirectly cuts her turn off to let her students focus ona new L2 word. Sharing a local understanding that this question is asked to many and unspecified participants, the studentsproduce candidate answers. The teacher repairs them, which is followed by the students’ repetition of what is repaired inunison. The pattern of Q–A sequence found in Phase 2 demonstrates that the participants are aware of the tacit rules in thisparticular setting such as the teacher’s cohorting strategy and, as a consequence, a solid structure of the participant roles ofteacher and student.

Phase 3: Expansion for individuals

The Q–A exchange discussed so far often expands in my data, as the vocabulary activity proceeds and the learners

become accustomed to its sequential shape. One of the cases appears in Excerpt 4, where the teacher repairs the students’pronunciation of a word three times within an episode.

S.H. Park / Linguistics and Education 29 (2015) 1–14 9

Table 4The expanded sequence of L2 classroom repair in Excerpt 4.

Line Speaker IRF Adjacency pair Repair

2 T Initiation First pair part3–4 S Response Trouble source5 T Feedback Repair6–7 S Another form of response (R2) Trouble source8, 10–11 T Another form of feedback (F2) Repair9, 12–13 S The other form of response (R3) Trouble source

E

123456789111111

1wabaist

tsppi

mtrtwio

tmfarm

aT

d

14 T The other form of feedback (F3) Repair15 Ss Repetition of feedback (F′) Second pair part

xcerpt 4

T: ((Moving the mouse pointer from one picture to another on the screen)) And then PA:⌠io::- Number se- nine is pa⌠io::. Ma: Pa-= Ha: =Pa::- ti- T: ((Enlarging the picture of patio on the screen using the mouse)) Pa⌠io:: Pa⌠io::. Ma: ◦Pati↑so. Ha: ((Looking at Mayike)) Eh? T: ((Looking at Haji)) Pa:⌠io. That’s outside too::, but on the first floor. Ha: Pati↑so. [Patis-0 T: ((Looking around the other students)) [First floor. Whereas ba:lco[ny on second1 floor, third floor, fourth floor.=2 Ha: [Pati↑so, Patos.3 =Patos.4 T: ((Looking at Haji)) PA::⌠io.s5 Ss: Pa⌠io.

In the segment above, the teacher neither provides a straightforward direction to say the target word aloud (e.g. Excerpt), nor cuts her turn off strategically to elicit co-completion (e.g. Excerpt 2), nor asks a direct question about the targetord (e.g. Excerpt 3). Instead, she produces a declarative, which results in responses from two students only (lines 3–4). The

bsence of responses from five other students can reflect two possibilities. First, the teacher’s declarative sentence might note received explicitly as a request for repetition of the target word, “patio.” Or, when we consider that the two students’ turnsre incomplete in lines 3–4, the others might have been hesitant to produce the target word owing to a lack of confidencen how to pronounce it. Note that the teacher flaps/t/(line 2), while Haji does not (line 4). Unlike the teacher’s cut-off, the L2tudents’ cut-off signals their “concern with the coming talk” (Iles, 1996, p. 37). Whichever is the case, the teacher repairshe students’ problem of hearing, understanding, or pronouncing by producing the target word twice more (line 5).

The teacher’s over-accented and stretched utterance in line 5, however, is not followed by the students’ production ofhe target word in chorus. What follows is Mayike’s new error, “patiso,” a nonsense word with an accent on the secondyllable and with a fricative/s/inserted between the second and the third syllables of the target word.4 As this turn was softlyroduced, it is likely that Haji, who sat next to Mayike, heard it but the teacher did not. Haji turns his head toward Mayike,roducing “Eh?” And the teacher’s eye gaze in line 8 demonstrates that she perceives Haji’s turn in line 7 as an other-repair

nitiator (for discussion of other-repair initiator techniques, see Schegloff et al., 1977).The teacher’s second repair in line 8 is twofold: a repetition of the target word as per usual, plus an explanation of the

eaning of the word (line 8). When Haji keeps mispronouncing the word (line 9), the teacher’s semantic approach to thearget form is built up in lines 10–11. Haji’s nagging trouble causes the third round of a repair sequence; this time the teachereturns to phonology, but as an even louder and more lengthened form. Two overlaps and a latch found in lines 9–13 intimatehat Haji does not listen to what the teacher says with care; it is even possible that he makes errors on purpose, playing onords by rearranging the phonemes and the place of accent. This possibility is supported by the fact that all the students,

ncluding Haji, produce the target form with no difficulty at last (line 15). Again, the teacher repair in line 14 simply consistsf a word, which seems more audible and more distinctive to the students than her prior explanation.

Besides, it is worth noticing that the teacher’s eye gaze specifies the speaker of trouble source in line 14 (for discussion ofhe effect of eye gaze in educational settings, see Lund, 2007). To track the course of her gaze move in this episode, the teacher

aintains the cohorting at base (lines 1, 5, 10), as examined in the analysis of Excerpt 2; such treatment, however, deviatesrom its usual route at odd times (lines 8, 14). That is, the recipient of the teacher turn switches between the entire class andn individual student in Excerpt 4. Such back-and-forth orientation of the teacher turn reflects the teacher’s real-time, dualole of leading a whole group of students and of coaching individuals whose language proficiency, background, and interestight vary.In view of the typical IRFF′ sequence discussed in Phase 2 through Tables 2 and 3, Mayike’s turn in line 6 in Excerpt 4 is

reply to the teacher repair in line 5 and at the same time a trouble source which triggers another teacher repair in line 8.

able 4 summarizes this cross-cut and expanded structure of L2 classroom repair presented in Excerpt 4.

Note that in line 15 all the students orient toward the teacher’s elicitation of repetition in the very first turn of the excerpt,espite the intervention of multiple troubles and repairs. It would not sound awkward even if the last line immediately

10 S.H. Park / Linguistics and Education 29 (2015) 1–14

followed the first line of Excerpt 4 without the sequence of repair in between. This structural connection is related to the factthat the participants share a common understanding of how the activity of vocabulary introduction and practice proceedsin this local setting. This context-dependent knowledge strengthens the association between the teacher’s first part and thestudents’ second part of the classroom adjacency pair, no matter how distant they are from each other by the embeddedtrouble-repair pairs of “R(n)-F(n).”

The multiply expanded sequence of repair in Excerpt 4 reveals that the teacher does not easily “let go” of her students’trouble. It is hard to imagine such a repairer in everyday conversation outside the classroom, who persists in correctingerrors until she reaches the second or third round of the sequence. What we normally do is to let the conversation move onwithout further repair once it fails, saying, for instance, “Ferget it” (Schegloff et al., 1977, p. 365). In contrast, the analysis ofExcerpt 4 elucidates how the participants tweak and complicate the general preference toward repair in communication, inorder to achieve the institutional aim and preserve the identity of teacher and learner in the L2 classroom context.

A question that arises here is at what moment, if ever, the teacher may give up her task of repair and just “let it go.”Indeed, the teacher will be bound to stop at some point in the face of further learner errors. What is the structural conditionthat often makes the teacher finally choose a laissez-faire attitude toward the endless learner trouble? Excerpt 5, anotherextract concerning the expanded L2 classroom repair in Phase 3, lays out an example of this condition.

Excerpt 5

1 Ss: Living room.2 (1.0)3 T: ((Opening the textbook on the table)) Oka:y, (.) now. (2.0) Let’s look at um- the4 six-5 Ab: ((Looking at the teacher)) Teacher.6 T: ((Turning her eyes from the book to Abdi)) Yes.7 Ab: It’s- uh- like- um- ee- li- liming-8 Ss: Li[ving room9 T: [Living room.=10 Ab: =Living- (.) RO::m.11 T: LIving- (.) ROOm.12 Mi: Living (.) room.13 T: Two group two words. Living (.) room. Uh-hmmm, (0.5) yes.14 Ab: Living, (.) RO:m.15 T: Living room. It’s- (2.0) the room [where you have a sofa::, and a cha:ir, and16 Sa: [<Living->17 T: maybe a tivi::, (0.5) [and you sit the:re and watch tivi maybe?18 Sa: [El, ai, vi, ai, en, ji. ((Looking at the handout))19 (2.0)20 T: ◦Living room.◦ <O:Ka:y.> (.) ((Pointing at a textbook page with her index finger))21 Let’s look at the six pictures on the bottom of this pa:ge. Number one is the:::

In the instance above, the teacher is about to proceed to the next activity. After the students’ reading-aloud of the lastpedagogical focus in the previous activity (line 1), a brief pause and the activity-transition marker “Okay” follow. The word“now” is another transition marker, which justifies the ensuing two-second pause for the participants’ getting ready for thenext activity. The teacher’s turn “let’s look at um- the six-” embarks the students on the new phase of the lesson. Such arole for this turn is affirmed later in line 21, where the teacher’s cut-off turn in line 4 is recovered after the inserted repairepisode (lines 5–20). Abdi’s summon interrupts the teacher’s ongoing turn (line 5), and the teacher reacts to it, saying “yes.”What is interesting in the following turns is that everyone in the class notices that Abdi’s stammering utterance with manycut-offs in line 7 is a form of question, or FPP which requests SPP, or a trouble source to be repaired. Multiple students andthe teacher correct it almost simultaneously (note an overlap between lines 8 and 9).

Abdi’s reproduction of the repaired (line 10) is more audible and complete than his first trial (line 7). It causes, however,another round of repair trajectory. That is, Abdi’s turn in line 7 is detected as a trouble source, which is repaired by others inlines 11 through 13. The linguistic focus of the teacher correction changes from phonology (line 11) to morphology (line 13).Abdi’s continuous mispronunciation (line 14) invites another teacher-correction, which changes its approach once again totouching the target word’s semantic level (lines 15, 17). The linguistic forms of three trouble sources vary little and providefew clues about what the exact trouble that Abdi addresses is. The teacher’s way of correction, per contra, evolves in accordwith different linguistic aspects. Of course, Abdi’s placing a short pause between “living” and “RO:m” in line 10 providesthe teacher with a cue that Abdi might be asking whether it is two separate words or not. This is why the teacher stopsemphasizing the word’s pronunciation and explains its morphological shape (line 13). Therefore, the teacher’s “LIving (.)ROOm” in line 11 and “living (.) room” in line 13 instruct entirely different linguistic aspects, although they sound the sameon the surface.

The teacher waits for Abdi’s signal that his problem has ever been resolved (line 13). Her gaze at Abdi implies it, in additionto a considerable amount of time for Abdi to produce any reaction while the teacher constructs her turn with confirmation

markers (“Uh-hmmm” and “yes”) and a pause. What appears in the next turn is, however, a problematic form as it is exactlysame as the one Abdi produced before the teacher-correction. To top it off, unlike the previous trouble source in line 10, thenew one in line 14 contains even less information about what Abdi might like to say. Nevertheless, the teacher once againmoves on to a different linguistic aspect (semantics) of the target word (lines 15, 17). After this repair, the teacher again

wtfi

stai

D

ritarolgr

totpao((at

emtar

ghu&wapi

qahwow

scts

S.H. Park / Linguistics and Education 29 (2015) 1–14 11

aits for Abdi’s reaction, allowing a two-second pause and producing “living room” for the last time. No reaction from Abdi;he teacher finally marks a topic transition by establishing “Okay” with an emphatic stress and an unhurried tempo. It isollowed by her instruction to look at the pictures, i.e. a recovered activity to which the participants return after the abruptlynserted repair episode from line 5 to line 20.

What Excerpt 5 tells us about why the teacher stops repairing at a particular moment is the silence. Cross-culturallytudied (e.g. Harumi, 2011; Tatar, 2005), the classroom silence is either a good or bad sign that the teacher receives fromhe students as a cohort and interprets to make decisions. In case of Excerpt 5, the teacher’s interpretation of the silence isrbitrary because of the language barrier. The episode might have been much longer with more pairs of trouble source-repair,f Abdi spoke English better or the teacher spoke Oromo.

iscussion and conclusion

Reframing the IRF convention in classroom discourse on the groundwork of CA, this paper demonstrates that teacherepair in the L2 class for low-literate learners is oriented to their state of literacy. The pedagogical priority for the teachers whether or not they are able to join the prepared activity by perceiving and reacting to her repair initiation and comple-ion, which become routinized as a classroom practice and create language learning opportunities. The teacher’s minimalnd straightforward way of doing repair, with forms of her single-word elicitation of student recitation and repair with nooom for the learner’s self-repair, is legitimate in this particular context. This finding departs from the general agreementn L2 repair that a delay of teacher repair is beneficial to language learners (Hosoda, 2002). In this study, the learners’ow literacy affects the length and structure of teacher repair, which helps them pay attention to and reproduce lan-uage input immediately and manageably and therefore explicitly acquaint themselves with target vocabulary items byehearsal.

In light of L2 vocabulary research, teacher repair in this paper plays a role in intentional vocabulary teaching withhe teacher’s direct and systematic planning and instruction (Hulstijn, 2011; Read, 2004; Rupley & Nichols, 2005). In thebserved class, teacher repair structurally channels the target words available to learners, while resolving the students’ry-marks (e.g. line 4 in Excerpt 3), cut-offs (e.g. line 4 in Excerpt 4), and prolonged silence (e.g. line 10 in Excerpt 1). Thisedagogic function of teacher repair as explicit vocabulary instruction facilitates the students’ identification of basic wordsnd their semantic relations, as they spontaneously repeat the teacher feedback aloud in unison in particular, a finish strokef the IRFF′ sequence. Hulstijn (2001) reports that “[i]t is the quality and frequency of the information processing activitiesi.e., elaboration on aspects of a word’s form and meaning, plus rehearsal) which determine retention of new information”p. 275). The teacher in this study ensures the quality and frequency of the L2 vocabulary input by elaborating differentspects of a word’s form and meaning (e.g. Excerpt 5) and inviting the students as a cohort to repeat it in each repairrajectory.

Note that two students, Haji and Mayike, dominate the production of the first turn in the IRFF′ quadriad throughout thexcerpts (e.g. lines 2 and 9 in Excerpt 2); they were the most communicative among the students in all the other days ofy class observation. The repetition of teacher feedback (i.e. the fourth slot of IRFF′) is then often the only chance at all for

he others to speak out in L2 during the class. In this quite reticent environment, teacher repair provides the learners with safe and workable interactional space for L2 production, where even those at rudimentary stages of L2 development canepeat the utterance of the teacher verbatim at least.

As the interview and conversation data reveal, such learners’ interactional dependence on the teacher’s utterance isrounded in the highly rule-governed and coordinated structure of classroom talk, including repair. Conversation analystsave investigated the structural regularity of talk-in-interaction that “get[s] participants to occasions of talk to do the work ofnderstanding the talk of others in the very ways and at the very times at which they demonstrably do that work” (Moerman

Sacks, 1988, p. 182). In the L2 classroom context in particular, the foreseeability of how turns will be constructed is linkedith the participants’ mutual understanding of “where they are within a social interaction” (Heritage, 2004, p. 104). Their

wareness that they are within a language learning space as a teacher and learners generates the space’s own communicativeotential. Ellis (2003) also makes point that the interactional environment for L2 learners is “governed by an “educational

mperative” which dictates the kind of discourse that arises” (p. 253).Admittedly, the rule-governed, regular pattern of classroom talk discussed so far through the repair practice raises the

uestion of (non-)authenticity of L2 classroom interaction, or the similarity (difference) between L2 use in the classroomnd that in its original communicative context. In L2 studies, the simplified and invented use of L2 in the classroom contextas often been criticized, as it ignores interactional complexity in real-life contexts (Roberts & Cooke, 2009). Few learnersill in fact meet a person who says “home” and expects them to recite it or one who repairs their linguistic errors tirelessly

utside the classroom. It is apparently paradoxical that they attend the class to learn English but use it differently from theay they are supposed to do outside the classroom.

Nevertheless, the present analysis implies that this paradox is a pedagogical scratch line where the low-literate adults

tand upon to obtain L2 exposure. Because their L2 talk is less authentic, L2 talk, including repair, can be fully oriented to thelass’ current pedagogical priority. Note that, to those with little to no literacy and/or experience of formal education, evenhe simplified, invented way of classroom talk is something to be acquired from the bottom up. The excerpts in this articlehow that the learners are in the process of gaining knowledge of invented teacher repair. They exhibit their understanding

12 S.H. Park / Linguistics and Education 29 (2015) 1–14

of the teacher’s treating them as a cohort in general, for instance, in Excerpt 5 where Minani and Safia constantly interruptwhile Abdi’s problem is repaired. Here the students do not exhibit yet, on the other hand, their understanding of whatthe silence means in this particular moment after the teacher repair. Overall, the students in my data are in the processof acquainting themselves with the mechanism of L2 classroom repair and how it is utilized as resource for languagelearning.

Likewise, the observed class’ non-authentic L2 talk and its structure are complicated enough to take time to be accustomed.In order to participate in the activity of vocabulary introduction, for example, the learners have to understand its differentphases – what the picture on the front screen is for, what the teacher expects by her turns “home” in line 11 of Excerpt1 or “number three is-” in line 1 of Excerpt 2, and so forth. The practice of teacher repair in this setting is therefore asophisticated and multifaceted task that encompasses both communicative and metacommunicative processes of languagelearning.

The last point that merits attention in the present analysis is the teacher’s decision-making in the repair trajectory.The teacher constantly adjusts to whom her turns are oriented between the whole class and individual speakers of thetrouble source. She tailors different linguistic approaches to confront misunderstandings, and at some point arbitrarilywraps it up. Relating these findings to the observed language barrier among the participants and a consequential dearth ofstudent feedback, this paper may join the discussion of the teacher’s recipient design based on the students’ L2 performance(Garfinkel, 1967). The muter the students are, the more agile the teacher is obliged to be; she may have to fill in morepigeonholes of interaction in the class for low-literate students than in other L2 classes, lest they be left awkwardly silent.Even though keeping her utterances concise and structurally simple in general, the teacher’s communicative behavior (i.e.shifting eye gaze, linguistically diversified explanation of vocabulary, a laissez-faire attitude toward failed repair) is adaptedto the particular addressees, the low-literate ESL learners.

To sum up, this article complicates the practice of repair as understood in the literature on the L2 classroom discourse byspecifically focusing on the subset of adult ESL learners who have a background of low literacy. Since this study is descriptiveand micro-analytic with a data set from a single classroom, the findings cannot yet be generalized. Future research is neededwith a larger amount of data from multiple classroom contexts to extend our knowledge of repair as a linguistic resource forlow-literate learners. That said, this paper can be an initial step to inquire whether our understandings of the L2 classroomrepair can be extended to all learners. Pedagogically, it is hoped that the present analysis contributes to informing theteachers who will teach low-literate immigrant adult L2 learners for the first time about how to adapt their repair strategiesto the learners’ literacy and linguistic status.

Notes

1. Correction, “the replacement of an error by what is correct” (Schegloff et al., 1977, p. 363), is a subtype of repair, in termsof its orientation to the problem of common understanding.

2. In the observed class, the vocabulary activity is twofold: introduction and review of vocabulary. This paper examines theformer only.

3. For discussion of repair along with the chronological phases of an action, see Egbert (1998).4. I consider “patiso” as a nonsense word based on the member-check with an Oromo-English interpreter who confirmed

that it was not an Oromo word.

Appendix 1. Participant information

Abbreviation Name First language

Ab Abdi OromoHa Haji OromoHu Hussein OromoMa Mayike SomaliMi Minani Kirundi

Sa Safia OromoTe Terefe (not appear in this paper) OromoS/S1/S2. . . A student unidentifiedSs All seven studentsT Teacher

A

R

BBCC

D

D

DD

E

EE

E

GH

HHHHHH

H

HH

HHIJ

JJKKL

L

L

LMMMM

S.H. Park / Linguistics and Education 29 (2015) 1–14 13

ppendix 2. Transcript conventions (Jefferson, 2004)

Transcript symbol Meaning

[ Overlapping utterances= Latched utterances(0.0) Length of silence by tenths of seconds(.) Micro-pause::: Prolonged sound↑ Shift into especially high pitchword Relatively high volumeWORD Especially high volume?/./, Rising/falling/continuing intonation◦◦ Soft sound- Cut-off> < Speedy utterances< > Slow utterances(()) Transcriber’s description

eferences

each, W. A. (1993). Transitional regularities for ‘casual’ okay usages. Journal of Pragmatics, 19, 325–352.igelow, M. (2010). Mogadishu on the Mississippi: Language, racialized identity and education in a new land. Malden: Wiley-Blackwell.azden, C. (2001). Classroom discourse. Portsmouth: Heinemann.ho, E. (2008). An examination of the use of repair strategies of elementary English as a second language (ESL) students by class types and grade levels (Ph.D.

dissertation). Texas A&M University.ehghani, A. P., Motamadi, A., & Mahbudi, A. (2011). Homework assignment affects EFL beginners’ vocabulary learning. US-China Foreign Language, 9(5),

331–338.ooley, K. T., & Thangaperumal, P. (2011). Pedagogy and participation: Literacy education for low-literate refugee students of African origin in a western

school system. Language and Education, 25(5), 385–397.oughty, C., & Williams, J. (1998). Focus on form in classroom second language acquisition. New York: Cambridge University Press.rew, P., & Heritage, J. (1992). Analyzing talk at work: An introduction. In P. Drew, & J. Heritage (Eds.), Talk at work (pp. 3–65). Cambridge, UK: Cambridge

University Press.gbert, M. M. (1998). Miscommunication in language proficiency interviews of first-year German students: A comparison with natural conversation. In R.

Young, & A. W. He (Eds.), Talking and testing: Discourse approaches to the assessment of oral proficiency (pp. 149–172). Philadelphia: John Benjamins.llis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.me, E. (2011). Cognitive and psycholinguistic skills of adults who are functionally illiterate: Current state of research and implications for adult education.

Applied Cognitive Psychology, 25(5), 753–762.rickson, F. (1982). Classroom discourse as improvisation: Relationships between academic task structure and social participation structure in lessons. In

L. C. Wilkinson (Ed.), Communicating in the classroom (pp. 153–181). New York: Academic Press.arfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs: Prentice Hall.all, J. K. (1998). Differential teacher attention to student utterances: The construction of different opportunities for learning in the IRF. Linguistics and

Education, 9(3), 287–311.ammerly, H. (1995). What visual aids can and cannot do in second language teaching. Mosaic, 2(3), 11–18.ardy, I., & Moore, J. (2004). Foreign language students’ conversational negotiations in different task environments. Applied Linguistics, 25(3), 340–370.arumi, S. (2011). Classroom silence: Voices from Japanese EFL learners. ELT Journal, 65(3), 260–269.ellermann, J. (2003). The interactive work of prosody in the IRF exchange: Teacher repetition in feedback moves. Language in Society, 32, 79–104.ellermann, J. (2009). Practices for dispreferred responses using no by a learner of English. International Review of Applied Linguistics, 47, 95–126.eritage, J. (2004). Conversation analysis and institutional talk. In R. Sanders, & K. Fitch (Eds.), Handbook of language and social interaction (pp. 103–146).

Mahwah: Erlbaum.osoda, Y. (2002). Analyzing Japanese native–nonnative speaker conversation: Categories, other-repair, and production delay (Ph.D. dissertation). Temple

University.osoda, Y. (2006). Repair and relevance of differential language expertise in second language conversations. Applied Linguistics, 27, 25–50.ulstijn, J. H. (2001). Intentional and incidental second language vocabulary learning: A reappraisal of elaboration, rehearsal and automaticity. In P. Robinson

(Ed.), Cognition and second language instruction (pp. 258–286). Cambridge: Cambridge University Press.utchby, I., & Wooffitt, R. (2008). Conversation analysis. Malden: Polity Press.uth, T. (2011). Conversation analysis and language classroom discourse. Language and Linguistic Compass, 5(5), 297–309.

les, Z. (1996). Collaborative repair in EFL classroom talk. York Papers in Linguistics, 17, 23–51.efferson, G. (2004). Glossary of transcript symbols with an introduction. In G. H. Lerner (Ed.), Conversation analysis: Studies from the first generation (pp.

13–31). Philadelphia: John Benjamins.ung, E. H. S. (1999). The organization of second language classroom repair. Issues in Applied Linguistics, 10(2), 153–171.uvonen, P. (1989). Repair in second-language instruction. Nordic Journal of Linguistics, 12(2), 183–204.asanga, L. A. (1996). Peer speech repairs in EFL classroom activities. TESL Reporter, 29(2), 1–20.asper, G. (1985). Repair in foreign language teaching. Studies in Second Language Acquisition, 7, 200–215.ehti-Eklund, H. (2013). Code-switching to first language in repair – A resource for students’ problem solving in a foreign language classroom. International

Journal of Bilingualism, 17(2), 132–152.evy, B. A., & Lysynchuk, L. (1997). Beginning word recognition: Benefits of training by segmentation and whole word methods. Scientific Studies of Reading,

1(4), 359–387.iebscher, G., & Dailey-O’Cain, J. (2003). Conversational repair as a role-defining mechanism in classroom interaction. The Modern Language Journal, 87(iii),

375–390.und, K. (2007). The importance of gaze and gesture in interactive multimodal explanation. Language Resource and Evaluation, 41(3–4), 289–303.

acbeth, D. (2004). The relevance of repair for classroom correction. Language in Society, 33(5), 703–736.cHoul, A. (1990). The organization of repair in classroom talk. Language in Society, 19(3), 349–377.ehan, H. (1979). Learning lessons: Social organization in the classroom. Cambridge: Harvard University Press.oerman, M., & Sacks, H. (1988). On ‘understanding’ in the analysis of natural conversation. In M. Moerman (Ed.), Talking culture (pp. 180–186). Philadelphia:

University of Pennsylvania Press.

14 S.H. Park / Linguistics and Education 29 (2015) 1–14

Moore, S. H. (2007). Appropriate assessment for low and pre-literacy AMEP learners. Sydney, Australia: National Centre for English Language Teaching andResearch.

Nakamura, I. (2008). Understanding how teacher and student talk with each other: An exploration of how ‘repair’ displays the co-management of talk-in-interaction. Language Teaching Research, 12(2), 265–283.

Nassaji, H., & Wells, G. (2000). What’s the use of ‘triadic dialogue’?: An investigation of teacher–student interaction. Applied Linguistics, 21(3), 376–406.Nystrand, M. (1997). Dialogic instruction: When recitation becomes conversation. In M. Nystrand, A. Gamoran, R. Kachur, & C. Prendergrast (Eds.), Opening

dialogue: Understanding the dynamics of language and learning in the English classroom (pp. 1–29). New York: Teachers College Press.Okada, Y. (2010). Repairing failed questions in foreign language classrooms. JALT Journal, 32(1), 55–74.Ong, W. J. (1988). Orality and literacy. London, UK: Routledge.Payne, G., & Hustler, D. (1980). Teaching the class: The practical management of a cohort. British Journal of Sociology of Education, 1(1), 49–66.Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn-shapes. In J. M. Atkinson, & J. Heritage

(Eds.), Structures of social action: Studies in conversation analysis (pp. 79–112). Cambridge, UK: Cambridge University Press.Pomerantz, A., & Heritage, J. (2013). Preference. In J. Sidnell, & T. Stivers (Eds.), Handbook of conversation analysis (pp. 210–228). Boston: Wiley-Blackwell.Pouwels, J. B. (1992). The effectiveness of vocabulary visual aids for auditory and visual foreign language students. Foreign Language Annals, 25(5), 391–401.Read, J. (2004). Research in teaching vocabulary. Annual Review of Applied Linguistics, 24, 146–161.Roberts, C., & Cooke, M. (2009). Authenticity in the adult ESOL classroom and beyond. TESOL Quarterly, 43(4), 620–642.Rolin-Ianziti, J. (2010). The organization of delayed second language correction. Language Teaching Research, 14(2), 183–206.Rupley, W. H., & Nichols, W. D. (2005). Vocabulary instruction for the struggle reader. Reading & Writing Quarterly: Overcoming Learning Difficulties, 21(3),

239–260.Sacks, H. (1987). On the preferences for agreement and contiguity in sequences in conversation. In G. Button, & J. R. E. Lee (Eds.), Talk and social organization

(pp. 54–69). Clevedon, England: Multilingual Matters.Sacks, H., & Schegloff, E. A. (1979). Two preferences in the organization of reference to persons in conversation and their interaction. In G. Psathas (Ed.),

Everyday language: Studies in ethnomethodology (pp. 15–21). New York: Irvington.Sagarra, N., & Alba, M. (2006). The key is in the keyword: L2 vocabulary learning methods with beginning learners of Spanish. The Modern Language Journal,

90(ii), 228–243.Schegloff, E. A. (2000). When ‘others’ initiate repair. Applied Linguistics, 21(2), 205–243.Schegloff, E., Jefferson, G., & Sacks, H. (1977). The preference for self-correction in the organization of repair in conversation. Language, 53, 361–382.Schegloff, E. A., & Sacks, H. (1973). Opening up closings. Semiotica, 8, 289–327.Seedhouse, P. (1996). Classroom interaction: Possibilities and impossibilities. ELT Journal, 50(1), 16–23.Seedhouse, P. (1997). The case of the missing “no”: The relationship between pedagogy and interaction. Language Learning, 47(3), 547–583.Seedhouse, P. (2004). The interactional architecture of the language classroom: A conversation analysis perspective. Malden: Blackwell Publishing.Seong, G. (2004). Conversational repair strategies of teachers of English as a second language (Ph.D. dissertation). University of Illinois at Urbana-Champaign.Sinclair, J., & Coulthard, M. (1975). Towards an analysis of discourse. London, UK: Oxford University Press.Strube, S. (2010). Conveying meaning: Oral skills of L2 literacy students. TESOL Quarterly, 44(3), 628–637.Tarone, E. (2010). Second language acquisition by low-literate learners: An under-studied population. Language Teaching, 43(1), 75–83.Tarone, E., & Bigelow, M. (2011). A research agenda for second language acquisition of pre-literate and low-literate adult and adolescent learners. In P.

Vinogradov, & M. Bigelow (Eds.), Low educated second language and literacy acquisition: Proceedings of the 7th symposium Minneapolis, MN, September2011, (pp. 5–26).

Tarone, E., Bigelow, M., & Hansen, K. (2009). Literacy and second language oracy. Oxford. UK: Oxford University Press.Tatar, S. (2005). Why keep silent? The classroom participation experiences of non-native-English-speaking students. Language and Intercultural Communi-

cation, 5(3–4), 284–293.van Lier, L. (1988). The classroom and the language learner. London, UK: Longman.

Waring, H. Z. (2009). Moving out of IRF (initiation-response-feedback): A single case analysis. Language Learning, 59(4), 796–824.Wells, G. (1993). Theories of activity and discourse for the analysis of teaching and learning in the classroom. Linguistics and Education, 5, 1–37.Wong, J. (2000). Delayed next turn repair initiation in native/non-native speaker English conversation. Applied Linguistics, 21(1), 244–267.Wrigley, H. S. (2007). Beyond the life boat: Improving language, citizenship, and training services for immigrants and refugees. In A. Belzer (Ed.), Toward

defining and improving quality in adult basic education (pp. 221–239). Mahwah: Lawrence Erlbaum.