FRAME ANALYSIS OF MICROTEACHING IN A COMMUNITY OF PRACTICE OF TEACHERS OF ENGLISH AS AN ADDITIONAL LANGUAGE IN SOUTHERN BRAZIL ANÁLISE DE ENQUADRAMENTO DE MICROTEACHING EM UMA COMUNIDADE DE PRÁTICA DE PROFESSORES DE INGLÊS COMO LÍNGUA

1 Doutor em Letras pela Universidade Federal do Rio Grande do Sul e professor de Língua Inglesa na Universidade Federal de Ciências da Saúde de Porto Alegre. 2 Doutora em Letras pela Universidade Federal do Rio Grande do Sul e professora de Língua Inglesa do Instituto de Letras da Universidade Federal do Rio Grande do Sul. Docente do PPGLET da mesma universidade e bolsista de produtividade do CNPQ. http://dx.doi.org/10.15448/2178-3640.2021.1.40296 BELT

Research has found teachers and student teachers benefit from participating in microteaching (Amobi, 2005;Metcalf, Hammer & Kahlich, 1996). However, little work has explained the minutia of what goes on as teachers or student teachers microteach by analyzing their discourse while microteaching.
This paper had its origin in a larger research project that investigated the professional development of undergraduate student teachers of English as an Additional Language 3 in a Community of Practice (Wenger, 1998) 4 generated by a program named Languages Without Borders at a large university in southern Brazil 5 . The study was affiliated with the paradigm of Practice Theory (Young, 2009;2010), relying on qualitative methods of data generation and analysis (Mason, 2002;Erickson, 1990) as well as on interactional sociolinguistics (Gumperz, 2005;Tannen, 2014). Although we did not originally set out to investigate microteaching, early in data generation it emerged as a central practice utilized by the community due to its pervasiveness in the teacher development meetings. This intrigued us to pursue a deeper understanding of how microteaching functioned in this community.
The purpose of this paper, thus, is to learn more about the discursive architecture of microteaching in the community investigated in this project. Inspired by Bell (2007), we drew initially on frame theory, using Goffman's (1974, p. 8) question -"What is it that's going on here?" -as a starting point for data analysis. This is the question participants in any sort of interaction must answer to make sense of a speech event.
While there can always be multiple responses to this question, there is often enough agreement on the definition of a situation so people can manage interaction -both interpreting others' actions and regulating their own. The construction of a tacit response to this question makes it possible for people to interact successfully.
In a Google Scholar search for peer-reviewed papers containing the expression "microteaching" in the title and "discourse analysis OR conversation analysis OR interactional sociolinguistics" among the keywords, published in between 2010 and 2020, 140 entries come back. However, we discovered that only a very small portion of the entries indeed make use of interactional sociolinguistics, conversation analysis or discourse analysis (Bell, 2007;Kim, 2006;Ryoo, 2016). Unlike previous research, the present work provides a dense description of the discourse of microteaching in a community of practice. By presenting such description, this paper aims to help pre-service teachers and teacher educators to understand what they are doing when they microteach, for teachers find microteaching stressful when they do not know what is expected from them (Bell, 2007;Ryoo, 2016).

Theoretical framework
In this section, we address pertinent literature for the present research. First, we explain the terms that are essential for the discussion.
Then, we review literature that has also addressed microteaching from the perspective of interactional sociolinguistics, conversation analysis or discourse analysis.

Interactional Sociolinguistics: setting the terms
Frame. The notion of framing comes from the work of Bateson (1972), who said that the meaning of an utterance or action cannot be correctly interpreted and responded to without the reference to a metamessage about the frame in which they were produced. For instance, an utterance may mean the opposite of what it says if it is "operating in a frame of play, irony, joking, or teasing" (Tannen, 2014, p.10). Goffman (1974, p. 21) expanded this notion by proposing that "when an individual in a Western

3/15
society recognizes a particular event, he tends […] to imply this response (and in fact employ) one or more frameworks or schemata of interpretation". Therefore, frames are structures of expectation (Tannen, 1979) which help participants navigate the practices in which they engage.
Goffman divided the frameworks into two categories: (1) primary frameworks and (2) keyed frameworks 6 . Primary frameworks may vary in degree of organization: some are so organized that they appear as a set of postulates or rules, whereas others do not appear to have any recognizable shape and only provide "a lure of understanding, an approach, a perspective" (Goffman, 1974, p. 21) regarding the event at hand. Primary frameworks allow people to "locate, perceive, identify, and label a seemingly infinite number of concrete occurrences defined in its terms" (p. 21). Thus, participants tend to apply primary frameworks to different situations, even when they cannot describe them. Consequently, primary frameworks are particularly important to provide an answer to Goffman's question, for several events fit within some primary framework.
Keying. The second category Goffman (1974) proposed is that of keyed frameworks, which is when primary frames are modified by signals that they should not be interpreted literally nor have their face-value meaning. Based on Bateson's (1972) account of an observation that he made in a zoo, in which he found that monkeys can play with one another, indicating awareness to metamessages that a certain action means play and not fight, Goffman (1974) described keying as the set of conventions by which a given activity, already meaningful in terms of some primary framework, is transformed into something patterned on this activity but seen by the participants to be something quite else (Goffman, 1974:43-44). This is what Goffman (1974) refers to as layering or lamination.
Layering/lamination. When no keying is involved, one interprets the activity in the light of the primary framework, and such activities are 6 I will henceforth refer to "secondary frameworks" for the sake of simplicity. usually named real or literal activities. However, a keying of literal activities on a stage would provide us with something that is not literal or not real in primary framework terms, but it is real as a keyed one. For instance, in a staging of Becket's Waiting for Godot, if one asks "What are they doing?" the answer is likely to be "they are actors pretending they are waiting." Thus, they are not actually waiting; they are pretending to wait.
Footing and contextualization cues. Footing can be understood as: 1. Participants' alignment, or set, or stance, or posture or projected self is somehow at issue.
2. The projection can be held across a strip of behavior that is less long than a grammatical sentence, or longer, so sentence grammar won't help us all that much, although it seems clear that a cognitive unit of some kind is involved minimally, perhaps a "phonemic clause". Prosodic, not syntactic, segments are implied.
3. A continuum must be considered from gross changes to the instance to the most subtle shifts in tone that can be perceived.
4. For speaker, code switching is usually involved, and if not this then at least the sound markers that linguists study: pitch, volume, rhythm, stress, tonal quality. 5. The bracketing of a "higher level" phase or episode of interaction is commonly involved, the new footing having a liminal role, serving as a buffer between more substantially sustained episodes (Goffman, 1981, p.10).
Therefore, a change in footing implies a change in the alignment participants of an interaction take up in the way they manage the production and reception of an utterance. In other words, it is another way to talk about a change in frame, which may signal that participants are changing what they are doing or even that they are performing different identities. Gumperz (2005) explained that the term contextualization cues refer to verbal signs which serve to construct the contextual ground for situated interpretation and, thus, affects how messages are understood. Contextualization cues represent speakers' ways of signaling and providing information to interlocutors and audiences about how they are using language at any point of an interaction. In this sense, they operate at various levels of speech production, including the aspects of grammar (phonology, morphology, lexicon, syntax) as well as (i) prosody (i.e., intonation, stress or accenting and pitch), (ii) paralinguistic signs (i.e. whispery, breathy, husky or creaky voice), (iii) markers of tempo, including pauses and hesitations; (iv) overlaps; (v), laughter; and (vi) formulaic expressions (Duranti, 1997).

Microteaching
In a study with 18 student teachers, Bell (2007, p.37) concluded microteaching is "a highly complex, layered (laminated) task for the participants. Within the same strip of activity their identities as students, classmates, and (future) teachers all compete for attention." In the recordings and questionnaires in which participants described their perceptions of the activity and explained how they approached the task, participants suggested they thought of microteaching in terms of performance or a classroom task more than properly teaching.
The author indicated that several verbal and nonverbal cues are used to contextualize what is going on during the micro-classes. That is, participants signal to one another how they should interpret their actions at every moment, as the frames by which the strips of interaction should be interpreted can change at any time. For instance, a participant may shift from the microteaching frame to that of a student teacher talking to peers or trainer in an educational activity. Ryoo (2017) conducted research in a preservice teacher education course offered at a college in southern Korea for English Education 7 MAXQDA is a software program designed for computer-assisted qualitative and mixed methods data, text and multimedia analysis in academic, scientific, and business institutions. It has been developed and distributed by VERBI Software based in Berlin, Germany. The emphasis on going beyond qualitative research can be observed in the extensive attributes function (called variables in the program itself) and the ability of the program to deal relatively quickly with larger numbers of interviews (https://en.wikipedia.org/wiki/MAXQDA). 8 Times New Roman 12, single space. 9 In this article, the original numbering of the lines have been kept as they appear in the original transcription in the database.
Majors pursuing a certificate to teach in secondary schools. Microteaching happened during regular class sessions in a college classroom; eight 20 to 30-minute microteaching sessions were taught by 24 participants, recorded and transcribed. Then, the researcher did emic analysis of the transcriptions for the description and interpretation of different situational frames evoked by the participants. Like Bell (2007), the author found that microteaching is a complex activity in which participants shift frames multiple times during the same speech event and, thus, constantly use contextualization cues to demonstrate to one another how actions should be interpreted. However, the author found that the dominant situational frame was that of teaching, although participants also framed it as a learning and a performative event. Ryoo (2016) understands the changes in frame also represent the performance in contexts of different identities. Therefore, there have been previous studies addressing microteaching from interactional perspectives. They have focused on the frames that participants use during a microteaching session (Bell, 2007) and own how the shift in such frames may indicate a different identity being performed in context (Ryoo, 2016).

Methodology
In the present study, data generated during microteaching sessions were analyzed pursuing the answer for the following question, inspired by Goffman (1974): "What is going on as participants microteach?" During microteaching observations, the first author generated field notes, took photographs, collected artifacts (e.g. lesson plans, handouts and classroom tasks), and produced audio recordings (Erickson, 1990;Mason, 2002).
All data were organized in a database on MaxQda 12 7 . The audio material was then transcribed, amounting to about six and a half hours and 51 pages of transcription 8 . Then, we engaged in initial, focused, and theoretical coding (Saldaña, 2009) 9 .

5/15
Participants consisted mainly of undergraduate or graduate students pursuing the teacher certification in EAL 10 . In pursuit of this certificate, they taught twelve hours a week (three classes) in addition to the lessons they still attended as students. They received a stipend funded by the federal government of Brazil and had to have at least a B2 level certificate of English language 10 In Brazil, this major will give you a lifetime certification to teach Portuguese, Additional Languages and their respective Literatures; it is called Letras. 11 Not really enforced. 12 We will use quotation marks to indicate that we are quoting a participant directly (except when quoting literature) and italics to indicate that it is a term we coined to name something participants did not mention.
proficiency. The coordinator (a tenured professor from the English Department) was responsible for organizing the LCs administratively and, above all, for pedagogically supervising the student teachers. Additionally, one Fulbright English Teaching Assistant (ETA) also participated in the microteaching sessions. Below there is a table that sums up the participants that appear in the data.

Findings and discussion
In this community, microteaching sessions took place in the pedagogical meetings during the first, third and fifth weeks of data generation. In each meeting, student teachers presented microclasses of about 20 minutes 11 . In these meetings, the micro-classes started after about fifteen minutes of "announcements" 12 and pressing "bureaucratic issues" of the coordinator. Differently from previous work (Bell, 2007;Ryoo, 2016), in this community participants were still in college, pursuing a teacher certification in EAL, but they actually had full teaching responsibilities in the program, such as planning and teaching classes as well as evaluating students. In other words, although they were theoretically pre-service teachers, the development meetings, including microteaching, felt like in-service teacher development.
LwB teachers' task was to microteach a class they had prepared for the students in order to exchange ideas and to get feedback from both the coordinator and other "more experienced peers" selected by the coordinator. During each microclass, the coordinator, Luisa (a grad student),  Helena A class about Academic English with a short reading passage, extensive vocabulary work and a short written exercise.

rd week Microteaching focus
Nadia A reading, speaking and writing class with two reading passages.

Mari
A speaking class about feminism.

th week Microteaching focus
Adam A reading and speaking class about cosmetic surgery.

Isabela
A reading and writing class about postcards and letters.

Antonia
A reading and writing class about formal e-mails.

Kelly
A reading and writing class about research articles.

7/15
In this case, it is an educational activity in which they pretend to teach to get feedback. Embedded in this primary framework, there is a keyed strip of interaction in which participants pretend to be in a class; the student teacher who is presenting the class pretends to be a teacher and the others pretend to be students.
After this introduction, Kelly indexes a change of frames from contextualizing the micro-class to teaching the micro-class by addressing her peers as students: "Okay, guys, so what is the idea today?
Okay, we are going to follow up our class from last week […]" (lines 564-572). In this way, Kelly provides participants with a verbal contextualization cue which signals contextualizing micro-class is over and actual micro-class has started. In other words, she shifts from addressing her audience as a peer to addressing them as a make-believe.
The secondary frame, a class performance, is embedded in the primary one, a technical redoing.
The discourse marker "okay" is often employed by the student teachers to indicate they are transitioning to a teaching frame, which is visible in the collection of micro-classes. Moreover, addressing peers as "guys" or "people" is used for showing that the secondary framework of microteaching as a makebelieve class has begun. Six out of eight microclasses have similar introductions followed by contextualization cues signaling the transition from introduction to micro-class (primary to secondary framework), which worked as an invitation for peers to start participating as students.
Kelly's peers start acting as students right away and begin discussing what an abstract is.
Everyone in the room (except for myself, Estevam, Maria Julia and Luisa, who are not expected to act as students) starts discussing 13 in pairs. After two minutes, Kelly mediates a whole-group discussion to define what an abstract is (lines 574-98), which her students define as "a summary" (line 576) and "an invitation to read your research" (line 580). Then, Pedro discusses the importance of an abstract for the research article (lines 589-96).
Next, Kelly asks participants to discuss the 13 At certain point, Maria Estevam, Maria Julia, and especially Pedro get carried away and participate as students, which is uncommon in the other micro-classes.

"parts of it, what constitutes [a research article]"
(line 606-7), which they do in pairs and groups.
After a couple of minutes, Kelly winds up again. which indexes that they are not going to do this in the micro-class. At the end of her last turn, she indicates the micro-class is over by saying "that's it" (line 763), immediately followed by applause.
Clapping suggests that the performance frame is always at stake, which is also consistent with previous studies (Bell, 2007;Ryoo, 2016 Adam starts his micro-class by introducing himself as the teacher for the afternoon (line 148-9). He then goes on contextualizing the microclass and talking to his audience as peers. For three minutes, Adam goes on and on explaining the micro-class. Perceiving that, Estevam jumps in and tells him to start (line 177), which he does immediately, in his following turn (line 177). He uses the discourse marker "all right" (line 178) and asks his audience a question, inviting them to participate. In our interpretation, participants not only negotiate the frames in which their actions should be interpreted, but also the identities that they are performing in these frames and through these actions, which was also claimed by Ryoo (2016). In these segments,

Estevam invokes her identity of a coordinator and
Adam of a pre-service teacher in a development activity, subordinated to the coordinator; however, only in two micro-classes does this happen.
Joking is another feature that Kelly's class does not encompass. At times, students say things that would sound unusual in a classroom or exaggerate the kinds of mistakes (linguistic or pragmatic) regular students would make.
Usually, these segments are followed by laughter, indexing that there is a joke going on, as we can see in the segment below. it as a joke. This kind of joke also supports the interpretation that microteaching is a complex and laminated activity. There is a primary frame in which they are peers doing a technical redoing and a secondary frame in which they are engaged in a performance, consisting of a make-believe class. In the latter, occasionally, there is a third lamination; that is, playfulness.
To conclude the section, it is important to look back into Kelly's microteaching event, which, as mentioned before, is prototypical. Using this event as a prototype, after having analyzed all others, is a way to look at the more generic elements of microteaching in the community. The following discernible compositional features were identified in this practice: 1. student teacher goes to the front of the room; 2. contextualizes the class (level and course) addressing others as peers; 3. shifts footing to begin micro-teaching, and addresses peers as students; 4. makes "parentheses", that is, changes from the secondary to primary frame, using contextualization cues to transition between frames; 5. brings micro-class to an end by transitioning back to the primary frame in order to explain to peers what would come next in a "real class"; 6. peers clap.
Moreover, there are two components that may or may not appear: (a) coordinator steps in, invoking the coordinators' identity; (b) participants make jokes. The figure below summarizes which of these compositional features are integral to each micro-class. The first ones represent the six main components (dark gray) and the last two (light gray) represent the other less systematic ones. The eight micro-classes that encompassed these compositional features "worked." The two classes that "did not work" according to the debriefing and feedback produced by coordination (Mari's and Ana Ricarda's) were precisely the ones that lacked many of the components which characterize the practice's pattern in the community. Hence, peer participation was minimal and there was no applause after micro-classes were over. This pattern is intuitive: there is no such thing as a manual for microteaching, and yet people usually do it in a patterned way. Not attending to the pattern may mean little participation or being considered a failure by the coordinator -as happened with Mari and Ana Ricarda.

Final Remarks
In this article, we have described the discursive architecture of microteaching in the specific community investigated in this project from the perspective of interactional sociolinguistics. We found that microteaching is a laminated discourse practice, wherein participants align to and perform different identities throughout the activity while they point to one another how to interpret their actions by using contextualization cues. Additionally, we found that microteaching is a practice with recurrent features, which has been demonstrated over the analysis. When participants did not observe such features while micro-teaching, it caused peers to fail to identify what was going on. When it happened, peers did not understand what was going on and did not know how to respond during the micro-class. The teacher educator, on her turn, thought of the micro-classes as unsuccessful.
Microteaching is, as pointed out in earlier studies (e.g. Amobi, 2005;Metcalf, Hammer & Kahlich, 1996), an important teacher development technique. Nevertheless, it can be awkward for the participants when they do not know what to do. This is also in our data.

11/15
As we set out to understand the interaction architecture of microteaching in this specific community, it is possible now to draw some situated conclusions about some possible pedagogical implications of this study for teacher development. In this sense, we would like to end this text by delineating some recommendations based on our data, which do not have the intention of generalizing the findings obtained here for all contexts. Quite the reverse, the idea is to share some learning tokens we, as teacher educators, consider important take-aways.
Firstly, it is important for teacher educators to be able to communicate what they expect from microteaching -the more information about the expectations regarding the structure of the micro-class, the better. For this, it would be a good idea to show an example of microteaching -acting out, from video tape, or even by using the transcript provided here. Secondly, it is a good idea for the trainer to show teachers exactly the components that he or she considers essential in the micro-class; for this, Table 3 may prove useful. Furthermore, pinpointing to teachers the necessity of demonstrating how others should interpret their actions with contextualization cues is also desirable.
As we navigate a new practice, sometimes explicit instruction of what we are expected to do, and how we are expected to do it can be lifesaving.
To conclude, we would like to state that we constructed the recommendations above by attending to the microteaching of a specific community. Therefore, they are not generalizable for all communities or to be taken in a prescriptive manner, though they may prove useful to other teacher trainers.