Repositorio Institucional de la Universidad Autónoma de Madrid https://repositorio.uam.es Esta es la versión de autor del artículo publicado en: This is an author produced version of a paper published in: Qualitative Research, 20.6 (2020): 854 – 873 DOI: https://doi.org/10.1177/1468794120909072 Copyright: © The Author(s) 2020 El acceso a la versión del editor puede requerir la suscripción del recurso Access to the published version may require subscription “This is the accepted version of the article may be posted in the author's institutional repository and reuse is restricted to non-commercial and no derivative uses” 1 Gatekeeping the interactional order: Field access and linguistic ideologies in CLIL-type bilingual education programs in Spanish secondary schools David Poveda, Universidad Autónoma de Madrid david.poveda@uam.es Frances Giampapa, University of Bristol frances.giampapa@bristol.ac.uk Ana María Relaño-Pastor, University of Castilla-La Mancha anamaria.relano@uclm.es Contact Author: David Poveda Facultad de Psicología Universidad Autónoma de Madrid Campus de Cantoblanco 28049 Madrid Spain e-mail: david.poveda@uam.es (January 2020) Gatekeeping the interactional order: Field access and linguistic ideologies in 2 CLIL-type bilingual education programs in Spanish secondary schools Abstract This article reflexively discusses field access as a continuous process in linguistic ethnographic fieldwork and illustrates how interactions generated during negotiations to establish a research collaboration, initial contacts with participants or data gathered to complement audio-visual recordings of naturally-occurring interaction can, in fact, become rich sources to answer research questions. The discussion is based on a critical sociolinguistic ethnography on the implementation of English-Spanish "bilingual programs" in a mid-sized city in central Spain. To build this discussion we propose a framework in which particular research stances held by participants become closely intertwined with particular research processes, spaces and techniques. Keywords: Linguistic Ethnography - Field Access - CLIL - Reflexivity - Research Devices 3 Introduction The starting point of this reflexive analysis is our revisiting of what, retrospectively, seemed a naive assumption: that as researchers we would be able to communicate easily our research project, present a work-plan and gain access to secondary schools in a way that would allow us to conduct fieldwork in a way that practically matched the outline of the research proposal. The discussion is based on a critical sociolinguistic ethnography (2015-2018) on the implementation of English-Spanish bilingual programs in the Spanish educational system. For this paper, we draw from fieldwork conducted in three secondary schools (two semi-private schools and one public state- run school) in a mid-sized city in central Spain. Ethnographic data includes 126 hours of classroom audio recordings in content subjects taught in English and English classes; 6 video recordings in these classes; 93 questionnaires with secondary students about the everyday use of English; 54 semi-structured interviews with different stakeholders (teachers, administrators, parents, policy makers and students); 12 class group discussions with around 300 secondary students attending the bilingual programs at these schools; 2 media and language landscape diaries of different workshops about the use of English in social media; photographs, website data and institutional documents of regional educational policies. These different data sets were triangulated following our critical ethnographic perspective to understand the role of language in social life in relation to wider socioeconomic, educational and cultural processes1. At the start we were optimistic that the project would unfold smoothly or within the usual 'indeterminacy' of ethnographic fieldwork. A number of factors supported this 4 initial attitude: (1) Several of the researchers involved in fieldwork had successfully completed linguistic ethnographic projects in complex urban schools in large metropolitan cities (Poveda, 2011; Mijares and Relaño-Pastor, 2011; Giampapa, 2011). (2) The senior members of the research team had expertise in working alongside participants and communities, thus understanding the ethical complexities of managing relationships in the field and also the diverse ethical processes regulated by universities and funding bodies. (3) The research project focused on a popular type of bilingual education program in Europe, Content and Language Integrated Learning (CLIL) (Nikula, Dalton-Puffer and Llinares, 2013) that seemed to have widespread support in the educational community. (4) The project was hosted in a department from the regional public university with close connections to local schools and, in addition, the university was involved in joint efforts in pre-service and in-service teacher training in the locality. The relationships forged from these on-going interactions set up the current project in ways that, supposedly, allowed for a negotiated entry to the various school sites. This was partially dependent on the school structures and their administration, and the way in which the bilingual programmes were realised within each of the school sites, as well as their openness to ‘outsider’ involvement (Bondy, 2013). In a sense, the researchers were both ‘insiders and outsiders’ (Mullings, 1999) and were positioned and self-positioned through the course of the study in various. This allowed for a different set of networked relationships that would open up or close down possibilities in the field. In previous work Giampapa (2011, 2019) notes these become ‘methodological rich points’ (Hornberger, 2013) for understanding the positionalities and identities that shift and emerge through fieldwork, which followed different paths 5 for each of the authors of this article: Frances self-positioned in relation to the setting as a Canadian born, Australian raised female from a Southern Italian background; but also as a non-Spanish speaker ‘outsider’, as a bilingual (English/Italian) speaker and inter-cultural knower, and as an experienced ethnographic traveller. These positionalities and researcher identities allowed for a different type of entry point and different types of relationships in the field. Having had a long standing relationship with the other researchers on the project and identified within the project structures as the ‘international expert’ allowed for the types of productive flows and extended absences in and out of the field sites. This was a subject position that was self identified but also marked by the regulatory conditions of the Spanish research funding system that structured and positioned researchers into different types of roles and relationships as part of the funding process. Ana María, a Spanish researcher with bilingual and intercultural experiences in the San Diego-Tijuana border (Relaño-Pastor, 2011), had developed previous relationships with the bilingual programme coordinator, the head of studies and some of the bilingual programme teachers. These relationships stemmed from a pilot study on language socialization conducted in 2011. Contrary to what was expected, when she embarked upon the current research, her role as the PI researcher was questioned and subjected to the gatekeeping practices described below. David was positioned as a Spanish researcher with extensive research 6 experience in the Spanish educational system. He is based at an institution in another locality and had an intermittent relationship with the various sites of the project as a 'commuter fieldworker' who was able to establish a good rapport with some of the participants during selected visits to the field and during different attempts at using digital devices and instant messaging platforms to communicate with teachers and students. However, in other instances these efforts failed and several of his field-work visits were misconstrued by participants in schools as a one-time special classroom activity or as 'meetings with an educational consultant'. As with any ethnographic research project, we expected that particular components of the work-plan would have to be modified or dropped, that research questions and relationships would transform during the research process or, even better, that new and unexpected pieces of data gathered during the project would enrich the original research plan. However, contrary to the optimism outlined above, moving forward the project involved surprisingly intense negotiations with multiple participants and institutional agents in the field. As part of field access work we navigated through multiple interests and 'investments' in the research project held by different social actors. These dynamics impacted the research plan and goals of the project (Mullings, 1999) more than we could foresee and required adjusting and re-assessing the work- plan at several points of the project. Team reflexive work (Creese, Bhatt and Martin, 2009; Creese and Blackledge, 2012) around the field encounters that led to transformations in the research plan also helped highlight how these very encounters, in fact, provided important insights (that is, data) relevant to central questions of the research project. In other words, documenting how teachers, school administrators, 7 parents, students and other agents framed access to their classrooms or linguistic practices or re-defined in their own terms the goals of the project helped us to understand substantial questions tied to, for example, the local linguistic ideologies around English-medium education or to how English is commodified (Heller, 2010) by local social actors (Codó and Patiño, 2018; Relaño-Pastor, 2015, 2018; Fernández- Barrera, 2017). In this paper we unpack these dynamics and hope to contribute to methodological debates in educational ethnographic and linguistic ethnographic research in relation to how field access and emergent relations become data spaces. We do this from a conceptual framing that offers a new reading of these field dynamics. Thus, our discussion is structured around two components. First, we identified a variety of research stances (cf. Jaffe, 2009), understood here as the way in which participants, and especially key participants -in a position to open access to a school, classroom, community of students or parents (see Lund, Panda and Dahl, 2016)- construe the research process and, consequently, the research relationship/collaboration they are willing to set up with the research team (Sánchez-Criado and Estalella, 2018). In addition, a research stance critically involves co-constructing an epistemic territory (Heritage, 2012) that structures how access to information and resources are negotiated. Second, these stances are closely intertwined with particular research processes, spaces and techniques. The connection between stances and research process can be understood in terms of the research devices (Ruppert, Law and Savage, 2013; Kullman, 2013) built during the project. In our analysis research devices are structured around two key features as discussed by Rupper, Law and Savage (2013). They are understood as an emergent arrangement that clusters together types 8 of research techniques, data forms and research issues (for example, 'interviews + discourses + administrators + program/policy aims' or 'observations + communicative practices + students and teachers + classroom discourse structures'). Research devices are generated through negotiations with participants and are, thus, teleological in that they are guided by the agendas and research expectations (albeit inconsistent, not completely formulated or implicit) of participants and researchers. As we hope to show below, by reflexively scrutinizing and turning to these devices and infrastructures (rather than directly to the 'data' that they are expected to 'produce') we also begin to answer our research questions2. In the following section we briefly discuss three distinct stances we identified during fieldwork as well as the research devices that emerge from these stances. We examine particular data instances that illustrate how participant stances and research processes unfold in the field and how different research positionalities (the participants’ as well as our own) are embedded in these processes. Three research stances in the study of CLIL-type bilingual education programs in Spanish secondary schools Conducting ethnographic fieldwork in schools -complex bureaucratic institutions- involves numerous interactional episodes in which the research goals of the study are communicated and substantial negotiations in relation to the research project take place. Participants do not come into these conversations as blank slates: without previous expectations or representations about what an educational research project involves or about the personal and professional implications that participating in a 9 research project might have. Rather, increasingly, participants often have robust representations about the impact of research on their daily lives and professional practices, on the role of 'expert discourses' in policy and practice and of the costs/benefits/risks that engaging in research can entail. In other words, research emerges as a social and indexical field (Silverstein, 2003) in which participants have to negotiate and create their roles, identities and positions (Bailey, 2007; Blackledge and Creese, 2014). In addition, it is often claimed that ethnographic research entails relationships, commitments and practices -that is, an ethnographic gaze (Crapanzano, 1977)- different to those of dominant educational research paradigms which require additional explicating in the field. As we presented our ethnographic approach and the practicalities it entailed for schools, teachers, students and families key participants began to both construe particular definitions of the research project and of their place in the process. This is what we called research stances above and we identified three stances among participants: (I) Research as reporting: In this first stance, key participants such as teachers or administrators, construed research as a 'discrete event'. From this perspective, volunteering to participate in a research project involves a few pre-established exchanges in which participants deliver relevant information to researchers. The epistemic stance (Heritage, 2012) of this research relationship is relatively closed and unidirectional. A series of research questions or instruments are designed and established by the research team and participants provide the required responses and information. In this context, there is little negotiation in relation to research objectives, 10 goals or relations but, more importantly, 'research as reporting' is encapsulated in particular types of data-gathering activities: research entails participating in research interviews or, at most, collaborating in administering classroom surveys. Complementarily, teachers and administrators who position themselves in this stance hesitate to engage with more sustained procedures or open data-gathering processes, such as participant observation or audio-visual documentation of practices, which they cannot control as closely. (II) Research as evaluation: In this understanding of the research process participants perceived that the main practical output of the project would unavoidably assess and portray them (whether at an institutional, classroom or individual level) in certain ways. The key feature of the research epistemic stance (Heritage, 2012) is the asymmetry between the research team (those who assess) and participants (who are assessed). Unpacking how 'research as evaluation' operated was especially relevant in relation to the implementation of English language bilingual educational policies. On the one hand, as it is often the case in educational research projects, some participants believed that their abilities and dispositions would be scrutinized and ranked against the standards of the research project and/or other participants/sites in the project. On the other hand, specifically, English bilingual educational practices operate in an educational policy system structured by assessment and accountability (Tiana, 2017). At an institutional level, schools and bilingual programs drew on their reputation, their ability to secure singular educational resources and credentials and were formally ranked within a three-tier system by the linguistic educational policies of the region at the time of the study. At an individual level, teachers and other actors had particular 11 professional biographies, linguistic experiences and (linguistic) credentials that positioned them at particular junctures of the language program or curriculum of their school. Within this logic, the research collaboration validated or casted doubt over the personal or institutional (self)-assessments that structured school programs or the professional trajectories of participants. (III) Research as an opportunity for professional development: This last stance involved a much more explicitly instrumental relationship between key participants and the research project. Again, in this relationship there is a potentially distinct epistemic stance (Heritage, 2012) that appears to be more reciprocal and symmetrical. Key participants, who in this stance were most often teachers, had certain 'leverage' in field negotiations as they facilitated or impeded access to their classrooms and students. In exchange for this access, some of the teachers we approached sought 'formal recognition' as collaborators in the research team and wanted to use this collaboration as a possible step into teaching in Higher Education. This generated interesting opportunities to build diverse research teams and establish researcher-practitioner collaborations but, as we illustrate below, it also opened the door to complex negotiations regarding what this 'collaboration' entailed. To illustrate these ways of understanding research we turn to the case of Charlie (pseudonym). Charlie was a secondary school English teacher with more than twenty years of experience and was the coordinator of the bilingual program in the public secondary school that participated in the study. He was also an adjunct instructor at the School of Education of the local university that led this research project and had collaborated with some of the members of the research team through teaching 12 university seminars on CLIL. Charlie's changing relationship with the project illustrates key features of each of these three research stances. During the course of the research project his stance towards the project oscillated back and forth between his availability to be interviewed ('research as reporting'), to his vacillation between the roles of evaluator of the teaching and language practices of his peers ('research as evaluation') and the particular way he enacted some of the duties of a team researcher ('research as an opportunity for professional development'). To begin, Charlie finally agreed to be interviewed jointly by the project leader (Ana María) and the only non-Spanish researcher of our project (Frances). As Frances does not speak Spanish the conversation took place in English. From our perspective, this was not a fortuitous decision and, rather, it allowed Charlie to co-construct an interview event in which the power dynamics of the situation and, particularly, Frances' positionality shifted (Rose, 1997; Mertkan-Ozünlü, 2007). The exchange gave Charlie an opportunity to situate, for the interviewer, the bilingual program of his school within local regional policies and provide a portrait of the regional educational system. The research speech event, in which a foreign scholar conducting research in a peripheral Spanish region interviews an experienced local educator, allowed for a reporting stance to emerge and for Charlie to construct and expert voice and display his "knowledge of policy during the interview" (Mason-Bish, 2019: 8). Excerpt 1 shows Charlie presenting his views on Spanish and regional educational policies, as well as the standing of his school while the interviewer consciously adopts a neutral attitude: Excerpt 1: Delivering information about the bilingual program (CH: Charlie / R: Frances / R2: Ana María) (...) 13 1 R: Yeah, what’s-what’s in the center for the school to-to kind of bring on the bilingual program, like is the region really pushing for? 2 CH: It is, but 3 R: yeah? 4 CH: I think Spain is like that (...) but (.) I think they’re taking the wrong approach to it because (...) It’s like (.) quantity over quality 5 R: Right 6 CH: So: it’s like 7 R: Ah, ok 8 CH: Our biggest politicians, want to boast about it (...) 9 R: yeah 10 CH: we’ve got like a thousand schools, which are bilingual 11 R: Yeah 12 CH: But it’s not true 13 R: Right 14 CH: So it’s like they’re pushing 15 R: Right 16 CH: Because of the figures 17 R: Right 18 CH: So that it’s nice they say ok, this is what we have 19 R: Yeah 20 CH: But our school, is like an exception in the sense that it’s got quality as well 21 R: Yeah 22 CH: So, it’s good because we’ve got like this kind of agreement with the British Council 23 R: Yeah, yes 24 CH: This is kind of special 25 R: Yes (...) The extract is rich in assessments, in the sense that Charlie also explicitly judged the logic behind regional educational policies ('they're taking the wrong approach', line 4) but as an evaluative frame emerges in the conversation, his own position in the interview also shifted. Potentially, Charlie's own linguistic competence as well as his 14 teaching practices and strategies could come under scrutiny in the interview, moving the exchange into an evaluative stance of his own standing as a teacher in the bilingual program. For example, Charlie provided a critical assessment of current policies ('quantity over quality', line 4) when referring to the number of bilingual schools implemented in the region, but he then re-situates this argument to evaluate his school as 'an exception' (line 20) in the local bilingual education market. From our perspective, how Charlie navigated this shift is closely tied to the linguistic identities (Bucholtz and Hall, 2005) that participants construct during the interview conversation (De Fina and Perrino, 2011; Giampapa, 2011; Relaño Pastor, 2011). Frances was positioned within the research team as the UK-based academic and native English-speaker and, thus, the interview provided an opportunity for Charlie to display his high command of English by resorting to very correct English ('Received Pronunciation'3) during the interview. This allowed Charlie to build a persona during the interview (Goffman, 1959) that was situated favorably in the evaluative frame regarding bilingual education and bilingual programs he himself had presented in the conversation. In addition, Charlie also built this linguistic and professional identity through comparisons between students: Excerpt 2: Evaluating students linguistic competencies and expectations (...) 1 CH: the problem is that bilingual students are good, so if you’re not bilingual by definition is that you’re gonna be bad, not only bad but of difficult background 2 R: Uhm-uh 3 R2: Ah, come on!- 4 CH: Yes, so I’ve got one of one of my students in the non-bilingual class lives in - is it squatter? 4 R: In a squat? 5 CH: Yeah 15 6 R: (That is) in a squat (...) 7 CH: It’s just like hh in the bilingual project you speak English, you ask them to do amazing things and projects and then you get there, the non-bilingual and you can’t do anything at all, it is impossible, it’s like "ok, I’m gonna survive" (...) 8 CH: Let’s do something (.) and it’s just impossible, it is I’ve got twelve students and their level is horrible and there is nothing you can do about it, I’ve tried everything (...) Charlie described the type of students that attend his school, categorizing them into 'bilingual' and 'non-bilingual' students. Yet, in the interview he explicitly disentangled how the 'bilingual/non-bilingual' distinction emerged as a categorization device (see Schegloff, 2007) that classifies students according to their second language skills and the instructional track in which they are enrolled and is also a label that encapsulates expectations about students' academic competencies and their social origin (line 1). Arguably, it is this part of the interview that illustrates better the shift in his role as a provider of information that the researcher solicits (research as reporting) to the role of the evaluator as an experienced teacher in the bilingual program. In Excerpt 1, Charlie presented his assessment of national/regional educational policies setting up a hierarchy of bilingual programs - while suggesting that he is legitimately entitled to be part of the type of bilingual program that sits at the top of the hierarchy. In Excerpt 2 he turned to the categorization practices of his own school and reflexively unpacked them for the interviewer. This allowed Charlie to quickly shift through two evaluative stances: one in relation to the categorization mechanism itself (noticeably, in collaboration with the interviewers who assess this mechanism, lines 2-3) but then applying it to assess his own students and the outcomes of his teaching effort (lines 7-8). 16 Finally, during the 2016-17 academic year Charlie's involvement in the project became much more instrumental and he joined the study as part of the 'project work- team'4 moving into the third stance outlined above. As we discuss further below, this potential window for collaboration generated tensions. One of the project components the research team wanted to develop with Charlie's students involved generating workshops with adolescents around their informal English media and language practices. In several conversations with the lead researcher, Charlie expressed his anxiety over visits to his class due to the reluctance of the school to be observed by external researchers. As an alternative, he suggested reframing this sub-study as a class activity he could control and execute fully and use as part of the final grading of his students. He organized the activity with his class using various website tools in which students uploaded images and information on their personal use of English outside the classroom. The resulting data shared for the project were several files with hyperlinks to the materials generated by the students, without any other ethnographic information about the processes that led to the generation of these materials, information on the identities and biographies of the students or a clear chronology of how these materials were generated. In short, this decision produced student materials difficult to interpret within our linguistic ethnographic project but revealed many things about the accessibility of classroom practices and of how teachers defined their collaboration with the research project. In summary, the brief discussion of our evolving relationship with one teacher illustrates how participants construed an understanding of the research process and the types of relationships they developed with researchers over time. This understanding may then evolve in different directions that we argue can be classified 17 within a limited set of fuzzy categories (McCloskey and Glucksberg, 1978) that reflect different overlapping stances, realized in particular ways by participants throughout the project. More importantly, each of these stances tends to generate types of research relationships and research infrastructures, what we defined as research processes. To restate the main argument of the paper: these junctures are often construed as field access negotiations that will allow subsequent access to research data (Bondy, 2013) but, in our case, they were instances where substantive interpretive data relevant to our research questions emerged. Figure 1 summarizes the analytical model we propose to connect research stances and approaches. In the following section we develop in more detail how research processes unfolded. Figure 1: Research Stances / Research Infrastructures / Research Processes Three research processes Gatekeeping 18 Research as reporting builds on a research relationship in which participants can be considered (and see themselves) in control of the research process, the information that is disclosed and the institutional spaces researchers may or may not access. This stance is especially visible in some school administrators (school principals, department heads, etc.) and, from another perspective, can be seen as a gatekeeping process (Erickson, 1975) in which key actors control access to what, from a linguistic ethnographic perspective, is often seen as the 'holy grail' of the research design: unmanufactured instances of everyday classroom interaction (Silverman, 2013). The impulse of an ethnographer might be to find ways to sidestep these barriers and negotiate the conditions under which 'actual' educational practices may be adequately documented through participant observation and/or audio-visual recordings. We attempted to do this with some success in some settings and failed to do so in others. As these fieldwork dynamics unfolded, we also began to consider how this reluctance connected to broader relational dynamics among different actors (schools, parents, students, administrators or local policy supervisors) involved in the implementation of bilingual programs in the region. Admittedly, gatekeeping as an analytical construct has been used to describe situations in which the actions of institutional agents can have a significant impact on the social trajectories of participants (Poveda, 2017) and it would be difficult to claim that this asymmetry captures the relationship between university researchers and educational administrators. However, the power of the concept also lies in unpacking how an apparent 'bureaucratic rationality' can be put into action to activate decisions and choices that are construed as procedural rather than based on the individual 19 preferences, desires or needs of social actors. In our study, administrators had reasons to be cautious about opening up their schools: problematic incidents could be uncovered, discrepancies between what was claimed about a school program and what actually happened inside classrooms could emerge or their school practices could be compared to those of other schools. Yet, many of these reasons cannot be openly put forward as a justification not to collaborate in a research project on a topic that is intimately connected to the mission of the particular schools approached for research. Thus, one alternative solution is to collaborate in research if it is reframed as engaging with research procedures and instruments, such as interviews, surveys, etc. that are seen as less compromising. As fieldwork progressed we also identified ways in which this stance emerged as part of a broader relational strategy between school administrators and other actors in the system, particularly in situations in which accountability played a central role. For example, in relation to educational authorities, the supervision of the implementation of bilingual programs was primarily limited to 'paperwork' exchanges. Monitoring a school program basically involved revising documents and program outlines to confirm that they meet policy guidelines and what, at the time of the study, were legislated as the minimum requirements a school had to meet to participate in the regional plurilingual education program. Excerpt 3: Interview with regional educational inspector (I: Inspector; R: Researcher Ana María. Originally in Spanish) 1 I: (...) we sit with the head of the service and we prepare a series of indicators which we will use to work, then we look at more things 2 R: Mm-hh 20 3 I: That we might see there, but that is the minimum we will go and check, so these indicators, and I am talking in general because they are almost always similar, on the one hand we look at the program document of the school, how they are written up and how they modify the school's educational project, the general annual programming, the annual report of course and each department's programming. What we found initially is that in most centers, not all thank God, but in most the linguistic program was something completely isolated and fixed (...) The goal here is not to critique how educational policies were monitored. Rather, we want to highlight how this was a relatively symbiotic arrangement between regional educational authorities and school administrators that allowed performing (cf. Youdell 2006) educational supervision while protecting key actors from more critical scrutiny. For regional educational authorities, this logic satisfied the demands from unions, the political opposition or the media that closely monitored possible administrative irregularities. For school administrators it allowed sufficient 'leeway' to sustain school- internal arrangements that, while in practice might not have been strictly in accordance to policy regulations, could be seen to conform to the guidelines set for administrative documents and responded to the demands of other actors in the school. A clear example of this leeway was the intrinsic value that families attributed to English 'native speakers' (i.e. instructors born and raised in an English-speaking country) regardless of the educational qualifications or language-credentials this staff had (Fernández- Barrera, 2017; Relaño-Pastor and Fernández-Barrera, 2019). In response to this demand, school administrators in state-run and semi-private schools found ways to arrange their resources to secure the presence of 'native speakers' in classrooms and to give their school an added value in the eyes of families. Staging 21 Eventually, fieldworkers were allowed to observe teachers' classroom practices. However, these encounters were tainted by an evaluative stance, in which in one way or another the language skills of participants were in the spotlight (primarily the language skills of teachers rather than their pedagogical practices). This led to an interactional dynamic and research process that we describe as staging, broadly speaking in Goffman's (1981) sense of the term (see Denzin, 2002). In our data, staging as an interactional arrangement has three features: (a) activity seems to be much more scripted than expected; (b) researchers are reframed as 'language experts' both in terms of their linguistic repertoire and professional knowledge and; (c) interaction unfolds around a participation framework (Goodwin and Goodwin, 2004) in which participants validate their linguistic practices and skills vis a vis researchers (see Excerpt 1). The following excerpt of a closing of a classroom activity illustrates a very explicit enactment of this evaluative framework: Excerpt 4: Closing the recording of a classroom observation (transcribed adapting Sacks, Jefferson and Schegloff, 1974) 1 T: (...) Ok professor, assistant ah:: ((raises his head and looks towards camera and back of the classroom where two ethnographers are sitting)) 2 R1: thank you ((outside the video frame)) 3 T: can we (.) can you say us (.) can you share a few words with us, what's your opinion about the students what's [your opinion about their proficiency (...) in English 4 R1 [((laughs)) 5 T: (...) what's your overall opinion about them? do you think they will succeed in life? EVEN if they have to migrate-let's hope-let's be positive probably in the:: let's say seven eight years time (.) Spain will be a bit more successful and they will have to - ssshh! (5) ((looks at students at his sides)) 6 R1: ((laughs)) 7 T: listen (don't play now) listen ((looks back at the camera/researchers)) 22 professor assistant (...) Screenshot 1: The teacher begins to ask the question to researchers (lines 1-3) Screenshoot 2: The teacher and class begin to listen to the researcher's answer (after line 7) Researchers Ana María and Alicia had pre-arranged to visit and record this class and conduct a focus group discussion around bilingual education. The excerpt above captures part of the closing sequence of this class. Once the group activity was concluded, the teacher (Federico, pseudonym) addressed the researchers -who were sitting next to the camera at the back of the classroom- and requested an explicit assessment of students' linguistic competencies (lines 1-3). Further, Federico later upgraded the relevance of the evaluation by connecting it to the future life trajectories of the students (line 5). This move re-organized the participation structure of the classroom by shifting the attention of the teacher and all students to the researchers (Screenshots 1 and 2). The question itself is potentially problematic, as it is basically a 'trick question', and was navigated by researchers as part of ongoing field relations. As framed, even if the two fieldworkers attempted to answer the question candidly it is not reasonable to think 23 they would be able to say something substantial about students' linguistic skills or future social trajectories based on a single visit to the classroom. More importantly, within a context in which researchers were negotiating access to the bilingual classes at this school and attempting to set up a research collaboration, the assessment sequence responded to the preferred and expected response (that is, overall positive to strengthen the collaboration). The staging of classroom activity emerged with some teachers who appeared confident about their linguistic skills and their involvement and place in the plurilingual/bilingual program of the school. It was also connected to their own professional and personal aspirations, which they considered may benefit from engagement with a university-led research group. However, we also identified forms of staging in group interviews with parents (in practice, 100% mothers), in which substantial parts of the conversation turned into an effort on the part of mothers to validate their own family linguistic policies (King, Fogle and Logan-Terry, 2008), and practices around English and their involvement in promoting their children's Spanish- English bilingualism (Relaño-Pastor, 2018). Finally, staging seemed to be relatively emergent as it unfolded in interaction during encounters that were specifically designed and planned by researchers, such as an appointment to conduct a classroom observation/recording or interview, and explicitly defined by researchers as non-evaluative. In these encounters not much previous consideration was given to the roles of participants beyond expecting participants to behave or respond naturally and acquiesce to the guidance of fieldworkers during the research process. In contrast, there were other research devices in which participants 24 (mostly teachers, in this case) pre-negotiated more explicitly the terms of their collaboration with researchers and the project. Co-designing The research processes discussed so far moved between a reluctant and guarded approach (gatekeeping) to a more open but monitored involvement (staging). However, some teachers did consider participation in the research project as an opportunity for development, embracing collaboration with researchers more openly. This collaborative relationship involved teachers asserting and negotiating particular roles in the research project and program. Teachers seeking to actively participate in the research project were also immersed in accreditation-oriented professional settings and, thus, actively requested that their collaboration was certified, that their names appeared in the research documents (including possible future publications) and that their research collaboration accrued towards their possible professional move into university-level teaching. Consequently, involvement implied having a voice in how research and data gathering practices developed, thus becoming to some extent co-designers of the research process. Under these circumstances, negotiations were complex as they took place within institutional and professional frameworks that created particular constraints regarding how 'collaborative' relationships could develop. Turning to the latter issue, it simply needs to be acknowledged that even within a context in which researchers and teachers come with the 'best intentions' into a potential collaboration, the labor conditions, schedules and time-frames of secondary school teachers and university 25 faculty made it increasingly difficult to find spaces to actively co-design a work-plan - more so in an educational policy context marked by austerity measures and a contraction of resources and available time in the work of both teachers and faculty. In this section we focus on the tensions that emerged as certain professional identities and research processes were put into motion within what researchers and teachers attempted to define as a collaborative framework. Specifically, examining this research process helped to unpack the tensions between two dynamics that co-existed during fieldwork. On the one hand, participants and researchers tended to co-create research infrastructures in which particular stances and processes were paired (see Figure 1), yet these stances and processes changed over time and evolved during fieldwork (for example, Charlie's changing relationships in the project). On the other hand, researchers and teachers were always positioned in some point of the field of social relations created by the research devices generated in the project. Within our framework, they had to assume certain research stances and engage in certain research processes. In addition, even if these positions were temporary and situated, the research infrastructures in which they participated had particular affordances where certain research dynamics were facilitated or inhibited. These frictions were made especially visible at transition moments of the project, when participants and researchers renegotiated their relationship more openly. A negotiation involves reconstructing the research devices in which participants engaged but in this transformation traces of other identities, stances or processes can emerge. As an illustration of this interplay consider the following. As a starting premise, researchers had to acknowledge that an evaluative stance was recurrently present in 26 many research encounters of the project. Maintaining an evaluative stance inhibited constructing a more collaborative research relationship that considered teachers as co-designers/researchers in the process. However, collaboration was also seen as an opportunity for professional development, something which was particularly appealing to those teachers who, given their language credentials and institutional position, could benefit more clearly from this investment. But given that it was the same actors who could be entangled in these dynamics, there were moments in which these two agendas -an evaluative and a collaborative stance- collided and tensions with the project or even among teaching colleagues emerged. From our perspective, this is so because two potentially incompatible research stances were displayed simultaneously. The following email exchange and subsequent conversations illustrates this dynamic. Excerpt 6: Declining an invitation to participate through e-mail5 For: Ana María Subject: Re: Hello/Meeting Hello Ana, [1] This morning we had a meeting of the 'British project' and I presented your project to the rest of the teachers. I am afraid that, at least for this year, we do not feel we have the energy (fuerzas) to participate. [2] The new LOMCE education law puts a lot of pressure on us, not only with more teaching hours but with more standard evaluations, which are constantly supervised by the inspection. [3] And if that were not enough, we have an Erasmus student who comes to class and Master students completing their internships in our classes. With this panorama, we as teachers in the school do not feel with enough energy to be observed in our classes once again. 27 [4] I also have to share with you that some teachers have expressed their discomfort with the fact that our school was approached "after the fact", without being informed of possible projects underway which would require our collaboration and in which, no doubt, we could participate more actively. In any case, as bilingual education is developing firmly in our province, I am sure you will be able to find another school in which to develop your project. Sincerely Charlie In the email, Charlie introduced two aspects that are critically relevant to understand the context of the project and the processes discussed in this section. In the first part (block 1), Charlie shifted his alignment towards a collaboration with the project. First, he positioned himself as the coordinator of the program with the British Council and limited his role as someone who presents the research proposal to the team of bilingual teachers (addressing the researcher in second person 'your proposal'). Then he shifted to the inclusive ‘we’ to index his identity as a teacher and affiliate with feelings shared by the teaching staff - which are repeatedly described metaphorically as 'energy' (fuerza, block 1 and block 3). The following parts of the message (block 2) elaborated on the conditions that supported their decision to decline participating in the project. This account adds features to the portrait of the teaching staff and school. On the one hand, they presented themselves as a school (like all others) affected by the increased surveillance of current educational policies under deteriorated labor conditions (block 2). On the other hand, Charlie presented the school as actively involved in a variety of voluntary programs such as the Erasmus student/staff exchange program or mentoring master students completing their teacher training, which suggested the school was committed to external collaboration and expanding 28 their relationship with other institutions in the system (block 3). At the same time, the last statement of this section hinted how research was being construed by the staff: as 'being observed' (block 3), that is, assessed and judged. This extended portrait of the school sets the stage for the final part of the message, potentially framed as an interpellation to the research project leader (block 4). Here, Charlie shared the negative feelings among the staff for what they perceived as being excluded from earlier stages of the decision-making process and the possibility of building an early collaboration to develop the research project. 'After the fact' (A posteriori) refers here to after the research bid was successful, implying that the school staff considered they should have also been involved in the preparation of the proposal and included in the grant application. This final assessment led to a string of emails and conversations with different teachers and administrators of the school that uncovered the multiple interests, representations of research and dispositions that co-existed during fieldwork. Immediately after reading this email, Ana María called Elsa (pseudonym), a music teacher of the bilingual program who, like Charlie, had a previous amicable professional relationship with part of the research team. The goal of this conversation was to try to get some insights into the email response, which at that point of the project came as a surprise to team members given the existing relationship between Charlie, Elsa and the faculty/team members at the local university. In the course of a long telephone conversation Elsa offered an understanding of the context of the email. She tried to explain some of the complexities in the relationships 29 among teachers, especially between those teaching in the bilingual program and those who do not teach in the bilingual education program. Elsa also helped clarify that the 'discomfort' expressed by some teachers at the end of the message (block 4) reflected Federico's feeling that the invitation to his classroom (Excerpt 4) was not sufficiently acknowledged in the project. Given this, she suggested that the best course of action to clarify issues would be to call Federico, who, at the time, was also head of studies of the school. Ana María followed her suggestion and called Federico. He candidly explained why teachers, himself included, did not feel comfortable with classroom observations, restating some of the arguments in the email sent by Charlie: pressures from practicum students, Erasmus students, and assessments (blocks 2 and 3). Federico also elaborated on the reasons for their discomfort explaining that they had felt excluded from the research process. He asked the researcher why they had not contacted the school earlier before asking for funding so that they could also be part of the application process. This led to a detailed explanation by Ana María of the different logistics and expectations between this project (funded through the Spanish National R+D Plan) and other projects funded by the regional government and educational administration, which are much more oriented towards implementing and assessing educational innovations and do require more active collaboration from schools at all stages of the proposal. In this context, Ana María stressed that there were still possibilities for collaboration but that these materialized once the funding was approved and the project was launched. These explanations seemed to ease some of the tensions and opened the door to different types of collaboration and forms of recognition with some of the teachers at the school. Federico continued to decline the 30 invitation to participate beyond the initial recording of his class due to his teaching and administrative commitments. Charlie developed the relationship described above in which he was interviewed and provided some materials from his classroom. Elsa co- designed with the research team a survey for her students, allowed field-workers to record her classes on a few occasions and participated in a series of research interviews about her work in the bilingual program. Conclusions The different processes and research devices with the associated stances we have discussed in this article have in common that they intervene between fieldworkers and those 'elusive' episodes of naturally-occurring interaction and English language practices in the classroom. However, rather than seeing these as watered down versions of ethnographic data or faulty alternatives to the core interactional order, we have attempted to show how these processes emerged as rich data sites to answer central questions in our research project. They revealed aspects of the linguistic ideologies surrounding learning and using English in the plurilingual/bilingual programs that are currently receiving so much support from Spanish regional governmental authorities. They point to how English use was tied to unfolding parental, student and teacher identities (Relaño-Pastor, 2018; Poveda, 2019; Fernández- Barrera, 2017). They show how English emerges as a commodity in the institutional image and social history of schools (Relaño-Pastor and Fernández-Barrera, 2019). In short, as a primarily methodological reflexive argument, we claim that linguistic ethnographic research may turn to field access and liminal spaces in the research process as data-rich sites (Hornberger, 2013). We also proposed a conceptual framework that might help in this process. 31 Acknowledgments We would like to thank the anonymous reviewers for their insightful comments and suggestions. Special thanks to High Towers teachers, who generously offered their time despite constraints and pressure to meet the demands described in this paper. Research reported in this article was supported by the project "The appropriation of English as a global language in Castilla-La Mancha secondary schools: A multilingual, situated and comparative approach", funded by the Spanish Ministry of Science and Innovation (Plan Nacional I+D+I 2014 / Reference: FFI2014-54179-C2-2-P). It has also been possible thanks to the funds granted to the Research Consolidated Group LADIC - UCLM, co-financed by the European Union and the European Regional Development Fund (Ref: 2019-GRIN-26973). Notes 1. The project followed the ethical guidelines of Spanish funding authorities, all participants were fully informed of the nature of the study and gave their written consent/assent to participate in the study and collect audio-visual data from them. In addition, the project sought approval from the institutional review board of the university hosting the study but at the time of the project this board was still not fully functional. 2. This discussion on reflexivity around research devices and collaborative infrastructures takes place within a project that is not designed as a collaborative ethnography as discussed in the literature we cite (Sanchez-Criado and Estalella, 32 2018, Kullman, 2013; Campbell and Lassiter, 2010). Research questions or strategies were not co-defined with/for participants and, rather, the project was framed as a critical sociolinguistic ethnography in which the roles of researchers and participants are supposedly to be more clearly delineated. 3. Received Pronunciation (RP) “refers to an accent in English regarded by many people as a ‘standard’ accent...also called ‘the Queen’s English’ or ‘BBCEnglish’. In the past, RP had high status in the UK, indicating an educated speaker…” https://www.teachingenglish.org.uk/article/received-pronunciation-rp (Accessed 27 January 2020) 4. 'Project work-team' is a category included in the Spanish R+D funding scheme that is explicitly introduced to formally acknowledge the involvement and participation of non-academics and professionals in research projects. 5. Emails were originally written in Spanish and have been translated by us. We divide them here into numbered analytical blocks to facilitate discussion. The author of the email is Charlie, the same teacher discussed in Excerpts 1 and 2 above. This email exchange took place nine months before (and in a different school year) he accepted to participate in an interview. References Bailey B (2007) Heteroglossia and boundaries. In: Heller M (ed) Bilingualism: A Social Approach. New York: Palgrave McMillan, 257-274. 33 Blackledge A and Creese A (eds) (2014) Heteroglossia as Practice and Pedagogy. London: Springer. Bondy C (2013) How did I get here? The social process of accessing field sites. Qualitative Research 13 (5): 578–590. Bucholtz M and Hall K (2005) Identity and interaction: a sociocultural linguistic approach. Discourse Studies 7 (4–5): 585–614. Campbell E and Lassiter L (2010) From collaborative ethnography to collaborative pedagogy: Reflections on the other side of Middletown project and community–university research partnerships. Anthropology and Education Quarterly 41 (4): 370-385. Codó E and Patiño-Santos A (2018) CLIL, unequal working conditions and neoliberal subjectivities in a state secondary school. Language Policy 17: 479- 499. Crapanzano V (1977) On the writing of ethnography. Dialectical Anthropology 2 (1): 69-73. Creese A and Blackledge A (2012) Voice and meaning‐making in team ethnography. Anthropology and Education Quarterly 43 (3): 306-324. Creese A, Bhatt A and Martin P (2009) Multilingual researcher identities: Interpreting linguistically and culturally diverse classrooms. In: Miller J, Kostogriz A and Gearon M (eds) Culturally and Linguistically Diverse Classrooms: New Dilemmas for Teachers. Bristol: Multilingual Matters, 215- 233. De Fina A and Perrino S (2011) Introduction - Interviews vs. ‘natural’ contexts: A false dilemma. Language in Society 40 (1): 1-11. Denzin N (2002) Much ado about Goffman. The American Sociologist 33 (2): 34 105-117. England K (1994) Getting personal: Reflexivity, positionality, and feminist research. The Professional Geographer 46 (1): 80–89. Erickson F (1975) Gatekeeping and the melting pot: Interaction in counseling encounters. Harvard Educational Review 45 (1): 44-70. Fernández-Barrera A (2017) Language appropriations, ideologies and identities in bilingual schools in Castilla-La Mancha (Spain). Bellaterra Journal of Teaching and Learning Language and Literature 10 (2): 41-58. Giampapa F (2011) The politics of 'being and becoming' as researcher: Identity, language and power in multilingual research settings. Journal of Language, Identity, and Education 10 (3): 127-131. Giampapa F (2019) An unfinished critical ethnographic journey: Collaborative reframing and repositioning of relationships in the field. Foro de Educación 17 (27): 167-184. Goffman E (1959) The Presentation of Self in Everyday Life. New York: Doubleday. Goffman E (1981) Forms of Talk. Philadelphia: University of Pennsylvania Press. Goodwin MH (1990) He-Said-She-Said: Talk as Social Organization Among Black Children. Bloomington: Indiana University Press. Goodwin C and Goodwin MH (2004) Participation. In: Duranti A (ed) A Companion to Linguistic Anthropology. Oxford: Blackwell, 222-244. Heller M (2010) The commodification of language. Annual Review of Anthropology 39: 101-114. Heritage J (2012) Epistemics in action: Action formation and territories of 35 knowledge. Research on Language and Social Interaction 45 (1): 1-29. Hornberger N (2013) Negotiating methodological rich points in the ethnography of language policy. International Journal of the Sociology of Language 219: 101-122. Jaffe A (2009) Stance: Sociolinguistic Perspectives. Oxford: Oxford University Press. King K, Fogle L and Logan‐Terry A (2008) Family language policy. Language and Linguistics Compass 2 (5): 907-922. Kullman K (2013) Geographies of experiment/experimental geographies: A rough guide. Geography Compass 7 (12): 879-894. Lund R, Panda SM and Dhal MP (2016) Narrating spaces of inclusion and exclusion in research collaboration – researcher-gatekeeper dialogue. Qualitative Research 16 (3): 280–292. Mason-Bish H (2019) The elite delusion: Reflexivity, identity and positionality in qualitative research. Qualitative Research 19 (3): 263–276. McCloskey M and Glucksberg S (1978) Natural categories: Well defined or fuzzy sets?. Memory and Cognition 6 (4): 462-472. Mertkan-Ozünlü S (2007) Reflexive accounts about qualitative interviewing within the context of educational policy in North Cyprus. Qualitative Research 7 (4): 447–459. Mijares L. and Relaño-Pastor A (2011) Language programs at Villababel High: Rethinking ideologies of social inclusion. International Journal of Bilingual Education and Bilingualism 14 (4): 427-442 Mullings B (1999) Insider or outsider, both or neither: Some dilemmas of interviewing in a cross-cultural setting. Geoform 30: 337–350. 36 Nikula T, Dalton-Puffer C and Llinares A (2013) CLIL classroom discourse: Research from Europe. Journal of Immersion and Content-Based Language Education 1 (1): 70-100. Poveda D (2011) Performance and interaction during 'reading hour' in a Spanish secondary school. Linguistics and Education 22 (1): 79-92. Poveda D (2017) The key situation revisited. EDiSO Working Papers n° 2: 1- 21. (http://www.edisoportal.org/workingpapers) Poveda D (2019) Bilingual beyond school: Students' language ideologies in bilingual programs in South-Central Spain. Foro de Educación 17 (27): 11-36. Relaño-Pastor AM (2011) Crossing language and identity as a critical border ethnographer in Southern California. Journal of Language, Identity and Education, 10 (3): 186-205. Relaño-Pastor AM (2015) The commodification of English in 'Madrid, comunidad bilingüe’: Insights from the CLIL classroom. Language Policy 14 (2): 131–152. Relaño-Pastor AM (2018) Understanding bilingualism in La Mancha schools: Emotional and moral stancetaking in parental narratives. Revista Española de Lingüística Aplicada / Spanish Journal of Applied Linguistics 31 (2): 578-604. Relaño-Pastor AM and Fernández-Barrera A (2019) The 'native speaker effects' in the construction of elite bilingual education in Castilla-La Mancha: Tensions and dilemmas. Journal of Multilingual and Multicultural Development 40 (5): 421-435. Rose G (1997) Situating knowledges: positionality, reflexivities and other tactics. Progress in Human Geography, 21 (3): 305-320. Ruppert E, Law J and Savage M (2013) Reassembling social science methods: 37 The challenge of digital devices. Theory, Culture and Society 30 (4): 22-46. Sacks H, Jefferson G and Schegloff E (1974) A simplest systematics for the organization of turn-taking for conversation. Language 50 (4): 696-735. Sánchez-Criado T and Estalella A (2018) Introduction: Experimental collaborations. In: Estalella A and Sánchez-Criado T (eds) Experimental collaborations: Ethnography through fieldwork devices. Oxford: Bergham, 1-30. Schegloff E (2007) A tutorial on membership categorization. Journal of Pragmatics 39 (3): 462-482. Silverman D (2013) A Very Short, Fairly Interesting and Reasonably Cheap Book about Qualitative Research, 2nd Edition. London: Sage. Silverstein M (2003) Indexical order and the dialectics of sociolinguistic life. Language and Communication 23: 193-229. Tiana A (2017) PISA in Spain: Expectations, impact and debate. European Journal of Education 52 (2): 184-191. Thapar-Björkert S and Henry M (2004) Reassessing the research relationship: Location, position and power in fieldwork accounts. International Journal of Social Research Methodology 7 (5): 363–381. Youdell D (2006) Subjectivation and performative politics - Butler thinking Althusser and Foucault: Intelligibility, agency and the raced–nationed– religioned subjects of education. British Journal of Sociology of Education 27 (4): 511-528.