2. Background

Coinciding with (and indeed, emerging from) neoliberalism as the current dominant version of capitalism in the world today, HE (higher education) institutions worldwide have undergone intensive processes of internationalisation over the past four decades (Law & Hoey, Reference Law and Hoey2017). In broad terms, internationalisation is about ‘interactions between national governments (such as formal agreements, conflicts, diplomatic relations), or … the to-ing and fro-ing of items from one nation-state to another (such as people/travel and goods/trade)’ (Vertovec, Reference Vertovec2009, p. 3). When the term is used in HE, internationalisation is understood to be an explicit policy adopted by a university, whereby there is an increase in the number of alliances and agreements with universities based in different nation-states around the world and an increase in staff and student-mobility programmes such as Erasmus. While there are the obvious benefits associated with these and other internationalisation activities, there is also, by now, a well-documented downside to internationalisation as part of the neoliberalisation of HE worldwide (Smyth, Reference Smyth2017), not least the extreme pressure on academics to follow new accountability regimes which value quantity over quality and in, some parts of the world, the conversion of degree programmes into cash cows for which students (especially non-nationals) are charged exorbitant tuition fees.

When it occurs outside of English-speaking countries, internationalisation often entails Englishisation, which may be defined as the use of English in HE contexts where previously local languages were used (Lanvers & Hultgren, Reference Lanvers and Hultgren2018). In practice, Englishisation means the introduction of English as the mediator of communication in a range of administrative, curricular and research-related activities. This is not an unproblematic process, as there is often the concern that English will come to supplant local languages as the mediator of research, culture and scholarship (what is known as ‘domain loss’). Nevertheless, despite this concern, and indeed others, there is in evidence in many contexts of a certain acquiescence to the new status quo and a degree of resignation (there is no alternative) with regard to the increasing presence of English in HE worldwide.

A key element of the Englishisation of HE institutions, and the one that most interests us here, is EMI. EMI refers to ‘the use of the English language to teach academic subjects (other than English itself) in countries or jurisdictions where the first language (L1) of the majority of the population is not English’ (Dearden, Reference Dearden2015, p. 2). As noted elsewhere (Block & Khan, Reference Block, Khan, Block and Khanto appear a; Moncada-Comas & Block, Reference Moncada-Comas and Block2019), this definition is useful as far as it goes, but ultimately it is far too general. Among other things, it engenders a good number of questions, such as possible distinctions with regard to different contexts and if English-mediated education in locations where the L1 is a variety of English might also be considered EMI. It is questions like these that have lead Baker and Hüttner (Reference Baker and Hüttner2017) to outline three general types of EMI programme: (1) Student-mobility programmes, which cater to students from a range of linguistic backgrounds (local and international); (2) Internationalisation-at-home programmes, which deliver an internationalised curriculum in English to local students who do not ordinarily use English in their day-to-day activities; and (3) Anglophone-context programmes, which are considered to be EMI due to their critical mass of international students and the emergence of an English as a lingua franca environments (based on Baker & Hüttner, Reference Baker and Hüttner2017, p. 502).

The context focused on in this paper involves the first two programme types outlined by Baker and Hüttner. I say this because while EMI has been introduced as a way of either attracting new students from abroad or accommodating those already on site, the presence of non-nationals is minimal in most classes and non-existent in others. Thus, of the three courses examined below, two were taught to groups of local students, all of whom were Catalan and Spanish bilinguals, while the other course had two exchange students whose attendance was sporadic. This means that in our research we were far more in the realm of ‘internationalisation at home’, with local students and lecturers who all shared the same multilingual profile (Catalan and Spanish, with English as an additional language), than ‘student mobility’.

As we see from the numerous publications that have appeared in recent years, whatever the context (see Dafouz, Reference Dafouz, Psaltou-Joycey, Agathopoulou and Mattheoudakis2014; Macaro, Reference Macaro2018; Macaro, Curle, Pun, An, & Dearden, Reference Macaro, Curle, Pun, An and Dearden2018 for surveys), EMI is a multidimensional and multiscalar phenomenon and its complexity increases when it occurs in multilingual settings. In an attempt to capture this reality, Dafouz and Smit (Reference Dafouz and Smit2016, Reference Dafouz and Smit2019) propose a language ecology-sensitive model for the study of ‘English-medium education in multilingual university settings’ (EMENUS), what they call the ROADMAPPING framework, which focuses on six interrelated components in any given EMI context.Footnote ² In adapted form, these six components look as follows:

1. 1. the roles of English in relation to other languages (which can be complex where English is entering an already crowded linguistic market);

2. 2. academic disciplines, both as broadly defined by international communities of scholars and in more localised varieties;

3. 3. management of the EMI curriculum, which means both language and content management;

4. 4. the different agents and/or stakeholders involved;

5. 5. the practices and processes constituting EMI; and

6. 6. broader, shaping and structuring discourse, practices and processes such as globalising forces and flows and the more specific internationalisation of higher education.

   (based on Dafouz & Smit, Reference Dafouz and Smit2016)

In this paper, the discussion will include the six components of Dafouz and Smit's framework, albeit variably and with differing strengths. I will return to exactly how this is the case in the conclusion to this paper.

One interesting development in the implementation of EMI worldwide is how in an increasing number of contexts it is being positioned as an opportunity for students to improve their English, and indeed to learn English (Macaro, Reference Macaro2018). This shift to the inclusion of language learning as an aim of EMI means that EMI is, following Moncada-Comas and Block (Reference Moncada-Comas and Block2019), ‘CLIL-ised’. This means that it becomes, at least in part, Content and Language Integrated Learning (CLIL), where CLIL refers to a dual-focused educational approach in which an additional language is used for the learning and teaching of both content and language’ (Coyle, Hood, & Marsh, Reference Coyle, Hood and Marsh2010, p. 1). The notion of CLIL-ised EMI, understood as the slippage of EMI into CLIL, and thus English language teaching territory, is what happens when EMI lecturers adopt behaviours that might arguably be positioned as being in the realm of language teaching, such as devoting attention and time to the organisation of academic texts, pragmatics, pronunciation and aspects of grammar and lexis. The notion of CLIL-ised EMI is foundational to my discussion in the second half of this paper on EMI lecturer identities.

Dafouz and Smit's framing of EMI as a complex phenomenon was foundational to the aforementioned research project exploring EMI in action in Catalan universities. Specifically, my colleagues and I were interested in the introduction of EMI in the multilingual setting of Catalonia, where English enters academic spaces in competition with two languages – Catalan and Spanish. There is by now a fairly solid research base about this context and indeed, the Spanish context more broadly (see Lasagabaster, Reference Lasagabaster2019 for a useful summary) and our aim was to insert ourselves into this body of scholarship. We thus drew on previous research and tried to fill a relative gap by carrying out detailed ethnographic fieldwork, which included interviews, audio/video recordings of classes, audio logs produced by both lecturers and students, policy documents, students’ written work and student presentation evaluation rubrics. We focused on what we elsewhere call ‘the secret life of EMI’ (Block & Khan, Reference Block and Khanto appear b) following our interest in a long list of issues relevant to our research domain. These include: how Catalan and Spanish compete with English in EMI classrooms; the real demand and need for English in the future lives of most students; whether or not lecturers are linguistically and pedagogically prepared for EMI; whether or not students are linguistically prepared for EMI; the different ways that lecturers and students resist EMI; and finally the extent to which EMI lecturers might be positioned as language teachers, given the CLIL-ised nature of EMI in the context being researched.

These issues are about being in the world, in this case the world of EMI; and as such they are about what we might term EMI identities. Space does not allow me to present everything we have discovered thus far about EMI identities, which means I will need to be selective. Thus, in the second half of the paper I will explore in detail the last issue listed here: how EMI lecturers orient to, or do not orient to, a possible role as English language teachers. Before moving to this discussion, however, it is necessary to discuss what I mean by identity for the purposes of this paper.

3. Identity and related constructs

As Moran (Reference Moran2015) notes, the turn to identity as a focal point of interest in humanities and social sciences began to take strength in the post-world-war-two era. It became a central concern via the social theory scholarship of scholars as diverse as Butler (Reference Butler1990), Hall (Reference Hall, Hall and du Gay1996), Weedon (Reference Weedon1987), Giddens (Reference Giddens1991) and Bauman (Reference Bauman2004), who wrote about the human condition in an increasingly globalised world at the end of the twentieth century. The broad approach to identity emerging from the work of these and other scholars has been assigned many different labels, such as ‘poststructuralist’, and ‘social constructivist’ (Preece, Reference Preece2016; Zenker, Reference Zenker and Callan2018). Whatever the label used, there seems to be a consensus around the notion that identities (in plural to indicate the multiplicity of the phenomenon) are socioculturally-constructed, ongoing narratives, which develop and evolve across different spatiotemporal scales, ranging from the micro, local and immediate, to the macro, global and long-term. These socioculturally constructed narratives are based on both self-inhabited and other-ascribed positioning, where, following Davies and Harré (Reference Davies, Harré, Harré and van Langenhove1999, p. 37), positioning is understood as ‘the discursive process whereby people are located in conversations as observably and subjectively coherent participants in jointly produced storylines’. Discursive processes may here be understood to be multimodal and not just linguistic, and they are constitutive of communicative acts which are the building blocks of the aforementioned ‘storylines’ (or narratives) and the people who populate them. One aspect of positioning which has been developed in detail by Rom Harré is the notion of rights and duties, which he sees as essential components of the ‘moral dimension’ of social psychology (Harré, Reference Harré and Valsiner2012). Rights and duties are key objects of negotiation in the interactions examined in the second part of this paper.

Positions emergent in discursive processes are often labelled according to demographic categories such as gender, race, ethnicity, nationality, sexuality, religion, age, profession and social class, although they may also be labelled according to locally constructed, situation-specific positions. In addition, while identities might carry with them a degree of permanence, or even transportability from one context to the other (Zimmerman, Reference Zimmerman, Antaki and Widdicombe1998), they are continually forged and reshaped during and within individuals’ engagements in activities and communication with others, with whom to varying degrees they share beliefs, motives and values. Some researchers have seen the structuring of such group-inflected and activity-led identity according to the notion of ‘community’, which Ken Hyland has defined as ‘places where participation involves communicating, and learning to communicate, in appropriate ways, and membership depends on displaying the ability to communicate as an insider: using language to represent ourselves as legitimate members’ (Hyland, Reference Hyland2012, p. 22). Hyland's definition of community is clearly aligned with the notion of ‘community of practice’ (CoP), which grew out of Lave and Wenger's (Reference Lave and Wenger1991) early attempt to understand identity as emergent at the crossroads of what we do and with whom we do it and community membership as achieved through socialisation processes which may be complex and time-consuming, depending on the context.

For the purposes of this paper, which focuses on EMI identities, legitimacy and group membership are linked to the notion of disciplinary identities. Hyland (Reference Hyland2012, p. 25) has written about the relationship between discipline and identity as follows:

There are several issues to bear in mind when thinking about disciplinary identities. Firstly, disciplines are constructed and shaped in (and by) what Foucault (Reference Foucault1989) some time ago termed ‘discourse formations’. Discourse formations are assemblages of articulated discourses which construct what in effect are the accepted and legitimate rules and regulations, practices, behaviours, and ultimately, the subjects of institutions and the practices constituting them. Discourses here are understood as acts of meaning making which may manifest themselves in a range of genres and styles through media ranging from the written and spoken word to other semiotic modes such as the visual media, music, clothing and ornamentation, bodily movement and so on. A key element in Foucault's understanding of discourse formations is how those in positions of power in institutions act as ‘authorities of delimitation’ (Foucault, Reference Foucault1989, pp. 41–42). These authorities of delimitation establish who is in and who is out with regard to categorisations based on ‘grids of specification’. The latter are ‘the systems according to which the different [group members] are divided, contrasted, related, regrouped, classified, derived from one another as objects of … discourse’ (Foucault, Reference Foucault1989, p. 42) and they are activated by the authorities of delineation via what Foucault calls ‘gaze’. The latter term arises in Foucault's (Reference Foucault2003) detailed analysis of how observation is a key element in the establishment and maintenance of the authority of gatekeepers in institutions. As Foucault explains, gaze is not just about taking in and documenting what is happening before the observer's eyes; it is also about categorising and shaping others according to dominant discourses of normativity. Foucault focused above all on the discursive formations of the medical and psychiatric professions, but his ideas apply to the context that concerns us here: HE, and more specifically EMI in a Catalan university. In short, there is an EMI gaze cast by the university administration on EMI lecturers, as can be seen in the university's language policy and statements made by those with responsibility for the introduction of EMI in the university (Block & Moncada-Comas, Reference Block and Moncada-Comas2019).

However, it should be noted that this EMI gaze has been observed to be more passive than active, and more default than proactive. Indeed, while the three lecturers discussed in this paper understood that EMI was something that all lecturers should be doing, ultimately they were plying their trade in something of a vacuum. I say this because although the university's official language policy was for an increasing number of classes to be taught in English across a range of degree types and disciplines, this policy provided very little information regarding the exact role that was envisioned for English in the present and future and it did not specify how EMI would be implemented. Indeed a perusal of the document reveals little more than a very vague reference to ‘the coordinated action of the Institute of Languages … and the Department of English and Linguistics … [which] must support teachers to facilitate teaching in English’ (UdL, 2013, p. 15; translated from the original in Catalan). Meanwhile, beyond extoling the virtues of internationalisation and Englishisation as part of it, those responsible for the organisation and ongoing administration of EMI showed little knowledge or awareness of what EMI entails in terms of support for staff (Block & Moncada-Comas, Reference Block and Moncada-Comas2019).

Nevertheless, it is not the more general EMI gaze that is my concern here; rather, my aim is to focus on an English language teaching gaze (hereafter, ELT gaze) that was cast onto EMI lecturers by researchers who perceived certain observed classroom behaviours as evidence that these lecturers might on occasion act as language teachers. English language teaching is understood in very general terms here as including practices that might be observed in more general English language instruction and those that occur in more specialised instruction, such as English for specific purposes and English for academic purposes. The key idea here is that it includes in its purview a range of practices, including the explicit presentation of linguistic features, the correction of learners’ errors, skills-based teaching (reading, writing, speaking and listening), attention to language functions and aspects of discourse (including disciplinary discourse), the teaching of vocabulary and the teaching of pronunciation.

Importantly, where there is a gaze, there is a wide array of possible responses by the object of the gaze. These range from compliance, acquiescence and acceptance, at one extreme, to refusal and resistance – understood as the ‘intentional, and hence conscious, acts of defiance or opposition by a subordinate individual or group of individuals against a superior individual or set of individuals’ (Seymour, Reference Seymour2006, p. 305) – at the other extreme. Resistance always exists as a counter to hegemony and uneven power relations and it may or may not be successful as an attempt to challenge and overturn such relations. This means that in any institutional setting, it is interesting not only to consider instances in which the gaze is, in effect, accepted and owned by the gazed-upon but also instances in which it is resisted. Resistance will be key focal point of our discussion in the second half of this paper, when we turn our attention to how EMI lecturers are subjected to the aforementioned ELT gaze by researchers, and how in response to this gaze, they position themselves in opposition to it, falling back on their STEM discipline identities, despite the fact that they at times engage in behaviour which betrays this exclusive self-positioning.

In the midst of EMI in action, it is worthwhile to consider issues around perceived and invoked sameness and differences, and authenticity and inauthenticity and other related binary relations that come into play. Authenticity is a particularly important concept related to identity, and in sociolinguistics it may be understood to be about involving ‘what sorts of language and language users count as “genuine” for a given purpose’ (Bucholtz & Hall, Reference Bucholtz and Hall2005, p. 601), where this counting is contingent on the behaviour of interlocutors in particular contexts. Thus, to ‘count as genuine’, it is not sufficient simply to claim a certain status unilaterally and without consideration for why, or with whom and in what context one finds oneself. Rather, all of these factors must be taken into account if one is to attain the desired status as genuine, and hence authentic. The process by which this occurs is ‘authentication’ (Bucholtz & Hall, Reference Bucholtz and Hall2005). In a sense authentication takes us back to the previous discussions of CoP. With regard to the latter, individuals aspiring to form part of an established CoP must over time undergo a process of acceptance by the group to which they will affiliate, moving from an outsider and peripheral status to an insider and central one (Lave & Wenger, Reference Lave and Wenger1991), in short, moving from relative inauthenticity to authenticity.

Finally, as my concern here is with the construction of identity in interaction, in interviews and in classroom-based exchanges, it is worthwhile to introduce two key constructs from the work of Erving Goffman. Firstly, there is the notion of ‘footing’, which refers to ‘the alignment we take up to ourselves and the others present as expressed in the way we manage the production or reception of an utterance’ (Goffman, Reference Goffman1981, p. 128). Footing is very much about image management in interactions – ‘the presentation of self in everyday life’ – developed in the earlier work of Goffman (Reference Goffman1959) and it is a useful way to understand another of his constructs – ‘frame shifting’. A frame is an interactant's ‘understanding of what it is that is going on’ and how ‘individuals fit their actions to this understanding and ordinarily find that the ongoing world supports this fitting’ (Goffman, Reference Goffman1974, p. 247). These two key constructs from Goffman's work allow us to develop a fuller understanding of identities as situated and emergent in interactions.

4. Data and analysis

In the sections that follow, I present data exemplifying how three EMI lecturers participating in our research self-positioned as discipline-bound professionals who resist the ELT gaze cast onto them by researchers. There is therefore a tension between lecturers’ self-ascriptions and the ascriptions cast onto them by others. The pieces of data presented here for discussion are taken from face-to-face interviews with participating EMI lecturers and transcriptions of their teaching. For reasons of space, I provide only the English translations of excerpts taken from interviews, which were conducted in either Catalan (with Isabel and Raquel) or Spanish (with Jaime). Translations were done by the author, a proficient user of the three languages involved. Transcription conventions can be found in Appendix 1.

Appearing in the data are two researchers and three EMI lecturers. The researchers are Gerard and Derek, both of whom are full-time academics who have a good deal of English language teaching experience. The three lecturers are Jaime, Isabel and Raquel. At the time of the study, all three had five to ten years’ of HE teaching experience and at least three years of EMI experience. Jaime is an agronomic engineering lecturer who teaches a B.Sc.-level course on pig farming. He is a Catalan/Spanish bilingual who prefers Spanish in his day-to-day interactions in a range of social and professional domains. Isabel is a mechanical engineering lecturer who teaches a B.Sc.-level course on installations. She is a Catalan/Spanish bilingual who prefers Catalan in her day-to-day interactions. Finally, Raquel is a veterinary science lecturer who teaches a B.Sc.-level course on animal biotechnology. She is a Catalan/Spanish bilingual who prefers Catalan in her day-to-day interactions. As regards the English level of these lecturers, we only had the information that they themselves provided. Using the Common European Framework of Reference for Languages as the measure, Jaime and Isabel claimed to have C1 level English (proficient) while Raquel claimed to have C2 level English (highly proficient). All researchers who came into contact with these three lecturers agreed that Raquel was the most competent user of English as well as the most confident. Jaime and Isabel showed considerably more modesty about their linguistic abilities and Isabel on occasion used what she saw as her deficiencies, such as difficulty pronouncing a key term, as a way of injecting humour into her classes (see Block & Moncada-Comas, Reference Block and Moncada-Comas2019 for a discussion).

5. Self-positioning as EMI lecturers but not language teachers

In this section I discuss each lecturer in turn, focusing on their ongoing alignments with their respective STEM EMI identities, and in particular, their acquiescence, conformity or resistance to the ELT gaze. I begin with Jaime, who, as mentioned previously, was teaching a course on pig farming. There were some 30 students on the course of whom two were exchange students from South Korea. However, these students’ attendance was sporadic and Jaime effectively addressed his students as Catalan/Spanish bilinguals.