1The ongoing trend towards the internationalisation of university teaching and research in education institutions in Europe is key to understanding emerging teacher identity (Cots et al.; Doiz et al.; Tange). Teaching subjects in English in higher education where English is not the majority first language (L1) of the population is referred to as English as a Medium of Instruction (henceforth EMI). Consequently, teaching academic content in English is likely to contribute to the emergence of new language identities.
2The research aim of the present study was therefore to explore how language identity was perceived by EMI lecturers as part of their developing teacher selves. Perceptions of language identity were thus explored in this ethnographic case study, carried out at a French engineering school (École des Mines), which explores the oral, visual and written responses that the fifteen participants shared with each other and with the researcher-trainer (the author of the present paper) when describing language identity in relation to teacher identity. The present study reflects on how professional EMI lecturer-identity is enacted through four main constructs: the auto-evaluation of competence, emerging bilingual identity, pedagogy (including managing interaction) and professional development.
3The results show that language identity was re-assessed and negotiated during the workshop. The combined objectives and outcomes of the study have led the way towards supporting French-speaking lecturers who wish to become English-medium lecturers in France. The present study provides support by suggesting recommendations for future EMI training schemes which wish to integrate EMI teacher identity as part of their development programmes.
4‘Internationalisation’ within higher education is a term used to describe increased student mobility, international masters’ courses and a growing population of overseas students attending French universities. Internationalisation also seems to imply the use of English as a global language of scientific communication, research, mobility and more recently, for teaching (Cots et al.; Doiz et al.; Tange). This means that teacher identity is undergoing a language shift in higher education in France. Teacher identity is generally understood as ‘the development of one’s awareness and understanding of self as a teacher’ (Johnson et al. 103-121). Language identity is understood as a social representation of the self and others through oral and written texts (Fairclough; Archer). In addition to the cumulative identities which shift according to need and context, teacher identity is a positioned identity enacting ideology and status in a community of practice (Wenger). Lecturer identity is understood as teacher identity specific to those teachers who have teaching and research responsibilities within higher education and who have a PhD qualification.
5Reflections on a developing sense of EMI lecturer identity can be accessed from an exploratory practice perspective (Allwright). In contrast to focusing on capturing the ‘linguistic needs’ of EMI lecturers (Gustafsson 2018), EMI professional development programmes, such as the one described in this article, can also aim to capture the EMI lecturer’s ‘identity needs’. In this case, practitioner research of this kind is in keeping with teacher development theory where teacher training is ‘largely discovered […] in teachers’ current practices, rather than invented ‘out of the blue’ for teachers’ (Allwright 110). As in Gustafsson’s EMI lecturer training study, teacher roles can be explored and the likely instruction scenarios enacted. Instruction scenarios are simulated teaching events where lecturers can develop their pedagogical skills outside of the classroom.
6The introduction of EMI courses in France in the 2000s (and especially since the Fioraso law in 2013 which encouraged EMI initiatives) was met with some criticism because of the worry that learning in a second language may reduce the ‘quality’ of higher education programmes. This is because both teaching and learning in a second language (L2) may reduce both the production and comprehension of academic discourse (Preisler; Roussel et al.). This criticism, however, has not been sufficient grounds for French higher education institutions to turn away from teaching in English. This may be due to the fact that both the students and the lecturers involved in EMI are already experienced users of English and that sustainable EMI courses involve willing teachers and students (see Salomone for failed attempts to ‘enforce’ EMI in an Italian university). Nevertheless, the question of quality is closely associated with the question of language competence in the EMI classroom (Taillefer). Competence is a contested notion involving attribution and measures of standard according to what it means to use language ‘appropriately’ (Kramsch; Canale and Swain). In terms of using English as a second language, competence is often considered in relation to a national (idealised) speaker model (Davies). The present study therefore invites EMI lecturers to reassess their language competence according to an English as a lingua franca (henceforth ELF) context (Jenkins 2015).
7Educational contexts in which English is used, be it ELF, as an additional language of instruction, have been described as bilingual linguistic policy (for the description of different types of bilingual policy within education, see García 2009). By adopting EMI within higher education, France is moving towards the use of English as a second language for learning. Recent definitions of bilingualism refer to the ‘ability to use more than one language’ (García 44) or the ability to use more than one language ‘on a regular basis’ (Grosjean 21). This study identifies a gap in current EMI literature which is shying away from explicitly defining EMI as a bilingual practice. The present study is in keeping with a more recent theory which has established that the functioning EMI classroom in France is, in effect, an English-French bilingual classroom in which both lecturers and students engage in the translanguaging practices associated with bilingual speakers (Reynolds; Creese and Blackledge; Wei).
8In line with current research on language identity within education, this project culminated in the following research questions:
9RQ1: How do EMI lecturers define their level of English language competence?
RQ2: Do EMI lecturers consider themselves to be bilingual speakers of French and English?
RQ3: Do EMI lecturers reassess their pedagogical approach when they teach their subject in English?
RQ4: Finally, how can future EMI courses be best developed and improved to adapt to EMI lecturers' needs?
10The answers to these research questions provide the groundwork for achieving a better understanding of emerging EMI teacher identity (discussed in 4.3).
11This section will give a brief overview of the EMI workshop and provide information regarding the participants.
- 1 All the participants of this study signed a complete assignment of copyright for their drawings and (...)
12The workshop consisted of fifteen lecturer-participants and one researcher-trainer (the author of the present article). The trainer was invited by a university administrator in charge of international development at the Ecole des Mines engineering school. The lecturer-participants (10 men and 5 women) all taught and did research at an engineering school in materials science. They were already familiar with using English for research publication and for presenting their research at conferences. They attended the EMI workshop to prepare themselves for teaching in English for academic purposes. They were not obliged to teach in English and they signed up for the EMI workshop on a voluntary basis. The participants were informed that the programme was part of a research project. They were given a research information sheet and consent form. Those who did not wish to participate in the programme were not included in the results but were welcome to attend the course. Pseudonyms were attributed to those who consented to being part of the research programme1. The participants were free to share some, none, or all of the data that was created during the workshop.
13The workshop was conducted from 9.00 to 12.00 for the duration of a week in July 2017. The daily sessions were followed by a group lunch so that the participants could practise talking about a variety of topics not necessarily related to their professional domain. In the afternoon the participants were given free time to return to their research activities and to prepare for the next day’s activities. This involved presenting images, objects and finally a short interactive lecture.
14As an introduction to the course the participants were asked to fill out a self-assessment form in which they situated their language competence on a CEFR scale of A1 to C2. They were then asked to tell the group who they were and explain why they wanted to teach in English. In pairs, they were given a list of terms and definitions which are currently used to in relation to EMI contexts (for example, Content and Language Integrated learning (CLIL), English for Specific Purposes (ESP) for its similarities and differences to EMI, and English as a Lingua Franca (ELF)). They were then asked which of the terms best suited the type of teaching they would be doing. Complementary to the introduction of key terms, the participants were invited to consider the term ‘bilingualism’ and whether they considered themselves to be bilingual. To help them to consider their identities in relation to the languages they spoke, they were invited to draw language portraits based on Busch’s method for collecting bilingual biographies.
15After the introductory stage, the participants were asked to simulate going for job interviews at a new university where they would be expected to teach their scientific discipline in English (speed dating format: one participant plays the role of the EMI lecturer-applicant, one plays the role of the Dean-interviewer). The objective was to reinforce the participants’ EMI lecturer identities because they had to present themselves as potentially competent and successful applicants.
Table 1. Five half-days EMI lecturer-training programme
|
9.00-10.00
|
10.00-11.00
|
11.00-12.00
|
12.00-14.00
|
Monday
|
Introductions. Presenting ourselves. Previous experience of using English professionally. Future needs.
|
Self-assessment of English competence and peer discussion.
|
Introduction to Key terms (EMI, CLIL, ESP, ELF) and discussion.
|
LUNCH in English
|
Tuesday
|
Written definitions of the term ‘Bilingualism’, discussion.
|
Individual Language. Portraits.
|
Analysis of language portraits and ‘re’ defining ‘bilingualism’.
|
LUNCH in English
|
Wednesday
|
3-minute individual presentations of a picture, graph or diagram.
|
Speed interviews.
Applying to be an EMI lecturer in a new university.
|
LUNCH in English
|
Thursday
|
10-minute individual presentations of an object to be used by your students (questions and interruptions from the ‘students’).
|
LUNCH in English
|
Friday
|
15-minute mini interactive lessons (questions from the ‘students’ and encouraging student interaction by including questions in the presentation).
|
LUNCH in English
|
16Most of the workshop consisted of opportunities for the participants to practise ‘being’ EMI lecturers. Starting with a three-minute presentation of an image, then a ten-minute presentation of an object (which could be potentially used by their students), to a final 15-20-minute interactive ‘mini’ lesson. The participants were encouraged to keep to time limits and to avoid reading from notes. This would help to ensure that the lecturers were aware of how long they were taking in their activities, ensure the smooth running of the institutional schedule and give them confidence and more room to manoeuvre away from notes. Presentations were expected to involve the other participants in some way, either by asking for oral responses or via the manipulation of physical objects, for example. The issues relating to interactivity in the mock lessons will be discussed further on. The participants were asked to explain what type of student they would be addressing (year and discipline, for example).
17Overall, the final lesson/lecture consisted in presenting the first fifteen minutes of a new course which included:
- the lecturer presenting him/herself as an expert of their field;
- the justification for the course being taught in English;
- the lecturer presenting him/herself as a speaker of English for international academic purposes;
- the presentation of the course objectives;
- the presentation of a course outline and evaluation procedures;
- an invitation for further questions from students.
- 2 Please contact author for original questionnaire as well as for details of the raw data.
18The workshop activities were designed to foster new declarations of language identity for EMI purposes. This meant that the participants were invited to tell the group how they defined their L2 competence, how they defined their own experience in relation to their bilingual identity and how they would present themselves to their students as EMI teachers. The ethnographic tools (Hammersley and Atkinson) used to collect the data were a) self-assessment English language competence forms, b) written and visual documents created by the participants, c) ethnographic notes taken by the researcher during the workshop and d) anonymous end-of-workshop evaluation forms (written feedback). The feedback questions were devised to assess whether the EMI course had prepared the participants for future or further teaching in English. The participants were asked what they had expected from the course and whether the EMI workshop had met with those expectations. Other technical and practical issues relating to the EMI course addressed whether the participants had appreciated the half-day format of the workshop, including lunch whilst speaking in English. They were asked to state which features of the course they had preferred or disliked and whether they had suggestions which could improve future EMI workshops2.
19The data collected explored the research questions and key themes relating to the field of EMI lecturer identity. The study could therefore provide information relating to the following themes:
1. Self-assessment evaluations of competence according to the Common European Framework of Reference (CEFR) scale ;
2. Written definitions of ‘bilingualism’ in relation to one's own language competence in English ;
3. Visual language portraits to discuss the language biographies of the participants (based on Busch’s methodology of using body maps to talk about multi-lingual identity) ;
4. Guidelines for further EMI training (based on the feedback received by the EMI participants).
20Data sets 1 to 3 were also parts of the course programme, with the objective of enabling open discussions about language, competence and identity.
21This section presents the data arising from some of the workshop activities.
22At the start of the workshop the participants were asked to self-assess their competence at English by filling in a CEFR document. This data set was used to answer the first research question relating to how EMI lecturer-participants defined their level of English language competence. The participants’ self-assessments of competence highlight how competence is socially situated above all.
23The CEFR document groups language competence into three main areas: understanding, speaking, and writing. This self-assessment form is used in other professional development courses (Test of Oral English Proficiency for Academic Staff (TOEPAS), CIP, Copenhagen University), where experienced users of English are deemed to have sufficient self-awareness about their strengths and weaknesses regarding their own competence as L2 English users. Nevertheless, there may be a difference between avowed language competence, (how one may label one’s own language competence), and language competence which is attributed by a third party (by an external examiner or the language trainer, for example). Professional development managers are keen to position participants into group levels to ensure similar ability in the EMI lecturer training courses they propose, specifying a minimum CEFR score when advertising their courses. Language trainers, however, may find that competence levels are not easily identifiable on an A1 to C2 scale. Other factors such as individual personalities, sociability, or confidence are not taken into account on the CEFR scale, nor indeed ‘willingness to speak’ (Yashima et al.). All these social factors nevertheless affect communicative competence.
24The participants of this present study identified themselves as being broadly B1 in the categories of writing and listening. The top C2 scores were self-awarded to self-competence at reading in English. This result was not surprising considering the extensive amount of reading in English the academic participants do for their research, placing their reading competence at ‘native-like’ and ‘expert’ in their specialist fields. In contrast to relative ease in reading, the second lowest score (B1 level) was attributed to ‘writing’. This result may appear surprising considering that the academics of the study largely publish in English. Nevertheless, this concurs with studies which show that writing research papers in English is more time-consuming and requires extra effort (Reynolds, Ferguson et al.). The self-assessment of written competence in English could be a reflection of affective work ‘load’ rather than actual competence (Reynolds). In addition, studies have shown that there is a discrepancy between what journal editors claim and what L2 English authors believe. Journal editors claim to welcome L2 authors to widen the scope of their journal’s research, whereas L2 authors believe that their articles are turned down ‘because of their English’ (Flowerdew). This impression of inferiority and rejection (as an identity trait of an L2 speaker) perhaps explains why a self-evaluation of written competence was estimated as average (B1). Having published over 10 research articles in English is surely proof of quality in itself, regardless of CEFR competence scales.
25Oral production (self-awarded an average of C1 by the participants) was perceived to be better than oral interaction (self-awarded an average of B2), which confirms results about difficulties with understanding questions at conferences, for example (Reynolds). Participant Antoine differed in this because he spoke four languages already and as a multilingual speaker, he felt that ‘understanding’ was one of his highest skills, reinforcing the argument that the more languages people speak, the more they are likely to feel they ‘understand’ others. We can enlarge the scope of ‘understanding’ to beyond language competence alone, to include social skills such as empathy, anticipation, and compassion which are not included in the current CEFR scale.
26For the purposes of the EMI lecturer development programme, the self-assessment of competence was an interesting point of discussion within the peer group. The participants were invited to study their peers' self-assessment forms as the forms were not only for the benefit of the trainer. Overall, the trainer, and the other lecturers would evaluate their peers’ language competence as being higher than the participants’ self-awarded score (for example: “Oh why did you put B1 for spoken? You are much better than that!”). However, comments could only be made regarding the general impressions they had gleaned of each other’s levels of spoken English. We may also have to consider social hedging and politeness when referring to our colleagues’ levels of English (i.e. it would be unlikely for a colleague to openly state that they believed their colleague’s English to be poorer than they had self-claimed). Some were unsure about where to place themselves on the scale and even looked over to their neighbour’s responses for inspiration; others needed more time and read the descriptors in detail before making a decision.
27The issue with asking future EMI lecturers to place their English competence according to a CEFR scale is that it situates them as L2 speakers of English rather than as experienced users of English in the workplace. More recent definitions of bilingualism would even favour the term bilingual to describe the language profiles of the participants of this study (Grosjean; García).
28The subsequent workshop activities were therefore aimed to answer the second research question: do EMI lecturers consider themselves to be bilingual speakers of French and English?
29The answer to this question was ‘no’ as the participants favoured definitions of bilingualism which focused on competence rather than usage. For example, when they were asked to define ‘bilingualism’, they referred to fluency, perfection and confidence:
Being bilingual is using, speaking and understanding languages perfectly, what is said explicitly and implicitly, and all kinds of specific, idiomatic expression. (EMI lecturer, François)
A bilingual is a person fluent in two or more languages. (EMI lecturer, Jean)
A bilingual is a person who is confident in listening, speaking and writing two or more languages. (EMI lecturer, Marie)
30We note, however, that the participants did not refer to ‘pronunciation’ or ‘accent’ in their definitions of bilingualism. This may be explained by the professional experience of the group who were attuned to different varieties of English in a global research setting. In answer to the more personal and nuanced question of ‘to what extent are you bilingual?’ none of the participants claimed that they were to no extent bilingual:
I am bilingual in the field of materials science, and not at all for everyday life. (EMI lecturer, François)
I do not consider myself as being bilingual. I understand way better than I speak, so it’s a bit unbalanced. I’d say I am 70% bilingual. (EMI lecturer, Pascal)
31Even if their own form of ‘specialist area’ bilingualism did not concur with their definition of bilingualism in general, this result shows a hesitant step towards claiming bilingual identity within their specialist disciplines.
32A further activity to encourage the EMI participants to situate themselves as multilingual speakers was the visual representation of language biographies. Visual methods are used in interviews and classroom interactions as points of reference to aid further talk. The drawings produced are not taken as fixed representations of identities but as representations of what the participants wished to talk about at that particular moment in time (Wheeldon and Faubert). As an adaptation of Busch’s method of using a picture of the body to represent the self, the participants of this present study were asked to visually portray, using colours or symbols, all the languages that they spoke, had spoken, would like to speak, or even understood, liked or just read. The participants then used their maps to describe their language biographies to the rest of the group and were keen to know more about their colleagues’ language biographies. As in past experiments where visual biographies have been used (such as Reynolds; Busch), the results show a rich and varied picture of language identities.
33Figure 1. Two examples of language portraits based on Busch as a tool for the EMI lecturer-participants to talk about their language biographies
|
|
Participant Jean
|
Participant Thomas
|
34Some participants chose to position the languages they knew to parts of their body (participant Thomas), whereas others stated that there was no relation between languages and their position on the body (participant Jean):
The different flags on my body mean the formal identity for the specific languages I speak, and cultural aspects. There is no relationship between the part of the body and the flags. (EMI lecturer, Jean)
- 3 In Arab culture the right hand is used for clean acts, and the left hand is used for ‘unclean’ acts (...)
Feet: languages of travel.
Right hand: Arabic used to survive in [name deleted] for two months. The right hand must say clean3.
Left hand: Italian for reading theatre texts.
Stomach: Czech for surviving in land areas where no one speaks English.
Heart: Multicultural family and heritage (food, drink, music, clothes, way of life)
Head: for work and interacting. I need to think in German and French before speaking in English! Sometimes I speak directly in English, lucky!
(EMI lecturer, Thomas)
35The experiment carried out during this present study invited the participants to take onboard their language biographies as a whole, including the notion of timescale (Lemke). The concept of language biographies, or histories, draws from Kehrwald who uses visual concept-mapping as an aid to encourage participants to talk about their present and past L2 learning identities. Language identity could therefore incorporate past or lost languages that nevertheless still make up a person’s language identity (Broudic). Participants could also represent languages that they would like to learn in the future, in keeping with Dörnyei and Ushioda’s concepts of ‘possible and desired selves’ as a way to encourage L2 learners’ motivation.
36The results of the language portraits were clearly at odds with how the participants had initially described their language identities (in relation to how they had positioned themselves as potential ‘bilinguals’). All of the language portraits featured at least three languages as integral to the participants’ identities. The multitude of colours used (one colour for one language) revealed a rich tapestry of multilingualism. This showed that when the question of multilingual or bilingual identity was posed outside a framework of competence (or in comparison with ‘native speakers’), then the participants did accept to claim various languages as being integral to who they were. The body portraits could also include languages which they had used or had had contact with in the past and which had left a ‘mark’ on their identities. This included languages spoken by loved ones and which were not necessarily spoken by the participants themselves.
37Below is an example of a body language portrait drawn by participant Juliette followed by her explanation of her drawing.
Figure 2. Detail of language portrait drawn by EMI lecturer-participant, Juliette
French is definitely my brain language. I've chosen the yellow colour for this, the colour of sun, the symbol of happiness, because it's easy for me to think in French, French is my native language. I always think in French, but I put some purple peas (because there is not a lot of English in my brain!) for English because sometimes I think in English, but it’s not natural for me!
My heart is in yellow and purple, because I like French, but I want to improve my English too. A little part of my brain is pink, for Spanish. I learnt Spanish 20 years ago, so I have sometimes some reminiscences! (EMI lecturer, Juliette)
38The objective of the activity was to draw attention to the participants’ already rich language backgrounds. The activity also encouraged them to represent language(s) as varying, unfixed elements of an evolving identity.
39In the introductory stages described above, time had been taken to consider language competence, the theoretical field and implications of EMI, and positioning in relation to multi-language use. Once these parameters had been set, the participants could then practise ‘being’ EMI lecturers.
40The fourth research question addressed whether EMI lecturers reassessed their pedagogical approach when they taught their discipline in English (rather than French). The following ethnographic observations during the workshop showed that the teachers had a preferred teaching style. The teaching style preferences were made apparent during the workshop activities described below.
41The first trials at practising teaching consisted of presenting an image which the lecturers would be likely to show their students during a class. During this activity, the lecturer‑participants were asked to involve the students in some way, by asking them questions for example. Overall, the lecturer‑participants stuck to a traditional lecture type presentation. When they did ask the ‘students’ questions, they would often wait a few seconds before answering the question themselves, showing a good command of the use of rhetorical questions. The lecturers were at ease with presenting a picture and then subsequently an object. After these initial practice stages, the trainer felt that the participants should be helped to deal with unscripted situations where students would interrupt and even disrupt the flow of their lecturers’ speech. The aim was to give the lecturer-participants further tools to deal with surprises, questions and interruptions, which is exactly the area they felt most uncomfortable with (Reynolds).
- 4 Please contact author for examples of such interruptions.
42Current trends in university teaching encourage ‘more active student participation and more interaction in the university classroom than is usual in the traditional frontal lecture (Gustafsson 3, Knapp). In this context, interactivity can be understood as a function of positioning within a community of practice (Wenger). Within this community of practice, we can refer to the identities of students who are learning how to belong to a community of scientists, for example. Interactions in the higher education classroom may include interruptions, questions and negotiations with a dominant speaker, here the ‘teacher’ (Bamberg). Within such a context where specialist knowledge is competed for and may be contested, the participants were told that they would be interrupted during their subsequent round of ‘mini’ lessons. The other lecturer-participants, who were playing the role of the ‘students’, were given cue cards which contained interruptions. The interruptions could be related to the content of their teacher’s lesson or could address other physical needs which the students may have4.
43It was not assumed that there was a right or wrong way to deal with such interruptions. The lecturer‑participants were in a position where they had to decide how to deal with the interruption and make their response or intention clear to the class. The lecturers were also expected to repeat or explain something they had just said. They were asked to deviate from the purpose of their present lesson, by being asked about the exam for example. They responded to such interruptions calmly, by either addressing the questions in detail, but by also proposing to answer a question later if required. The lecturer-participants also had to deal with the physical issues which may arise in a classroom (such as a student requesting to leave the room). All the lecturers were able to react quickly to any physical needs of the students and answered students' questions clearly and patiently before returning to the subject of their presentation. As experienced lecturers, the participants had dealt with critical incidents in the past (Gustafsson 9), but had perhaps not practised dealing with such interruptions in English.
44Those participants who were playing the role of being the students were also given permission to interrupt the ‘teacher’ at any time which may not have been part of their own politeness code with respect to asking questions, or being asked questions. For example, during our observation of classroom interactions, it was noted that some students preferred to ask the teacher questions at the end of the class, sometimes even queuing up at the end, rather than interrupting the lecture. The students were also shy about asking questions in front of their peers. The interruption exercise gave the participants the opportunity of playing both student and teacher roles and was an opportunity to share their combined experiences of classroom interruptions. Faced with some of the chaotic interruptions (such as a student being stung by a wasp) the lecturers focused less on language and more on all of the parameters of a social setting.
45In terms of revisiting pedagogy when there is a shift in language, cultural aspects need to be taken into account (Derivry). When teaching in English in France, a lecturer may need to give guidelines concerning question time. For example, a lecturer may tell her/his students to interrupt her/him at any time, or only when she/he asks for questions. To this end, the lecturer-participants were subsequently told to tell their students if they could interrupt them at any time or if they should wait until the end of the class. If the lecturer announced that she/he would take interruptions at any time, it was advised that she/he should then appear to welcome such interruptions when they came. Such messages are essential when classes contain mixed populations of students who have experienced different educational backgrounds (Derivry).
46Practising giving ‘mini’ lessons in English in front of their peers gave the participants the opportunity to reassess their teaching styles and experiment with deviations from their preferred practices.
47The post workshop questionnaire (see Appendix 2) aimed to obtain feedback from the lecturers who had attended the EMI teacher-training workshop. The subsequent analysis of the feedback provided information relating to the fifth, and final, research question concerning how future EMI courses could best be developed and improved to adapt to EMI lecturers' needs.
48The results concerning the participants’ feedback to the EMI-training were as follows.
49The participants described that subsequent to the EMI teacher-training course they were more accepting of their own limitations and were more prepared to work with the English they had. Most (86%) of the participants felt that the EMI teacher-training course had helped them to prepare for further teaching in English. The two participants who did not feel that the workshop had helped them explained that they were not comfortable about their own competence in English when they compared themselves to their workshop peers. These two participants would have preferred to have been given ‘set phrases’ to help them with teaching in English (‘Phrase-banks’ such as the Academic Formula List created by Simpson-Vlach and Ellis may be useful in this case). One participant did not like the ‘workshop’ style of the course and would have preferred to have received a formal lecture presentation from the trainer and to have been provided with vocabulary lists.
50The participants appreciated receiving individual feedback from the trainer after their presentations. They learnt how to trigger interactions by including questions throughout their presentations and prompt students for interaction with themselves, by openly stating that they would welcome such interruptions. They enjoyed being with and learning from their peers.
51The participants all agreed that the location of the workshop was very important to the well‑being of the group. Most of the workshop was held in a large room overlooking a garden. One session was held in a small overheated room which impacted negatively on the group. Concerning the communal lunch, they appreciated having the choice about whether to attend the lunch in English or not. They appreciated the half-day format of the course so that they could attend to other business and prepare for the following day during the afternoons.
52According to the questionnaire responses, 80 % of the participants had already experimented with teaching in English prior to the workshop (either by having taught a few hours in English or through PhD student supervision). One may consider, therefore, why they felt they needed further training. This can be explained by Soren’s research, which showed that EMI lecturers were sometimes thrown into EMI teaching, with the famous example of lecturers receiving emails from the Dean informing them that they would be teaching their course in English come September. The participants of this study, (and possibly many other acting EMI lecturers), may have been hesitant to claim EMI identity. They therefore welcomed EMI training of this kind, even if it had perhaps come a little late. These highly motivated lecturers welcomed teacher-training as part of their professional development and well-being within an increasingly ‘international’ workplace.
53Finally, the participants suggested the following improvements to the workshop: a final round-table feedback discussion and a lower number of participants were recommended. They felt the number of participants (15) was too high.
54Firstly, concerning professional development aims, one of the objectives was that the participants would be made aware of the debate concerning the types of ‘acceptable’ English which they and other users of English are entitled to use in academia (Jenkins). The inclusion of these theoretical considerations in the workshop impacted on the participants. In the context of this EMI training session, these were also the objectives of the course which may well differ in EMI teacher-training courses which would encourage participants to align themselves with idealised ‘native speaker’ role models, for example. The following feedback from an EMI workshop participant shows that these objectives were met:
To become aware that EMI and ELF are concepts that have been theorized has enabled me to realize that I don’t have to be fluent at English to give a scientific lesson in English. (EMI lecturer, Vanessa, in answer to the question: “Did this workshop help you to be better prepared to teach in English?”)
55Secondly, the study revealed that the participants made clear distinctions between areas of their language competence: deeming their reading competence to be very high (C2) and writing and listening to be lower (B1). Thirdly, the participants did not consider themselves to be bilingual speakers of French and English because of their own definitions of bilingualism. Their definitions were idealised, unattainable, and modelled on perceptions of perfection. They did nevertheless define themselves as having multilingual identities. This is because their multilingual language portraits showed biographies of language use rather than language competence.
56The research questions regarding a reassessment of pedagogy within an EMI context showed that teaching, and especially lecturing, styles were to some extent ingrained. This can be explained by the ‘propensity for a theoretical rather than an applied or practical approach to learning and study [in France]’ (Taillefer 15). Nevertheless, the participants appeared to enjoy experimenting with different approaches, especially when validated by both their peers and trainer. The study showed that participating in an EMI workshop made them feel more reassured about their own language competences and more likely to experiment with actively engaging in improvised communication with their ‘students’ in English.
57An emerging definition of EMI teacher identity is addressed below, as are the recommendations for future EMI teacher training schemes. The recommendations arise from the results regarding the participants’ needs.
58A first step towards being an EMI teacher is training to become an EMI teacher. As cited in the literature review, the development of one’s awareness as a teacher is integral to being a teacher (Johnson et al.). It is partly through further training that lecturers can re-assess this awareness of teacher-self. In short, giving training opportunities to lecturers of this kind is identity making. How then can EMI teacher identity be defined subsequent to this study describing a teacher-training event? Firstly, EMI teacher-identity is central to teacher development because a re-assessment of pedagogical practices seems to occur when there is a shift in language use. Secondly, this shift in teaching practice remains innovative and relatively rare in France. The proportion of study programmes provided in English in French higher education was estimated at 3.4% in 2014 (Wächter and Maiworm). However, Campus France estimates that English taught programmes have increased by 464% since 2004 (Campus France 56).
59EMI teachers are aware of the challenges associated with teaching in a second language. Nevertheless, they retain expert authority in their fields and negotiate languages with their students because they do not present themselves as ‘teachers of English’ but as (Science) teachers who teach in English (Airey). This double positioning was highlighted by the participants of the present study. Although the EMI lecturer-participants of this present study and other studies (Reynolds, Soren) do not claim to be experts of the English language as a whole, they will nevertheless claim and defend expert knowledge of the language they will use when speaking or writing about their specialist fields. Finally, this distinction between specialist and general language use also explains why the participants acknowledged bilingual competence and practices in their specialist areas, without necessarily acknowledging being bilingual speakers of French and English.
60This present study stems from an example of a French higher education setting where university administrators are committed to furthering teacher-development programmes to improve their academic staff’s English language skills. Firstly, this study invites further EMI training to be made available to both practising and future EMI lecturers. Professional development schemes of this kind are something they deserve and are entitled to. Providing EMI training also signals that the institution is considering and acknowledging the specificities and difficulties associated with teaching in another language.
61This study recommends that future EMI trainers define the parameters of the course with the educational developers that manage EMI lecturer trainers and lecturers. The participants of the present study were not assessed for their level of English competence prior to the course. However, two participants, who claimed that their English competence was not as high as that of their peers, felt less confident after the workshop. This could appear as a lack of institutional preparation, but also reflects the experimental hitches relating to this first EMI training endeavour.
62On the whole, the feedback revealed that the participants did not have any clear expectations concerning the EMI course, generally stating that they hoped it would ‘help them to teach in English’ or that ‘they didn’t have any expectations’. The novelty of the EMI course may explain why the participants did not have specific expectations. Future EMI training programmes may benefit, therefore, by asking lecturers to complete a pre-workshop questionnaire which would pre-assess their expectations. However, we may hope that after attending this first EMI workshop at Ecole des Mines, the participants and administrator will have a clearer idea about what their needs may be in the future.
63Once the course content has been agreed on, the participants should be invited to define what EMI teaching means for them and to present themselves as EMI lecturers. An EMI trainer would benefit from an understanding of how English is used for academic contexts and be aware of varieties of English used in the international workplace. Once teacher identity has been addressed, future EMI lecturer trainers should create room for as much teacher practice as possible. In this way the lecturers will address teaching styles and will hopefully gain greater empathy with student experience because they will have experimented with ‘playing the role of the student’ during the EMI workshop. In the past, Danish and French English-speaking academics have described being more comfortable with talking about their scientific discipline (Soren; Reynolds) than when conversing over dinner for example. For this reason, EMI workshops could therefore combine formal and informal language by including a shared lunch or tea breaks where the participants will have to practise talking about topics outside of their specialist areas. Finally, depending on need and institutional policy regarding teacher certification, EMI lecturer certification such as the TOEPAS could be used as a benchmark to work towards, for example.
64The present study reported on the data created during a teacher-training course held at an Engineering school in France. The training involved practising teaching in English for academic purposes where English is not the local language. Considering that English is being used as a lingua franca for international educational purposes, the training also addressed how EMI impacts on teacher-identity. The results show that the participants considered themselves as being intermediate to advanced users of English depending on the activity they were taking part in. They did not consider themselves bilingual despite their bilingual language practices. This belief contradicted the participants’ rich and varied multilingual language biographies (as depicted in their language portraits), showing that language identity shifts according to context and purpose.
65Concerning pedagogical issues which arose from the study, the participants tended to favour a frontal lecture approach. Nevertheless they adapted swiftly and easily to other teaching styles including increased interruptions and questions from their students. They revealed their experience as lecturers by managing even ‘chaotic’ classroom situations. They enjoyed playing the role of both the teacher and the student in a safe peer-to-peer environment. The feedback from the participants was generally positive, even if they would have preferred to have worked in a smaller group.
66The study invites further studies in the field of EMI to incorporate frameworks for EMI lecturer-identity based on the specialised, and language-aware contexts within which they are evolving.
We would like to thank the editors of the journals and of this volume, Linda Terrier, Danielle Joulia and Claire Chaplier for their careful editing and reading. Thank you to the anonymous referees for their helpful comments in preparing this text for publication. Finally, thank you to Stéphanie Roussel for her suggestions and support.