EMI Lecturer Trainers: Reflections on the Implementation of EMI Lecturer Training Course

Universities in non-native English-speaking countries strive to increase the number of international students using EMI as one of their internationalisation strategies, “being caught up in the rush to offer both undergraduate and postgraduate programmes through the medium of English” (Macaro et al. 2018, 37). This has resulted in an unprecedented demand for EMI lecturers to acquire both improved English language proficiency and new teaching methodology skills (Gay et al. 2020). The research, however, evidences the lack of published materials for the development of EMI lecturer training courses (Costa 2015; Dafouz 2018; Gay et al. 2020). Moreover, the literature on hands-on practice with respect to the experience of those involved in EMI lecturer training is scant. Thus, here, the authors offer an outline of an EMI lecturer training course and provide an analysis of the experience of eight EMI lecturer trainers from three leading Russian universities that conducted EMI training courses in the academic years 2017-2019. Semi-structured interviews and preand post-course surveys show that EMI trainers encounter difficulties which are psychological, methodological and linguistic in nature, these being low English language proficiency of EMI lecturers, their insufficient knowledge of pedagogical strategies, their lack of self-reflection and feedback and their resistance to active learning techniques and a student-centered approach. The findings of this study will be useful for EMI lecturer trainers since the study highlights potential challenges and practical advice on how to increase training efficacy.


Introduction
Globalization has had a tremendous impact on tertiary education all over the world, prompting many universities in non-native English-speaking countries to develop strategies aimed at raising the quality of the education they offer and increasing their market presence in the international education arena, with English Medium Instruction (EMI) becoming the driving force of internationalization and a prevalent phenomenon in tertiary education (Doiz et al. 2012;Macaro et al. 2018). Galloway et al. (2017, 8) summarise the reasons for designing and implementing EMI programmes, which have been widely studied in the literature (e.g., Costa 2015, Doiz et al. 2012;Macaro et al. 2018;Ruiz de Zarobe & Lyster, 2018). They include "gaining access to cutting-edge knowledge and increasing global competitiveness to raise the international profile; increasing income; enhancing student and lecturer mobility; enhancing the employability of graduates' international competencies; improving English proficiency; reflecting developments in English language teaching (ELT); using English as a neutral language; offering EMI for altruistic motives" (Galloway et al. 2017, 4).
English Medium Instruction (EMI) programmes are increasing all over the world (Wächter and Maiworm 2014;Macaro et al. 2019a;Farell 2020). There was a sharp rise in the number of EMI programmes in Russian Higher Education Institutions (HEIs) from 122 in 2015 to 3160 in 2019. The Russian Ministry of Education supports innovative development in tertiary education through the 5-100 initiative, launched in 2012, which aims to raise the research and academic profiles of the leading Russian universities and to increase student and staff academic mobility. By 2020, 21 universities had joined, each having developed new strategic plans. One of the criteria for inclusion in the initiative is the number of EMI programmes the university can offer to international students; therefore, the departments have to internationalise their education programmes and the academic staff have to acquire new skills to be able to provide quality EMI programmes. Another high-priority project of the Russian Ministry of Education "Developing the export strategy for the Russian education services" states that by the year 2025 exports of Russian education services should increase, with the number of international students rising from 220,000 in 2017 to 710,000 in 2025 (Ministry, 2020). Thus, for Russian universities EMI is a top-down initiative, deeply rooted in national and supranational policies for the internationalization of tertiary education.
University administration often delegates the responsibility to develop EMI teacher training courses to university Language Centers, where mostly English as a Second Language (ESL) teachers work, or Chairs of Foreign Languages (Fortanet-Gómez 2020). However, both language specialists and ESL teachers are not usually aware of what EMI is as this is not part of the professional training they received at university. Language specialists and ESL teachers themselves need EMI training to be able to teach others and a well-designed staff training is crucial to the success of an EMI training course (Wilkinson 2012;Jiménez-Muñoz 2020). Although, pedagogical guidelines for effective EMI teaching and learning do exist (e.g., Educational Quality at Universities for inclusive international Programmes (EQUiiP); Transnational Alignment of English Competences (TAEC) Project EMI Handbook), there is little or no EMI content in initial teacher education programmes, which is concerning given the rapidly growing number of EMI programmes worldwide (Galloway et al. 2017;Jiménez-Muñoz 2020, 124). ESL teachers should know particular EMI skills to teach international students, be competent EMI researchers to create the courses that are relevant to a particular EMI context and satisfy the needs of EMI content teachers. The literature on EMI is vast and "while the theoretical base for the development of EMI training programs is growing, the body of available research regarding the implementation of planned EMI initiatives is limited" (Ploettner 2019, 264). There is also only limited literature (Dafouz & Smit 2020, Gay et al. 2020) that outlines the competencies needed and the challenges the ESL teachers meet when they start their EMI training career.
This case study employs observation and semi-structured interviews as well as pre-course and after-course surveys, to analyse the experience of EMI trainers from three Russian universities delivering EMI training courses for in-service academic staff. The data sheds light on how EMI teacher trainers reflect on the challenges they meet while implementing EMI training to the academic staff of their universities. The paper contributes to an empirically-based understanding of how EMI teacher training courses can be designed and implemented and what challenges EMI teacher trainers should be ready to cope with.
The research seeks to answer the following questions: 1. What are the perceived shortcomings and prevailing challenges for EMI teacher trainers in designing and implementing an EMI training course? 2. What are the potential professional changes that EMI trainees are resistant to?

EMI Staff Selection and Preparation
With the increase in EMI education programmes, one of the many challenges that university administration experience is selecting staff to deliver such courses. There are several approaches to choosing EMI lecturers. One is to assign those lecturers who have a high level of English proficiency (Werther et al. 2014), albeit that what the acceptable level of English for EMI content teachers is remains a topic of debate. Macaro et al. (2018) consider there is no definitive universal benchmark for the level a teacher needs to be able to teach in English at national or international levels. In Spain in 2015, 43% of universities required their teachers to demonstrate a B2 level of English within the CEFR, 44% required C1, and 13% required C2 (Costa 2015). For Nordic universities, the only acceptable level is C1, while in Russia, the reality is that language proficiency for EMI content teachers ranges from B1 to C1. Moreover, the meaning of "ability to teach" can also mean different things to different people as the term "qualified" has been left undefined in EMI literature. "Qualified" might mean that an EMI teacher has a high level of language proficiency or that he has a certificate of an EMI course completed (Dafouz 2018). This approach works for the universities where a high level of English proficiency is an evaluation criterion for academic staff assessment scheme or tenure track. Such language requirements can be included in the language policy of a university. This strategy works well for the leading Russian universities in Moscow and Saint-Petersburg, which have a high concentration of leading researchers who collaborate with international scientists and develop networks of international academic contacts. The second approach is organizing EMI certification of the academic staff without providing EMI training courses. Several European universities have certification procedures: Test of Oral English Proficiency for Academic Staff (Copenhagen University -Center for Internationalization and Parallel Language Use) (Center 2020); Test of Performance for Teaching at University Level Through the Medium of Instruction (University of the Basque Country). However, Klaassen and Räsänen (2006, 249) advocate the creation of a universal assessment tool that can be disseminated both in Europe and worldwide. Moreover, a large number of universities after the certification is made will see a high demand for developing EMI teacher training initiatives to prepare lecturers to work efficiently in international classrooms. In Russia few universities have certification of EMI skills and knowledge and the assessment criteria to evaluate the readiness to become an EMI lecturer are mostly focused on language proficiency.
The third approach is to design and implement EMI training courses for EMI staff which are an integral part of CPD. The demand for EMI teacher training is huge universally and is well-documented in the literature (Gustafsson 2018;Trent 2017;Werther et al. 2014;Lasagabaster & Douz 2018) with both academic staff and students on EMI courses complaining that the EMI staff lack expertise in the areas of English language proficiency, including English for Specific Purposes (ESP) and English for Academic Purposes (EAP), pedagogical skills, EMI-specific micro-skills and intercultural awareness. These knowledge and skills gap lead to the reduced self-assurance of EMI lecturers when delivering their courses and reduced quality of teaching. As the demand for EMI courses exists in universities, the third approach is widely adopted by many European and Eastern universities, and it may take a variety of forms, as will be shown later in the paper. For regional Russian universities the third approach prevails as there is a lack of EMI staff with high language proficiency. It usually takes two or three years of general and academic English training courses before a lecturer is ready to take an EMI training course, for which B1 is the minimum language requirement.

Design of EMI Training Courses
The practice of EMI teacher training is evolving, however, few case studies are described in the literature. One of the reasons, as Galloway et al. (2017) explain, is that EMI teacher training courses are heavily context-dependent and there is no one-size-fits-all approach. Nevertheless, there are two recent projects-EQUiiP and TAEC-that offer general recommendations on what to include into an EMI training course. EQUiiP project (2016-2019) is an Erasmus+ partnership consisting of seven European Universities. The project partners are Aarhus University, Denmark; the University of Bordeaux, France; Universidad Complutense de Madrid, Spain; the University of Copenhagen, Denmark; the University of Freiburg, Germany; Glasgow Caledonian University, United Kingdom; the University of Groningen, The Netherlands. The project has produced a programme providing support for educational developers and lecturers aiming to ensure quality in international and intercultural classrooms in higher education. The TAEC project is a collaboration between the following partners: University of Copenhagen, Denmark; Maastricht University, the Netherlands; Universidad de Lleida, Spain; Universita Degli Studi di Torino, Italy; Faculty of Humanities and Social Sciences in Rijeka, Croatia. The purpose of the project is to develop a common framework for EMI quality assurance and support, which will help the partners adapt local EMI training and certification language assessment instruments for transnational uses. The projects' results may serve as guidelines for an EMI trainer, but they do not give a detailed structured description of classroom activities that can be used by an ESL teacher making their first steps in an EMI course design. It is still a challenge to decide on the content of the course, the skills to be developed for EMI trainees, and the learning outcomes to be achieved.
There is no general agreement on the content and length of EMI training courses. A study of the current practice at Spanish universities showed the length of EMI teacher training courses to be quite diverse: 25% report courses lasting between 1 and 15 hours, 36% between 15 and 30 hours, 25% between 30 and 60 hours, and 14% longer than 60 hours.  Table 1 that is based on the comprehensive research of Costa (2015) and the information from the websites of universities. Some universities focus their EMI training courses on methods of teaching and pedagogy, while some concentrate on English language training, and others consider EMI conceptualization and EMI strategies as the key skills that EMI academic staff should acquire. What most universities agree upon is that an EMI course should be a hands-on experience and provide practical teaching skills in an international classroom. As few universities in Russia offer EMI training courses for their staff, the only option for Russian EMI trainers is to study the best practices of European and Asian universities, where the EMI agenda has been developing since the 1990s.

EMI Lecturers' Resistance to Change
The third challenge that an EMI teacher trainer should be prepared for is that academic staff are adult learners with professional experience, particular values and deep-seated attitudes to teaching and learning. They should be taught using the principles of andragogy (Knowles 1980;Volchenkova 2015). They are adults with well-developed critical thinking skills and they need strong argument made by an EMI teacher trainer to change their current practice (Knowles 1980;Volchenkova et al 2017). This may be an issue, coupled with a lack of motivation to increase English-language proficiency that stems from the fact that improving language skills is time-consuming (Bradford 2016). Moreover, John Airey (2020, 343) notes "In my dealings with physics lecturers, I struggle to convince them that they should view themselves as teachers of disciplinary discourse. However, for content lecturers, content is king. EMI is simply seen as a pragmatic means to a content-related end. In such situations, it is not surprising that content lecturers have been reported as insisting that they do not teach language (Airey 2012)." Thus, EMI content lecturers may not feel responsible for adjusting their own language to their students' level of English or to helping students develop their linguistic proficiency in order to operate successfully in their academic subject. This misconception that may be held by content teachers should be challenged in EMI teacher training, as it is the task of an EMI teacher trainer to persuade the lecturers to take responsibility for the students' language ability, learning outcomes and social skills. Over the last 20 years, language proficiency has become the responsibility of education providers (Räsänen 2008) as higher education institutions need to prepare graduates for the global market, where employers value proficient first and second language use, disciplinary expertise and global-networking skills (Yang 2017). As such, "a paradigm shift is needed" and EMI content teachers should "become responsible for the development of students' language proficiency" (Jiménez-Muñoz 2020). The main pedagogical implication is that EMI trainers should teach language scaffolding as part of an EMI course.

The Study
The study seeks to identify the prevailing challenges for EMI trainers in designing and implementing an EMI training course in Russia. The EMI contexts of the three universities have some similarities but also some differences. On the one hand, the approach to EMI at SUSU, ChelSMA and ITMO has been a top-down one, so academic staff have to deliver courses in English at the demand of the university administration, i.e., the lecturers did not choose to teach EMI courses, and the main selection criteria for the staff was their proficiency in English. All three universities have an internal language assessment procedure that is aligned with the CEFR. The average level of English proficiency of EMI lecturers ranged from upper-intermediate On the other hand, each university's approach to EMI training course design and implementation was different. At SUSU, a research-driven approach was used that helped to shape the course and integrate it into a CPD programme called "Lingva" which aims to raise the English language proficiency of the those teaching STEM (science, technology, engineering and maths) subjects and social sciences. The results of the SUSU research were reported at two ICHLE (Integrating Content and Language in Higher Education) conferences (2017 and 2019), and used as the basis for the textbook "English for Researchers: English Medium Instruction" and a document of recommendations for the universities involved in teaching international students. The EMI training course at SUSU has been conducted since 2016, with all the trainers using the same EMI textbook (EMI textbook was developed at SUSU by Volchenkova and Bryan 2019) and having participated in a set of workshops on how to teach EMI. At ITMO the university's Foreign Language Training Center has been providing workshops and training courses on EMI since 2017. The EMI course is a stand-alone language course delivered for Physics and IT lecturers. The course has been shaped and developed on the basis of feedback from a series of EMI trainers' internships at American universities. ITMO trainees had the opportunity to take a three-week course on EMI methodology at Boston University after they had completed a basic EMI course at ITMO. ChelSMA does not have a language training center. The department of foreign languages was responsible for EMI training. In 2019 they brought in both the teaching materials and an EMI trainer to provide the EMI course for the academic staff.

Course participant profiles
The EMI trainees were 113 university instructors, associate professors, and professors of various fields of expertise from 3 leading universities in Russia: 79 content teachers from SUSU (69.9%) (the participants at SUSU taught a wide range of subjects from mathematics to philosophy); 18 content teachers (doctors) from ChelSMA (15.9%); 16 content teachers (IT specialists) from ITMO (14.2%). The number of EMI participants are listed in Table 2. The participants' teaching experience ranged from 5 years to 30 years, most had Doctoral degrees in their subject and when they started their courses, none had taught or studied abroad. At the beginning of the EMI course, the trainees could not name any effective EMI teaching and learning strategies. 65 respondents (57.5%) perceived the idea of introducing English taught courses in their departments as positive, seeing it as the main reason international students would be attracted to their university. However, 78 respondents (69%) perceived EMI as a simple direct transfer of their content knowledge from Russian into English, without taking into consideration the potential pitfalls of such a process, or the multicultural issues that may arise in the classroom. Safipour et al. (2017) showed that, although language proficiency is one of the main barriers for the integration of foreign and local students in EMI programmes, the main hindrance seems to be cultural differences. Thus, EMI trainees should pay a considerable amount of attention to the basic guidelines for teaching international students: raise awareness about the types of challenges foreign students face; provide examples of the kinds of issues that may affect students; offer possible suggestions.
Before the EMI course, each of the three universities had had an internal language assessment procedure by giving their trainees the Cambridge placement test that is aligned with (CEFR The intermediate level (B1) of English in speaking, writing, listening, and reading, with the receptive skills rated significantly higher than the productive skills. Even upper-intermediate users (B2) of English admitted they had difficulties of self-expression and stumble on both grammar and vocabulary issues every time they used English. Thus, it is essential to develop CPD schemes for content teachers to increase their level of proficiency in English with an EMI course being part of the scheme and not an isolated course. Our results match with the recent findings of Martinez and Fernandes (2020, 138) that offer to organize English without Borders support (this is a program which gives the opportunity to study the language free of charge, in online courses, for example) at universities for both content teachers and students for EMI not to turn into an elitist programme.

Trainers
Eight EMI trainers were from the three universities: SUSU (4), ChelSMA (1), ITMO (3). Their teaching experience ranged from 10 years to 30 years. As indicated in Table 3, half of the trainers had received their PhD degree in Linguistics and Pedagogy. All of the trainers fulfilled the TKT (Teaching Knowledge Test) qualification, two of them are certified CELTA (Cerificate in Teaching English to Speakers of Other Languages) teachers.

Instruments
Since the study was conducted with the participants of EMI teacher training courses, the background information on the trainees is given. This data is important because it shapes the EMI trainers' perception on organizing a successful learning process and producing the desired result. The data for this study was collected via a pre-course questionnaire (Appendix A) and Likert scale (Appendix B), and an after-course evaluation survey (Appendix D). The reflections of EMI teacher trainers on the training they provided were collected and analyzed after the courses were completed via semi-structured interviews (Appendix C). 8 EMI teacher-trainers from 3 Russian universities were interviewed.
The pre-course questionnaire and a Likert scale were distributed among the participants before the EMI teacher training course. The questionnaire consisted of 18 questions (Appendix A) related to teaching experience, experience in EMI, and their ideas about the influence of EMI on the teaching process. The Likert scale survey used 40 statements to ascertain participants' perceptions of the possible benefits and drawbacks of EMI, the context of EMI at their university, the motivation of the academic staff to develop EMI courses. The after-course survey asked participants to assess the course design and course content, to describe the skills they had acquired and give ideas on how to improve the course.
The pre-course questionnaire aimed to gain knowledge on the participants' background: 7 questions were on teaching experience, 5 questions on their experience (if any) and willingness to teach subjects in English and their attitudes to it, 2 questions were on the potential EMI training and 4 questions aimed to ascertain participants' views/ideas about the influence of EMI on the teaching process. The Likert scale survey was designed to explore initial EMI trainee's' motivation and cognitions of what EMI is.

Pre-course Survey
The majority of the EMI trainees expressed concern about using English as the only tool to deliver the content of their subjects and admitted that they felt they had deficient English language proficiency for EMI instruction. Only 25 (22%) out of 113 respondents considered their English language proficiency high enough to conduct lectures and seminars in English.
It must be noted that EMI trainees did not have a clear understanding and needs for EMI introduction, though 32 respondents (28.3%) strongly believed that it was the responsibility of the department to choose which courses were suitable for EMI instruction, and 81 respondents (71.6%) were uncertain about whose responsibility it should be. Similar findings in EMI participants' motivation can be found in the works of Fortanet-Gómez (2020) and Llinares & Mendikoetxea (2020).
The participants on the training courses had little or no experience of delivering EMI courses (only 16 respondents (14%) had had one or two years of experience teaching in English to foreign students being visiting professors at universities abroad). They noted that they needed to learn teaching strategies to increase their range of options for providing knowledge, and they needed to improve their language skills in both classroom and general English. The major findings of the pre-course survey/questionnaire (Appendix A) can be found in Table 5. The university's need to introduce EMI has become the motivation for staff to adjust their methods of teaching. Though few of the participants had EMI experience they realized how beneficial it could be for their career at their university as the Key Performance Indicator (KPI) they had to reach included publication activities, establishing international academic contacts and participating in international research teams, and teaching in EMI could substantially improve their standard of English and provide them with the necessary skills to improve their KPIs.

Post-course Survey
The post-course evaluation survey had 10 questions and was answered by the EMI trainees after the final examination. The participants were asked to evaluate the structure of the course and the usefulness of the topics studied, as well as to give recommendations on how the course could be improved. There were also questions that addressed participants' opinions about the difficulty of the course and the challenges they had encountered during the course.
The questionnaire results reveal overall satisfaction with the course structure. SUSU and ChelSMA trainers taught the material using the same EMI textbook (EMI textbook was developed at SUSU by Volchenkova and Bryan 2019). ITMO trainers designed their own EMI programme which is a stand-alone language course for IT specialists and physicists. The structure of the course is similar though. It includes such units as EMI pedagogy and methodology, EMI strategies, Micro-teaching, Intercultural communication, Academic English. The content of the EMI course was not generally perceived as difficult. For 62.5% of participants with a high level of English (B1, B2) it was easier to complete the various tasks and prepare for lectures and seminars that they have to give as part of the course, that is why they had more positive feedback than other participants.
In terms of satisfaction with the different topics covered on the course, 97 respondents (86.6%) said that the first topic on EMI conceptualization consisted of too many unnecessary descriptions and clarifications about what EMI means, and was considered redundant. 102 respondents (90%) pointed out that interactive learning techniques and practice-oriented sections were extremely useful, and 60 respondents (53%) appreciated the information on the polite feedback design for effective international communication. The detailed review of the usefulness of the different EMI units is shown in Table 6. Three main findings were identified from the post-course evaluation survey: the EMI trainees appreciated the usefulness of new pedagogical and EMI strategies; they acknowledged the double value of the course as the new knowledge can be used for their professional development; and they noted that developing EMI content for classes was time-consuming.
"I learned a lot of new tools and approaches to teaching during this course. I am going to apply these strategies working with Russian students." (participant 2) "I enjoyed the implementation of studying techniques during lessons. On the other hand, I'd prefer to have not as much time-consuming homework as we were given." (participant 4) As for recommendations on how to further improve the course, 72 respondents (63.7%) suggested paying more attention to specific aspects of English in teaching: e.g., common grammar and vocabulary for their specific discipline. 68 respondents (60%) suggested adding a lesson on How to make presentations.

EMI Lecturer Trainers: Reflections on the Implementation of EMI Lecturer Training…
Alicante Journal of English Studies, Issue 34, 2021, pages 185-219

Semi-structured Interviews with 8 EMI Teacher Trainers
As the focus of our study is the experience of EMI teacher trainers, 8 EMI trainers were asked to fill the semi-structured interview through online Google forms after the course had finished.The questionnaire comprised 8 questions: 3 related to the trainer's background, their experience in EMI teacher training and their university context; 2 questions were on the typical myths about EMI among the trainees and the challenges that EMI teacher trainers encountered while delivering an EMI course to content teachers. We also asked about the beliefs of content teachers that were most difficult to change and the methods that worked best with content teachers.

The myths of EMI trainees
The myths that most content teachers share on EMI implementation are about the concept of interactivity in EMI classes, professional expertise and intercultural awareness. One of the typical myths of content teachers is that organizing highly interactive classes is next to impossible. Moreover, many lecturers consider devising interactive tasks as a waste of time. Three of the trainers expressed this idea in their interviews. For example, one of the trainers admitted that: "Content teachers consider EMI lessons to be quite difficult to organize, and using interactive techniques every lesson is time-consuming and unproductive for many of them." (Trainer 5) This can be explained by the fact that in Russia class interaction does not usually occur during lectures, where students are passive listeners, especially in technical departments. Traditionally the interaction in the classroom is considered from the standpoint of communicative leadership of the teacher. This type of teacherstudent interaction came from Soviet times and is still deeply rooted in the consciousness of both teachers and students.
The new forms of cooperative learning are developing slowly in Russia, mainly due to teachers' reluctance to change their style of teaching. So many efforts were made by the trainers to actively engage the trainees in classroom interaction by asking questions, to activate schemata and check comprehension.
Researchers (Airey 2011, Hellekjaer 2010, Klaasen and de Graaff 2001, Morell 2018 claim that effective lecturing behavior (when the lecturer uses explicit instructions, scaffolding, actively involves students in the discussion, etc.) is a necessity for information comprehension in second language instruction, and that learning outcomes depend on lecture structuring and the use of interaction supported by appropriate non-verbal behavior and well-prepared visuals. Therefore, EMI trainers should encourage content teachers to ask audienceoriented questions (Morell 2020) based on students' personal experience to support their understanding in lectures of any discipline.
Another misconception, that many content teachers share, is that they have enough teaching experience to work with any audience and they do not need any innovative pedagogical techniques. Simply translating their lectures into English, without the need for any adjustment to the audience, is enough. Thus, the trainer has to explain to a content teacher that in EMI content understanding is a sensitive issue and the teacher should be able to provide adequate scaffolding, check comprehension and give feedback to achieve the desired outcomes for students studying the subject in a second language.
The third myth that is actively supported by the majority of Russian content lecturers is that international students should be treated like Russian students. Content teachers often do not have time to learn about the culture and educational background of international students. Moreover, most of the EMI teachers do not consider it necessary to study the cultural peculiarities of their international students to help create a comfortable atmosphere in classes.

The beliefs of content teachers.
The myths mentioned above are supported by the beliefs of content teachers described by the trainers in their semi-structured interviews. The trainers referred to the issues of lesson preparation, trust in authorities, and the evaluative nature of observation by senior management. In interviews, the trainers reported the unwillingness of the EMI trainees to spend much time on lesson planning. As two of the trainers noted: "They believe that too much preparation doesn't pay financially." Trainer 6 "Planning is always bureaucracy and not helpful for teachers." Trainer 8 At many Russian universities no incentive scheme has yet been developed to compensate lecturers in this respect. Moreover, to make the lecture interactive requires extra planning and the preparation of handouts that makes planning an EMI lecture extremely time-consuming, doubling or even tripling the workload of the academic staff (Doiz et al. 2012;Dafouz 2018). Thus, content teachers prefer using the materials they have in Russian and translate them into English as the easiest ways to deliver an EMI course. The goal of the EMI trainer is to show participants that word-for-word translation does not convey the appropriate meaning of their lectures, and may lead to misunderstanding of the key terminology and the overall message. Our results match with the recent findings of Méndez-García and Luque Agulló (2020).
The other issue encountered by the trainers was the deep distrust in highly directive university management, which has to comply with national standards and generally lacks flexibility in terms of education programme design. One of the trainers mentioned that: "Content teachers don't believe that their authorities will allow them to make the necessary changes in their syllabi and curriculum to adapt the program to international students." Trainer1 The most critical issue in content teachers' beliefs revealed by the trainers is their negative attitude to peer observation (Göker 2016). EMI trainees referred to their experience of being observed by their colleagues or heads of the departments several times during the academic year. These observations were performed in order not to help a person to become a better specialist, but to check if teachers followed their syllabi without any variations. Few content teachers treat observation positively. One trainer noted that: "EMI trainees feel that teacher observations are evaluative and not just formative, so they impact on their financial status." Trainer 3 The first aspect that should be clarified here is that peer observation is not common practice for Russian universities as it is mostly done by the heads of the departments when the university administration decides whether to extend a lecturer's contract or not. Thus, the fear of content teachers comes from their working experience. Secondly, peer observation is rarely conducted in Russian universities, and the content teachers need practical skills to give feedback properly following the sandwich model (where you give the person being observed a comment that identifies an area where they can improve 'sandwiched' between two positive comments that pick out where they have done well) and not to be judgmental while criticizing their colleagues.

The methods that work best with content teachers.
The trainers highlighted the methods that work best with content teachers. Content teachers appreciate pair and group work as they are eager to share their experience and knowledge. One such techniques is TPS (think-pair-share). This is a collaborative learning strategy where students work together to solve a problem or answer a question about an assigned task. TPS requires students to (1) think individually about a topic or answer to a question; and (2) share ideas with classmates. This activity not only stimulates discussion, but also encourages teachers to share views with a partner/group, thus, building their confidence and creating a cooperative environment.
Content teachers understand the material, that is being presented to them during the course, much better if experiential learning techniques are used with them. Micro-teaching sessions (several 20-minute lectures and practical classes conducted by the trainees and observed by their peers and the trainer during the course) were a good way to apply this. Peer observation and constructive feedback enriched content teachers' practice as all of them received feedback regarding lecture content and delivery.
The third aspect that trainers recommend is to ask content teachers to reflect on their own lectures delivered at micro-teaching sessions and turn this practice into a professional habit. Peer observation works best, as there are usually two or three teachers from the same field of study in an EMI group. They can see each other's mistakes in terms of content, and the EMI teacher trainer can give feedback on the content delivery. Thus, the participants can learn both from each other and from an EMI teacher trainer, practicing coaching, giving feedback and reflecting on the advantages and drawbacks of the lecture given.

The challenges for the trainers
The answers to the question on the challenges the trainers encountered can be divided into three areas: language issues that content teachers face, resistance to changing their teaching style, and lack of trainers' expertise in the subject areas of content teachers. The lack of participants' proficiency in English is one of the most recurrent topics in the trainers' interviews. As discussed above, an EMI course should be taught to the lecturers with B2-C1, but the reality of Russian universities, as well as many European universities (Costa 2015), is that lecturers often start teaching EMI courses when they have a lower CEFR level, often causing difficulties to the EMI teacher trainers. One trainer stated that: "EMI is perceived too challenging and time-consuming by the trainees. They think that they are supposed to simplify the content and have impressive language competence to provide clarity and accuracy of expression." Trainer 5 In addition, EMI trainers note that the most serious issue observed through classes was language simplification both at the lexical and syntactical levels through the adoption of a more direct and concise style. As lecturers' answers showed that language simplification was caused by their limited ability in spoken English.
The lecturers explained that they were more accustomed to reading and writing in English. Valcke and Båge (2020) have similarly found that EMI lecturers have more experience in written than oral communication.
Based on the materials that trainees prepared for the micro-teaching sessions, an interesting analogy was observed. The comparison was made between the software-aided presentations used in the EMI trainees' native language and those used in the EMI courses. The first difference is the number of slides, with those in English outnumbering those in Russian. Secondly, the slides written in English tended to have more text as compared to those in Russian, as if there was a need to make some concepts and issues more explicit. Moreover, English slides served as cue-cards for content teachers, enabling them to read the information, because producing a spontaneous coherent narrative in English remains a problem for them (Gibbons, 2015). EMI trainees often justify the amount of text on the slides in terms of the low abilities of their international students, and seem afraid to confess that it is a way for them to get their own language support and to ensure they deliver the key content to the students. Moreover, EMI trainees should keep in mind that many of their students may experience more difficulties in English than others and thus be able to provide language scaffolding assistance. So, it is essential to improve the English language proficiency of both staff and students through English language training courses.
Another issue noted by seven of the eight trainers is the resistance of EMI trainees to changing their teaching styles. One of the trainers stated: "Sometimes they are resistant to change and to the novelty of teaching techniques suggested. Resistance to active learning techniques and student-centered approach. Determining learning objectives and splitting big goals into sub-skills goes very slowly." Trainer 7 EMI trainees are experienced university teachers and to change their perspective on teaching methods is a challenging task. Before the EMI training, 67 EMI trainees (59%) firmly believed that no other changes needed to occur when they change the language of instruction from Russian into English. It was quite a challenge to encourage EMI trainees to teach more interactively. Their unwillingness to switch to student-centered lecturing and employ different EMI strategies in the classroom additional efforts from the trainers. This opinion, however, changed after completing the EMI course. Trainees adopted a different approach conducting and organizing lectures (9 events of Instruction by Gagne is a communicative strategy that is intended to further deepen the students' learning process. As each step is completed, learners are meant to become more interested, engaged and invested in the learning process), seminars using various pedagogical techniques, such as a flipped classroom, scaffolding (Think-Pair-Share, project work, pre-teaching vocabulary) and interactive strategies (picture prompt, word cloud guessing, empty outlines).
The third issue raised by the trainers was their lack of expertise in the content areas of their trainees. Most of the trainers stated that the area of content teachers' expertise was difficult to understand since many of the trainers simply lacked knowledge of teaching such content and they did not have the opportunity to observe their trainees in real-life scenarios.
Some of the ways to tackle this issue were discussed by Lasagabaster & Doiz (2018);and Trent (2017). At least an EMI trainer can observe trainees' practical classes or lectures to develop a better understanding of their course structure and style of lecturing. This will help to relieve stress and avoid conflicts between participants and trainers during the EMI class. Moreover, it will allow trainers to design an individual learning trajectory for each member of academic staff and provide them with transferable skills that they can also use with L1 students.

Conclusions
The main objective of this study was to identify the perceived shortcomings and prevailing challenges for EMI teacher trainers to design and implement an EMI training course. To achieve this goal, we analyzed the reflections of EMI trainers on the main challenges that they had encountered: selection of participants for EMI training; design of the course; EMI content teachers' resistance to change. We collected the data via semi-structured interviews with teacher trainers from three Russian universities. The data on teacher trainees was analyzed via a precourse and after-course survey. What follows is a summary of the findings and their pedagogical implications.
Since the approach to EMI at the three Russian universities involved in this work (SUSU, ChelSMA, ITMO) has been a top-down one, the academic staff have to deliver their courses in English at the demand of the university administration, whose selection criterion was the proficiency in English of lecturers not their interest in EMI. The results showed that the average level of English of the EMI content teachers in this study was CEFR B1/B2. Throughout the course, the EMI trainees encountered linguistic problems and are still in need of the language support. Thus, an EMI course may be better not as a stand-alone course, but rather combined with English language training as part of CPD in English. In line with the recent findings of Martinez and Fernandes (2020, 138) it is essential to develop CPD schemes for content teachers to increase their level of proficiency in English before they are enrolled in EMI training.
The analysis of the semi-structured interviews with the trainers revealed the misconceptions and beliefs of content teachers that interfere with the successful EMI course implementation. Many of the trainees doubt that interaction with and between students can provide a better understanding of lecture content. This is likely because they are content teachers and are not familiar with the theories and practice of second language acquisition and second language instruction. In line with Llinares and Mendikoetxea (2020) and Morell (2004Morell ( , 2007Morell ( , 2020, the trainers tried to persuade participants that audience-oriented questions, especially referential questions, will enhance interaction that will not only promote student engagement, but also allow for negotiation of meaning. The pedagogical implications here are: 1) EMI trainers should demonstrate the practical value of interactivity as a scaffolding technique in mastering the discipline's content in EMI; 2) recommendations on how to make EMI interactive need to be created for different disciplines as each discipline has its limitations and possibilities in terms of interactivity.
Another misconception, that many content teachers share, is that they do not need any innovative pedagogical techniques since they are all experienced teachers. Moreover, they consider that to give a word-for-word translation of their lectures into English is enough. An EMI trainer should explain that providing adequate scaffolding, checking comprehension and feedback will help to achieve the desired outcomes for students studying the subject in their second language. Word-for-word translation does not convey the appropriate meaning of their lectures, it leads to misunderstanding of the key terminology and may affect the overall perception of the given material. As stated by Norte Fernández-Pacheco (2018) and Morell (2018Morell ( , 2020, developing students' and teachers' multimodal competence (the ability to understand the combined potential of various modes for making and eliciting meaning) has proven to be instrumental for improving comprehension and expression in language and content learning and teaching contexts.
The challenges that EMI trainers encountered fell into three broad categories: language issues, resistance to changing their teaching styles and their lack of expertise in the content areas of EMI teachers. The lack of EMI trainees' proficiency in English is one of the most recurrent topics in the trainers' interviews. EMI lecturers were more accustomed to reading and writing in English and their speaking skills needed more practice. Our findings are in line with Valcke and Båge (2020) stating that EMI lecturers have more experience in written than oral communication. Thus, it is necessary to improve the English language proficiency of the academic staff through English language training courses and the content of EMI courses should include language scaffolding for EMI lecturers, and they can then use the strategies in their own EMI courses. EMI trainers stated that EMI trainees' resistance to changing their teaching styles, their unwillingness to switch to student-centered lecturing and employ various EMI strategies caused additional difficulties for them. However, most of the EMI trainers noted that they succeeded in changing these attitudes of EMI trainees and the latter obtained a better understanding of how to organize lectures and use various pedagogical techniques (flipped class, scaffolding, Think-Pair-Share) and interactive strategies (picture prompts, empty outlines, word clouds, etc.). The pedagogical implications for this are 1) EMI trainers should be ready to meet with the resistance of the EMI trainees and devise methods to cope with their unwillingness to change; 2) EMI trainers could use video recorded examples of interactive lectures from previous EMI courses to demonstrate the advantages of student-centered lecturing.
Most of the EMI trainers stated that the area of content teachers' expertise was difficult to understand since many of the trainers simply lacked knowledge of discipline teaching and they did not have the opportunity to observe their trainees in real-life scenarios. At least an EMI trainer can observe trainees' practical classes or lectures to develop a better understanding of their course structure and style of lecturing.
Thus, to design and implement an EMI training course an ESL teacher will certainly experience most of the challenges described in the research. The combined efforts of university administration, EMI trainers and EMI trainees are needed to make EMI teacher training successful. The university administration has to select academic staff with a set of specific pedagogical skills and an adequate level of English-language proficiency; an EMI training course should be research driven and universities must assign ESL language teachers to develop an EMI teacher training course, considering their awareness in EMI and their research interests. To reduce academic staff's resistance to change, university administrations need to develop a language policy, incentives for participation in EMI and provide support in terms of CPD schemes for the academic staff.
The study is not without its limitations. One is that it describes the Russian context of EMI teacher training and its findings cannot be generalized to other contexts. Another limitation is the number of respondents. The sample of EMI trainers is not enough to make broad conclusions on the pitfalls of EMI teacher training. The third shortcoming of the research is that only one of the EMI trainers had a research interest in EMI and was capable of thoughtful reflection on the challenges encountered. All the other EMI trainers had valuable but purely practical reflections on their experiences.
Further research is needed to analyse and compare case studies from other universities in order to develop general recommendations and ascertain the practices that work best with content teachers from different disciplines. It is more difficult for me to have a discussion related to my specialist field in English than in Russian 37 I feel less confident when I teach in English than in Russian 38 I feel I am a less successful teacher when I teach in English than in my mother tongue 39 My professional authority suffers/will suffer from teaching in English 40 I feel that I am more efficient as a teacher when I teach in English