Teaching in English Is Not Necessarily the Teaching of English

Described as a “galloping” phenomenon now considered “pandemic” in proportion, the use of English as the lingua franca medium of instruction (EMI) at higher education institutions (HEIs) across the globe is today considered the most significant trend in educational internationalisation. Japan is no exception and a growing number of the nation’s universities are increasingly offering classes–and even entire courses–in English. Seen by some as a panacea for jump-starting the nation’s stagnant internationalisation profile and improving overall English language skills, this paper firstly explores the theoretical background and rationale behind the trend to utilize EMI based on a review of the literature. Secondly, questionnaire data and feedback from Japanese students taking such classes at two, second-tier universities are analyzed to help shed light on attitudes and ascertain the issues as well as highlight some limitations and problems involved with EMI classes. Finally, it concludes with practical recommendations for greater language support activities and warns of the implications of naively equating EMI alone to an automatic improvement in English language ability in the Japanese context.


Background
A recent combined report from the British Council and Oxford University states that "there is a fast-moving worldwide shift from English being taught as a foreign language (EFL) to English being the medium of instruction (EMI) for academic subjects" (Dearden, 2014, p. 2).Growing out of the bilingual education movement of the 1950sin Europe and Canada (Barnard, 2013), the pace of implementation today is rapidly advancing, being referred to by some as a potential "pandemic" (Phillipson, 2009) or "revolution" (Dafouz & Guerrini, 2009).In 2002 there were over 800 EMI programmes operating in Europe alone (Doiz, Lasagabaster, & Sierra, 2012).Today there are well over 6400 (ICEF Monitor, 2013) and a growing number in various other traditionally English-as-a-foreign-language countries around the world (Björkman, 2011).French universities, for example, are now legally allowed to offer a limited number of classes in English and Italy's Politecnico di Milano began teaching all of its graduate classes in English in 2014.In short, the move to use English as the lingua franca of higher education globally is seen as the single most significant current trend in internationalising higher education (Parr, 2014).
In Asia too, universities in countries like China, Taiwan, Korea Vietnam, Indonesia and Malaysia are increasingly adopting English as a way to gain access to cutting-edge knowledge, enhance national competitiveness in innovation and knowledge production (Hu, 2007) and, naturally, as a strategy for improving graduates' English proficiency (Ali, 2013;Hamid, Nguyen & Baldauf, 2013;Wong, 2010).Japan has also increasingly begun to experiment with EMI classes, although originally for slightly different reasons.The initial impetus to increase the number of classes offered in English at HIEs being traced back to the statement below issued in 1997, relating to foreign students: In order to lessen the burden of having to learn Japanese for prospective international students, and to provide to many high-achieving foreign students an opportunity to study in Japan, it is important to establish educational programs using foreign languages (primarily English).It is also important for Japanese students, in addition to foreign students, to participate in these programs (Cited in Tsuneyoshi, 2005, p. 67).
More recently, the government and other private institutions (such as the English Language Education Council) have strongly urged universities to offer between 10-30% of their academic courses in English (Brady, 2008).Such moves have accompanied the Ministry of Education, Culture, Sports, Science and Technology (MEXT)'s flagship project for internationalisation, the so-called "Global 30" project (in Japanese: kokusaika kyoten seibi jigyō) that primarily aims to support a small core of select universities to increase their ability to accept and educate international students in English.The purpose of the project, according to MEXT (2009), is as follows: [The G30 Project] aims to create an environment which is conducive to attracting foreign students by offering a level of education attractive to them and fostering advanced (kōdona) human resources capable of actively working together in competition with international students, in order to strengthen our nation's (waga kuni no) international competitiveness in higher education (Author's translation).
The human resources referred to here are Japanese nationals, and thus the goal for EMI in Japan is seen as the increased accommodation of foreign students first, and-almost as a kind of natural spin-off-the internationalisation of Japanese nationals too (i.e.developing their English skills).From a mere six departments at five universities and 101 courses at 57 graduate schools in 2006 (Burgess, Gibson, Klaphake, & Selzer, 2010), the plan is to add an additional 33 undergraduate courses and 124 graduate courses to make 300 courses in total.In 2009, out of 22 applications, 13 universities were selected (seven national and six private, see Note 1) and will receive between 200 to 500 million yen annually for 5 years to increase the number of international students they accept, the number of foreign faculty employed, the number of Japanese they send abroad and the establishment of branch offices overseas.However, the 13 universities chosen were all comparatively international and progressive from the outset and, given their size and prestige, their selection came as little surprise.

Objectives of the Study
Despite the increasing spread and interest in EMI, an appropriate theoretical and pedagogical framework is still lacking (Keuk & Tith, 2013).Because of the relative nascent nature of EMI in Japan, there is a particular absence of research into the method of delivery and success or impact of EMI courses there, especially in relation to second tier universities.This study aims to contribute in some way to this body of research.Researchers have already addressed some aspects of the G30 and their EMI to date (Yonezawa, 2010;Rivers, 2011;Burgess et al., 2010).The focus of this article will rather be on what Igami (2014) refers to as universal (or non-elite) universities, or what I prefer to term second tier institutions (as opposed to the 'elite' G30 members and other 'special case' universities, see Note 2); those just below the top ranks.While second tier universities rarely have the resources to extend their global reach by opening overseas branch campuses or either the name value or financial means to attract substantial numbers of international students, they recognize the potential value of EMI courses and classes as a way to both accommodate greater numbers of international students and (perhaps more importantly) assist in the acquisition and improvement of English language skills amongst Japanese students.It is the problems associated with this aspect of EMI in these institutions in Japan that is the focus in this article, which sets out to initially question whether the growing trend to offer more classes in English in Japan really can lead to improvements in Japanese students' English skills.
The remainder of this article is organized as follows.Firstly it briefly introduces and explores the background and rationale behind the trend to utilize EMI in Japanese higher educational institutions as a means of aiding in internationalisation and its supposed ensuing English proficiency requirements.Next, while concurrently drawing upon the results of a small case study of Japanese and international students who had taken EMI classes and some teachers involved in instruction, problems and limitations revealed in the Japanese context of second-tier universities are examined and discussed.The discussion is divided into broad categories that mesh with issues and concerns reported in literature and other sources in different settings.Finally, it concludes by providing some practical recommendations for greater language support activities and warns of the danger of naively equating English medium instruction classes alone with an automatic improvement in English language ability in the Japanese context.

The History and Demand for EMI in Japan
The benefits of bi/multilingual education at the tertiary level have been the subject of much empirical research.Institutionally, things like the opportunity for the promotion of student and lecturer mobility (Symon & Weinberg, 2013), raising research profiles and visibility in rankings (Kassteen, 2013), and increasing institutional income from greater student numbers (Whitsed & Volet, 2011;Symon & Weinberg, 2013) have been outlined.
At the level of the individual student, Kassteen (2013) succinctly summarizes some of the main benefits including: improving cross-cultural understanding and global awareness; enhancing academic progress in other subjects; fostering creative thinking; enriching and enhancing cognitive skills and emotional development; helping students score higher on standardized tests, and enhancing career opportunities.In Europe, according to a report from the Economist Intelligence Unit (2012), "Even when recruiting for jobs in their home market, almost one-half of all companies say that prospective candidates need to be fluent in a foreign language, and a further 13% say that multilingual ability is a key selection criterion" (p.12).Similarly, in Japan a recent trend in HEIs is the fostering of so-called 'global jinzai' (globally-competent human resources), one of the key components of which is said to be foreign language skills (Gurōbaru jinzai ikusei suishin kaigi, 2011; Keidanren, 2011) and hence again we see the link being made between EMI classes and language acquisition.In Brown's (2013) study other institutional push factors noted behind the EMI trend in Japan were a desire to catch up with and/or distinguish themselves from other schools (including those in the G30), prepare students for study abroad or internships, and adding authenticity to language teaching.Yet, in spite of such a wide variety of benefits, "the aim of improving students' English proficiency is not generally addressed" (Symon & Weinberg, 2013, p. 22), and certainly not in terms of how may be improved.
With the proliferation of EMI courses in universities around the world, one concern that is increasingly being voiced is the danger of the English bias and the detrimental impact it can have on national languages (Brock-Utne, 2013;Cason & Rodriguez, 2013).In Israel, Senegal and Venezuela, for example, EMI has not been allowed in public education for this very reason (Dearden, 2014).Such an argument, however, is still rare in Japan and this is generally because the collective English proficiency level is considered to be comparatively low (Tsuneyoshi, 2005;"Japan ranks 26th", 2014;Björkman, 2011), certainly sufficiently removed from any such threats to the dominance of the national language at present.Put differently, the perceived poor English proficiency level of Japanese university students has created an increased interest in the potential of EMI to alleviate this problem in Japan and the consequential growth in number of institutions already offering or preparing such opportunities.The latest available figures from MEXT show an almost 5% jump in the number of universities offering classes in English in the past 2 years (Table 1).Source: MEXT (2013).
As Table 1 shows, most of the growth is in the private sector (where the majority of second tier universities are).
Regardless of the overwhelming active support of the Japanese government, big business along with pressure from society and parents (Hamid et al., 2013), if not carefully prepared and supported, there is an inherent danger in the naïve and overly simplistic assumption that merely equates classes taught in English as leading automatically to increased overall language proficiency.While it should be noted that other surveys in different settings have found a correlation between increased exposure to EMI and increased English proficiency (Wong, 2010), such 'success' is often in elite institutions (where students already posses a high degree of linguistic ability) and/or in settings more amenable to linguistic opportunities immediately outside the classroom, neither of which are pertinent to the present study.

Data Collection
This research employed a mixed-method approach in which questionnaires and interviews were conducted over a period of two and a half years with students enrolled in two different classes at two private, second-tier universities in Western Japan.Both classes were 15-week semester long content courses (i.e.not English language classes) in which international exchange students as well as domestic students with a certain level of English proficiency (a minimum requirement of TOEFL 450 for one and an overall IELTS score of 4.0 or 600 TOEIC for the other) were eligible to enroll.Both courses formed part of a study programme for international students at the respective universities.Questionnaires in the form of brief, open-ended feedback forms in both English and Japanese were used to allow respondents to express their opinions without being influenced by the researcher (Foddy, 1993).Comments and feedback from teachers involved in teaching other classes in the programme have also been included in the results, which are interspersed in the following discussion so as to provide further insight into the issues encountered in the delivery of classes and impact on language learning for Japanese.

Participants
A total of 89 responses were collected overall from Japanese students and 26 from international students at various times during the semesters (the first class, a number of times during the semester and in the final class).
The Japanese students (56 female and 33 male) were enrolled in a variety of faculties (classes are open to students from various faculties in both universities) as well as years (7 first year students, 39 second year, 29 third year and 14 fourth year).The international students (16 female and 10 male) were from a variety of ethnic backgrounds but with the majority being native English speakers (42%) from the United States, Australia, England and Canada.The remainders were from France, Germany, China, Norway, Spain, Portugal and Thailand.
A number of other discussions were held with Japanese students enrolled in other EMI classes in the same programmes although they did not complete any feedback forms.All discussion interviews held were unstructured and informal, generally taking place before or after classes, although discussion times were also held during classes on several occasions.Discussions were conducted in both English and Japanese with randomly selected students who were assured their identities would remain anonymous.

Linguistic Issues
As mentioned, in Japan EMI classes are increasingly being utilized in order to raise the English proficiency of university students.Certainly that was the main reason reported by the Japanese students in this study as represented in Table 2.As can be seen, students (n=89) were asked to select up to three reasons for taking the classes.The total number of responses was 190 with the most frequent response given being to improve English language ability.However, as this paper questions, the idea that merely taking a content class taught in English will lead to substantial linguistic gains is dubious.Some of the most important factors affecting the successful implementation of EMI in Japan are those of a linguistic nature, starting with the level of English language proficiency students possess.Insufficient ability has been identified as a major impediment to successful EMI courses in numerous previous studies (e.g., Beckett & Li, 2012;Doiz et al., 2012;Webb, 2002).As a consequence, research has shown that students lacking ability to keep up tend to participate less in the classroom (Webb, 2002), have considerable difficulty comprehending lessons (Tsuneyoshi, 2005;Hellekjaer & Räsänen, 2010), take longer to complete their courses (Tsuneyoshi, 2005), and are more likely to drop out (Selzer & Gibson, 2009).
Similar findings were recorded in the present study.On average, 34% of students who initially enrolled in the classes failed to complete them and either gave up or officially withdrew.While exact data on the reasons was not obtained, anecdotal evidence suggests that insufficient language ability was the main factor.The following comments received from two male students who pulled out of the class (but remained in other classes taught by the instructor) are indicative.
At first I thought it would be a good way to listen to and practice English, but it was too hard for me.I couldn't understand.
It was above my level.It was too difficult to listen in English and take notes.
Regardless of the fact that both of the classes had entry requirements in place, clearly these are insufficient in terms of functioning as effective gate-keeping mechanisms.During the courses, the majority of the Japanese respondents admitted that their image of the class turned out to be vastly different to the reality (Table 3).When asked about the strategies the students employed to keep up with the class, activities such as note comparison with foreign students, downloading lecture slides, recording and listening to the lectures again, keeping vocabulary logs and asking Japanese classmates, were mentioned.Further, one of the instructors specifically made efforts to support the Japanese students through activities such as providing translated documents, simpler explanations or links to readings in Japanese.
Thus, with regards to the linguistic benefit that was reported by students at the completion of their courses, just over half reported that they felt either "considerable" (18%) or "some" (33%) improvement, 24% no change, and the remainder were unsure.This was greater than Symon and Weinberg (2013)'s findings in which less than half of all students reported any real or perceived improvement in language abilities but less than that reported in Chang's (2010, p. 75) study (in which "most students" reported an improvement).From this brief discussion to date, one conclusion that can be drawn is that whether the provision of regular and organised structural language support was provided or not affected students' levels of satisfaction and comprehension.The students who utilized strategies or assistance offered by the instructor reported both higher levels of satisfaction and/or greater feelings of improvement and confidence.A clear correlation between students who claimed to have not utilized any of the support assistance and low or failing grades was also observed.In the class without regular linguistic support, comprehension and satisfaction scores were lower as were the Japanese students' grades in general.In short, without sufficient availability and utilization of specialized language support, EMI alone appears to lack the ability to confer linguistic benefits.

Making International Friends
After the expectation of being able to improve their English ability, the second most sort-after goal according to student responses was "to make friends with foreigners".In other international settings, increased importance is being placed on fostering such intercultural dimensions within the classroom (Whitsed & Volet, 2011) which is clearly an important function of such classes not only for furthering intercultural experiences but also for providing additional extra-curricula language learning opportunities.For students in this survey, making international friends was certainly seen as an extension of the linguistic goal as it was hoped that they would "become conversation partners" and thus lead to extra practice opportunities occurring outside of the classroom (soto demo hanaseru kikai).
However, while for international students the opportunity to meet students from other countries, take classes and converse in English together makes their entire experience more accessible, doubt has been cast on this function when it comes to local students.In a European setting, Cason and Rodriguez (2013) refer to the danger of what they call "Erasmus bubbles" forming whereby European students taking part in the Erasmus (European Community Action Scheme for the Mobility of University Students) programme gravitate to interacting solely with one another and not local students in the host country.A similar situation was witnessed in this study also.Of the 28 international students questioned, only 6 mentioned any regular, out of class interaction occurring (despite of their desire for it as well) and a further 3 saying sporadic relationships had formed but were short-lived.
It would appear that real engagement comes when courses have been carefully prepared to provide the correct tools and environment to allow it to happen.In other words, universities need to ensure that both international and local students have been equipped with the opportunities, skills, content and reason to allow for such exchange.Assuming it will naturally occur spontaneously is wishful thinking.

Double-Sided Dissatisfaction
While the Japanese students noted that the classes were generally either "very" or "extremely" difficult, the opposite was reported from the international students.Tsuneyoshi (2005) writes that one criticism received from Western students in Japan was that courses were not "Westernized enough".In the present study too, similar feedback was received.In response to an open question about the course overall, the following comments were received from international students: At times it felt like the class was held-up to clarify information for the Japanese students.
I would've liked more challenging material, more readings and homework, for example.

The assessment was not what we get back home, probably because of the different levels in the class.
This point is related to the question of at which level to pitch class content.If, as expected and desired by the majority of international students, the class adopts a curriculum and delivery method on par with that found in institutions in Western institutions, for example, Japanese and other non-English native or near-native speakers will oftentimes be disadvantaged.In one of the Japanese participant's words: I wasn't able to concentrate for long as it was too difficult for me.When there were words I didn't know or the speech was too fast, I often stopped listening and sometimes even fell asleep.
When they (international students) asked questions or made comments or jokes, I was completely lost.I wanted the teacher to translate it all for us.

Teaching Issues: Who?
Of all the issues impacting on the success of EMI in terms of the impartation of language skills, the question of who teaches and how are the most important.Globally, the selection and retention of suitable teachers is a pressing task.According to Tsuneyoshi (2005), one of the biggest challenges in continuing a programme with lectures in English "was finding faculty who were both willing and able to present lectures in English" (p.80, emphasis added).Similarly, it was reported that in 83% of the countries where EMI courses are taught, the lack of qualified teachers was a large complaint (Dearden, 2014) with serious implications for teaching quality.While it appears that some instructors volunteer to teach EMI classes, others are often nominated to do so merely because they had spent time abroad or spoke English well (Dearden, 2014).In Japan's case, there also appears to be a clear shortage of both those willing and able to teach EMI classes effectively.
One possible reason for this may be related to teaching style.University teachers in Japan have been criticized for being closed-minded, indifferent to internationalisation or lacking international sense (Jinzai kyōsō kokkyo, 2012) and this general apathy towards internationalisation impacts on their attitude and desire towards teaching classes in English as well.In discussions with other instructors in this survey the following comments were received:

It [EMI] is a positive move, but I can't do it. My English isn't sufficient I feel.
We need more classes taught in foreign languages, but we [all] are too busy.
In Japan, even in the home language, questions have been raised about the quality of teaching.As an editorial in Japan's largest English daily newspaper recently lamented: "Unfortunately … the teaching methods at most universities, … remain mired in one-way, teacher-centered approaches that do not help students acquire confidence, communication skills or a broader understanding that they need for engaging in international situations" ("Too many inward looking students", 2013).Such generalizations tend to reflect widespread sentiment.There is an underlying belief that merely translating Japanese classes into English constitutes an EMI class.However, an insufficient focus on the quality of the language of instruction naturally leads to a decline in the overall quality of education (de Wit, 2011).This no doubt stems from the widespread belief that teaching is merely conveying information and the means of communication is of little relevance.Yet the process itself is much more complex than that.Merely possessing "the ability to read widely and write at length in a second language does not necessarily transfer to effectively explaining key concepts to students in such a way as to make the lectures comprehensible" (Barnard, 2013, p. 4).In the absence of clear guidelines and standards, institutions are struggling to obtain, place and keep the most qualified teachers in EMI classrooms.Certainly forcing those not willing or capable is an unproductive act, to say the least.
Another reason can be attributed to what Phillipson (1992) refers to as a "native-speaker fallacy".In order to be seen as 'international', many institutions and leaders believe that native speakers are the ones who should be teaching since the classes are English.Kyoto University, for example, plans to hire100 new foreign teachers over the next 5 years to teach half of its undergraduate general liberal arts classes in English ("Kokusaika ni futatsu no kabe", 2013).However, this approach fails to take into consideration whether the original Japanese content was up to standard in the first place (as mentioned above) and whether the foreign teachers are qualified and experienced to teach in a Japanese context.
A third possible reason why teachers may be unwilling to teach EMI is that many do not consider themselves language teachers (Doiz et al., 2012).As Dearden ( 2014) asks from her survey, when EMI teachers do not feel that teaching English is their job and they lack responsibility for their students' English, how can the students understand the classes?"If subject teachers do not consider it their job to improve the students' English, whose job is it?"(Dearden, 2014, p. 6).However, in a traditional Japanese university context where a clear demarcation between so-called specialist content teachers (senmon) and general liberal arts (ippan kyōyo) teachers oftentimes still exists, a certain stigma remains making the question more difficult to answer.Today still, in the majority of cases, language teachers fall into the category of kyōyo teachers and stereotypical perceptions see them held in lower status to their content counterparts.Thus, feelings of pride and status perhaps impact on the desire of content teachers to be willing to teach in English.Arguments have also been made by Japanese content teachers against teaching in English (Saito, 2013) on more pedagogical grounds.However, these tend to miss the point of EMI courses in Japanese universities.Resolving this superfluous dichotomy of teacher types is essential for Japanese universities to ensure their EMI classes meet both linguistic and content goals.
A final issue contributing to the lack of enthusiasm from teachers and related to the previous point, is a lack of faculty and institutional support and recognition of the added workload required to prepare EMI classes.Teachers need to be specially trained and assisted in their preparations and this additional burden needs to be properly compensated by institutions."Unless this dimension is explicitly recognized in the process of evaluation and promotion of the faculty, most of those drawn into internationalization initiatives will be haphazardly self-selecting faculty with a personal interest in the international dimension of education" (Hawawini, 2011, p. 10).Under the present university system, most departments and faculties create their own reward structures and opportunities or benefits for professional development that rarely value international learning contributions.Consequently, as Stohl (2007) claims, given this situation it is hardly surprising that relatively few faculty members participate in international education activities beyond such things as attending international conferences.
In the present study, the teachers involved in both institutions comprised of native English speakers, native Japanese speakers with good English ability and two non-English, non-Japanese native speakers with near, or native-like, English proficiency.All had willingly agreed to teach the courses.Remuneration was the same for the EMI classes as it was for any other classes taught with no special concessions made for extra preparation time or the like.At one institution, a number of native Japanese teachers had been approached and refused to teach classes in English stating a "lack of time to prepare" and the "enormous workload" that would be required.This coincides with the findings of Hawawini (2011) above and others (Symon & Weinberg 2013, Dearden, 2014;Alidou, 2004) that unless due consideration is given to supporting the teachers deemed best suited for EMI classes, there is little way of guaranteeing the quality of the education provided.

Teaching Issues: How?
Intrinsically related to the topic of who teaches, is the question of how to teach, which in turn impacts on whether linguistic gains are made or not.As mentioned, contrary to general perception, merely teaching a course previously taught in the national language in English alone is unlikely to achieve significant linguistic benefits.
As Symon and Weinberg (2013) fear: inadequately prepared EMI instructors may believe they only need translate their course from L1 to L2 in order to teach effectively in English and that no other modifications or adaptions are necessary.When reporting on their experience of EMI in Israel, Symon and Weinberg (2013) write that "if the only exposure to English is during lectures, and if assignments and examinations can be submitted in their first language, then improvement in English will be minimal" (p.23).
The question of how to teach EMI classes is complex and involves issues such as how much (if any) of the first language should be used and, if so, in what way and when (in class, in handouts or other supplementary material or text books, in presentation slides or assessments?), the content (how much to teach? using the same as in the first language or a more watered-down version?) and assessment criteria and methods.Because EMI in Japan is still in a relatively infant stage and literature is still sparse, the emphasis is primarily placed on content acquisition and any language development is seen more as a bonus (Symon & Weinberg, 2013).Yet, this is counter to the alleged goals of EMI created by the administrating institutions.
In this study, all instructors estimated that English was used for approximately 90% of the time overall, and almost 100% of the time when teaching.However, content was reportedly adjusted so as to be more understandable to the Japanese students.One instructor mentioned such activities as preparing slides with Japanese translations of specialized words, providing brief reading handouts (in English and/or Japanese) which summarized the main points, regularly stopping and checking understanding, and including small group discussion opportunities, with international students in the class acting as facilitators.The other instructor provided minimal assistance unless specifically asked by individual students.
Assessment policies were varied and ranged from holding only end of term tests in English to more complex forms of assessment throughout the term (written reports and presentations).However, one instructor confided that grading had to be scaled "considerably" in the Japanese students' favour and some questions were written either in both English and Japanese or with Japanese glosses on difficult words.Overcoming the language discrepancies between Japanese and international students was clearly a large concern.
I explained in detail what was expected, but it was clearly well beyond the Japanese students' ability to write short essays in comprehensible English that actually answered the test questions.They often either didn't have enough time or wrote an insufficient amount.It made grading an extremely challenging task.

Another commented:
It is unrealistic to try to test Japanese students and international students in the same way.In terms of linguistic ability, they just cannot be compared.

Structural Issues
A final issue deemed important-yet often overlooked in literature-is of a structural and administrative nature.
For EMI courses to be able to contribute to the English linguistic development of students, systems and structures need to be put in place to support both staff and students alike.On most second tier Japanese campuses there is a lack of staff competent in dealing with international education either linguistically or conceptually, representing a dangerous void.Institutions must realize that for EMI courses to be successful, they need to be considered a fully-fledged, official part of the educational experience and not merely be regarded as 'add on' frills for appearance.It is, therefore, "necessary to sensitise all stakeholders to the importance of bilingual academic literacy and to the necessary shift in role perceptions" (Barnard, 2013, pp. 12-13).Fostering such perceptions can be done through faculty and administrative development programmes but also requires the appointment of permanent administrative staff to support and develop courses from a non-academic point of view.Administrative staff in Japanese universities are regularly rotated every 3-5 years irrespective at times of abilities and personal preferences (Lassegard, 2014) meaning appropriate staff are not always available.In both of the universities involved in the study reported here, neither employed dedicated staff for the EMI programmes and those who were involved had neither experience nor expertise in any form of international education.Given the specialized needs and care required for EMI classes, such an approach represents a lack of understanding and dedication to the overall concept.

Implications, Considerations and Conclusion
As the preceding discussion has attempted to show, the implementation of EMI classes in Japan's second tier universities is a growing phenomenon with potential, but still very much a work in progress requiring more preparation and support to be of linguistic benefit to the majority of students.Based on the issues and problems highlighted in this small study, several practical recommendations are highlighted here for future consideration and research.
Firstly, universities and instructors need to abandon the assumption that simply teaching in English means students will automatically pick up the content and improve their language skills at the same time.In fact, there is a danger that in some cases neither could happen (Lei & Hu, 2014).Thus, a different approach to content instruction in English suitable to the goals of Japan's second-tier universities needs to be constructed, perhaps along the lines of the content and language integrated learning (CLIL) method.Regarded as the most developed form of EMI (Symon & Weinberg, 2013), CLIL, or dual-medium instruction (Barnard, 2013) encompasses both the teaching and learning of content and language (Marsh, Mehisto, Wolff, & Frigols, n.d.) into courses.In support of CLIL, Brüning & Purrmann (2014) write it "certainly could be a good starting point and helpful as it has the potential to combine LI…, L2 and subject matters" (p.319) by fusing the best of language education together with the best of general education (Georgiou, 2012).Adopting such a dual focus requires the careful reconceptualization of class content and assessment and consideration of how to include international students into such an arrangement.This kind of groundwork appears to be lacking presently in Japan, yet is a fundamental essential.
Secondly, with regards to the actual method of instruction for students in Japan's non-elite universities, an English only environment can have serious academic repercussions especially when their English language ability is not advanced enough or they are not sufficiently supported linguistically.Thus, instructors should aim to incorporate intercultural discussion opportunities and activities as well as linguistic enhancement activities throughout a course.For example, adopting a dual-medium model in which the texts are written in English with summaries and abstracts written in L1 (Barnard, 2013).A similar approach was used by one instructor in the study and results were markedly more positive from Japanese students.In this way, international students are also able to participate as well as benefit from intercultural discussions and, if they are studying Japanese, use the same process in reverse (for example receiving summaries and study material in Japanese).Similarly, in order to firmly support and assist Japanese students to be able to comprehend and keep up with the classes, techniques such as scaffolding or sheltered content instruction should be employed regularly (Echevarria & Graves, 2006;Gibbons, 2002;Mohan, 2001).While the targeted usage of Japanese in classroom may contradict the direction even Japan's Ministry of Education, Culture, Sports, Science and Technology is taking (such as insisting that English be taught in English at high schools and even junior high schools in the future, MEXT, 2014), it allows for the integration of international students, caters to the needs of less confident Japanese as well as enlarging the potential instructor base.Overall, as Tan and Ong (2011) write, there needs to be a reconsideration of in-class linguistic practices, in-school testing practices and the linguistic accommodation measures used to provide optimal support for those students who are less proficient in English.
A third point in need of consideration is the issue of entry requirements.One option is the creation of gradual step-up classes to allow for students to work their way up into higher-level content classes.The first level could be a mandatory general introduction 'tester' class providing students with an opportunity to appreciate what is required as well as instill some of the requisite study skills.Those who pass and agree with the goals and learning style would be able to continue on to more challenging EMI classes.Another possible method is the introduction of tutorial-style smaller classes to encourage discussion of content topics as well as provide an opportunity to gain practice in language-related techniques (such as how to ask and answer questions, write reports, make presentations, etc.).This could be done in tandem with writing centres, language teachers or other language assistants as well.However as these calls for much wider involvement in courses as is usually the norm in tertiary education, it also involves a commitment of resources and a new approach to understanding from the institutional point of view.
Fourthly, and related to the previous point is the issue of students' recognition and understanding of their own proficiency level and its limitations.Generally speaking Japanese students tend to undervalue their abilities in accordance with cultural and social norms.Even still, however, there is a lack of appreciation of the gap which exists between conversational English ability and that required to comprehend, question, and internalize a university lecture in English.Thus, while researchers such as Wong (2010) claim that students in Hong Kong are in favour of English only instruction as a means of improving proficiency, this preference "is positively related to their own English proficiency" (p.127).Similarly, Chang (2010) cautions that as students' English language proficiency influences their level of comprehension we need to carefully assess whether teaching in English actually is leading to a general overall improvement and not just the perception of such.In Japan, there is still much work to be done in terms of both understanding the requisite skills for learning in English, objectively evaluating individual proficiency, and increasing motivation for usage within (and without) the classroom (as would be expected in Hong Kong, for example).
Finally, and related to the above point, universities that are serious about implementing EMI classes for their students, need to assess their requirements and invest in such courses appropriately.This involves investing in not only language support for students (Chang, 2010), but also such activities as pedagogical training in EMI techniques for teachers along with infrastructure and support staff and mechanisms.EMI instruction requires an additional skill set of both theoretical and practical training.As Kyeyune (2003) writes: Teachers need to break away from teacher domination of classroom talk and the emphasis of subject content and adopt a language-or skills-based, communication-oriented bilingual approach to teaching.This approach recognizes student participation and teacher responsibility especially for bilingual language support for learning (p.183).
Similarly, Aguilar and Rodríguez (2012) argue that training which has been specially adapted to university teachers is necessary so lecturers can overcome their reluctance to methodological training and unlock the potential for EMI to be realized.Presently such professional development programmes appear to be ignoring the need for substantial EMI pedagogical components (Dearden, 2014).At the same time, institutional recognition opportunities (salary scale adjustments, promotions, development opportunities, etc.) need to be firmly in place as well to ensure those who are willing and able to teach the classes can and do teach them.While appreciation and understanding of the pedagogical requirements and complexities are gradually becoming clearer, institutional backing is essential in the success of such courses.Institutional support also means universities need to commit to the employment of dedicated administrative and language support staff to enable programmes to function properly and appropriate infrastructure, such as writing centres and self-access language facilities are essential requirements (Symon & Weinberg, 2013).
While small in scale, the results of this preliminary exploration paint a picture similar to that gradually emerging in institutions throughout Japan.With a predicted 67.3% (Dearden, 2014) of institutions worldwide expecting an increase in EMI in the future, universities need to seriously consider how best to create and organise such classes.Universities in Japan not involved in the G30 project or other Good Practice (GP) initiatives initiated by MEXT, are also equally likely to increase their number of EMI courses.While a large part of the aim of these courses is to contribute to the improvement of their students' English language abilities while at the same time opening the door to potentially greater numbers of international students, as has been explained, substantial preparation and full, institution-wide commitment is essential to enable EMI to live up to the lofty expectations placed upon it.

Table 1 .
Number of Japanese universities offering undergraduate classes in English

Table 2 .
Japanese students' reasons for enrolling in EMI classes (n=89)

Table 3 .
Japanese students' attitudes towards, and impressions of, the difficulty of EMI classes (n=76)