THE USE OF ENGLISH MEDIUM INSTRUCTION IN MULTILINGUAL CLASSROOMS IN JAPANESE LANGUAGE TEACHING

. Purpose. This article examines the use of English medium instruction (EMI) for teaching Japanese as a foreign language (JFL) by non-native speakers with a focus on multilingual classrooms. It also explores teachers’ positive and negative beliefs about using EMI in classroom settings for JFL instruction. Methods and procedure. 274 non-native Japanese language teachers from around the world (57 countries) voluntarily participated in a survey, answering a questionnaire on Google Forms and Jotform. Both qualitative and quantitative methods were employed. The research instrument was piloted before the main study and was found to be effective and adequate to elicit the desired data. The research questions aimed to identify whether there were any relationships between the use of EMI and the multilingual classroom. Findings. The results showed a clear correlation between the extensive use of EMI and the multilingual character of JFL classrooms. Translanguaging turned out to be a common practice adopted by non-native Japanese teachers regardless of the primary language of instruction. The article concludes that EMI can be a valuable tool for JFL instruction in multilingual classrooms. Based on the investigation of the teachers’ beliefs, the results showed a changing positive attitude towards English employed in JFL


INTRODUCTION
Recent studies justify using the first language (L1) and assert its positive impact in foreign language classrooms (Hall & Cook, 2012;Masuda, 2019).One common reason for the persistent avoidance of L1 in foreign language education is that students typically come from diverse cultural and linguistic origins and do not share a common first language.The growing status of English as a global language has nonetheless influenced some changes worldwide, and the issue of using it instead of L1 as a medium of instruction in multilingual classrooms has arisen.The study presented in this article was conducted on a supranational level and sheds light on the current state of EMI implementation in foreign language teaching, which, unlike other academic subjects (e.g.law, medicine and engineering), has been a rare field of academic inquiry so far (Bruen & Kelly, 2017).The study of English as a language to mediate the learning of the target language (Japanese) received very little attention in the past.
The field of applied linguistics is experiencing a transition when research on second language acquisition is gradually giving way to the study of multilingualism, encompassing the acquisition of third languages (Sanz & Cox, 2017).Multilingualism has an impact on foreign language learning.Teachers should understand the unique characteristics of multilingual learners to aid their language acquisition process (Alba de la Fuente & Lacroix, 2015).
In this article, a multilingual classroom is defined as one that comprises students from various cultural and native-language backgrounds (regardless of their level of proficiency) and where there is no one common mother tongue for teachers and all students.Multilingual classrooms are growing in number for various reasons, including geopolitical conflicts, war and environmental catastrophes.People are seeking better employment opportunities and higher living standards.
Although the prevalence of multilingual classrooms has been a significant trend in the United States, Canada and Australia for many years, it has recently spread to several EU countries (Brutt-Griffler, 2017).Overall, the earlier forms of EMI in Europe emerged due to an increasingly mobile European academic student and staff body (Aizawa & Rose, 2019).In a European Commission report entitled "Language Teaching and Learning in Multilingual Classrooms" (2015), the multilingual classroom is referred to as "a challenge" that education authorities in many parts of the EU have faced.Classrooms in tertiary education have become more linguistically diverse because of educational opportunities.Higher education establishments are interested in attracting more international students because of the accompanying financial benefits.Such classrooms have also become the norm rather than the exception in Europe due to globalisation, increased European mobility and international migration.These factors have increased the physical movement of people across national borders and influenced the JFL classrooms as well (Luchenko & Bogdanova, 2023).
EMI has become a rapidly expanding global phenomenon in all types and stages of education.Nevertheless, it remains a relatively new field of interest in academic research.In particular, the empirical studies that have collected data at the global (Dearden, 2014) or European (Brenn-White & Faethe, 2012) levels are very limited, and some studies have been conducted at the level of two countries (Galloway et al., 2010) or one country (Kapranov, 2021;Vural & Ölçü Dinçer, 2022).The highest proportion of investigations can be classified as "case studies of one institution".Thus, research in EMI is still young, despite the growing interest in the phenomenon (Macaro et al., 2018).
The language or medium of instruction may be the mother tongue of students, the official or national language of the country, an international language such as English or a combination of all of these (Peyton, 2015).In some countries, the term "medium of instruction" is also known as "the language of learning and teaching" (Heugh et al., 2019, p. 10).This article uses the notion of "English as a medium of instruction", which is defined as the "use of English to teach academic subjects (other than English itself) in countries or jurisdictions where the first language of the majority of the population is not English" (Macaro et al., 2018, p. 37).While L1 is a learner's native dominant language (mother tongue), L2 is the second or additional language that necessarily has a lower level of proficiency than L1.
In the study on the use of English in learning Japanese, Turnbull (2018) uses other terminology, such as "English as a lingua franca" (ELF), denoting communication held in English between speakers of different mother tongues, the majority of which are non-native speakers of English.Two paradigms, "World Englishes" and ELF, share similar ideologies when English can no longer be considered the property of native speakers, who constitute only a small minority of those who use the language.Therefore, the vast non-nativespeaking majority of English users are thus understandably entitled to their own ways of using English (Fang & Widodo, 2019).
According to Turnbull (2018), "very little research has investigated the role of ELF in other language learning environments, such as those in which Japanese is learnt as a second language in Japan".Turnbull's findings suggest that learners generally welcome English language use.He concludes that "learners seek security and comfort in what they already know, with ELF easing the gap between their L1 and their developing Japanese skills" (Turnbull, 2018, pp. 131-132).
Moreover, Bruen & Kelly's (2017) findings indicate that non-native speakers of English consider themselves to have an advantage over native speakers of English in studying JFL at an English-medium university.This is mainly due to non-native English speakers' extensive linguistic repertoire, which makes them experienced language learners.
Research studies on this subject have investigated how EMI affects students' motivation (Kojima, 2021).Other research findings showed no difference in whether students engaged in full or semi-EMI programmes; both were beneficial in motivating students and in linguistic outcomes (Ament et al., 2020).
This article undertakes a worldwide study on the use of EMI in teaching JFL and focuses on teacher-based linguistic practice.The study presented herein was entirely dependent on the profile of the teacher in question rather than on official programmes taught in English provided at different institutions.Partially, this can be explained by the fact that introducing EMI in subjects such as foreign languages is not common practice.Firstly, teaching using direct or indirect methods is somewhat controversial and contentious.Furthermore, the decision of whether to use a first language or English as a medium of instruction (or a mixture of both) can mostly depend on the teacher.
During the research, it was hypothesised that EMI had become helpful, particularly in multilingual Japanese language classes.This was the main correlation that was sought during the data analysis.
Therefore, the given empirical research aims to investigate the state of the implementation of EMI at different educational stages worldwide in Japanese language teaching by non-native speakers with a focus on multilingual classes.The article also aims to shed light on teachers' beliefs regarding the positive and negative aspects of EMI implementation.

METHODS
This research, which includes non-native Japanese teachers, was conducted within the scope of larger-scale work that also involved native Japanese teachers.During the project, we conducted a survey using a questionnaire sent to the official email addresses of Japanese-language institutions worldwide (The Japan Foundation, 2021).Each email contained a cover letter informing participants of the research aims and anonymity (in Japanese and English) with two Google Forms or Jotform links (for native and non-native Japanese teachers).All the teachers were asked to participate regardless of whether they used English or not.Additionally, the invitation to participate in the survey was shared with JFL teachers through professional networking groups on Facebook.
In the study frame, we identified the countries to be included in the current research as those where the majority of the population's native language is not English.Countries where English is the first language were excluded from further analysis.For this reason, we did not consider six responses received from Canada (n=2) and the United States (n=4).
The pilot survey was devised and conducted in August 2023 during O. Luchenko's training for Japanese language teachers at the Japan Foundation Japanese-Language Institute in Urawa, Japan.The survey was compiled after extensive consultation with Japanese language specialists at the institute.As a result, the questionnaire was tested by thirty-three representatives from twenty countries who responded to the pilot Google Forms.These responses were also included in the overall analysis.
Due to the presumably different teaching methods between native and non-native Japanese language teachers, these findings were presented separately.The present article has a sample size of 57 countries and 274 teachers.Participants in the present study were from countries and regions across the globe: Argentina, Armenia, Azerbaijan, Bangladesh, Belgium, Brazil, Bulgaria, Chile, China, Côte d'Ivoire, Croatia, Czech Republic, Denmark, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Honduras, Hong Kong, India, Indonesia, Italy, Japan, Jordan, Kazakhstan, Kenya, Kyrgyz Republic, Lithuania, Madagascar, Malaysia, Mexico, Mongolia, Morocco, Myanmar, Nepal, Norway, Philippines, Poland, Portugal, Romania, Slovakia, South Korea, Spain, Sri Lanka, Switzerland, Taiwan, Thailand, Turkey, Turkmenistan, Ukraine, Uruguay, Uzbekistan, Venezuela and Vietnam (The Japan Foundation, 2021).
The survey was split into three focus sections.The first section was devoted to the teacher's profile and background, focusing on their education and teaching experience.It asked about the respondents' country of origin, age, level of formal education, whether they had studied foreign teaching methods or linguodidactics, their form of employment, work experience, native language, Japanese language proficiency, knowledge of other languages and English language proficiency.
The second section, entitled "Teaching Environment", addressed questions on the country of teaching, the educational stage and institution where Japanese was taught, the subjects and levels taught, the status of the Japanese language in the institution (e.g. as a compulsory or as an elective subject) and whether the classroom was multilingual.
The third section was devoted to the usage of English as a language of instruction.The findings of the former two parts are primarily presented in a separate article; however, the data from those sections were used to find possible correlations for the current research.The questionnaire for non-native Japanese language teachers contained twenty-seven closed and open-ended questions, eight of which were exclusively used for the analysis in the present article.
Both quantitative and qualitative methods were used.For the quantitative analysis, the MS Excel statistical program was used as a tool to calculate the obtained data.The method of descriptive statistics was employed to summarise and present the data.As for the qualitative method, the method of subjective interpretation was applied.Two questions about the teachers' beliefs included predetermined variables for the possible positive and negative attitudes towards EMI as well as space for "freedom of expression".Short answers to the open-ended questions were analysed with a subjective interpretation and were added to the number of predetermined ones based on their similar meaning.Long answers and those that did not fit any predetermined variant were analysed through the content analysis relevant to the study's overall purpose.They are presented separately as additional comments in this article.

RESULTS
The participants were asked to identify the primary language of instruction in their Japanese language classroom.As the findings showed, only one-fifth of the teachers (21.90%, n=60) indicated the target Japanese language, while the majority (64.23%, n=176) stated L1, with only 13.87% (n=38) specifying English as the primary language of instruction (Figure 1).Being aware that the question regarding the primary language was not supposed to provide a comprehensive understanding of the state of EMI, we asked it to identify teachers' general preferences.In Japanese classrooms where most students are non-native English speakers, EMI is not used strategically as L1 but it is considered to be an additional help.Students are engaged in multilingual practices, switching between L1, English and Japanese.
As Turnbull (2018) states: "English plays a unique role in Japanese language classrooms as the 'universal' language that students are often expected to know.Many classrooms have adopted multilingual practices through the use of English in the teaching of Japanese" (p.133).This study addressed teaching the Japanese language through the full or partial use of English.The different degrees of intensity in EMI implementation received little attention in the past.Some researchers have differentiated it as full and semi-EMI instruction in their research, not in the context of one lesson but rather in the context of a course of study in different groups of students (Ament et al., 2020).
Essentially, EMI implies a monolingual approach to teaching and learning.However, as Dearden (2014) indicates, 76% of her respondents reported having no written guidelines specifying whether English should be the only language used in EMI classrooms or whether code-switching was forbidden, allowed or encouraged (p.25).
To investigate this in detail, the teachers were further asked about their experience with EMI and to estimate how much they used it in the teaching process (Figure 2).As the academic subject in question was a foreign language (L3), we knew that the percentage of English as the language of instruction could be partial; for the current research, 30% of use could be regarded as sufficient.When further discussing the variable of EMI use in JFL classrooms, the employment of English from 30 to 100% of instructional time will be considered significant.
Figure 2. The degree to which English is used as the instructional language in the Japanese language classroom (n=274).
Although most teachers suggested that the native language was the primary language of instruction, further inquiry revealed the role of English as a medium language.The majority of the teachers (52.92%, n=145) claimed that English was employed between 30 and 100% of the instructional time in their classroom; however, very few participants (6.57%, n=18) claimed that English was always or frequently used.The vast amount of the substantial usage of EMI (46.35%, n=127) thus falls into the category of "occasionally or sometimes" (30-50% of instructional time).
Interestingly, the analysis showed that among those respondents who stated L1 as the primary language of instruction, almost half of the teachers (47.73%, n=84) used English as an additional help from 30 to 50% of the time (Table 1).Over one-third of the participants (38.33%, n=23) who stated Japanese as the primary language of instruction used English as an additional language to the same extent.We cannot confirm the existence of a completely English-based learning environment in the classrooms where English was claimed as the primary language of instruction.English was used by most of the participants (52.63%, n=20) between 30 and 50% of instructional time.
Our research showed that translating or code-switching between Japanese, English and L1 in JFL classrooms is a common practice.Based on the teachers' comments, this phenomenon is mainly observed when the meaning of words (subject-specific English terminology) or grammar patterns are explained and when students' comprehension is checked.Translanguaging has several potential educational advantages.Two of them, mentioned by Baker (2001), a leading scholar in the field, are "promoting a deeper and fuller understanding of the subject matter" and "helping in the development of the weaker language" (p.281).Galloway et al. (2020) highlight that the EMI approach does not have to be a monolingual endeavour: "It is hoped that the flourishing research of language use in the EMI classroom will showcase the valuable use of translanguaging" (p.410).
Teachers in most multilingual countries use code-switching when explaining concepts and information to students in the classroom, but they feel guilty about doing so (Heugh et al., 2019).In the comments to the questionnaire, many participants were willing to know the results of the present study.We can suppose most of them wanted to validate their own teaching methods by comparing them with other teachers' practices and beliefs.
The research results showed that over 40% of all 274 JFL teachers taught in multilingual classes (Luchenko & Kovinko, 2024).The findings demonstrated a clear correlation between the use of EMI in multilingual classrooms where English is used by the vast majority (68.47%) of the respondents (n=76) in a range between 30 to 100% of the instructional time, which is a quarter (26.14%) higher than that in non-multilingual JFL classrooms (Table 2).Those teachers who confirmed using English to a certain degree (from 10% and more) were asked to express the advantages and disadvantages of the EMI approach.There were five predetermined options with the possibility of multiple answers.The question about the advantages of EMI received 195 responses and comments out of 211 teachers.Sixteen teachers refrained from answering the question (Figure 3).

Figure 3.
Advantages that teachers find in using the EMI approach while teaching Japanese (n=195).
As can be seen from Figure 3, the most common function for employing English in Japanese language classrooms was to present English equivalents of grammar patterns and of words or word combinations.Among the received qualitative comments, five teachers believed that English helped to differentiate the meaning of words and grammar patterns; this is why they used it for comparison to give an additional variant when the meaning is unclear in L1.One participant commented in favour of EMI: "It is used whenever required to bring a better understanding of the grammar."Another positive comment was: "Students understand the nuance of the Japanese language better when comparisons are made in English and Japanese."This positive feedback confirms the beneficial function that was earlier mentioned: "The use of the dominant L2 allowed students to build bridges between the vocabulary and grammar of Japanese, English and their native language to learn new content effectively" (Turnbull, 2018, p. 143).The participants' comments suggested that even those teachers who prefer to use only Japanese resort to EMI for clarification in order to build the connection between the known and unknown.
Two teachers mentioned that they mainly used English to explain loan words (gairaigo of English origin written in katakana).Two other opinions highlighted the advantages of EMI in multilingual classes, especially when the students' Japanese level was not high enough, namely, beginners and elementary (shokyū).
In favour of catering to the students' needs, one respondent stated: I use English to lessen the cognitive task demanded of the learners when they must memorise new words and grammar patterns.Also, since I am not a native speaker of Japanese, it is easier for me, as an instructor, to communicate in a language that is comfortable in a learning environment.
The idea of reducing the cognitive load on learners was supported by another comment: "Students feel comfortable if English is initially used in the classroom at the beginner's level."The importance of the capacity to understand students' needs is crucial for teachers.As Lin & Lo (2018) highlight in their study on the spread of EMI programmes in the Southeast Asian context: "As the teacher's knowledge base also includes knowledge of learners, effective EMI teachers should have the ability to empathise with their students" (p.92).
Although generally in favour of the direct teaching method, one participant commented: In today's universities, class time is limited, and the number of students exceeds thirty per class.The senior faculty members with long tenures frequently use "translation" as a learning tool.We face the practical problem of curriculum design in which subjects such as reading and writing are taught as separate classes.Under these circumstances, the frequency and proportion of EMI will inevitably increase somewhat.
As Figure 3 shows, almost half of the teachers favoured EMI for its benefit of developing two foreign languages simultaneously for students.This positive attitude was closely followed by the advantage of being helpful in multicultural classrooms and promoting students' mobility.The results showed that "improving English ability" is one of the top positive aspects of implementing the EMI approach, even though it is not a primary goal in an L3 foreign-language class.As Goya (2020) concludes: "Considering the recently globalised world, it is inevitable that people will need to communicate through a common language that is usually different from their native language" (p.79).
Over one-third of the respondents also stated the availability of better-quality Japanese-English language resources and the lack of those in L1 in various countries as reasons for adopting EMI to a certain extent in teaching Japanese.When asked if they used Japanese-English textbooks or supplementary teaching materials, most teachers (56.57%, n=155) answered positively.One participant mentioned recommending English-based textbooks to those students who have difficulty catching up.Another participant used a lot of what was referred to as "interesting" supplementary English materials available on the Internet.Some respondents used resources in English to prepare for their lessons rather than instruct students.Interestingly, our findings show that 44.19% (n=57) of those who stated that they "almost never" or "seldom" employed EMI in their classroom (n=129) used Japanese-English textbooks in their practice.
One of the predetermined statements, which turned out to be the most debatable, was that learners concentrated more on foreign languages when EMI was employed.Only one-fourth of the teachers supported this idea.
Conversely, the question about the negative aspects of EMI received 198 responses and comments.Thirteen teachers refrained from answering this question.
As Figure 4 demonstrates, more than half of the respondents indicated the first two statements as their primary concern, namely, insufficient English language proficiency of students as well as the absence of direct equivalents to some words and grammar patterns.Most research studies that focused on instructors' views regarding EMI have found that inadequate English language proficiency among students is a significant challenge for teachers conducting EMI classes.Borg (2016) found that beginner or elementary English levels were insufficient for studying an academic subject in English and argued that the effective implementation of EMI requires intermediate or upper-intermediate levels of English proficiency.
Having analysed the participants' qualitative comments, we can conclude that the students' poor English proficiency prevented many respondents from using EMI to the desired extent.Such comments were received from France, Italy, Taiwan and some Spanish-speaking countries.This overall situation regarding the level of language proficiency contrasts with that in other commentaries from Romania, Slovakia, Turkey and (expectedly) India, where the teachers stated their students' level to be sufficient (B1-B2).Nonetheless, students' relatively low level of English did not prevent the teachers from employing it occasionally.One participant commented: "I think it can be beneficial to students here and there.That is why I push English words along with their Japanese equivalents."Besides the disadvantages mentioned in the figure above, some participants commented that it was better for them to explain grammar in their native language due to existing similarities with Japanese or simply because it is easier than doing so in English.
Although English often functions as a medium of instruction in India, starting from secondary education, it is not necessarily the primary language of instruction for Japanese lessons.One comment from a participant stated: Typologically, English is a subject-verb-object language, whereas Japanese and languages in India are subject-object-verb languages.They share many language universals.Word order and many phrases are closer to that of Japanese.Many cultural concepts are similar; therefore, students are encouraged to think in their respective Indian languages while translating or forming sentences in Japanese.
Similarly, in his study on L3 syntactic transfer selectivity and typological determinacy, Rothman (2010) concludes that typological proximity is the most decisive factor determining multilingual syntactic transfer and that L3 transfer is driven by the typological proximity of the target L3 measured against the other previously acquired linguistic systems.Furthermore, in the participants' qualitative comments, the role of the native language and the teacher's general preference for it were highlighted based on the similarities between L1 and L3: "Teaching Japanese to students who speak Finnish as their native tongue is very rewarding because there are so many similarities in the languages.The students understand the structure of the Japanese language easily, and their progress is very quick."Using the native language instead of EMI is also favoured in universities where Japanese is studied as a major.As students are supposed to become specialists in translating, it helps to develop their translation skills.
Although many teachers preferred to use a mixture of three languages (the target Japanese, L1 and English), some comments mentioned four languages being used.They divide explanations so that grammar is taught through the vernacular languages, while English is used for memorising words (primarily nouns).This was followed by another concern: "Confidence in speaking Japanese is reduced if English is primarily used from the beginning.Students tend to translate English into Japanese in their minds.The transition to speaking reasonably in Japanese will take many years."Another comm ent supported the idea that direct translation from English has disadvantages: "Students tend to rely more on translating sentences in English into Japanese to express themselves, which often results in students using incorrect Japanese." There was a concern regarding the explanation of grammar: "The English explanation of such grammar patterns as a causative form (shieki), giving and receiving (yarimorai), and transitive/intransitive verbs (jidōshi/tadōshi), can be a bit difficult for students."One teacher did not find it helpful to explain Japanese grammar via English "because the languages are so different".Another respondent assumed: "Students can become over-dependent on English as a language medium, especially at N2 [upper-intermediate] and N1 [advanced] levels."One teacher stated that "learners' Japanese proficiency is slowed down tremendously."Two participants could not find any specific disadvantages but said teaching only in Japanese would be better, even with beginner students, adding that clarification could be made with the help of gestures and picture cards.
Interestingly, among all the responses, there were no concerns that English might pose a threat to the home language and substitute it, as this was one of the concerns for rejecting EMI in some countries mentioned in other studies (Dearden, 2014).

DISCUSSION
Carrying out this large-scale research remotely via email had a number of challenges, such as the participants' misunderstanding of the key notions of EMI and multilingual classrooms, restricted access to Google Forms in China and, on top of that, a low response rate.The latter issue necessitated sending emails twice as a reminder to all the countries.Out of 141 countries and regions implementing Japanese-language education overseas, we contacted as many as 122 countries and all the registered Japanese language education institutions in those places (The Japan Foundation, 2021, 2023).
From the beginning of our research, in August 2023, when the pilot survey was conducted, we became aware of its complexity.During that period, if inconsistencies were found in the responses to the questionnaire, we had the opportunity to ask the corresponding participants for subsequent short interviews in order to identify flaws in the composed questionnaire and to minimise the possibilities of a misleading interpretation of the questions (as was the case in understanding the term "multilingual class").
Due to the discussion around the research topic, we were interested in learning teachers' attitudes and perceptions towards EMI practice.We included open-ended questions to find out the benefits and disadvantages of EMI as well as the overall beliefs of the respondents, who were also allowed to openly express their opinions on the use of EMI in Japanese language classrooms in the space provided at the end of the questionnaire.We received extensive feedback in the form of commentaries from the Japanese language teachers.
As a result of our study, we identified some positive attitudes towards EMI among Japanese language teachers.The majority of the non-native Japanese language teacher respondents held the belief that it was helpful to provide equivalents of grammar patterns and words in English.Therefore, English is used to enhance input morphologically, syntactically and lexically.Promoting students' mobility and being helpful in multicultural classrooms were also two of the top-mentioned benefits.Approximately the same ratio of the teacher respondents believed there were also instrumental advantages to studying through the English medium and that it was beneficial for students to develop two foreign languages simultaneously.Compensation for lack of resources in L1 was also mentioned as a positive motivation for using EMI.One more long-term positive impact on content learning was found: code-switching between three languages (L1, English and Japanese) minimises the misunderstanding of the content students attempt to learn.The likelihood of poorly understood content is higher when it is learnt through Japanese only.
Overall, the respondents had mixed opinions regarding the usage of EMI in Japanese language classrooms.Interestingly, the participants from countries traditionally speaking English as a second native language, such as India, stated that they found it more beneficial to use their first native language (i.e.Hindi or Marathi) for instruction because of more grammar similarities; indeed, English was seen as "just a mode of communication".On the other hand, the participants who teach in countries where students have a less proficient command of Englishsuch as Brazilpointed out its usefulness as L2 in learning Japanese as L3.Remembering their student experience, one B1-level participant commented that "it was easier to study Japanese through English because it helped to comprehend the grammar better."These positive attitudes towards translanguaging discursive practices were mentioned in the previous studies (Baker, 2001;Turnbull, 2018;Galloway et al., 2020).
The primary language of instruction in foreign language teaching has always been a topic for discussion.The analysis of the views expressed by the participants showed an interest in the results of the current study.Moreover, some teachers were willing to know whether their methodology in EMI usage was approved by others or not.
A limitation of the study is that the sample could not be carefully chosen and consisted of participants who voluntarily answered the questionnaire.It cannot be regarded as being completely representative of the overall population of non-native Japanese teachers.As the present study aimed to explore teachers' perspectives, further empirical research is required to investigate students' perspectives in different geographical areas.

CONCLUSIONS
The significant number of the responses from different geographical settings allowed us to draw conclusions that might not otherwise be possible in smaller-scale studies.We could find the use of EMI by non-native Japanese language teachers in all areas of the world, although there are a few countries where it is less prominent.We proceeded from the fact that the implementation of EMI in JFL classrooms was mainly driven by teaching staff rather than institutional-level policy.
Based on the data analysis, we determined strong correlations between the extent to which EMI was employed and the multilingual classroom.Our findings suggest that EMI is particularly helpful in multilingual classrooms where most teachers employ it substantially.The results showed a significant difference in the amount of instructional time that English was employed in multilingual and non-multilingual classrooms.Even though the teachers are non-native English speakers, most of them support using English because it helps students to better understand the third acquired language.English began to be perceived as a transitional instrument or a bridge connecting the native language and the target language.
In JFL classrooms, the target Japanese language skills are developed through translanguaging (interpreting, translation and code-switching), which is a widespread method that instructors employ to promote a more profound and fuller understanding of the subject.This is what distinguishes the EMI approach in JFL classrooms from the monolingual EMI approach employed in other academic subjects.

Figure 1 .
Figure1.The primary language of instruction in classes of non-native Japanese teachers (n=274).
of good teaching materials in English Benefiting students' mobility and multicultural classrooms Developing two foreign languages at the same time Providing equivalents of grammar patterns and words

Figure 4 .
Figure 4. Disadvantages that teachers find in using the EMI approach while teaching Japanese (n=198).
distracted from learning Japanese itself Learning two foreign languages at once can be challenging Teachers must be proficient in both English and Japanese No direct equivalents for some words and sentence patterns Students' level of English can be insufficient for using EMI Advanced EducationISNN 2410-8286 (Online)

Table 1 .
The degree to which English is used with different primary languages of instruction (n= 274)

Table 2 .
The degree to which English is used in multilingual and non-multilingual classrooms (n=274)