‘So, only relying on English is still troublesome’: a critical examination of Japan’s English medium instruction policy at multiple levels

ABSTRACT

Drawing on neoliberal ideology as a theoretical lens, this study critically examines how neoliberal ideological assumptions shape the interpretation and implementation of TGUP (Top Global University Project) as an English-medium instruction (EMI) policy in the Japanese context at multiple levels of government, universities, and multilingual international students. We examined documents describing TGUP policies published both by the government and TGUP universities, and then contrasted these policies with interview data drawn from conversations with three key stakeholder groups (international students, EMI faculty, and administrative staff) in four focal universities. The findings reveal the conflicts of policy making at different levels and challenge the English-only ideology in EMI practices. By taking a critical language policy approach, the study offers a nuanced understanding of how neoliberal ideology constitutes a policy-to-practice gap in language policy and planning. We argue that EMI should not be reduced to English-only medium of instruction, but rather treated as an alternative approach to the internationalization and globalization of education, and understood as a learning space where multilingual students enhance their learning by relying on their shared language repertoires.

KEYWORDS:

• Top Global University Project (TGUP)
• English medium instruction (EMI)
• Japan
• Chinese students
• neoliberal ideology
• internationalization of higher education

Introduction

English-medium instruction (EMI) has become popular in a global climate where English is regarded as the most widely used academic language (Jenkins and Mauranen 2019). EMI is defined as ‘the use of English language 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 2018, 19). Driven by the internationalization of higher education (IHE), universities around the world see EMI as a strategy to increase their standing in global university rankings as well as to attract international students and their attendant monetary benefits (De Costa, Green-Eneix, and Li 2022; Macaro 2018). In Europe the number of EMI programs increased by 1115% from 2001 to 2004 (Wächter and Maiworm 2014), and similar growth has occurred in Asian countries such as Saudi Arabia (Phan and Barnawi 2015), Nepal, India, and Pakistan (Sah 2022), China (Song 2021; Yuan and Li 2021), and South Korea (Piller and Cho 2013; Shin and Park 2016). In particular, EMI has been expanding rapidly in the Japanese context, with almost 40% of Japan’s universities offering undergraduate EMI courses in 2015 (MEXT 2017), and it has attracted much scholarly attention (Aizawa and Rose 2019; Aizawa et al. 2020; Bradford 2019; Galloway, Kriukow, and Numajiri 2017; Galloway and Ruegg 2022; Rose and McKinley 2018). Over the last two decades the Japanese government has issued a series of higher education internationalization policies such as Global 30 (G30) and the Top Global University Project (TGUP) (MEXT 2014), which can be seen as a major impetus for the booming of EMI programs in Japanese universities.

Despite the wide spectrum of research examining the implementation of EMI policies and programs, De Costa, Green-Eneix, and Li (2022) pointed out in their recent review that a central question remains underexplored, namely, how the implementation of EMI policies at the institutional level translates the aims and objectives of national language policies and subsequently affects stakeholders such as students and teachers. In addition, many previous studies have adopted an uncritical stance towards the development of EMI, valorizing it as a ‘success of transnational ventures’ (Kosmützky and Putty 2016, 22). These studies tend to overlook the hidden ideology and asymmetric power relations when EMI is implemented as a top-down language-in-education policy (De Costa, Green-Eneix, and Li 2022; Kuteeva 2020), and thus position international students, who are often non-native English speakers, as vulnerable English language learners (Ou and Gu 2021). Market-driven neoliberal ideology is a key underlying force boosting the proliferation of EMI, particularly in non-Anglophone contexts (Phan and Barnawi 2015; Piller and Cho 2013; Tollefson and Tsui 2018), but scant research has been conducted to look into how such key neoliberal constructs as language marketization, commodification, and instrumentalism (Holborow 2015; De Costa, Park, and Wee 2019, 2020; Kubota 2011, 2016) influence policy implementation.

Given these issues, we use Japan as a case in point to critically examine the implementation of the country’s most recent EMI initiative, the Top Global University Project (TGUP). Using the neoliberal ideology as our theoretical lens, we aim to unveil how the neoliberal interpretation and implementation of TGUP as an EMI policy at the national and institutional levels subsequently influence multilingual students’ learning experiences at the individual level. The present study aims to add to the empirical findings to language-in-education planning in multilingual and transnational contexts at multiple levels and to provide insights into EMI policies from a political economic perspective. Findings may also generate implications for language policymakers to make informed decisions for more effective EMI planning and implementation in non-Anglophone contexts.

Literature review

Neoliberal ideology as a language ideology

Neoliberalism as a political economic concept features the ‘philosophy of sustaining entrepreneurial and competition-seeking practices under the umbrella of free markets’ (Phan and Barnawi 2015, 546). Neoliberalism is explicitly laissez-faire in nature (Piller and Cho 2013), aiming to construct an unregulated market with the minimum governmental regulation needed to ensure the mobility of goods and services for maximum economic profit (Holborow 2015). From a linguistic standpoint, neoliberal free marketization views English as the most efficient language for commercial, scientific, and cultural exchange in cross-national contexts (Heller and Duchêne 2016; Shin and Park 2016; Piller 2015).

Together with globalization, neoliberal marketization has permeated education through the agenda of internationalization and framed English as the most useful language for knowledge communication (Zheng and Guo 2019; Piller and Cho 2013; Tollefson and Tsui 2018). As a social discourse, neoliberal ideology reinforces an instrumental approach to language learning and constructs language as a commodity in the global knowledge economy (Kubota 2011, 2016). This trend is particularly prominent in Japan, where internationalization has been articulated as a crucial component of Japan’s social, cultural, and economic reform (Hashimoto 2010, 2011). Since the 1990s the government has launched initiatives aiming to develop individual Komyunikeshon noryoku [communication competence] (Hashimoto 2011). English competence is valorized as a means to promote economic competitiveness in the neoliberal linguistic market (Kubota 2011; Kubota and Takeda 2021), and individuals are regarded as bundles of skills (Urciuoli 2008) whose language competence, usually English, is seen as linguistic capital that can be converted to economic value (Heller and Duchêne 2016). Neoliberal commodification turns individuals into entrepreneurial selves (Li and Zheng 2021; De Costa, Park, and Wee 2019, 2022) who shoulder the responsibility and obligation to improve their English proficiency to enhance their self-worth.

Under the pressure of neoliberal internationalization, universities in Japan face fierce competition in global ranking mechanisms which favor knowledge production in English (Zheng and Guo 2019; De Costa, Green-Eneix, and Li 2022). To increase Japanese universities’ competitiveness in the global rankings, over the past 20 years the Japanese government has launched a series of initiatives to create a Westernized system of higher education with a heavy emphasis on English proficiency (Bradford 2019; Hashimoto 2011; Galloway, Numajiri, and Rees 2020). However, neoliberal knowledge production and distribution in English has also been perceived as a new form of colonization that hampers knowledge production in local languages, including Japanese (Qiu and Zheng 2021; Kubota 2016; Phipps 2019; Piller, Zhang, and Li 2020; Sah 2022; Song 2021). In short, the status of English in Japan is inextricably linked with the neoliberal valorization of communicative competence in English in the global neoliberal market (Kubota and Takeda 2021). This prioritizing emphasis on English implicates a monolingual orientation to IHE in Japan, and thus warrants the adoption of a neoliberal perspective to examine the implementation of Japan’s EMI policy.

EMI research in the Japanese context

Japan’s Ministry of Education, Culture, Sports, Science, and Technology (MEXT) has launched a series of policies and strategies such as the Go Global Japan project and the Global 30 (G30) Project, in order to promote the internationalization of its higher education; these have been well summarized by previous scholars (e.g. Aizawa and Rose 2019; Bradford 2019; Galloway, Kriukow, and Numajiri 2017; Macaro 2018; Rose and McKinley 2018). As a result of these policies, around 800 universities, more than one third of all Japanese universities, were offering undergraduate EMI programs by 2015 (MEXT 2017). The most recent EMI policy, launched in 2014, is the Top Global University Project (TGUP). As MEXT’s most heavily funded initiative to date at $77 million USD, TGUP selected 37 Japanese universities: 13 universities were designated as Type A (aiming to rank among the world’s top 100 universities) and 24 as Type B (expected to lead the internationalization of Japanese society). TGUP has led to a significant increase in the number of EMI programs in Japanese universities (Rose and McKinley 2018), meaning that the English-medium language policy is intricately intertwined with the globalization of Japan’s higher education.

Rose and McKinley (2018) compared TGUP to previous Japanese IHE policies and found that TGUP is ‘a cumulative effort of prior international projects’ issued by MEXT that also features a new understanding of language policy and planning (Rose and McKinley 2018, 120). While previous internationalization projects in Japan, such as Global 30 (G30), mainly targeted English medium education, TGUP has been reworded to emphasize education in all foreign languages. In addition, TGUP features the dual goals of achieving the internationalization of higher education in Japan at both the domestic and international level. In a follow-up study, Aizawa and Rose (2019) focused on one university and examined the meso-level TGUP goals in combination its micro-level practices, revealing a discrepancy between policy interpretation at the meso- and micro-levels.

With regard to the implementation, researchers have stressed the language-related challenges faced by students enrolled in EMI programs (e.g. Aizawa and Rose 2019; Aizawa et al. 2020; Bradford 2019; Galloway, Kriukow, and Numajiri 2017). For example, students perceived their teachers’ insufficient English abilities as an obstruction to effective teaching in the EMI classroom (Galloway, Kriukow, and Numajiri 2017), and they experienced challenges in listening, speaking, reading and writing in their EMI learning (Galloway and Ruegg 2022). Although not many EMI instructors deemed ‘native’ English proficiency to be essential to successful EMI teaching, a native-English speaker is still seen as an ‘ideal’ EMI instructor (Galloway and Ruegg 2022), suggesting a tendency towards ‘over-problematiz[ing] linguistic issues in EMI without considering pedagogical solutions’ (Bradford 2019, 715).

Despite the fruitful findings of EMI research in Japan, there seems to be an absence of any critical stance in these studies. As already pointed out by Bradford (2019), overemphasizing English-related language difficulties may hinder faculty engagement in EMI teaching. However, a critical perspective is necessary in order to closely examine how ideological assumptions such as native-speakerism and language commodification affect key stakeholders’ EMI learning experiences. In addition, despite the recent reconceptualization of EMI as a multilingual site (Blair, Haneda, and Bose 2018; Galloway, Kriukow, and Numajiri 2017; Yuan and Li 2021), previous research has seldom framed international students as multilingual speakers, or discussed how multilingual resources may be capitalized to enhance EMI learning (Bradford 2019). Considering that TGUP policies aim to attract international students as a response to Japan’s stagnant economy and aging population (Hashimoto 2011; Kubota 2011; Macaro 2018), neglecting to view international students as multilingual speakers may overlook their difficulties in gaining equal access to educational resources in the context of transnational higher education.

To address these two points, the present inquiry aimed to answer the following overarching question:

RQ: How did the neoliberal interpretation and implementation of Top Global University Project (TGUP) EMI policies at the national and institutional levels subsequently influence the individual level?

Methodology

Following the three-level conceptualization of Japan’s EMI policy implementation proposed by Rose and McKinley (2018) and Aizawa and Rose (2019), the present study also considered the macro-meso-micro levels of EMI policy implementation by drawing on three main sources of data: (1) policy documents collected from the official TGUP website; (2) policy documents collected from the public websites of all 37 TGUP universities; and (3) semi-structured interviews with 13 international students, two international student management administrators, and one professor from four focal TGUP universities, as illustrated in Table 1. We also collected the students’ reading/lecture notes to further facilitate our understanding of their learning experience.

Table 1. Data sources and data collection at the macro, meso- and micro-levels.

Macro-level (national)	Meso-level (institutional)	Micro-level (classroom)
MEXT’s TGUP policy documents collected from the TGUP website (131,875 words in total)	policy documents from the public websites of all 37 TGUP universities semi-structured interviews with two international student management administrators from focal TGUP universities	semi-structured interviews with 13 students and one professor in EMI degree programs at four focal TGUP universities reading/lecture notes of Chinese students

The study was conducted in two phases. During the first phase, we collected documents (in both English and Japanese) relevant to the TGUP initiative and EMI programs from the official TGUP website. We then collected documents (in both English and Japanese) from the public websites of all 37 participant universities, comprising more than 131,875 words of documentation in total. We analyzed these documents in terms of how aims and objectives related to the internationalization of TGUP were outwardly communicated. The data were coded using qualitative content analysis, also known as thematic qualitative text analysis (Schreier 2014), a technique that has also been used in similar policy analysis studies (e.g. Rose and McKinley 2018).

The second phase of the study narrowed down to focus on the actual implementation of EMI policies. Semi-structured interviews with students, faculty members, and administrators were conducted to investigate how TGUP universities implemented EMI policy at the meso- and micro-level. We recruited 13 international students from four focal TGUP universities located in Tokyo through snowball sampling. The four universities included both types of TGUP universities (Type A and Type B). They all have a long history of EMI education, and offered a wide variety of EMI programs attracting a large number of international students. Our participants were enrolled in various disciplines and came from different educational backgrounds, which increased the diversity of our data (see Table 2 for more information). To triangulate the policy document and student interview data, we also interviewed other key stakeholders from the focal TGUP universities, including two Japanese administrators responsible for international student management and one senior Japanese professor teaching EMI courses (see Table 3 for more information).

Table 2. Student participant demographics

Participant	Gender	Year	Discipline	Language proficiency
Ai	Female	M2	Intercultural Communication	Chinese English (TOEFL 94) Japanese (N2)
Bao	Female	D1	Linguistics	Chinese English (TOEIC 920) Japanese (N1)
Che	Male	M3	International Relations	Chinese English (IELTS 7.0) Japanese (N2)
Deng	Male	D3	Architecture	Chinese English (IELTS 6.5) Japanese (elementary)
Fang	Female	M2	International Culture	Chinese English (TOEFL 110) Japanese (N1)
Gao	Female	M2	Intercultural Communication	Chinese English (IELTS 7.5) Japanese (N2)
Hu	Female	D3	Medicine/ Anatomy	Chinese English (TOEFL 70)
Jin	Male	M2	Chemistry/ Systems Engineering	Chinese English (TOEFL 88) Japanese (N1)
Kun	Female	M2	Economics	Chinese English (TOEFL 95) Japanese (N1)
Li	Male	D1	Economics	Chinese English (TOEFL 100)
Ma	Male	M2	Economics and Finance	Chinese, English (TOEFL 110) Japanese (N1) Korean (advanced)
Ni	Male	D3	Linguistics	Chinese English (TOEIC 920) Japanese (N1)
Pang	Female	D4	Applied Mathematics	Chinese English (IELTS 6.0) Japanese (N2)

Notes: M = Master program, D = Doctoral program; N1: highest level of the Japanese Language Proficiency Test (JLPT); N2: second-highest level of the JLPT

Table 3. Administrator and professor participant demographics

Participant	Gender	School
Administrator T	Female	International Culture and Communication Studies
Administrator S	Male	Economics
Professor K	Female	International Culture and Communication Studies

It should be noted that all our student participants came from China. The recruitment of Chinese international students enrolled in Japan’s EMI programs was motivated by two reasons. First, Chinese students constitute the largest group of international students (40%) in Japan (JASSO 2019), so a close examination of their experiences in EMI programs in Japan may generate plenty of insights for a better understanding of the implementation of EMI policy in the Japanese higher educational setting in general. Second, as Chinese scholars with academic experience in Japanese higher education, we felt that we could mobilize our own knowledge of both the Japanese community and the Chinese educational context; our familiarity with both contexts might position us as compatriots and encourage the participants to share their experiences candidly. Furthermore, our shared background might facilitate researcher-participant co-construction of knowledge (Talmy 2010) and help us to gain a more in-depth understanding of the participants’ experiences in Japan’s English-medium instruction programs.

We used pseudonyms for all the participants. Face-to-face interviews with the 12 student participants were conducted in an office or in public spaces such as coffee shops; one participant, Ni, was interviewed over WeChat, a popular social media platform in China. All the interviews were conducted in Chinese Putonghua, the participants’ and interviewer’s mother tongue, and were audio recorded on site. Individual interviews lasted between 23 and 70 min, with an aggregate of 552 min. The core prompts during each interview were: (1) the participant’s background information; (2) the institutional context of the participant’s EMI program; (3) the participant’s linguistic challenges; and (4) the participant’s overall evaluation of their EMI programs (e.g. goals, language requirements, language support). Online interviews with a faculty member and two administrators were conducted in English. The core prompts for the administrators centered on the universities’ language requirements during the recruitment process and institutional language support. The core prompts for the professor were mainly about her attitudes towards the EMI program, the language difficulties faced by international students in her classes, and the support she offered to help international students overcome language barriers.

We adopted a qualitative approach to analyze the interview data. We first transcribed the interview data verbatim, and then confirmed the transcripts with the interviewees to ensure their trustworthiness. We then analyzed the data using an iterative process of inductive and deductive reasoning (Merriam 1998) using NVivo 12.0. We first categorized the data into ‘English learning’, ‘foreign language learning’, and ‘Japanese learning’, and then referred to the key neoliberal constructs of ‘bundle of skills’, ‘language, commodification’, ‘language marketization’, ‘language instrumentalism’, and ‘native-speakerism’ for pattern coding (Miles, Huberman, and Saldaña 2014). A Kappa’s alpha of .83 was obtained for inter-rater congruence.

Findings

In this section we present the findings from the three data sources under the following main themes: (1) multilingual vs. monolingual orientation; (2) conflicting views of the local language in EMI; (3) language learning as self-worth improvement; and (4) students’ strategic responses: only replying on English is troublesome.

Multilingual vs. monolingual orientation.

Multilingual re-orientation at the national level

MEXT’s TGUP policy at the national level has set up a total of 16 goals relating to three aspects: internationalization, governance, and educational reform. An analysis of MEXT’s TGUP policy shows that the Japanese government carefully reworded the language aspect of internationalization as ‘foreign languages’ instead of ‘English’ at the national policy level.

1. Among the 10 goals relating to the internationalization of TGUP, three explicitly emphasized the importance of foreign language education or education through foreign languages: Goal 5: ‘Increase the number of subjects taught in foreign languages’; Goal 6: ‘Increase the number of students enrolled in degree courses conducted in foreign languages only’; and Goal 7: ‘Increase the number of students who meet the standards of proficiency in foreign languages’.¹ Only one goal mentions English: Goal 8: ‘Develop English syllabi’.

2. In the three goals relating to governance, the only goal to mention language proficiencies was Goal 3: ‘Increase the number of administrative staff who meet the standards of proficiency in foreign languages’.

3. Among the three goals relating to education reform, Goal 2: ‘Introduce external examinations such as TOEFL in undergraduate entrance examinations’ explicitly designates a standard English proficiency test to measure international students’ language capacities. This point echoes the previously reported position that measurable English skills are an important indicator of individuals’ value in the neoliberal education market (Heller and Duchêne 2016; Holborow 2015).

Unlike the previous Global 30 policy, which was widely criticized for its overemphasis on English to achieve internationalization (Rose and McKinley 2018), TGUP shifts its language focus from English to foreign languages and aims to build heterogeneous communicative competence to achieve internationalization. This policy reorientation indicates a multilingual turn at the national-level policy making. However, there still seems to be an emphasis on English proficiency.

English-monolingual orientation at the institutional level

Policy analysis of the TGUP participant universities’ project-related documents showed that 11 of the 37 universities participating in TGUP stated in their policy documents that they would like to promote linguistic and cultural diversity. They underscored foreign language education rather than English education, using ‘foreign language’ and ‘diversity’ as key words when stating the objectives of their language policies. MEXT worded the TGUP policy to encompass all ‘foreign languages,’ and the multilingual emphasis in these universities’ policies responds directly to MEXT’s macro-level call and demonstrates how institutions’ language beliefs come to be influenced by and aligned with national interests (Block 2017).

In particular, heterogeneity was mostly emphasized by Type A universities seeking a leadership position in the international knowledge community. For example, the International Christian University (ICU) claimed that they would have ‘no ‘English only’ or ‘Japanese only’ program[s]. ICU promotes true globalism,’² a statement that challenges the neoliberal ideology wherein English specifically represents globalization (Jenkins and Mauranen 2019). Guided by this multilingual belief, ICU employed a ‘two plus one’ language teaching model, involving English and Japanese plus another language. Similarly, the Tokyo University of Foreign Studies stated that ‘[m]ultilingual global human resources refers to personnel who have deep knowledge of not only English but also a highly diverse range of languages, cultures, and communities that make up the real society and are able to take effective action in the era of globalization.’³

In contrast, 17 of the 37 TGUP universities (mostly Type B universities) tended to conflate the internationalization of higher education with English education by increasing the number of EMI programs offered and increasing the presence of English on campus, despite MEXT’s reorientation towards ‘foreign languages’ at the national level. For example, Kumamoto University sought to teach 50% of its liberal arts courses in English,⁴ while the Tokyo Institute of Technology claimed that ‘[b]y 2019, most graduate courses will be conducted completely in English.’⁵ Their interpretation of the goal of TGUP seems to encapsulate the market-oriented philosophy of neoliberal ideology in Japan’s domestic internationalization. It seems that this emphasis on EMI for domestic students in order to educate them to become global talents (jinzai) was a consequence from the G30 project (Bradford 2019), which is also associated with the position of Type B universities in the TGUP plan as universities to lead the internationalization of Japanese society (MEXT 2014).

Although multilingual and monolingual orientations co-exist in the meso-level interpretation of TGUP policy, the actual implementation of TGUP initiatives is almost entirely monolingually English. We found that although a few Type A universities such as the University of Tokyo offered some degree programs in languages other than English, most of the TGUP universities exclusively provided EMI programs where English proficiency was positioned as a key factor in students’ academic success. This emphasis on English is clearly illustrated by Administrator T in the following episode.

Excerpt 1:

All of our EMI courses are conducted only in English, and students whose mother tongue is not English also have to finish assignments in English. Therefore, international students will need to prepare themselves to meet the English requirements. They have to pass the English proficiency test such as TOEIC test in order to get into the program. (Administrator T)

Administrator T’s comments indicate that neoliberal English monolingualism has penetrated institutions, reflecting the hegemonic position of English in the global knowledge production. TGUP institutions choose to use standardized English tests (the TOEIC test in the present case) to measure individuals’ linguistic skill bundles in order to determine the individual’s eligibility. This practice is inherently contradictory to the national-level expectation for a society featuring ‘diversity, flexibility and broader perspectives’ (MEXT 2014).

Ideological conflicts at the individual level

During the interviews we found that teachers and students contested the English monolingual orientation at the individual level. The institutional over-reliance on standardized language tests provoked a backlash from some international students, including Gao:

Excerpt 2

Many students used their TOEIC scores to apply for EMI. The program should have requirements for applicants’ writing and speaking abilities. The current recruitment criteria are problematic. (Gao, M2, Intercultural Communication)

Standardized English tests do not capture students’ overall language capacity (Blair, Haneda, and Bose 2018). In particular, the TOEIC test that is popular in Japan does not include speaking or writing sections. As a consequence, ill-prepared students or students with insufficient English might encounter particular difficulties in learning disciplinary content in English (Aizawa et al. 2020; Song 2021). The negative experiences reported by two participants illustrate this point:

Excerpt 3

We Chinese students are not willing to participate in classroom discussion. We generally think we [their English skills] are not as good as [English] native speakers. So, we are a bit self-abased. As a non-English native speaker, I feel inferior [to native speakers]. For example, if we read the same article, I am slower. (Che, M3, International Relations)

Excerpt 4

I don’t think my English is good enough for me to talk about my research project or join in the class discussion. I often fail to understand those students [from Anglophone countries] because they speak very fast, and they use words and sentence patterns different from what I have learned. (Hu, D3, Medicine/Anatomy)

Hu believed that her lower speaking and listening abilities placed her at a considerable linguistic disadvantage in the EMI context, and Che’s deficit thinking shows that he might be falsely linking EMI education with assumptions about native speaker-based education, where the asymmetrical power relationship between students whose L1 is not English and their native-English-speaking peers can lead the former to position themselves as vulnerable learners (Ou and Gu 2018). This sense of inferiority as a non-native English-speaking student illustrates the impact of the native-speakerist ideology upon students’ EMI learning experiences (De Costa, Green-Eneix, and Li 2022).

However, the English monolingual orientation was openly challenged by Professor K, who taught in EMI programs in international culture and communication studies. She made the following comment:

Excerpt 5

Some students misunderstand that high English proficiency (often just conversation level) means a high academic level. Often, native or near-native English speaker students perform very badly in class because they are over-confident in their English and don’t study. (Professor K)

From Professor K’s perspective, native English proficiency does not necessarily equate to academic achievements in EMI; in fact, it can negatively influence students’ performance. Excerpt 5 implies criticism against the conceptual conflation between EMI and English learning, a misconception revealed in Bradford’s (2019) study. Students who have ‘often native or near native’ English proficiency may equate content learning with English learning, and this perceived privileged position may actually hinder their academic learning in EMI programs. In other words, the English-first neoliberal ideology not only positions non-native English international students as vulnerable, but also potentially threatens native-English international students’ sustainability learning. However, it should be noted that the degree of critical awareness demonstrated by Professor K may be exceptional; in general, EMI instructors in Japanese universities still deem ‘native’ English speakers to be ‘ideal’ EMI instructors (Galloway and Ruegg 2022).

Conflicting views of the local language in EMI learning

One of the neoliberal paradoxes of imposing an EMI policy is the decline of local knowledge in knowledge production (Kirkpatrick 2011; Kubota and Takeda 2021), as English is assumed to be the default language of scientific communication (Zheng and Guo 2019; Lillis and Curry 2010).

Devaluing Japanese at the national and institutional levels

Our macro-level policy analysis reveals no mention of the role of Japanese in achieving the internationalization of higher education. At the meso-level, only a few universities explicitly accentuated the importance of Japanese when studying and working in Japan. For example, the Nara Institute of Science and Technology (NAIST) stated that ‘Japanese language acquisition and communication skills are necessary for finding employment at Japanese companies in Japan or overseas’,⁶ while the University of Tokyo also emphasized that the highest level of educational research at the university is carried out in Japanese.⁷

However, except for the few elite universities such as the University of Tokyo and Waseda University who promised to offer Japanese language and culture courses, most institutions appear not to believe it is valuable for international students to learn Japanese, as illustrated by an administrator from one of our focal universities:

Excerpt 6

If international students need help with their Japanese, they can go to the university’s Japanese center. They can take Japanese language courses there. But please note that the credits earned in these courses won’t be taken as student’ credits for graduation. (Administrator S)

This finding that the international students’ Japanese learning might not be recognized by their university suggests that institutions may devalue international students’ Japanese learning. This institutional position on local languages aligns with the ‘benign neglect’ (Wright 2016) of the Japanese language at the national level. In the social space constructed around EMI learning, English is prioritized as the most valuable language of internationalization, whereas Japanese, the local language, is restricted to the local level.

Huge need for Japanese at the individual level

The findings also reveal a huge need for Japanese on the part of international students in order to enhance their EMI learning. Gao commented that ‘[i]f students’ Japanese is insufficient to stay in Japan in the future, the program will be pointless.’ Another example comes from Fang’s experiences. While Fang’s teachers mostly used English in class, they preferred Japanese when communicating with international students after class:

Excerpt 7

Especially after our ゼミ (seminar), if you go to the teacher’s office to discuss your dissertation, Japanese is usually used. But the class is all in English. (Fang, M2, International Culture)

Academic learning does not only take place in the classroom; after-class communication is equally important in international students’ academic socialization. Therefore, Japanese should be accorded equal value with English if international students are to survive and thrive during their EMI learning in Japan, or else they run the risk of finding themselves marginalized during their academic socialization.

Professor K also stressed the importance of Japanese for international students enrolled in EMI programs. When we asked her if international students need Japanese language proficiency, she said:

Excerpt 8

In theory ‘No’, but in practice ‘Yes’. Everywhere in the world there are more resources and information in the local languages, so it is very sad that if students only rely on English information. They can’t go to domestic conferences, seminars, or courses if they don’t understand the local language. We get a lot of useful information by communicating with local people, so I would say it is very important to understand the local language. (Professor K)

The above excerpt demonstrates the discrepancy between theory and practice at the institutional level. Professor K pointed out that a lack of local language proficiency may limit international students’ access to the academic resources afforded by the local people and context. Overall, these findings reveal the deeply ingrained conflict between English, the supranational language, and Japanese, the local language, in transnational higher education. They also contradict the neoliberal framing of the universal functionality of English in academic communication, suggesting that an overemphasis on English and the resulting devaluation of Japanese can restrict international students’ equal and effective participation in EMI learning.

Language learning as self-worth improvement

The above two sections have revealed an overemphasis on English proficiency that generates language challenges for international students, and the lower value placed on Japanese learning that restricts their equal access to resources afforded by the local context. This section reports findings suggesting that language learning has been framed as individual efforts to improve self-worth, as a neoliberal consequence of the EMI policy implementation.

Insufficient language support at the national and institutional level

In our analysis of macro-level policy documents, we found that despite its emphasis on ‘internationalization’ and ‘foreign language education,’ MEXT provides few instructions to guide the implementation of TGUP at the national level, thus implying the a ‘benign neglect’ (Wright 2016) approach to multilingual students.

At the institutional level, as shown in Excerpt 1, international students ‘need to prepare themselves to meet the English requirements,’ which undermines the assumption that students bring their own linguistic (especially English) capital to EMI programs upon enrollment. Learning English is a crucial component of students’ self-improvement as well as an individual responsibility, which implicates the ideological orientation of neoliberal entrepreneurship whereby one needs to constantly enhance one’s linguistic capital in the competitive education market (De Costa, Park, and Wee 2019).

However, not every student enters the EMI learning site with equal amounts of English capital. For students who find they need to improve their English proficiency, the resources afforded by the institution are rather limited:

Excerpt 9

Interviewer: Have you tried to choose some courses on academic writing, say, in English or Japanese?

Kun: Yes, I wanted to choose English academic writing, but the course was too popular. Every time hundreds of students wanted to register for the course, but in the end there were only around 14 positions in this course. (Kun, M2, Economics)

The quality of the institutional language support is also less than satisfactory. Jin complained that his academic English writing class was not beneficial in improving specific areas of his academic English proficiency, and he described it as ‘useless’:

Excerpt 10

It [the Academic English writing course] is useless. We just wrote whatever we want. Anyway, he [the teacher] didn’t know what our specific area was, so he just randomly checked the grammar. Our major task was to give presentations, two times per person, and we passed. (Jin, M2, Chemistry/ Systems Engineering)

Furthermore, some universities set up language support in EMI programs not as a service to meet students’ language needs, but rather as a means of generating revenue. Ma noted that his university charged students an additional fee for Japanese courses, which he believed was a deliberate move to encourage students to focus on English:

Excerpt 11

The university wants to be international, so it wants to Englishize everything. If you want to learn Japanese, you are going against the university’s wishes. The university may discourage you from learning Japanese through its price policy. What choice do I have? I have to learn Japanese by myself. (Ma, M2, Economics and Finance)

The ‘price policy’ mentioned by Ma suggests that some institutions are viewing TGUP as a means to earn a profit. In these institutions, neoliberal ideology has molded their TGUP implementation to suit the market ethos. A neoliberal emphasis on the instrumental value of language and the commodification of language learning collectively imposes extra burdens on international students, both financial and emotional, and might negatively position them by decapitalizing their linguistic repertoires (Li and Zheng 2021; De Costa, Park, and Wee 2019; Kubota and Takeda 2021).

Language learning as one’s own responsibility

As a neoliberal consequence of poor language support, students have to rely on their own efforts to enhance their language proficiencies, such as by making friends with proficient English speaking students. For instance, Li achieved substantial progress in English by drawing on his social network resources. He explained:

Excerpt 12

I met a Vietnamese girl at my part-time workplace. She is enrolled in an undergraduate English program at our school. We eat Lanzhou noodles together every Wednesday. She is not good at Japanese, so we speak English to each other. Most of my English practice opportunities come from her. My English became much better after I met her. (Li, D1, Economics)

From an alternative perspective, Professor K pointed out that language support was available in her university:

Excerpt 13

The university offers special conversation orientated courses in a few different L2 languages, and the writing center helps students to learn how to write essays/theses in English and Japanese. So if students utilize these facilities fully, they should be able to tackle their language problems if not overcome them fully. (Professor K)

However, when we probed Professor K about how she would help international students to tackle their language-related challenges, she commented that ‘it may simply be their personal problems’, and ‘I’m not too worried about their language problems’. Although Professor K also problematized EMI policy implementation, as shown in Excerpt 5, she seems to be inadvertently influenced by the neoliberal discourse of language learning—that is, challenges are the international students’ personal problems, and it is their own responsibility to capitalize on learning resources—thereby valorizing ‘individual self-reliance and accountability for developing human capital’ (Kubota and Takeda 2021, 460). The logic behind this interpretation can be attributed to the ratification of language marketization and commodification (De Costa, Green-Eneix, and Li 2022; Kubota 2011; Shin and Park 2016), wherein individuals’ language skills are seen as their own linguistic capital, and the onus of enhancing these skills is firmly placed on individuals themselves.

Students’ strategic responses: only replying on English is troublesome

The preceding sections have delineated the neoliberal consequences of EMI policy implementation for key stakeholders. The participants found themselves grappling with different languages in their EMI learning. For example, Hu described her painstaking efforts:

Excerpt 14

I watched the teaching video again before submitting the report. I took a snapshot of each slide, typed out the content of the slides and used Google to translate the content [in Japanese] into English or Chinese. Only then did I start to understand what the teacher had tried to say. Then I could write a report. So, only relying on English is still troublesome. (Hu, D3, Medicine/ Anatomy)

This frustrating experience shows that in the de facto multilingual context in Japan’s EMI programs, ‘only relying on English is still troublesome,’ as Hu put it. These students’ experiences attest to the failure of the neoliberal emphasis on English in academic communication.

Despite their language related challenges, our findings show that the participants also strategically utilized their multilingual repertoires and communicative strategies to enhance their EMI learning. For example, Pang recounted the following episode:

Excerpt 15

In the seminar, I often mix the two languages [Japanese and English] … [At academic conferences] [i]f someone speaks English, I will speak English. If he is Chinese and is willing to speak Chinese, then I will speak Chinese. If he is Japanese but speaks English to me, I will also speak English. Many students in academic conferences can’t speak English, in which case, I will speak Japanese. In short, I try to maximize communicative efficiency. (Pang, D4, applied mathematics)

Pang’s multilingual practices bridge institutional English policies and her linguistic repertoire, creating an orchestrated space where diverse and multiple skills and resources can be used for meaning-making (Li 2018). In addition to oral communication, our participants also used multiple languages to take notes during their academic learning (see Figures 1–2).

Figure 1. Pang’s translanguaging notes

Figure 2. Bao’s translanguaging notes

In the above examples the students flexibly crossed language boundaries, exercised their agency, and engaged in fluid language use to meet their situated academic needs. They not only used different languages to summarize important points; they also deployed other symbolic resources like pictures, as seen in Pang’s notes. In Bao’s notes she used her mother tongue, Chinese, to summarize the key ideas in paragraphs. In this sense, the students’ ‘fluid and emergent’ (Kubota and Takeda 2021, 479) multilingual practices contest the English-only monolingual ideology.

Discussion

Many universities in Japan are caught in a nexus of conflicting language ideologies, influenced by both the neoliberal promotion of English in higher education and the government’s multilingual re-orientation in the de jure EMI policy. Previous studies argue that a lack of detailed information in the TGUP initiative may have led to inconsistent interpretation and implementation at different levels of language-in-education planning (Aizawa and Rose 2019; Rose and McKinley 2018). From a neoliberal perspective, we argue that discrepancies between the policy as implemented at different levels can be seen as reflective of an ongoing conflict between government regulation and the neoliberal free market (Shin and Park 2016). Despite the pronounced multilingual orientation stated in the macro-level policy, most institutions continue to prioritize English, largely driven by the neoliberal belief that English is essential for universities’ internationalization (Phan and Barnawi 2015; Piller and Cho 2013). This is consistent with the observation that language policies in higher education invariably collapse at the meso-level implementation stage because institutions are reluctant to obey macro-level guidance (Linn, Bezborodova, and Radjabzade 2020).

This study adds to our current understanding of the consequences of the neoliberal ideology foregrounding English in EMI learning for multilingual international students. Institutions predominantly view the ‘mighty force of English’ as an easy solution to intercultural communication in EMI programs (Kubota and Takeda 2021; Macaro et al. 2018). However, prioritizing an English-only orientation in the EMI context can bring more problems than benefits. The exclusive use of standardized English language tests to assess students’ language abilities fails to capture international students’ multilingual abilities and their communicative competence in other languages (Kubota and Takeda 2021). This may jeopardize TGUP’s goal of attracting and developing global talents (MEXT 2014) and negatively affect the overall sustainability of EMI programs (Bradford 2019). The English-only assumption and its attendant challenges may negatively position international students ‘towards other important issues related to power relations, group dynamics, social integration, and learning’ (Kuteeva 2020, 296). In particular, the neoliberal emphasis on English might cause negative consequences for native speakers by impeding their learning potential and constraining their integration into the local community. The sense of inferiority as non-native English-speaking students and the myth of equating English proficiency with academic achievement in EMI learning, as revealed in the study, illustrate some of the consequences of the neoliberal ideological assumptions underlying the EMI policy.

The primary issue revealed in our study is that TGUP participant universities, as powerful institutions, overlook the fact that foregrounding English and devaluing Japanese actually prevents millions of international students from accessing inclusive education. In practice, asymmetrical power relations between monolingual institutions and multilingual individuals result in the marginalization of foreign language education in spite of the TGUP’s overall ethos of global language education, and jeopardize TGUP’s objective of attracting overseas students and cultivating a domestic pool of global talents. Our findings align with previous arguments that neglecting local language proficiencies in EMI contexts causes difficulties in meaning making, and may lead to uneven consequences in participating in transnational education (Sah 2022; Song 2021). Wright’s (2016) claim that a benign neglect approach is negative for linguistic minorities in language planning, we further argue that devaluing the local language disfranchises international students from gaining equal access to educational resources in the Japanese EMI context.

Consistent with the finding of an absence of support mechanisms in EMI implementation in most Japanese universities (Bradford 2019; Galloway and Ruegg 2022), the neoliberal perspective adopted in the present study has revealed that when national- and institutional-level policies appear to be ambivalent about providing language support, the responsibility falls onto the shoulders of individual students. Language learning requires resources including time, effort, interest, and environmental support, but under the laissez-faire neoliberal discourse the enhancement of language skills, regardless of whether this relates to English or the local language, becomes one’s own responsibility for self-improvement (De Costa, Green-Eneix, and Li 2022; Kubota and Takeda 2021). Neoliberal language learning is not independent from other structural conditions, such as an entrepreneurial individual’s ability to invest in time and financial resources (De Costa, Park, and Wee 2019). Thus, the neoliberal ideologies and norms that underpin individuals’ everyday linguistic practices may ultimately accentuate and reproduce existing systemic inequities.

Stakeholders at the micro-level of EMI seek to create sub-local multilingual spaces where they can employ their entire linguistic repertoires. The observed disjuncture in language planning between the macro- and micro-levels therefore creates a space for stakeholders as social actors to take ownership and initiate bottom-up language planning activities. The flexible language competence that is desirable for effective study in EMI programs in non-Anglophone contexts can be difficult to measure using standardized English tests (Blair, Haneda, and Bose 2018; De Costa, Green-Eneix, and Li 2022; Kubota 2011), since it includes traits like ‘strategic competence and communicative dispositions’ and ‘fundamental qualities that promote interpersonal and intercultural understanding’ to meet social expectations (Kubota and Takeda 2021, 479); this is evidenced by Pang’s experience in this study. By reconsidering micro-level linguistic and pragmatic constraints, policymakers can account for the fluid and heterogeneous features of EMI programs in non-Anglophone countries.

Conclusion

Our critical examination of Japan’s TGUP at multiple levels strengthens the existing literature by revealing a neoliberal tension between national, institutional, and individual levels of EMI policy interpretation and implementation. Echoing previous studies (Aizawa and Rose 2019; Galloway and Ruegg 2022; Rose and McKinley 2018), the TGUP policy downplays the role of English in the internationalization of higher education, but this multilingual reorientation has trickled down to only a few elite universities which emphasize linguistic and cultural diversity in their policy rhetoric, while most still continue to promote an English-only ideology to realize internationalization. We interpret these discrepancies as consequences of the English-first neoliberal ideologies underpinning the EMI policy implementation, which not only position non-native English international students as vulnerable learners, but also potentially obstruct native-English international students’ learning experiences. On the other hand, the devaluation of local languages carried by neoliberal ideologies also risks disenfranchising international students’ equal access to educational resources in the Japanese EMI context.

We hope that our study will inspire future researchers to consider how a multilingual ideology can be translated into language planning activities, and how multilingual learners can take advantage of EMI to obtain social and economic benefits. It is difficult to generalize this study’s findings to other contexts as it is not a large-scale study; for instance, further studies might explore the experiences of international students from more diverse backgrounds. Nonetheless, by offering a nuanced understanding from a neoliberal perspective of the linguistic reality at the national, institutional, and individual levels, our study adds insight to ongoing bottom-up explorations of EMI endeavors. By examining the implementation of EMI policy in Japan, our study may also generate implications for similar non-Anglophone contexts, where language-in-education policies are implemented in a top-down manner and where English proficiency is prioritized in the national foreign language education landscape.

Perhaps our most important insight is that EMI is not necessarily English-only. EMI should not be reduced to English medium of instruction, a neutral and natural tool of knowledge transfer; it is ‘rather a whole process of English as lingua franca socialization situated within the global-scale neoliberal competition for symbolic and cultural capitals indispensable to transnational mobility’ (Song 2021, 353). That said, in the contemporary neoliberal era EMI can be a means for Asian countries to compete with Anglophone universities as more accessible alternatives for global talent development (Tsou and Baker 2021). Simultaneously, however, we need to realize that using English as the only language in EMI classrooms will result in linguistic obstacles and even neoliberal injustice, which may further lead to inequity in knowledge acquisition and dissemination. Therefore, we call for future research to explore how language diversity may generate learning opportunities for multilingual students to learn from other multilingual students, thereby creating a space that allows students to contribute to enhanced learning by drawing on their shared language repertoires.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 MEXT’s TGUP Website: https://tgu.mext.go.jp/en/about/index.html

2 International Christian University TGUP website: https://tgu.mext.go.jp/en/universities/icu/index.html

3 Tokyo University of Foreign Studies TGUP website: https://tgu.mext.go.jp/en/universities/tufs/index.html

4 Kumamoto University TGUP website: https://tgu.mext.go.jp/en/universities/kumamoto-u/index.html

5 Tokyo Institute of Technology TGUP website: https://tgu.mext.go.jp/en/universities/titech/index.html

6 Nara Institute of Science and Technology TGUP website: https://tgu.mext.go.jp/en/universities/naist/index.html

7 University of Tokyo TGUP website: https://tgu.mext.go.jp/en/universities/u-tokyo/index.html