English medium instruction, identity construction and negotiation of Teochew-speaking learners of English

ABSTRACT

The promotion of standard languages as mediums of instruction as well as the worldwide spread and popularity of English have generated various issues related to attitudes towards and ideologies underpinning different languages, language practices, and language teaching and learning in different contexts. With the promotion of the English-as-a-Medium-of-Instruction (EMI) policy in Chinese higher education, it is worthwhile to investigate how multilingual speakers perceive English in EMI programmes and its relationship to their first languages (L1s). This study investigated how Chinese university students from a minority language group perceived the effectiveness of EMI and how they constructed and negotiated their identity. The findings from the interview and focus group data collected from a group of Teochew speakers, representative of minority language students, revealed that the participants’ L1 was marginalised in comparison to the dominant use of English and Putonghua in academic settings, although they held mixed attitudes towards EMI. We argue that an unquestioned embrace of EMI in higher education does not benefit linguistic diversity (i.e. the use of other languages in academic settings) but would endanger the L1s of minority language students. Accordingly, we call for language policymakers in particular and society in general to take an inclusive multilingual perspective.

KEYWORDS:

• English as a medium of instruction
• minority language learners
• identity
• attitude
• Teochew

Introduction

The English language has functioned as a global language for intercultural communication, which has also given rise to issues of language planning, pedagogy and assessment, as well as concerns regarding how English should be used and taught in various settings internationally (Fang 2020; Jenkins 2014). Notably, with its power and dominance, it has increasingly been taken for granted that English should be adopted as the de facto language of instruction for knowledge access purposes, in particular in higher education (Dafouz and Smit 2016; Macaro 2018; Rose et al. 2020). The prevalence of English-medium instruction (EMI) has created concerns about the privileging of English as the language of knowledge (Aguilar and Muñoz 2014; Lei and Hu 2014) and the role of mother tongue and other minority languages in both education and other settings (Huang and Fang 2021; Phyak 2021; Sah 2020).

The power and dominance of the English language in academic settings may marginalise minority languages for knowledge dissemination and reinforce the coloniality of English language teaching (ELT) (De Costa 2020; Kumaravadivelu 2006). The commodification of English has created an unequal relationship of languages, with a neoliberal labour market and a neo-colonial ideology underpinning the use of languages (Heller 2010; Holborow 2018; Tupas 2015). Here, from an uncritical and generalised perspective, the invisible hierarchy of languages has prioritised English (i.e. Anglophone English) or some national languages as the centre for power and knowledge dissemination, while local languages are often relegated to a marginal position in many settings (Haukås 2016; Putjata and Koster 2021). The hegemonic representation and ideology of Anglophone English have also created a circumstance of ‘unequal Englishes’ (Tupas 2015).

Taking a critical sociolinguistic viewpoint, the multilingual paradigm has argued the need to view English from a complex and dynamic perspective and thus urged the reform of ELT to challenge native speakerism (Holliday 2006; Hu 2021). EMI, if viewed from the multilingual perspective, should also be divorced from a native English model when implemented in various non-Anglophone settings (Fang and Baker 2018; Hu 2009; Jenkins and Mauranen 2019). However, there remains a gap between the promotion of EMI at the policy level and its implementation through bottom-up EMI practices that need to be bridged to further our understanding of the complex relationship of language ideology, power, and hierarchy (Tsui and Tollefson 2004; McKinney 2017). This is particularly true of minority multilingual students with respect to how they draw on language resources to perform their multiple identities through language use. Situated in a Chinese multilingual context, this study explores how Teochew-speaking minority language students in higher education perceive the effectiveness of EMI and how they construct and negotiate their identity in both academic and social settings.

Literature review

Language policy and planning

This paper adopts a critical view of language policy as the ‘mechanisms for creating and sustaining systems of inequality that benefit wealthy and powerful individuals, groups, institutions and nation-states, as well as for resisting systems of inequality’ (Tollefson 2013, 27). To understand language policy, it is necessary to give attention to language beliefs or ideology, language management, and language practice (Spolsky 2004). Language policy on the choice of instructional medium does not only concern linguistic practice but is also motivated by ideological and socio-political factors (Blommaert 2010). Previous studies have revealed the complexity of language polices in relation to national and minority languages, where in many circumstances the minority languages are marginalised (Huang and Fang 2021; Li 2019; Wang and Curdt-Christiansen 2021). For example, Wang and Curdt-Christiansen (2021) found that Chinese parents played a key role in the intergenerational transmission of dialects. These parents used dialects to communicate with the grandparents but switched to Putonghua, the national language, when they spoke to their children. This unidirectional medium of language use created a Putonghua-only environment for the children at the home domain, despite the multilingual family environment.

Previous research has also found that prevalent official language policies are largely hierarchical (top-down) in order to promote the mainstream language and subscribe to a ‘one language, one nation’ ideology at the expense of so-called dialects or non-dominant varieties of an official language (Cooke and Simpson 2012; Manan, David, and Channa 2019; Putjata and Koster 2021). However, in the Chinese context, we also need to understand that promoting Putonghua is a means to ensure political unity rather than eliminating regional dialects (Wang and Curdt-Christiansen 2021). Still, the promotion of Putonghua and EMI in Chinese higher education has generated some debates and concerns regarding linguistic diversity in today’s multilingual world as well as the marginalisation of minority languages (Fang and Xie 2019; Gao 2017). In particular, there is a need to examine, from a multilingual perspective, minority language students’ perspectives on language management and practices in relation to EMI, the use of Putonghua and minority languages for identity negotiation and construction.

Critical multilingualism and English as a medium of instruction

English has become a global lingua franca for international communication. More recently, it is positioned within multilingualism as a multilingua franca (EMF) (Jenkins 2015). In this view, English speakers do not blindly follow the native norms for intercultural communication; rather, without a fixed norm, people (re)negotiate meaning and (re)construct their identities in and through the communication process. The dynamic and complex nature of the English language also requires a new understanding of language contact and a paradigm shift in ELT. For instance, Fang (2020) proposed an approach named Teaching of Pronunciation for Intercultural Communication (ToPIC) arguing the need for language learners to ‘re-think and re-evaluate learning models and targets’ (104) for the recognition of their identity construction when learning English. Language use from the multilingual perspective is unexpected and complicated, and consequently ELT needs to go beyond native speaker ideology in due recognition of the current landscape of English (Kirkpatrick 2014; Piller 2017).

Furthermore, language policy and practice have been influenced by the commodification of education, particularly through English in the job market (Heller 2010). Heller (2010) has pointed out that language is often commodified through a technical skill and as a sign of authenticity from a neoliberal perspective. At the language policy level, the so-called standardised variety is often represented as the sole legitimate one (Piller 2016) at the expense of other varieties in an attempt to consolidate the nation-state in language policy and planning. The tension, however, also lies in national and supranational language education policy (Huang and Fang 2021; Tupas 2004), particularly in the choice of instructional medium, as reflected in the implementation of monolingual language policy in multilingual contexts (Piller 2016). It is noted that language policy through the commodification of English language education may create further existing inequalities between elite and non-elite universities and students, which leads to the ‘de facto marginalization of minority languages’ (Gao and Wang 2017, 222). Policy on medium of instruction is under-researched because there is still a lack of consensus regarding the role of English in EMI (Macaro et al. 2018). Against this backdrop, questions regarding ‘what English and whose English in EMI’, and whether and to what extent translanguaging is allowed in EMI should be further explored (Jenkins 2014; Kuteeva 2020).

The spread and dominance of the English language in higher education has led to the popularity of EMI for language learners in both Anglophone and non-Anglophone settings. As mentioned earlier, the spread and dominance of English is in turn a function of internationalisation (Dearden 2014; OECD 2014). This paper views EMI from a multilingual perspective: on the one hand, it ‘focuses on English-medium education because of the particular role that English plays both as an academic language of teaching and learning as well as a means of international communication’ (Dafouz and Smit 2016, 399); on the other hand, it also recognises the complexity and diversity surrounding languages and pedagogies due to the multilingual nature of international higher education through EMI. Similar to many contexts, EMI has been promoted in higher education in China for graduate competitiveness and university ranking, as well as to ‘facilitate student and staff mobility and credit transfer’ (Kirkpatrick 2014, 4). Indeed, the adoption and promotion of EMI in higher education should be viewed not only from the linguistic perspective but also in sociocultural and socio-political terms.

EMI should be examined more critically as a top-down language policy (Hu and Lei 2014). Although EMI is generally believed to promote internationalisation and students’ competitiveness in the global job market (Macaro et al. 2018), the language-in-education policy may neglect the plurality of languages and cultures. The implementation of EMI is critiqued as being ‘not purely about educational efficacy but also about social, political, and economic participation, social equality, and human rights’ (Tsui and Tollefson 2004, 17). In particular, there remains a tension of EMI in universities attended by multilingual students, where a key local language also plays a role in higher education. Such tensions also exist in communities where minority languages are spoken. In China, for example, Putonghua is the default language of instruction except in English language classes, though the Ministry of Education in China has recently required the adoption of EMI (Hu and McKay 2012) in Chinese higher education, while minority languages in the Chinese context have been largely marginalised.

More recently, however, EMI has invoked controversies regarding its effectiveness in enhancing both English language and content learning (Fang 2018a; Hu and Duan 2019; Lei and Hu 2014), and concerns about its potential negative impact on the loss of local linguistic and cultural identities (Guo and Beckett 2007; Niu and Wolff 2003). From a multilingual perspective, however, EMI ‘would allow students to become better prepared to use English as an academic lingua franca, by making them more aware of the language used as the medium of instruction, and by increasing their confidence as multilingual users of English’ (Rose and Galloway 2019, 196). To summarise, EMI policies have greatly affected ‘the culture and values associated with the mother tongue’ (Tsui and Tollefson 2004, 7), particularly in ethnolinguistic minority contexts. There is complexity regarding linguistic and cultural values, as well as national and local/minority identity construction through language (Fang 2018b; Gao 2017; Gu and Patkin 2013).

EMI, language choice and identity negotiation

Medium-of-instruction policy has been found to be involved in various language-mediated activities and conceptions, including ideological constructs, discourse practice, ethnolinguistic diversity and even social conflicts because of a lack of equal access to EMI learning resources (Tsui and Tollefson 2004). Ethnolinguistic contexts can be very complex in that minority language speakers such as Teochew speakers in mainland China can be members of the ethnic majority (i.e. Han Chinese). However, minority language groups, regardless of whether they are ethnic minorities, can offer ‘theoretical and practical tools for understanding the sociocultural worlds that are inter-related with linguistic practices’ (Black and Falconi 2017, 479). For minority language students, the process of language choice and practice would consciously or unconsciously represent and reflect their linguistic and social identities. Language use is a process of demonstrating people’s linguistic and cultural capital (Bourdieu [1982] 1991; Phillipson 2019), and a language realises its symbolic power (Kramsch 2020) in many ways, including its adoption as the medium of instruction.

It is important to recognise that language use is inextricably intertwined with power and ideology. As a top-down policy, EMI has been found to fail to achieve better learning outcomes compared to mother-tongue instruction (Hu and Duan 2019; Lei and Hu 2014). The decision to adopt EMI thus needs further investigation. The stakeholders may not be cognizant of linguistic diversity and the need for equitable access to resources through multilingualism and thus may subscribe to a monolingual language ideology in EMI courses (Fang and Xie 2019). For instance, Baker and Hüttner (2018) reported students’ negative views on using other languages in EMI courses in three universities in the UK, Thailand and Austria. In a similar vein, Song (2021) pointed out the unequal knowledge production and epistemic (in)justice embedded in the ‘semi-Americanised academic norms of the EMI programmes’ (353). The uncritical and blind adoption of EMI represents a neo-colonial hegemony concerning local languages and minority identities in higher education, especially in the Global South (Abrar-ul-Hassan 2021; Pennycook and Makoni 2020; Sah and Li 2018). With respect to the relationship between EMI and local identities, particularly for minority language groups, a question remains about how EMI could protect or undermine the right to mother-tongue education (Botha 2016; Song 2021). A further concern has to do with how the linguistic identity of ethnolinguistic minority people can be recognised in competition with the national language and even English.

Minority language individuals’ experience with EMI and use of other languages are conceivably linked to their construction and negotiation of academic and social identities (Norton 2013). Norton (2013) defined identity as ‘how a person understands his or her relationship to the world, how that relationship is constructed across time and space, and how the person understands possibilities for the future’ (45). Recent research has shown that identity is complicated and fluid in relation to language contact (Pan et al., 2021; Liu, Zhang, and Fang 2021; Sah and Li 2018). For instance, geolinguistic dimensions figure prominently in folk metalinguistic awareness, or what Albury and Diaz (2021) termed perceptual multilingualism, concerning the relationship of an official language and dialects (fangyan, 方言) in popular discourse. Regarding identity construction for minority language speakers, Li (2019) investigated Shanghainese participants’ cross-generational use and knowledge of English loanwords and found that many words were replaced by Putonghua, the national lingua franca in China. Li argued that the replacement of English loanwords in Shanghainese was a process of ‘imposition’ by monolingual Putonghua migrants, and that lexical attrition suggested that ‘the regional cultural heritage is at the risk of being eroded and homogenised’ (734). In this case, local ethnolinguistic and cultural identities are mediated or even challenged by the dominant regional and national lingua francas. Therefore, the tension among English, the national lingua franca and the local minority language makes it necessary to investigate various inequalities historically and socially constructed through language policy and language choice at various levels (Phillipson 2019; Tupas 2020).

In a similar vein, Sah and Li (2018) revealed the privileging of EMI in a school in Nepal, regardless of resource availability and EMI-readiness of the stakeholders. The ideology of EMI as linguistic capital has been entrenched in parents, teachers and students; the rosy picture of EMI, in reality, could lead to the lack of content learning, low proficiency in both national language (Nepali) and the global lingua franca (English), as well as the loss of students’ L1. In this case, the embrace of EMI could create an ‘identity crisis’ for the ethnolinguistic multilingual students. Zhang (2018) revealed the complexity of language ideology and management at national, institutional and individual levels by exploring China’s EMI policies in higher education. Zhang found a mismatch between language policies on the one hand and language ideology and management on the other. He argued the need to balance policies regarding EMI and Chinese as a medium of instruction. While it is useful to consider language policies at the national or global levels, it is even more important to consider the use of minority languages by ethnolinguistic students at the level of language practice, for example, the adoption of translanguaging during instruction or in language use (Fang and Liu 2020; Muguruza, Cenoz, and Gorter 2020; Sah and Li 2020).

This paper takes the perspective that EMI should be examined in order to understand the complexity of language contact from a sociocultural perspective, and that EMI should not be implemented in a top-down manner without considering ethnolinguistic complexity. Drawing on qualitative data collected in interviews and focus groups, this paper attempts to answer the following two research questions:

1. How do Chinese students from a minority language group who are taking EMI courses perceive the effectiveness of EMI?

2. How do these minority language students perceive their L1 and other languages in relation to their construction and negotiation of social and academic identities?

Methodology

Research site

This study examined how a group of Teochew speakers taking EMI courses at a local university perceived the effectiveness of EMI in relation to their identity negotiation and construction as minority multilingual speakers. The focal community is located in Southeast China where Teochew is mainly spoken as a heritage language, along with Putonghua and other languages/dialects, in people’s daily lives. As a Southern Min dialect, Teochew is traditionally spoken in the Chaoshan area, which encompasses Shantou, Chaozhou and Jieyang, three key cities in Guangdong province, China. Teochew has a long history that has influenced Southeast Asian cultures and languages, as numerous Chaoshan people migrated to Southeast Asia to establish businesses by ‘the old Red-Top Sailing Merchant Boats’ (坐红头船过番) during the Ming and Qing dynasties (Wen 2012). It is viewed as one of the most conservative Chinese dialects preserving various features from ancient Chinese (Wen 2012), and also acts as a migrant language for people in Southeast Asian countries such as Singapore, Malaysia, and Thailand. As an intergenerational dialect of immigrants, Teochew has long been linked to local identity.

As one of the three better known dialects (along with Cantonese and Hakka) in Guangdong province, Teochew is spoken by approximately 10 million people in the Chaoshan area (Eberhard, Simons, and Fennig 2021). However, as a minority language, Teochew has seen a decline in intergenerational transmission (Huang and Fang 2021), mainly due to Putonghua’s status as the primary language of instruction since kindergarten and the implementation of EMI in higher education (Feng and Adamson 2018; Wang and Curdt-Christiansen 2019). As a minority language, Teochew does not enjoy an equal social status today and is increasingly regarded as a low variety because Chaoshan citizens have gradually shifted from bilinguals to diglossic users (Huang and Fang 2021; Li 2019). The use of Teochew has been in decline across generations due to the social and cultural contact with other dominant languages such as Putonghua and English (Huang and Fang 2021). It is therefore necessary to investigate how people of this minority language community perceive the relationship of their L1 (Teochew), the regional lingua franca (Putonghua) and the global lingua franca (English), along with other linguistic resources they possess.

This study was conducted at a university in the Teochew community. The focal university has more than 17,000 students in 12 colleges, and more than 40% of faculty members have overseas education backgrounds and/or work experience. According to the university website of admissions, speakers with Teochew as their L1 account for approximately 12% of the university students. In recognition of Teochew, the university also offers a course titled ‘Teochew Dialect and its Culture’. At the same time, both in line with the Ministry of Education’s promotion of EMI and for global competitiveness in higher education, the university has been actively participating in the current trend of offering a growing number of content courses through EMI. That a sizeable proportion of the faculty members have had overseas academic experiences has provided the basis for offering EMI courses university-wide. Besides, some of the EMI courses are taught by expatriate lecturers from both Anglophone and non-Anglophone countries, including the UK, the US, Italy, Spain, France, among others.

Participants

The selection criteria are as follows: (1) participants spoke Teochew as their L1; (2) they were all taking EMI courses at the time of research. As this study did not take English language classes as EMI, no students majoring in English were recruited. A total of 12 participants were purposively recruited and consented to participate in this study (see Table 1 for the profiles of the participants). We sought to find out those participants’ perceptions of the effectiveness of EMI and their attitudes towards the use of Teochew, Putonghua, and English in relation to their identity construction and negotiation. Given the relatively small sample of participants involved, we do not intend to generalise our findings to other Teochew-speaking students or other minority language contexts in China.

Table 1. Profile of the interviewees.

Name	Gender	Year of study	Major	Length of interview
HWQ	F	First Year	Statistics	48’47”
YYL	F	Fourth Year	Journalism	53’05”
HZM	F	Second Year	Accounting	36’19”
LC	F	Second Year	Finance	37’52”
ZMH	F	Fourth Year	Finance	37’14”
ZHZ	F	Third Year	Business Management	26’23”
XPZ	F	Third Year	Advertising	46’07”
LBY	M	First Year	Medical Engineering	27’24”
SXT	F	Second Year	International Trade	32’17”
ZDL	F	Third Year	Finance	25’02”
ZYY	F	Second Year	Journalism	53’07”
XZJ	M	First Year	Medical Engineering	45’16”

Data collection

Purposive snowball sampling (Miles, Huberman, and Saldaña 2014) was used for data collection. The potential participants were contacted through WeChat, a popular social media in China. After posting recruitment information on the first author’s WeChat moment, he was contacted by two students regarding the criteria of participation. Both met the selection criteria and were asked to recommend more participants. Through snowball sampling, 12 participants were invited to participate in semi-structured interviews. The interview questions centred on ‘EMI experiences’, ‘evaluation of EMI’, ‘benefits and struggles of EMI’, ‘language use in EMI courses’, and ‘the sustainability and future development of Teochew’. After the interviews, two focus groups were further conducted with 11 of the students (6 in the first focus group and 5 in the second focus group). Leveraging on their highly interactive nature, the two focus groups were conducted to scope out the participants’ use of their L1 in academic, family and daily contexts and unpack factors influencing their L1 use. All the interviews and focus groups were conducted online and were audio-recorded. The interviews, lasting between 25 and 53 min, were conducted in Putonghua to ensure that the participants expressed their ideas freely and comprehensively (Mann 2011). The participants chose to use Putonghua primarily during the interviews because Teochew was traditionally regarded as a low variety by diglossic users in academic contexts. The two focus groups were also conducted in Putonghua and lasted 75 and 85 min, respectively.

Data analysis

The recordings were transcribed by a student assistant and checked by the authors. Qualitative content analysis was employed to analyse the transcripts of the interviews and focus groups (Schreier 2012). The first author coded the data via NVivo. First, preliminary codes, for example, ‘attitudes towards EMI’, ‘perceptions of the effectiveness of EMI’, ‘language practices’ and ‘identity construction’, were deductively generated from the relevant literature and the research questions formulated for this study. These deductively derived codes served as sensitising categories for identifying broad patterns in the data. Second, the preliminary coding scheme was then used to code the data in an iterative manner, with the preliminary codes being revised and new codes being added to better capture the data. Third, the inductively derived codes, such as ‘accent’, ‘language requirements’, and ‘language support’, and the refined deductively derived codes were compared constantly and integrated to yield four major themes: ‘attitudes towards EMI’, ‘language ideologies and practices in EMI courses’, ‘Language requirements and practices in EMI’, and ‘Identity negotiation and construction of minority language speakers’. The identified themes were shared with the participants for member checking. The coded extracts were translated into English and double-checked by a professor who majored in translation studies to maintain the accuracy of the translation.

Trustworthiness

The first author worked at the focal university and shared Teochew as the L1 with the participants. As a minority language speaker himself, certain subjectivity would remain during the data collection process. This was balanced through reflexivity by establishing rapport with and making the study purpose explicit to the participants. Furthermore, while it is not possible to generalise the findings of this study, it is hoped that through rich descriptions of the participants’ experiences and the research context, the findings can be transferable to other similar dialect-speaking areas (Lincoln and Guba 1985). The data were primarily collected and analysed by the first author, with the second author serving the roles of auditing the processes and providing alternative perspectives. Throughout the processes, both authors endeavoured to be reflexive and explicit about their roles and exchanged and responded to their interpretations of the data iteratively (Lincoln and Guba 1985).

Findings

Attitudes towards EMI

The students held mixed and ambiguous attitudes towards EMI when they negotiated the relationships of their L1, Putonghua and English in various contexts. EMI as a language policy was imposed on the students by the focal university, and they simply had no choices regarding the EMI courses. Only two participants (XPZ, YYL) expressed positive attitudes towards EMI. XPZ explained that ‘I like English and I believe that we need EMI courses for students’ competitiveness in this international world; EMI motivates my learning’; but she also added that ‘the effectiveness of EMI really depends’. YYL believed that ‘we need EMI courses because they are more original’ but she also pointed out the difficulties in communicating with lecturers and challenges in understanding course content.

The other participants expressed their reservations about the top-down EMI policy being implemented regardless of their English proficiency (although there was an English placement test) and their having to take EMI courses right after they entered the university (Macaro et al. 2018). HZM confided that ‘we had two EMI courses in the first semester, Microeconomics and Management. I didn’t really understand why we needed such EMI courses, and some of my classmates bought the textbooks with Chinese translation [to cope with the EMI]’. Several other students (e.g. HWQ, SXT, LC, and ZMH) indicated mixed attitudes towards the EMI policy. On the one hand, they expressed their concerns when they had to take EMI in the first semester: ‘I felt frightened when I was told that the course was lectured in English when I entered the university’ (HWQ); ‘we didn’t have an option – we had compulsory EMI courses taught by the same teacher. We had to translate everything into Chinese all the time, but in the end I felt that I learned nothing’ (ZMH). On the other hand, some (e.g. HWQ, SXT) saw a need for EMI and found themselves to be able to cope with it over time. HWQ commented that ‘I gradually accepted EMI and found it easier for me to write academic papers in English’. SXT also reacted similarly: ‘I had to preview before the course, and I felt that I could accept the model’. Notably, she had her concern too: ‘in EMI we only learn something quite superficial because the teacher doesn’t teach much content knowledge indeed’. In sum, the implementation of EMI was perceived by the students as a top-down policy that involved no negotiation between teachers and students. In some cases, EMI was implemented simply because the lecturer did not have a shared language with the students, and English became the de facto medium of instruction.

The students also experienced difficulties in communication through English and saw English as a ‘gatekeeper’ controlling knowledge acquisition (Jenkins 2014). For instance, LC confided:

Although it may be more systematic and formal to use the original resources, I felt frustrated in my attempt to understand the lecture because I am not familiar with the teachers’ accents, let alone the content itself. I may be too clumsy because the teachers would not understand me either. Those students whose English level is higher don’t really have such problems.

Another participant, XZJ, also commented on a similar situation: students good at English were able to finish their assignments easily and achieved higher scores in exams, while students with limited English proficiency lost many opportunities and their interest in learning the subject content:

Some classmates who struggle in learning English are not able to meet the teachers’ expectations when doing their homework and thus receive low scores. Gradually, they are discouraged and question their potential or even experience a feeling of guilt. It is a vicious circle in learning.

The English language was considered a great obstacle to the EMI courses. For instance, ZMH shared, in the focus group, the difficulties she encountered:

It was difficult in an EMI class environment, and I always forgot what I had learned. I was not sure what the teachers wanted to express either. I identified two reasons: first, it is not a language that we commonly use in our daily lives; second, the teacher doesn’t use English in his/her daily life either and skips some key knowledge when teaching us the content.

Regarding such ambivalent attitudes towards EMI, ZHZ drew an apt conclusion: ‘the effectiveness of a lecture doesn’t really depend on the language used – it depends on how the lecture is prepared and delivered’. The complexity of the students’ attitudes towards EMI was also related to the lack of language requirements and stemmed from specific language practices in the EMI courses.

Language requirements and practices in EMI

Both the interview and focus group data revealed that there were few prerequisites for taking EMI courses. More often than not, compulsory courses were simply made EMI, and every student had to take them. Exceptions were a few electives, which required students to pass certain English courses before they could attend them. There was little formal language support after class for students taking EMI courses, as confirmed by ZDL: ‘We don’t have any language support apart from college English courses’. Another student (LC) was also disappointed about the lack of language support for those who were taking the EMI courses.

Many participants reported that despite their language problems, many EMI teachers would focus on content instead of language, though some teachers had their language preferences.

My teachers didn’t correct my pronunciation or grammar; they told us that we learned the subject through English. Anyway, I felt that my teachers had their own accents too. […] My teacher did not require us to always use English during class, but he told us that we would obtain a higher score if we use English on PowerPoints and during our presentation. (HZM)

My teacher didn’t really correct grammar mistakes if we were able to convey what we meant, but of course we shouldn’t make too many mistakes if our writing or presentation is to be intelligible. But my teacher didn’t allow us to use other languages except English. (YYL)

Notably, the students also reported their difficulty in understanding both the language and content of the lectures:

I remember a course taught by a Spanish teacher. The course was about theory of game, so it was rather difficult. The teacher’s English was not fluent, and we had difficulty understanding her. So she had to draw pictures on the blackboard in order to explain a concept. (ZMH)

In terms of language use, some participants (e.g. HWQ, ZDL) would speak only English regardless of how difficult it would be for them because they did not want to violate the nature of the course. Other participants adopted translangauaing as the coping strategy in their EMI courses, as revealed by ZYY and ZMH:

When doing homework I first wrote in Chinese and then used machine translation. Of course I needed to read and revise again before submission […] because it was too difficult for me to express my ideas directly in English. (ZYY, Focus Group)

Especially when I don’t know how to express a term in English I would resort to Chinese. There is also a local teaching assistant to help us translate what we don’t understand. (ZMH)

Language practice in the EMI courses was rather complicated because various linguistic and multimodal resources were involved. For minority language students in this study, Teochew also served as a resource for communication.

I used Teochew and Putonghua to help me process some information for a mathematics course. Sometimes Teochew even functioned as the primary tool for me to understand some formulas. (HWQ)

I think Teochew sometimes helps me to understand a concept which can’t be explained in either English or Putonghua. I think it is deep-rooted. (XPZ, Focus Group)

In summary, there was a lack of language requirements and support for students to take EMI courses. Linguistic accuracy did not seem to be the main focus in many EMI courses, where the lecturers were reported to focus more on the content. Language practices in the EMI courses were complex as the students took their initiative to leverage their linguistic repertoire and multimodal resources in language use, although imbalanced in actual practice. Therefore, as revealed in the following section, the students experienced various levels of identity (re)construction through EMI.

Identity negotiation and construction of minority language speakers

As was evident in the students’ responses, there was a hierarchy of language use in the EMI courses and in their daily language practices. In the EMI courses, English was regarded as the dominant language in academic settings, while Putonghua was the most frequently used after class on or off campus. Teochew, and other minority language varieties, did not enjoy the same status as English and Putonghua. This pecking order is evident in the follow excerpts from a focus group and an individual interview:

XPZ:

People seldom use Teochew in academic settings, unless people have a project related to the dialect.

ZMH:

I feel the same.

HWQ:

Teochew is barely used in academic settings even when people share the same first language. I don’t know to whom I shall use Teochew for academic discussion.

I use Putonghua at school and Teochew at home with my family members. […] We seldom use minority languages, such as Cantonese or Teochew in academic contexts. Instead, we use Putonghua or English. (ZDL)

Obviously, Teochew, as the participants’ L1, was marginalised in academic settings. It should be noted that Teochew was once a dominant language in primary education because ‘teachers used Teochew to teach when I was in primary school but now it’s replaced by Putonghua’ (ZMH). As mentioned, the use of Teochew in the local community has shifted to a diglossic situation where it is used mainly in informal situations. The students even expressed their worries that their identity as a Teochew speaker may eventually be lost:

We mainly use Putonghua, then English in the courses, and then other dialects. Teochew has been marginalised in our daily lives. If we don’t teach Teochew it will no longer pass on as an intergenerational language. People may be able to understand it but they may not be able to speak it. (XPZ)

Despite her understanding of the importance of family language planning in maintaining the sustainability and use of Teochew, XPZ believed that Teochew would gradually lose its status. However, she was comfortable with her multilingual identity: ‘I choose to speak Teochew depending on context. I don’t struggle much with Teochew’. Another participant shared XPZ’s view about language choice in relation to identity:

I use Teochew with my dad because he can’t speak Putonghua well. But I’ve been used to speaking Putonghua since I was young, though I switch to Teochew when going shopping and ordering food in canteens. It’s more cordial to use Teochew sometimes. (LC)

The diglossic use of Teochew is salient in LC’s description of her language choice. Nonetheless, LC stressed the importance of her identity as a Teochew speaker and maintaining Teochew through family language practices: ‘As a speaker of Teochew, I don’t want to see a situation where people around me don’t know the dialect in the future – but it seems that many young people today could not speak Teochew well or fluently’.

In the focus group, LBY also discussed his identity negotiation and construction in relation to the use of Teochew.

Although some children are used to speaking Putonghua and English, and don’t speak Teochew, I feel proud of myself to be able to speak Teochew well. I push myself to speak it because I have gradually realised the importance of speaking one’s dialects well. I’m sure other people feel the same. It’s important for a language, even when it’s spoken by a minority group, to be passed on from generation to generation.

In sum, many of the participants still regarded Teochew as a valuable linguistic resource and as a core part of their identity, although they also recognised the existence of a linguistic hierarchy. They felt a sense of solidarity when speaking Teochew and were keenly aware of the importance of intergenerational efforts to maintain this dialect (Huang and Fang 2021). However, the current situation is that Teochew has gradually lost its function in academic settings and is regarded as the low variety in the diglossic use of the languages under examination.

Discussion and implications

The findings revealed the complexity of EMI and its adoption as a top-down policy without considering students’ actual needs. To some participants, EMI was a welcome policy for their international competitiveness; to others, EMI seemed to dislocate their multilingual resources in academic settings. There was an invisible linguistic hierarchy for language choice and practice among the minority language students. They negotiated and constructed their identities, consciously or sub-consciously, through their language use in both academic settings and daily life.

In relation to our first research question, EMI was not perceived by most of the participants to be effective or in much demand. These participants did not see the need for EMI and felt compelled to take the EMI courses. Many of them were not really ready for EMI and therefore had to resort to translanguaging to cope with the demands of learning subject content through a language not well mastered (Fang and Liu 2020; Paulsrud, Tian, and Toth 2021; Wang and Curdt-Christiansen 2019). Their linguistic resources were downgraded and marginalised in an EMI discourse. Another relevant finding concerns the lack of language requirements and the limited language support available for students taking EMI courses, in contrast to the self-access support services and academic writing support reported in previous research (Galloway and Ruegg 2020). Therefore, there is a need to examine more closely how the English proficiency of teachers and students positioned them and facilitated or inhibited learning in EMI courses. Our findings suggest that the classroom discourse exhibited many multilingual characteristics because few teachers adopted native standard norms in the classroom and many did not even give much attention to the formal aspects of language. Similar to what Galloway and Ruegg (2020) found in their study, the teachers that the participants mentioned in this study focused more on content than linguistic correctness. These patterns of language practices in the EMI courses are consistent with the call for EMI programmes to move beyond native speakerism (Fang 2018a; Rose and Galloway 2019).

As regards our second research question, the minority language students were found to experience complex identity negotiation and construction through the use of the languages at their disposal. On the one hand, they seldom used Teochew in academic settings and regarded it as the low variety in a diglossic situation. The dominant languages were English in their EMI courses and Putonghua in their other courses, though translanguaging was often employed to cope with language difficulties. Here, a hidden linguistic hierarchy was at work (Lippi-Green 2012). Although Teochew was once used in academic contexts as a medium of instruction for some students (e.g. ZMH), today’s students (e.g. ZPZ, ZDL, LC) are more used to speaking Putonghua in their daily lives. On the other hand, some participants (e.g. LC, LBY) were worried about the loss of Teochew as their L1 as they showed resistance to mainstream academic and popular discourse, contesting the underlying ideologies and championing for Teochew in its own right (cf. Albury and Diaz 2021; Huang and Fang 2021). Although it did not have the same social status enjoyed by English or Putonghua, the use of Teochew communicated a sense of solidarity for the participants (e.g. HWQ, HPZ, LBY) because the language was an ethnolinguistic heritage they felt proud of and made them who they were (Huang and Fang 2021). The ambivalent and complex relationship of the participants to Teochew was evident in a perceived contradiction: while Teochew as a minority language has been greatly marginalised in academic settings and showed a decline in daily use, the participants still insisted on the importance of maintaining the language as part of their identities.

Based on the findings of this study, we would offer some implications regarding language policy, management and practice in academic contexts to promote the sustainable development of minority languages from a multilingual perspective. First, language-in-education policy such as EMI should be implemented from a bottom-up perspective instead of top-down imposition without giving due attention to the real needs and challenges in the classroom. Second, EMI programmes should legitimate translanguaging practices that involve students’ L1s (including minority languages) and other linguistic and multimodal resources. From the perspective of multilingualism, there is a need to go beyond the monolingual ideology in EMI and embrace the important role of translanguaging in ‘not only facilitating students’ disciplinary learning and their English development, but also increasing their flexibility in making use of whatever language resources available’ (Wang and Curdt-Christiansen 2019, 335). Third, as the students in this study saw Teochew as playing a key role in their identity construction, it would be beneficial to design courses of local minority languages and culture, such as the one offered by this university. It is also important to revitalise minority languages beyond educational settings, such as providing some activities to promote the use of minority languages on social media or applications for the younger generation to reflect on their own linguistic practices (Huang and Fang 2021; Li 2019). Indeed, some conflicting goals of language policy and practice regarding official languages and minority languages should be reconciled in both classroom and social settings including family language policy (Cenoz 2019; Fang and Liu 2020; Wang and Curdt-Christiansen 2019). Although the traditional ‘E’ in ELT has been argued to be implicated in a colonial ideology, and there have been calls for minority languages to be recognised in language education (Tupas 2015, 2020), we should also in turn critically evaluate the effectiveness of multilingual pedagogy.

Conclusion

Before making any tentative conclusion, we reiterate that based on a small-scale qualitative study, this paper does not aim to generalise the situation of EMI and minority language use beyond our research context and participants. Our findings need to be interpreted in a way that recognises the complexity of language policy and language practice. It should also be noted that data in this study consisted of only self-reports and that classroom observations should be conducted for triangulation purposes to depict a more comprehensive picture of language ideology and practice in EMI courses.

The caveats notwithstanding, our study took an initiative to understand the complexity of language use involving dominant and minority languages by exploring students’ pedagogical and sociocultural experiences in EMI courses. As mentioned, the critical multilingual paradigm challenges native English ideology and calls for English to be approached from a dynamic and fluid perspective in relation to other languages. This paradigm also proposes that language policy, ideology and practice should be contextualised and understood in terms of the larger agenda they serve. In this view, and as we have demonstrated, EMI in the Chinese context supports essentially a neoliberal agenda for two reasons. First, EMI is perceived to bring along various economic benefits in a free job market. Second, although EMI may detract from, rather than consolidate, Putonghua’s position of power as the national language, it does promote Putonghua-English bilingualism, which is officially characterised as being crucial to successful international communication and competition in the twenty-first century.

Thus, critical multilingualism provides a vantage point from which linguistic diversity can be viewed not only at the societal level but also in academic contexts (Jenkins and Mauranen 2019). To recognise linguistic diversity, translanguaging strategies involving the use of minority languages should be adopted not only to enable people to communicate effectively but also to promote their linguistic rights for their identity construction in both social and academic settings (Fang and Liu 2020; Sah and Li 2020). This paper argues the need to recognise ‘multilingual speakers’ socio-cultural, political, and historical experiences living with multiple languages in the periphery contexts’ (Phyak 2021, 229). It is hoped that further studies can extend the conversation to promote a critical multilingualism-oriented EMI discourse and integrate translanguaging practice in language education policy to meet the various needs and goals of different stakeholders.

Disclosure statement

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

Additional information

Funding

This work was supported by the Ministry of Education Humanities and Social Sciences Fund (Youth Fund Project) - ‘Investigation of the Interaction Mechanism between Family Language Planning and Cultural Identity’ [Grant number: 21YJC740012].