1. Introduction
The invited colloquium, “Researching Multilingually to Rethink EMI Policy and Practices”, organized by Xuesong Andy Gao (University of New South Wales) and Yongyan Zheng (Fudan University) took place on 18 March 2023, at the annual meeting of the American Association for Applied Linguistics (AAAL) in Portland, Oregon, USA. This colloquium consisted of six paper presentations, followed by comments from a discussant.
English Medium Instruction (EMI) refers to the use of English to teach academic subjects other than English itself, in regions where the majority's first language (L1) is not English (Macaro, Reference Macaro2018). EMI is increasingly used in contexts where English is not used widely. Although instructors and students in diverse EMI contexts are expected to utilize an expansive linguistic repertoire to meet academic goals (De Costa et al., Reference De Costa, Green-Eneix and Li2021), applied linguists are still confronted with critical questions, including how the implementation of EMI policies can be refined to ensure that EMI practices lead to the desired educational outcomes (i.e., student gains in learning both English and subject content simultaneously), and how to ensure that the rise of EMI education does not undermine efforts to sustain the linguistic ecology and multilingualism of the local educational contexts.
We invited applied linguistics researchers from different linguistic and cultural backgrounds to contribute to this colloquium and to argue that EMI practices need to rely on the multilingual resources that students, teachers, and researchers have when designing, implementing, and investigating EMI policies and practices. To this end, this colloquium adopted the framework of researching multilingually (RM-ly). RM-ly is described as an approach that researchers can take to conceptualize, understand, and make decisions about the generation, analysis, interpretation, and reporting of data when multiple languages are involved, along with the complex, negotiated relationships between the researcher and the researched (Holmes et al., Reference Holmes, Fay, Andrews and Attia2013, Reference Holmes, Reynolds and Ganassin2022). Holmes et al. (Reference Holmes, Fay, Andrews and Attia2013, Reference Holmes, Fay, Andrews, Attia and Zhu2016) established the RM-ly framework, which revolves around three dimensions: intentionality, spatiality, and relationality. Intentionality refers to the awareness researchers develop in multilingual research, and their ability to recognize opportunities and reflect on the uniqueness of their investigations. Spatiality concerns understanding the multilingual characteristics and context of each research subject, the linguistic resources of the researcher, and potential representation. Relationality involves building and sustaining relationships with participants and stakeholders and identifying linguistic resources that facilitate these processes. RM-ly challenges the monolingual mindset. Instead of merely adding one language to another – yielding dual monolingualism (Prinsloo & Krause, Reference Prinsloo and Krause2019) – it is imperative to explore synergy between multiple languages across all stages of the knowledge-generation process.
We expanded the RM-ly approach to the EMI learning and teaching contexts, aiming to examine: (1) how students’ English language proficiency and subject content knowledge can be enhanced; (2) how EMI researchers can maximize opportunities during their research process, stimulating self-reflection and raising their multilingual awareness, thereby empowering applied linguists, students, and teachers in underrepresented contexts to be heard; and (3) how to ensure educational equity as the expanding disparity in individual students’ access to learning resources leaves the less privileged at a disadvantage. To address these issues, the colloquium featured six paper presentations that tackled EMI policy and practices from various perspectives (e.g., indigenous, sociocultural theory, and critical) across different educational contexts (e.g., secondary schools, higher education).
2. Papers
2.1 Researching EMI policy and practice multilingually: Reflections from China and Turkey
Jim McKinley (University College London), Heath Rose (University of Oxford), Kari Sahan (The University of Reading), and Sihan Zhou (The Chinese University of Hong Kong) opened the colloquium by sharing their position on “responsible EMI research” and recognizing the inherently multilingual nature of EMI studies. They then outlined the RM-ly framework through a lens of “researcher reflexivity”, which entails the various values, views, and knowledge that researchers bring with them into the contexts that they study (Johnstone Young et al., Reference Johnstone Young, Ganassin, Schneider, Schartner and Walsh2022). Through the critical exploration of the three dimensions of the RM-ly framework, and guided by the methodology of Ganassin and Holmes (Reference Ganassin and Holmes2020), they introduced a new analytical study (McKinley et al., Reference McKinley, Sahan, Zhou and Rose2023) that explored the challenges and affordances of RM-ly in two previous EMI research projects conducted in China and Turkey.
For the project in China, the four-person research team comprised L1 English speakers and Chinese (L1)–English bilinguals. In evaluating 93 EMI policy documents and interviewing 26 policy stakeholders in eight Chinese universities, they used both languages throughout the research process, from collection and analysis of data to final dissemination of findings in reports and at conference presentations. Using a translanguaging approach for the fieldwork, they avoided conventional translation or interpreting methods. The approach allowed the researchers to develop increased awareness of multilingual possibilities. For the Turkish study, the third presenter, an L1 English and second language (L2) Turkish bilingual, reflected on her assessments of policy documents and interview data in both languages. The study involved conducting observations of EMI classrooms, creating research and coding frameworks to analyze the multilingual data, and observing frequent usage of both English and Turkish. The reflections on these two EMI studies demonstrated the benefits of the multilingual approach the researchers took at each step of the research process, from the creation of a multilingual research team to the bilingual presentation of the research findings.
In conclusion, the presenters offered six recommendations for conducting EMI research in an RM-ly informed manner. These are: (1) multilingual researchers and/or research teams are necessary for conducting effective EMI research, as EMI is in itself a multilingual phenomenon; (2) all decisions throughout an EMI research project should be underpinned by a multilingual mindset; (3) outsider EMI researchers should strive to minimize any perceived linguistic or cultural power differences between them and their participants; (4) researchers should maximize the utilization of translanguaging during data collection in EMI research; (5) the outcomes of the research should be of benefit to the EMI contexts being researched; and, to facilitate this, (6) research dissemination should be not only in English but also in the main languages of the stakeholders in the EMI context.
2.2 Capitalizing multilingual resources for EMI in the borderlands between China and Myanmar: Myanmar students’ transnational experiences
Yongyan Zheng (Fudan University) presented the findings from an ethnographic study conducted with Jia Li (Yunnan University). The study looked into ethnic minority learners’ EMI learning experiences in borderland institutions in Myanmar (Li et al., Reference Li, Zheng and Yan2023). First, Zheng introduced the framework of borderlands theory (Anzaldúa, Reference Anzaldúa1987). This perspective refers to border culture as a new consciousness that tolerates contradiction and ambiguity, and that foregrounds the knowledge the multilingual speakers possess and bring to the mainstream classroom. Zheng and Li's study sought to reveal the challenges that ethnic minority students encountered when they were engaged in EMI learning in Myanmar, how they overcame these challenges, and under what kind of structural constraints. To answer these questions, they focused on four participants purposefully selected from a large pool of 76 that were involved in a large-scale ethnography. A qualitative approach was adopted to analyze multiple sources of data.
The findings showed that the students’ and their teachers’ insufficient English proficiency and lack of subject-specific vocabulary knowledge constituted the first challenge. In particular, the bilingual model of using English and the Myanmar language (the national language) in EMI instruction made it more difficult for the ethnic minority students to comprehend course contents. To deal with this challenge, the students resorted to their multiliteracies and local community resources, and agentively exploited digital and technological resources afforded by their borderland positions. However, it was found that an array of structural factors, such as financial stress, poor educational infrastructure, and bleak employment prospects in their local context, still constrained their learning.
Zheng concluded that the challenges encountered by these students were similar to those reported in other contexts, but what is unique in this study is that these students found alternative ways of dealing with these challenges, particularly by resorting to their borderland repertoires. This finding is important as it suggests that ethnic minorities in the marginalized areas, such as on the borderland, could also become active contributors to knowledge construction in the EMI setting. The finding also problematizes the use of nation-state as a frame of reference for EMI research, as borderlands are ambiguous areas without a clear-cut boundary, and border dwellers become jugglers between different linguistic and cultural traditions. From this perspective, it is dangerous to stick to a one-size-fits-all model in implementing EMI education. It is imperative to identify the specific types of multilingual resources that ethnic minority students on the borderland may actually need. At the end of the presentation, Zheng called for more research to be conducted in similar peripheral contexts to valorize alternative ways of learning and thinking, and to enhance marginalized learners’ equal participation in EMI education.
2.3 More like TMI: Translanguaging “en privado” among college students in an EMI program in Spain
Josh Prada (University of Groningen) presented a study he conducted that examined the nature and role of student translanguaging “en privado” (in private) within an EMI program (bilingual track) in teacher education at a Spanish university. Against the backdrop of the rise of EMI around the globe, Prada began the presentation with an overview of the general picture of EMI education in the Spanish university system. By quoting statistics and reviewing previous studies, Prada showed that more than 1.5 million students were involved in EMI education, and the majority of EMI programs were taught by non-native speakers of English. The complex interplay between different languages (Spanish, English, and other languages) and content learning, coupled with vague EMI policies, has given rise to ambivalent attitudes to EMI among both teachers and students. Prada went on to differentiate several key concepts, namely translanguaging, Ubuntu, trans-semiotizing, corrientes (i.e. translanguaging currents), and showing-telling, and then proposed the overarching research question: “What do college students do with their linguistic and broader semiotic repertoires in an EMI classroom in order to make meaning and sense of course content when en privado?”
To answer this question, Prada examined the strategies employed by Spanish college students to interpret and create meaning within an EMI classroom setting en privado. Through individual interviews with six students, Prada connected translanguaging practices to a range of learning activities, including in-class note-taking, at-home concept exploration, in-class small group discussion and peer collaboration, and composition writing. Beyond the initial examination of grassroots literacy practices, his analysis also revealed how students linked their covert translanguaging practices to the development of their identities as emerging multilinguals, although perceptions of this association varied among students. In particular, Prada highlighted the connection between translanguaging and emergent feelings of empowerment, agency, and security, thus highlighting the socio-emotional aspect of translanguaging in EMI learning.
In conclusion, Prada emphasized the significance of students’ translanguaging en privado despite their acknowledgement of teachers’ promotion of English-only behaviors. The students utilized extra resources in Spanish and English as well as multimodal resources, flowed between these resources to make meaning, and ultimately translated them into English for coursework purposes. Prada encouraged researchers to focus on how students bring their home literacies and digital skills into their learning in ways that bridge meaning-making activities outside and inside the classroom.
2.4 Translanguaging and transknowledging: EMI-STEM teachers in higher education
Mingyue Gu (The Education University of Hong Kong) started the fourth presentation by providing a comprehensive review of the roles of translanguaging in EMI learning. Gu established the connection between translanguaging and knowledge co-construction in EMI and Content and Language Integrated Learning (CLIL) contexts and explained that translanguaging enables the recognition of knowledge plurality and facilitates knowledge production ecologies situated in the local context. She then described the study she conducted with co-author Wanyu Ou (Chalmers University of Technology), which aimed to observe moment-to-moment knowledge construction in EMI classrooms, to understand the role of translanguaging in two-way knowledge exchange, and ultimately to contribute to the theorization of transknowledging – that is, the moment-to-moment knowledge co-construction process entailing “two-way exchange of knowledge systems” (Heugh, Reference Heugh, Harding-Esch and Coleman2021, p. 44).
Gu defined knowledge as “historically, ideologically, and socio-culturally situated and mediated by languages and semiotics”. She explained the combined theoretical framework of translanguaging and transknowledging, in which the systematic use of translanguaging pedagogies and culturally responsive pedagogies collaboratively facilitates the two-way knowledge exchange and production between L1 and L2 knowledge systems. She then explained that the study was embedded in an interdisciplinary STEM institute affiliated with a top-tier comprehensive university in southeast China and that she recruited four teacher participants. Data were collected through individual interviews, class observations, field notes, and video recordings.
Next, Gu elaborated on the four case participants’ translanguaging strategies. Jerry employed a strategic approach to English-dominated translanguaging, actively using multimodal resources and visualized verbal explanations. Ye regarded teaching as embodied performance enacted through the coordinated use of linguistic and non-linguistic resources. May considered translanguaging a response to knowledge construction flow, and adjusted her language use based on monitoring the students’ moment-to-moment comprehension of content knowledge. Zhao used translanguaging as a process to validate the L1 knowledge system by legitimizing the students’ linguistic repertoires and related knowledge in L1.
Gu concluded her presentation by discussing the contributions of the transknowledging perspective. She pointed out that while all the teachers participating in the study were aware of the importance of mobilizing students’ knowledge in L1 to scaffold knowledge construction based on L2-mediated (EMI) learning materials, and of the differences in L1- and L2-related epistemic frameworks, they were not aware of the possibility of establishing a new epistemic framework based on students’ exposure to different knowledge systems. Thus, a new perspective striking an ideological balance between L1 and L2 knowledge systems would help EMI teachers maximize the scaffolding potential of students’ L1 knowledge system, and ultimately help transform EMI into a space for all parties to experience and exercise multiple value systems, subjectivities, pedagogies, affections, and changes.
2.5 Developing Vietnamese teachers’ capacity for CLIL instruction
Hang Thi Thu Nguyen (VNU University of Languages and International Studies) presented the next co-authored paper, which reported on a multinational research project on teachers’ professional capacity for CLIL instruction in Vietnam. Nguyen first reviewed the conceptual framework of CLIL by emphasizing its dual focus on language and content and then described the educational landscape of CLIL policy and implementation in Vietnam. To contextualize her team's study, she provided a historical review of Vietnam's educational policies. The Vietnamese government initiated a national project to implement CLIL in several selected high schools, vastly increasing the demand for qualified CLIL teachers and professional development programs.
Nguyen reported on the results of a qualitative study that employed the Cultural-Historical Activity Theory (CHAT) framework (Engeström, Reference Engeström2001) and examined Vietnamese teachers’ professional development paths and capacity-building strategies relating to CLIL teaching. They collected data through online interviews with 52 CLIL practitioners. A thematic analysis of the interview data showed that, despite the teachers’ high level of enthusiasm for improving their teaching practice through professional development, they were restricted by limited exposure to CLIL pedagogy in their initial teacher education, inadequate English-language competence, limited teaching resources (including teaching materials and support), and an absence of community engagement and quality professional learning opportunities. In particular, Nguyen pointed out a lingering misconception of CLIL that oversimplified it as a mere language teaching of the subject. This misunderstanding might, in the long run, threaten to undermine the ultimate goal of CLIL as being dual-focused on both language and content.
Nguyen concluded the presentation by discussing the implications of using the CHAT perspective to investigate CLIL teachers’ professional learning. Her team's study contributes to the advancement and refinement of CHAT as a conceptual tool for understanding teachers’ professional development and learning. She also pointed out that future professional development needs to focus on creating opportunities for all stakeholders including teachers, managers, and policy makers to engage in critical reflection on their shared practices and work together to develop teachers’ capacity for CLIL teaching readiness.
2.6 Between a multilingual ethos and an English pathos: How multilingual scholars navigate the international academia
Sílvia Melo-Pfeifer (University of Hamburg) concluded the colloquium by presenting her study on multilingual researchers’ practices in multilingual academic life, including publishing, engaging in scientific collaborations, and instruction (Melo-Pfeifer, Reference Melo-Pfeifer2023). She anchored the study by offering a refined understanding of EMI: EMI is not just a process situated in a specific institutional context. Rather, she claimed, it should be explored as lived by scholars, as ecologically entrenched in their multilingual repertoires, their research profiles, and the multilayered facet of their academic lives. She engaged in a critical discussion of the three components of the concept EMI, by exploring the specificities, controversies, and dilemmas of what is meant by “English”, “Medium”, and “Instruction”. By discussing these issues, she showed the nuances, changes, and challenges that take place in researching EMI multilingually.
Her study addressed two research questions: (1) how do multilingual researchers make choices related to the different tasks they are called to perform in academia (teaching, researching, publishing, etc.)? and (2) how do they perceive and react to the role of English in multilingual higher education?
Melo-Pfeifer's study analyzed the reported practices of 25 multilingual European scholars researching multilingually and working on a project entitled “Evaluation of Competence in Intercomprehension”. These 25 individuals represented 14 partner institutions from eight countries, and each had more than 15 years of experience in academic multilingual collaboration. A questionnaire including both closed and open questions was distributed to all 25 members to investigate the linguistic ideologies and linguistic research dynamics characterizing the group. Every participant reported actively using at least four languages in addition to English to carry out various tasks within their institutions (including, for example, Catalan, French, German, Italian, Portuguese, Romanian, and Spanish), and French was the main language of publication. These scholars, the analysis showed, opted for an English–French bilingual mode in academic presentations, actively used nine languages in daily academic life, and reported only using English in instruction when combined with other European languages (e.g. Portuguese or French).
In discussing the findings, Melo-Pfeifer highlighted the struggle that the research community has in finding a balance between the (perceived as mandatory) use of English (called “English pathos”) and other languages in both knowledge production and knowledge dissemination about multilingualism (called “multilingual ethos”). She found out that the linguistic profiles of the scholars and their research fields (e.g. working on intercomprehension between Romance languages and multilingual education) contributed to their strong ideological promotion of multilingualism and implicit rejection of English. Ironically, though, these scholars mostly used French as the single language for academic communication, reproducing the pattern of using a lingua franca in higher education, substituting English with another language. In conclusion, Melo-Pfeifer called for an ecological and holistic understanding of “EMI as lived” by multilingual scholars, interconnected with their multilingual repertoires, their research domain, and the multi-layered domains of academic life. Finally, she proposed changing the perspective of EMI to “English as (one) medium in academic life”, which acknowledges the need to remain open to the creation and maintenance of translanguaging spaces in instruction where English and other linguistic resources co-exist. She made the point that the “E” in EMI can go beyond English, and the “I” can go beyond Instruction.
3. Discussion and conclusion
Angel Lin (The Education University of Hong Kong) assumed the role of the discussant for the symposium, summarizing the prominent themes emerging from the researchers’ RM-ly-informed studies of EMI policies and practices. These themes included RM-ly practices in conducting EMI studies in various contexts; the cultivation and application of borderland repertoires; the process of transknowledging in EMI-STEM classrooms; inherent conflicts within diverse goals or activity systems; and the tension between a multilingual ethos and an English monolingual ethos. Lin employed the apt metaphor “紧箍咒 (band-tightening spell)” to describe the restrictive monolingual mindset frequently encountered in the field of language education and EMI research. Originating in Chinese culture, this metaphor refers to the stringent control or constraints exerted by external forces. Lin also discussed the immense challenges facing EMI research. These include breaking boundaries to leverage the transformative power of translanguaging across various EMI contexts and challenging deep-seated ideologies, practices, and attitudes. Lin stressed the need for future research to examine subjectivities and emotions, inviting a deeper exploration of the desires, fears, and struggles of both researchers and participants. This introspective focus could lead to future research offering richer, more nuanced insights into EMI policies and practices.
In summary, the contributions to the colloquium exemplify the conceptualization, execution, and introspection integral to RM-ly-informed EMI research. The integration of RM-ly principles and practices enriches the EMI research landscape, delivering advantages to both researchers and their research contexts. The colloquium lays a solid foundation for the progressive transformation of EMI policies and practices, steering them further towards an educational paradigm that values dynamism, humanity, and equity. We plan to continue addressing these important lines of thought in EMI research through a special issue of Language and Education.
