Revisiting the ‘E’ in EMI: students’ perceptions of standard English, lingua franca and translingual practices

ABSTRACT

Conceptualizations of English as standard, as a lingua franca, or as part of translingual practice form part of the discourses surrounding its use in EMI. While researchers generally agree that the ‘E’ in EMI should not stand for native varieties of standard English, the stakeholders’ perceptions of English call for further research. This paper addresses this gap by examining students’ conceptualizations of English in an EMI programme at a Swedish university. Drawing on interview data collected from local and international students, the analysis focuses on students’ conceptualizations of English in connection to their positionings. The analysis shows that all three above-mentioned conceptualizations are present. The tensions in the students’ conceptualizations of English and positionings point towards issues related to power relations, group dynamics, social integration, and learning. The analysis shows that translingual practices in EMI contexts are not always associated with empowering the students by allowing them to resort to their L1s to fill gaps in their English. Translanguaging can also function as a mechanism of exclusion and reinforcement of language standards by a group of ‘elite’ translinguals. The idea of what is acceptable English in EMI is not static and can move along the standard – non-standard continuum.

KEYWORDS:

• Language ideologies
• conceptualizations of English
• English as a lingua franca
• translanguaging
• linguistic diversity
• multilingual university

1. Introduction

Over the last decade, English-medium instruction (EMI) has been subject of extensive research in different branches of applied linguistics. This year has seen a further increase in the number of studies conducted in various educational and geographical contexts published in monographs (e.g. Macaro 2018), edited volumes (e.g. Murata 2019) and special journal issues (e.g. Pecorari and Malmström 2018). An expanding phenomenon, EMI is still being defined and re-defined as an object of study. Drawing on different definitions, Pecorari and Malmström (2018, 499) describe four characteristics of EMI settings, three of which are concerned with language: ‘English’ is central to defining EMI settings since it is ‘the language used for instructional purposes’, although ‘not itself the subject being taught’, and it is also a second or additional language ‘for most participants in the setting’. As far as language and content integration are concerned, the use of English in higher education has been described as a continuum ranging from pre-sessional English for Academic Purposes (EAP) courses to full EMI or Integrating Content and Language in Higher Education (ICLHE) programmes (e.g. Schmidt-Unterberger 2018). Every EMI context ‘has its own characteristics’ (Doiz, Lasagabaster, and Sierra 2013, 219) and constructs its own ‘language regime – its own set of rules, orders of discourse, and ideologies – in which linguistic resources are assessed differently’ (Busch 2012, 520).

The diversity and variability of English language uses at the international university has also been subject of substantial research (e.g. Mauranen 2012; Canagarajah 2013; Jenkins 2014). As English became the medium of instruction in ‘countries and jurisdictions where the first language of the majority of the population is not English’ (Macaro et al. 2018, 37), the prevalence of language norms associated with education in the Anglophone countries was put into question, particularly by researchers studying English as a lingua franca and translingual practices. Following a dramatic increase in the number of international students at universities, the standards of English have also been questioned in officially Anglophone universities (e.g. Murray 2017). Applied linguistics researchers tend to agree that language tests and other forms of assessment based on standard English norms are not the most adequate ones to measure students’ language skills (e.g. McNamara and Shohamy 2016) or to deliver EMI courses.

At universities outside the Anglophone world, English is used primarily as an academic lingua franca by L2 users of English. A recent definition of English as a lingua franca (ELF) views it as ‘multilingual communication in which English is available as a contact language of choice, but is not necessarily chosen’ (Jenkins 2015, 73). This shift towards multilingualism brings ELF research closer to translanguaging, although the two approaches do not share the same conceptual basis (cf. Pennycook 2010, 85; Canagarajah 2013, 68 on Lingua Franca English). Translingual practices involve the mixing of two or more languages, varieties or codes to fulfil a communicative purpose (e.g. García and Wei 2014). In university contexts, English plays a key role in such practices to the extent that Canagarajah (2013, 68) even conceptualizes English ‘as a translingual practice.’ This perspective on language use challenges the assumption that sharedness and uniformity of norms are required for communicative success. The outcomes of translingual practices may vary as they ‘will approximate “standard language” in some occasions and vernacular in others, both mediated by the diverse other language repertoires which students bring with them’ (Canagarajah 2013, 188).

Over the last decade, sociolinguistic research has challenged monolingual norms and conceptualizations of languages as static entities (e.g. Pennycook 2010). At the same time, it can be argued that standard written English is easier to define compared to its spoken varieties (e.g. Crowley 2003). Standard written English is often associated with the language of literature and viewed as an accepted norm which does not belong to any specific community and needs to be learnt in educational settings by everyone, including native speakers. This position has been popular among EAP and Academic Literacies practitioners and researchers (e.g. Tribble 2017).

Conceptualizations of English – be it standard, lingua franca, or translingual – form part of the discourses surrounding its use, which are central in English-medium education in multilingual university settings (Dafouz and Smit 2016). What kind of English is expected in EMI may seem secondary compared to some more practical concerns of such programmes. However, this question has repercussions for EMI, as different conceptualizations of English may affect teacher-student interaction, student-student interaction, group dynamics, assessment practices, and so forth. These conceptualizations form part of language ideologies, which are socially situated and connected to questions of identity and power (Woolard 1998). Together with the ‘baggage’ of linguistic resources (often conceptualized as languages x, y, and z) and lived experience of language, language ideologies also form part of an individual’s linguistic repertoire (Busch 2012, 2017).

This article focuses on an EMI context in Sweden, one of the pioneers and major providers of EMI in Europe (e.g. Hultgren, Gregersen, and Thøgersen 2014). The research carried out on EMI at Swedish universities has focused primarily on the challenges experienced by Swedish L1 students and teachers when switching from Swedish to English (see Kuteeva 2019 for an overview). Some studies took a more holistic approach, including international students who do not have Swedish as their L1, but these were usually in a relative minority. For example, Söderlundh’s (2012, 2013) ethnographic studies of six EMI courses were important in showing the central role that the Swedish language plays in the officially English-medium courses and in conceptualizing EMI as a local practice (cf. Pennycook 2010). The majority of students and teachers in Söderlundh’s studies were L1 speakers of Swedish, and the rest were primarily exchange students. In that context, English was constructed as a language of exchange students, whereas Swedes were associated with English and Swedish (Söderlundh 2013). At the same time, varieties that index native or near-native knowledge of English appeared to be ‘the only truly, non-negotiable mobile linguistic resources at the international university’ (Söderlundh 2013, 129).

The question of statuses attached to different native varieties and translingual practices has not been sufficiently addressed, not only within the Swedish context but also elsewhere. While researchers have argued that the ‘E’ in EMI should not stand for (native) varieties of standard English (e.g. Smit 2010; Galloway and Rose 2015), the stakeholders’ perceptions of English have received less attention, with the exception of few studies (e.g. McCambridge and Saarinen 2015). This article aims to address this research gap by examining students’ conceptualizations of English in one of the first EMI undergraduate degree programmes at a Swedish university.

2. Background and study

The study reported in this article was conducted as part of a larger case study focusing on the use of English in an EMI programme in Business Studies, a subject highly suitable for EMI (Macaro et al. 2018). In the year when the data for this study were collected, the programme attracted approximately 80 students, both local and international, in roughly equal proportions. Here I draw on interviews with students at the end of the first year of their EMI programme. This dataset proved to be particularly rich in providing insights into students’ conceptualizations of English, as studies on self-reported language elicit the participants’ perceptions of practice and are suitable for researching the ways in which languages are talked and written about (e.g. Hultgren, Gregersen, and Thøgersen 2014).

Several students volunteered to be interviewed about their experience of using English and other languages in the EMI programme. Informed written consent was obtained. As with all self-selected participants, we may assume that these students were generally confident in their language uses and/or had some opinions they wished to share with the researcher. A more purposeful selection of five students was meant to reflect the composition of student population in the EMI programme and its multilingual setting, with more than 15 L1s among students. Both genders were included, 3 males and 2 females. For reasons of anonymity, the year of data collection is not provided, and the students are given unisex pseudonyms. Also, for anonymity purposes, the proportion of genders is swapped in the presentation of the findings below (3 females, 2 males) and assigned randomly to individual participants. The analysis does not focus on gender issues, and this swap was made solely for the sake of clarity in the presentation of results and in order to avoid referring to all participants using the same pronoun. English language proficiency was not measured and does not play a role in the analysis either, although it can be noted that all five study participants would be placed somewhere between the middle to higher range in their programme. Due to the competitive requirements of the given EMI programme, lower English language proficiency was relatively rare. Table 1 summarises relevant information concerning each participant’s L1, previous experience of English-medium instruction, and knowledge of languages other than their L1.

Table 1. Profiles of interviewed students.

Student	L1	Previous EMI experience	Other languages
Alex	Swedish	International Baccalaureate (IB) programme in Sweden	English
Robin	Swedish	None	English
Sam	English/Swedish	School in the US, IB in Sweden	English, French, German
Pat	Hindi	EMI school and university in India	English, Japanese
Kim	German	Boarding school in the UK	English, Spanish, Swedish

The semi-structured interviews were conducted in English with each participant by the present author, who was not involved in the teaching of their programme in any capacity and had not met the students previously. Since the researcher is not a native speaker of either English or Swedish, and all study participants had used English as their working language for several months before the interview, it is unlikely that the choice of English as the interview language resulted in a power-infused relationship. The students were asked what languages they used in their studies and for what purposes, how they perceived their own English in relation to fellow students and teachers, and how they viewed ‘good’ English. In addition, the students were asked to give examples of successful and challenging communicative situations in their EMI programme. The interviews lasted between 30 and 45 min, were audio-recorded and manually transcribed using simplified transcription (Appendix).

Data analysis proceeded in two stages, first focusing on the content related to students’ conceptualizations of English and then delving deeper into how these conceptualizations were expressed through students’ positionings in their interview accounts. The analytical concept of ‘positioning’ (Davies and Harré 1990) as adapted in bilingualism research proved very useful in the context of the interviews, since positioning is viewed as ‘the process by which selves are located in conversation as observably and subjectively coherent participants in jointly produced story lines’ (Blackledge and Pavlenko 2001, 249). Positioning is further divided into interactive, which assumes one individual positioning the other (e.g. a student commenting on their peers’ English), and reflective, which refers to the process of positioning oneself (e.g. a student talking about their accent or the kind of English s/he uses in written assignments). It is noteworthy that subject positions are not stable entities, as people are constantly involved in producing selves and others, so the positionings discussed below are limited to the context of the interview events.

The first round of analysis focused on the content and identified relevant passages connected to students’ conceptualizations of English. Several rounds of close readings were required to group relevant excerpts under three focal categories: ‘standard English’, ‘English as a lingua franca’ and ‘translingual practices’. The three main categories included subthemes that could be traced across different interview accounts. Thus, ‘standard English’ includes the perceived prominence of British English in the EMI programme, a variety of native and non-native English varieties among students and staff, and the importance of correctness in academic writing. ‘English as a lingua franca’ consists primarily of students’ accounts of communicative strategies (or lack thereof) to ensure mutual understanding in group work and other student interaction. It also covers the perceived importance of clarity and understandability. ‘Translingual practices’ includes students’ accounts of mixing English and Swedish in different ways and the attitudes expressed towards such practices and resulting outcomes.

In order to ascertain the reliability of the data analysis, the results were discussed and verified with a colleague who had been involved in the teaching of this EMI programme for several years and is experienced in qualitative research on academic uses of English. In order to demonstrate patterns in the students’ conceptualizations of English, the findings for each theme are presented in parallel order, i.e. starting with general trends and then focusing on individual students. This order of presentation also makes it easier to unveil students’ positionings in relation to other actors involved in the EMI programme.

3. Results: students’ conceptualizations of English in their EMI programme

3.1. Standard English

The data analysis has identified several references to major varieties of English (e.g. British, American, Australian, Indian, etc), usually in connection to what the study participants deemed to be ‘good’ or ‘proper’ English. Overall, all study participants perceived British English to be the most prestigious variety in the context of their EMI programme. This preference may not be surprising given the fact that for many decades Swedish schools had taught British English, although this practice changed at the turn of the millennium as English became to be perceived as more international and at the same time gained a more local dimension in Sweden (e.g. Hult 2012).

Before joining the EMI programme, Alex had been taught British English in the IB programme. She positioned her English relatively high, particularly in comparison to the teachers: ‘it felt that … you knew more English than the teacher did even though they knew more about their subject’. Robin was taught English as a foreign language at a regular Swedish school and recollected that: ‘it’s always been British with the words and the pronunciation … It feels more proper I don’t know why … yeah, I would like to speak as British as possible’. Sam is an L1 speaker of American English but had adapted to using British English in the IB programme in Sweden: ‘I’ve really shifted myself into writing academically in British. It does feel more correct for me now’. When asked to elaborate on what he means by writing academically in British, Sam mentioned spelling and the referencing system and went on to confess:

(1)

I do feel that there is a slight kind of stereotype against Americans … British seems to be a little bit more proper, a little bit more formal, whereas Americans- our accents sound a little bit more laid back … So, even during my high school years with the IB programme, I voluntarily chose to start- I’d actually have to sit in my essays and edit them afterwards, when I’d finished writing them, just so I could shift it to British. Now it comes naturally to me … I don’t notice it anymore, but I do remember that I did try and actively try and change my English- or academic English to British. (Sam)

Having worked for several years in multinational companies in Sweden and another country in East Asia, Pat is rather perplexed by this liking for British English outside the UK:

(2)

I observed that people have liking for British English and I mean, when I worked with some people from the UK or from Britain … I can see that- I mean of course it is their native language so they are more- they know more vocabulary and they have more ways to express their thoughts in proper English but why people who are not from UK have tendency or like- for liking British English? That I never understand. (Pat)

But, similarly to Sam, Pat also gave in to accommodate: ‘I’m now interacting with more university students … they try to use British English … I started using that- like British English accent as well and in a way I like it.’

However, this perceived higher status of British English, expressed both in reflective (e.g. Alex, Robin) and interactive positionings (e.g. Sam, Pat), did not correspond to the reality of linguistic practices of students and teachers in the EMI programme. It is natural that, in a multilingual setting with over 15 L1s, various kinds of Englishes would be used by students and teachers alike, particularly in spoken interaction. Kim gave a fairly accurate description of their educational setting:

(3)

British English is- like people recognize it but I don’t think is as widely used as an accent people would take on when they learn English. We have like one girl in the class she speaks with an Australian accent and a girl that speaks with a South African accent and they really stand out because it’s not so common. Most people pick up this American, Swenglish, whatever their mother tongue is accent. So yea, I think that’s more common. But I wouldn’t say it’s proper English if you speak American, it’s just the one that people are most used to in a way. (Kim)

It is interesting to note that some native varieties of English (South African, Australian) were perceived by Kim as ‘standing out’. Kim described the kind of English she speaks as an ‘island in the Atlantic ocean’, i.e. a blend of British and American. When contrasting her everyday English with the kind of English used in academic contexts, Kim described the latter as more ‘proper’, meaning more sophisticated.

Four students, with the exception of Pat, mentioned an increased importance of correctness in academic writing. Alex, Sam and Kim also reported acting as editors and proofreaders of other students’ writing and aligning it with what they perceived to be a standard English norm. As the excerpt below demonstrates, the students in this EMI programme produced academic texts of somewhat varied quality, which often included non-standard forms. The use of such non-standard forms in academic writing was not perceived as acceptable by students with higher English proficiency. For example, Alex described herself as a ‘perfectionist’ in writing and talked about ‘fixing’ collaboratively written texts:

You can see the level of differences in some people’s English especially when it’s written … I’ve kinda fixed it so it looked like- it looked better … you notice this when your English is not on par with the teammates. (Alex)

3.2. English as a lingua franca

Different accounts of using English as a lingua franca emerged in connection to describing instances of how communication unfolded during group work. Interviews with Robin, Pat and Kim contained most mentions of such situations. Since the boundary between ELF – at least in its current conceptualization – and translingual practice is somewhat blurred, this section focuses mainly on the data excerpts that refer to communication between speakers who have other L1s than English or Swedish. Two students have reported using what can be described as typical ELF strategies (e.g. Mauranen 2012), e.g. exemplification (Robin), clarification (Robin), repetition (Pat), and rephrasing (Pat). On the other hand, Kim described instances of failed communication in lingua franca contexts, which she attributed to varying degrees of English proficiency among the groups she had interacted with, as discussed below.

Robin seems to have a fairly positive experience of interacting in English with his peers. He described a supportive environment in which students try to help each other to fill lexical gaps and to ensure successful communication. He positioned himself on par with other peers and reported actively engaging with international students when asking for help:

(4)

it feels like we’re kind of on the same level and we try to help each other if we don’t find the words and- so it feels- good … sometimes it could be hard to even explain the word, you know and then you could turn to a Swedish student but I try to ask the international ones first. (Robin)

Robin was also positive about group work with students with different L1s:

If I’m mostly with international students we try to- I try to explain the words or try to give an example like I think I mean- you mean like this- but if it’s Swedish it’s very easy because someone usually knows how to express it if you’re stuck.’

Pat mentioned no misunderstandings or failed communication with students and teachers who have Swedish as their L1 because she had spent several years living in Sweden and working with Swedes at a multinational company. When it comes to international students with other L1s, there were more misunderstandings but Pat did not put all the blame for such misunderstandings on others and positioned herself as responsible for ensuring mutual understanding:

(5)

It could be because … I couldn’t understand what they said or maybe I said something which they didn’t understand or listen to right … I’m a bit cautious when I say something to people … and I make sure that what I’m saying, or what I think I’m saying, if it is meaningful and rational, then they get what I am trying to say. Maybe I have to repeat it once or twice. (Pat)

Apparently, Pat had developed various communication strategies through interacting in English as a lingua franca in professional and academic settings before joining this EMI programme: ‘I myself have changed a lot. I used to speak like Indian English but when you go out and you see that you are not- I mean people are not able to understand you, you have to adjust yourself’. By interacting in English with people from different L1 backgrounds, Pat had developed strategies such as repetition and rephrasing:

(6)

I worked with people from Latvia or Estonia … maybe I need to ask two or three times that is this what they mean … I mean when they say something, they said something, some sentence and then I interpret it and I try to put it in my own lang- in my own words and then I say ‘do you mean this?’ and they say ‘no’ and then I put it in another way and they say ‘yeah maybe’- so in that way. (Pat)

However, despite appearing to be an effective communicator, Pat believes that it is better if at least one of the parties speaks their L1 but she fears that learning Swedish would make her English worse. Ideally, she would prefer interacting with Swedish people in Swedish ‘because if two persons speak … non-native language, they have restricted vocabulary so they are restricted in their conversation’.

Kim reported using English as a lingua franca in the hall of residence with all students, including L1 speakers of Swedish but in the EMI programme she tried to speak Swedish with L1 Swedish classmates. She also mentioned a couple of incidents involving international students in the EMI programme, where the stakes were higher, and a misunderstanding resulted in a lower grade:

(7)

for example, we have a Chinese girl in our class, and she has a bit of problems with English, so she doesn’t understand what I’m telling her so when we, we were in the same group and she didn’t understand what I wanted from her so we had a misunderstanding in the group work which resulted in a worse grade. (Kim)

Apparently, the Chinese student was supposed to be the last person to add her part and submit the assignment but did not do it and the group missed the deadline. When asked to elaborate on why this misunderstanding took place, Kim explained:

(8)

Yeah, because we were sitting in the group and it was like yeah, so you’re the last person to add your part so just like, you know, let the spelling thing go over it once and then if it’s all right just send it in. That’s what we said and she was like ‘Ah, ok’. She didn’t really say obviously that she didn’t understand what we said and so it didn’t turn out the way it was supposed to. (Kim)

From the description above, it appears that the students in the group did not consider that the allocation of tasks and responsibility requires mutual understanding between all group members. Thus, the student with what they considered the lowest English proficiency was given a seemingly straightforward task but did not actually understand the instructions. The Chinese student nodded in agreement displaying an ‘illusion of understanding’ (Pavlenko 2018). This incident points towards a lack of communicative strategies required for communication with L2 speakers. Kim’s reflective positioning as a fluent and competent speaker of English and the interactive positioning of others as linguistically incompetent did not lead to successful communication.

On another occasion, Kim had to work in a mixed group involving two students from China, one from Lithuania and one from Iran, which s/he described as ‘a bit messy sometimes’. Again, this was due to varying levels of proficiency in English, which required Kim to slow down and explain or repeat some things during the preparation stage and then to correct the written text produced by different group members (cf. Alex, section 3.1). As a result, ‘there’s been like a dynamic in the group for those that are on the same language level to stick to each other because it makes working just a lot easier’ (Kim). Similarly, following just one incident of working in a mixed group which included two students from China, Sam had avoided working in such groups in the EMI programme: ‘since then I have sort of tried to find at least people who speak Swedish or English.’ This group dynamic leads to separation of students on the basis of their language proficiency. Even though language is not central in an EMI programme, as definitions of EMI assume, it still becomes an important factor which inhibits learning and creates segregation across the student population.

In high-stakes situations such as those involving examination and written assignments, all students stressed the importance of clarity and understandability, especially as far their teachers’ language uses were concerned. For example, Sam suggested that some teacher training would be in order: ‘when you have a student that speaks one language and a teacher that speaks another language … sometimes there’s different approaches to English and they might not really understand’ (Sam). Pat underscored the importance of clarity even more strongly: ‘I think for me now it is more like the way the person express his thoughts in English and whether he or she is able to communicate clearly’. Kim talked about the importance of clarity and precise expression in writing, for example in connection to formulating exam questions by their teachers.

Overall, it appears that Pat, Kim and Sam perceived using English as a lingua franca to be potentially wrought with problems, particularly if it involves people with different L1s and lower levels of English proficiency. Alex, Sam and Kim positioned themselves as language regulators in such mixed groups, responsible for editing the final version and fixing English. Only Robin reported actively seeking interaction with students who do not share L1 Swedish. As a result of only one experience of failed communication, Sam deliberately chose to do group work with Swedish L1 students.

3.3. Translingual practices

All study participants commented that most translanguaging was taking place between English and Swedish. Both students and – less often – teachers may translanguage, usually in connection to specific terminology. A great deal of translanguaging took place in student group work involving the majority of Swedish students, as one would expect considering the local context and the findings of previous research (e.g. Söderlundh 2012, 2013). Interestingly, the study participants’ attitudes towards translingual practice vary: two students (Alex and Sam) perceive it to be helpful but associate it with speaking an inferior variety of ‘Swenglish’; Robin finds it helpful if all people involved understand both languages but otherwise sees it as problematic; Pat reports having experienced feelings of exclusion, whereas Kim feels an urge to learn the local language Swedish as soon as possible in order to join the English-Swedish translanguaging community.

Alex, Robin and Sam (L1 users of Swedish) reported that when doing their written assignments, they tried to stick to English only: ‘it’s really foreign to me to use Swedish really so like when you write stuff down you write in English’ (Alex). When discussing their group assignments in Swedish, they often used English terms related to business studies. Alex admitted that using a native language in discussions ‘is always easiest’ but it becomes ‘Swenglish’ because many disciplinary concepts were learnt in English: ‘you know lots more words in English than you know in Swedish so you tend to use Swenglish’ (Alex). Alex also reported some teachers asking for students’ support in Swedish to fill lexical gaps in English (cf. Söderlundh 2012, 2013) or to provide a Swedish equivalent to an English term, which she found to be pedagogically sound: ‘we’ve had like people who are really good teachers who said the word in Swedish as well.’

Since translanguaging between English and Swedish quickly became common practice in student interaction, it started affecting non-Swedish-speaking students who could not keep up with the discussion. Alex admitted that switching to Swedish was often unintentional and went unnoticed unless someone called attention to it:

(9)

… we tended to sometimes digress and just started speaking Swedish because it felt like faster and … you remember that ‘Oh, now I’m talking in Swedish again’ so you mix it with English again so I think that you tend to sometimes not to think about when you- when you use the different languages. (Alex)

Sam had a similar experience and admitted that the choice between English and Swedish was more random depending on who started the conversation and in what language. When it comes to group work, Sam deliberately chose to work with L1 speakers of Swedish to avoid any misunderstandings which could potentially lead to lower grades. In such groups, ‘we usually switched to Swedish or sometimes we even had full study sessions in Swedish and all we would do is read in English and then go back to Swedish’ (Sam).

Robin reported more varied experiences of group work and translingual practices in the EMI programme. He noticed translanguaging taking place when doing group work with Swedish classmates and admitted that it was helpful for speakers of Swedish: ‘If it’s Swedish and you’re Swedish then it’s good. But it- that it’s understandable for everybody. Sometimes it’s not.’ He described strategies used by others to ensure understanding when students unfamiliar with Swedish were part of the group: ‘Mostly we correct each other. It’s like- do you mean this? Yeah, yeah of course’.

Pat and Kim do not have Swedish as their L1 but they reacted to the above-mentioned translingual practices very differently. Pat admitted that, even though students who switch to a language other than English probably do it unintentionally, she did not feel included in these translingual practices and did not always ask for clarification in English. Excerpt 10 mentions other students speaking Russian but Pat then clarified that she was referring to Russian as an example of a language other than English, and that English-Swedish translanguaging was the most common. Pat found such practices uncomfortable and did not position herself as part of translanguaging groups:

(10)

if I am the only person who is- or how I shall put- if we are three or four people and three people know Russian or they in some way- and I am the only person who don’t know Russian then the conversation will probably end up speaking all these three persons Russian and I was like standing there probably- it depends … if I see that I have something in that conversation for me to learn or experience then I will probably ask them to speak in English but otherwise I will just [..] (Pat)

During the interview, Pat expressed an interest in learning Swedish but, despite several years of living in the country, had never mastered the language in the fear that this would affect her English in a negative way. It is therefore unlikely that Pat started participating in English-Swedish translingual practices during this EMI programme.

Kim adopted a different approach to working with her Swedish-speaking classmates by learning the language: ‘I always happen to be in groups where I’m the only foreigner so we end up speaking Swedish’. Since Kim has German as L1, adding Swedish to her linguistic repertoire proved to be possible by sheer immersion: ‘I was just like thrown into the cold water and I learn by doing so to say.’ When asked about her experience of such group work, Kim replied: ‘That works really well. It’s just like some person says … skriv så här [write this] and then they say it in English and we’re like okej ska jag göra? [okay shall I do that? ].’ By learning Swedish, Kim was able to gain access to groups which she perceived to have higher proficiency in English and therefore fewer problems related to their studies. By doing so, Kim could also position herself as a member of the group with a perceived higher status. From the comments below, it appears that speaking Swedish in this EMI programme allows for peer communication on a different level:

(11)

Kim: … sometimes like even during the lectures when you’re like commenting on- on something, I usually do that in Swedish. It’s not necessarily English. It’s like switching back and forth. It really depends, so [..]

M: So, during a lecture in English, you comment with your classmates in Swedish

Kim: It’s like- yeah, if the person next to me like says something it’s mostly in Swedish so we like, comment on it. But I mean of course if it’s a classmate that isn’t Swedish, it would be English but-

M: So Swedish is more like a student language like-

Kim: It’s like our secret hidden language. ((laughter)) I don’t know, yeah, it became like that in a way actually.

It may seem odd to have Swedish as a ‘secret hidden language’ in Sweden but, in the context of this EMI programme, it appears to exercise this function in some cases as it indexes belonging to a specific group.

Like Alex and Sam, Kim also reported mixing Swedish and English at random but, instead of referring to it as ‘Swenglish’, she talked about ‘pidgin’ and ‘creole of English and Swedish’. When asked to elaborate whether this is a problem, Kim replied that it made her English worse, confirming the fear expressed by Pat in Extract (6): ‘I used to be better at English, I feel, before I came here but- yeah, if you think about it, it’s bad because you study in English to improve your English and then it gets worse but I guess it just comes naturally with mixing it up’ (Kim).

4. Discussion and conclusion

What conceptualizations of English emerge from the students’ accounts? Standard English, English as a lingua franca, and English as part of translingual practices are present, although to different extents and in connection to different kinds of positionings. Alex and Sam seem to align themselves with standard language ideology and position their Englishes above their peers. At the same time, both are flexible enough to adjust their language uses, or ‘standards of English’ in Sam’s case, and to participate in English-Swedish translingual practices, even if they involve the use of what they call ‘Swenglish’. Alex, Sam, and Kim also function as language regulators for their peers in different assignments prepared in groups. Robin and Pat position their language uses on par with their peers and appear to be the most attuned to communicating in English as a lingua franca with students from different L1s. Pat in particular positions herself as an effective communicator with other L2 users of English, regardless of their proficiency level. At the same time, she also admits that communication works better if at least one of the speakers uses his or her native language, which is not a very positive view of ELF. As an L1 user of Swedish, Robin participates in English-Swedish translingual practices but at the same time expresses an awareness of how such practices are only helpful for those who know Swedish. Pat is the least attuned to translingual practices and often feels excluded when they take place. The conceptualizations of English in positionings expressed by Kim are interesting in that they align both with standard language ideology, although detached from any national variety of English (neither British nor American), and with English as a ‘tool’ or a ‘lingua franca’ (Kim). Kim is the most multilingual and most translingual of the five study participants, which means that she can accommodate her English to different contexts and also quickly acquire a new language in order to participate in English-Swedish translanguaging practices in the EMI programme. At the same time, Kim seems to lack communicative strategies for using English as a lingua franca with students whose language proficiency is lower than what she considers to be good and acceptable. Unfortunately, students who experienced serious language-related problems did not volunteer to be interviewed, and their voices have not been part of this study.

A glimpse into students’ perceptions of language uses in their EMI programme shows that these are not simply about language – the tensions in the conceptualizations of English and students’ positionings point towards other important issues related to power relations, group dynamics, social integration, and learning. For example, Kim is in a privileged position to join the ‘elite’ group of Swedish-speaking students and to be accepted by them. Pat seems to be less integrated, despite having the longest experience of EMI and of using English for professional purposes, particularly in lingua franca contexts. The findings of this study resonate with a call for further research into ‘how to accommodate translanguaging in a highly linguistically diverse EMI context where use of the L1 will actually exclude students who do not share that particular L1’ (Hultgren in Coleman et al. 2018, 716). Although research on translanguaging – and more recently on English as a lingua franca (e.g. Jenkins and Mauranen 2019) – has been promoting such practices among students and teachers, the question of student diversity in EMI has not been sufficiently addressed. The latest university language policy document for Nordic countries states that: ‘If an examiner permits the use of two or three languages because he/she understands them, students whose repertoires do not include these languages will be unfairly excluded’ (Gregersen et al. 2018, 40). This raises the question whether resorting to the local language in a linguistically diverse EMI setting is always pedagogically sound. Judging from the findings of this study, English-Swedish translingual practices are certainly helpful but they also tend to favour local students. The same local students are often in charge of regulating English language uses among their peers, and are therefore in a particularly advantageous position. This dynamic raises the question whether students taking an EMI programme in a particular country should be invited to learn the local language, at least to some extent, even if they might feel that this makes their English worse (e.g. Kim).

Native language ideologies are strong in the students’ conceptualizations of language, both with regard to English and to Swedish. Native varieties of English seem to be valued by the study participants but they do not enjoy the same status (e.g. South African and Australian stand out) and even native speakers of English such as Sam need to make adjustments – however minimal – in order to feel fully integrated. Judging from the students’ accounts, post-colonial varieties of English hold a lower status in this EMI programme compared to local English in Sweden (cf. Kuteeva 2014). Thus, native (or near-native) speakers of British and American English may take on expert roles in group work and lectures but their authority is not always clear-cut. However, we can see how different conceptualizations of English in students’ interactive positionings contribute to assigning different statuses to their peers in the given EMI context. In one of the interviews not included in this study, a content lecturer in the EMI programme even expressed irritation with a student who tried to avoid taking academic English seminars on the grounds of being a native speaker of English: ‘I was a bit annoyed when he said … that he was an expert in English … come on, who is an expert that you don’t ever need to train?’

This study has barely scratched the surface of the complex reality in students’ conceptualizations of English and how they affect their language uses in EMI contexts. However, combined with insights from previous research carried out in similar EMI settings (e.g. Bolton and Kuteeva 2012; Söderlundh 2012, 2013; Kuteeva 2014), the findings point towards a need to go beyond fixed approaches to conceptualizing English in EMI contexts and view them along the continuum of standard versus non-standard language uses. Both native and non-native Englishes can approximate to, and deviate from, standard language, and linguistic practices and associated norms tend to vary across formal and informal contexts: what works for learning (e.g. code-meshing in academic writing) can be penalized in examination or high-stakes contexts. Following Bakhtin (e.g. 1981, 270–273), I argue that language uses in EMI contexts are shaped by both centripetal forces and centrifugal forces, pulling simultaneously towards language unification (i.e. adherence to standard), and towards heteroglossic variation (i.e. non-standard language uses and translingual practices).

The findings of this study have a number of pedagogical implications. While it is true that adherence to a single standard variety of English is not required and even practically impossible in any EMI context, some kind of linguistic norms – however negotiable – are often appreciated by both L1 and L2 users of English, particularly when it comes to written assignments, exams or instructions. In such high-stakes contexts, conceiving of English as ‘variably variable’ (Jenkins and Mauranen 2019, 4) might be confusing for students or teachers involved in EMI programmes. Clarity and understandability, both in spoken and written English, are valued in EMI contexts, and these are often associated with some form of standard, as standardization ‘leads to greater efficiency in exchanges of any kind’ (Milroy 2001, 534). In this regard, equating the English used in lingua franca contexts with accepting a great variety of non-standard forms at all times seems unhelpful for EMI, due to different perceptions of what is acceptable in writing versus speaking, for example. In my view, the main contribution of ELF research to EMI is related to identifying the communicative strategies required for ensuring understanding among different stakeholders (cf. Canagarajah 2013 on Lingua Franca English), which is not to be taken for granted even by L1 or advanced L2 users of English. Students and teachers alike need to be attuned to others’ Englishes and take the responsibility for mutual understanding (e.g. asking for paraphrasing high-stakes information). According to Macaro (2018), the importance of comprehension checks, requests for clarification, and confirmation checks and signals had also been established in previous SLA research.

At the same time, translingual practices – especially those involving English and the local language of a given EMI context – have been and will continue to be common place among students and teachers due to their spontaneous nature and close connections to local language practices. Many language teachers who share L1(s) of the student population have been resorting to the use of these languages in their teaching, and the same applies to content teachers in EMI contexts (cf. Macaro’s 2018 discussion of bilingual or multilingual teachers). While it is true that the outcomes of translingual practices may ‘approximate “standard language” in some occasions and vernacular in others’ (Canagarajah 2013, 188), as translinguals can navigate different varieties, registers and codes, it is not only the outcomes but also the processes involved in producing those outcomes that matter. Educators involved in EMI should be aware that, while clearly helpful in some contexts, translingual practices are not desirable in others, as they can result in excluding all those who do not share the required linguistic resources. The analysis of the students’ conceptulizations of English in relation to their positionings shows that translingual practices are perfectly compatible with standard language ideologies, as the students involved in such practices (e.g. Alex, Sam, Kim) act as language regulators for their peers whose English they consider to be below the required standard. In other words, translingual practices in EMI contexts are not always associated with empowering the students by allowing them to resort to their L1 or another linguistic or semiotic resource to fill gaps in their English. Translanguaging can also function as a mechanism of exclusion and reinforcement of language standards by a group of ‘elite’ translinguals.

To conclude, I return to the argument that each EMI context constructs its own language regime; the idea of what is acceptable is not static and can move along the standard – non-standard continuum. This dynamic calls for further research and should be kept in mind by both teachers and students involved in EMI programmes as they try to navigate along the sliding ‘double standards’ of language use.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Maria Kuteeva is Professor in English linguistics in the Department of English, Stockholm University, Sweden. Her research and publications have focused on the role of English in multilingual university settings, academic discourse analysis, university language policy, and academic writing in English as an additional language. Her work has appeared in Applied Linguistics, Journal of Second Language Writing, English for Specific Purposes, Journal of Multilingual and Multicultural Development, and other peer-reviewed outlets.

Appendix: Transcription conventions

The transcription has been simplified and some punctuation has been added to aid comprehension.

- break off

(()) extra-linguistic content

[ ] context explanation

[..] faint

… omitted text from quote