WHAT STUDENTS HAVE TO SAY ABOUT EMI: EXPLORING UNIVERSITY STUDENTS’ PERSPECTIVES ON CHANGING THE LEARNING/TEACHING LANGUAGE TO ENGLISH

This study documents the perceptions and self-expressed experiences of eleven Catalan university students as they shift from learning in their first language(s), Catalan and/or Spanish, to learning in English. The paper presents a qualitative analysis of students’ reflections in relation to: how disciplinary knowledge is transmitted and learned in English-medium instruction (EMI); the multilingual dynamics employed in the EMI class by both lecturers and students; the English-language teaching and learning events; and perceptions of students ’ own and their classmates’ performance. It outlines the main foci identified through thematic data analysis (Airey, 2011; Saldaña, 2013) of students’ responses to a set of questions, considering their personal opinions regarding the implementation of EMI subjects as part of the internationalisation process of the university and offering t he students’ general views and concerns about content subjects taught in English. The findings suggest that students have a generally positive perspective on EMI subjects and they assign a pedagogical value to the L1(s). Yet, the paper also reports several students’ concerns in relation to EMI implementation. Overall, these findings may lead to a better and deeper understanding of EMI in the context under study. differences between through English and L1, transmission/learning of disciplinary knowledge; instances of English language teaching/learning; and 4) plurilingual strategies, the multilingual dynamics, adopted make for


INTRODUCTION
It is well-known that processes of internationalisation have permeated higher education (HE) institutions in recent years (Ennew & Greenaway, 2012;Law & Hoey, 2017). One of the key pillars of internationalisation is the introduction of Englishmedium instruction (EMI) in HE institutions. EMI is understood as the teaching of disciplinary academic subjects in English in contexts where English is not used for everyday activities (Macaro et al., 2018). Universities usually include EMI within their internationalisation policy as a way to attract international students and to facilitate communication when students have diverse linguistic backgrounds (Dearden & Macaro, 2016).
However, progressively, EMI has also been positioned as an opportunity for students to improve their linguistic skills (Arnó-Macià & Mancho-Barés, 2015). The application of EMI with this objective is based on the assumption that language learning takes place simply via exposure to content in English due to EMI's immersive nature (Dafouz, 2018), or as Turner (2011: 21) calls it "a pedagogy of osmosis" (see Mancho-Barés & Arnó-Macià, 2017). For this reason, it has been claimed that in some settings EMI becomes partially CLILised, meaning that EMI is not only adopted for content delivery, but also as a means for students to improve their English proficiency (Block & Moncada-Comas, 2022;Moncada-Comas, 2020;Moncada-Comas & Block, 2019).
The term captures how EMI, which in principle was about the use of English as a lingua franca, is translated into a CLIL-like practice. The term Integrating Content and Language in Higher Education (ICLHE) has actually been used in tertiary settings to refer to those courses that include both content and language learning outcomes. In contrast, EMI is often perceived as a space for disciplinary content learning only, where the development of academic and disciplinary language remains unlikely and serendipitous. By introducing the label CLILised EMI, the aim is to foreground the importance of language in the EMI context, suggesting that there exists "a continuum of approaches" (Airey, 2016) between ICLHE and EMI. Therefore, CLILised EMI combines two ideas: 1) EMI emerges in a lingua franca context, where the primary objective is the transmission of disciplinary knowledge, and 2) EMI may also include and integrate language learning objectives to promote the development of disciplinary literacies.
This shift in the language of education at tertiary level has turned researchers' attention to key stakeholders involved in EMI: lecturers, their linguistic and pedagogical competences/practices, and students, who experience the complexity of learning disciplinary content through English as a Foreign Language (EFL). Therefore, students' perspectives, experiences and opinions about EMI need to be considered in light of the additional layer of difficulty afforded by the language of instruction. This study attempts to obtain a more detailed picture of EMI students' concerns and perceptions at a Catalan HE institution. Although generalisations cannot be easily drawn, it is assumed that this paper will encourage EMI

WHAT STUDENTS HAVE TO SAY ABOUT EMI: EXPLORING UNIVERSITY STUDENTS' PERSPECTIVES ON CHANGING THE LEARNING/TEACHING LANGUAGE TO ENGLISH
Vol. 10(2)(2022):  policymakers to consider the student perspective so as to maintain the quality of EMI in a context where its implementation is relatively recent and carried out in a non-systematic way (Mancho-Barés & Arnó-Macià, 2017).

BACKGROUND: RESEARCH ON EMI STUDENTS' EXPERIENCE
This article documents and focuses on EMI students' experiences and general attitudes towards EMI within the Southern European (SE) context. In many southern parts of Europe, EMI has received little planning and has suffered from under-resourcing (Block & Khan, 2021), given lecturers and students' often inadequate levels of English. Compared to the rest of Europe, EMI arrived in Spain at a later stage and both lecturers' and students' language competence is often regarded as insufficient for EMI (Aguilar, 2017;Costa & Coleman, 2010;Dafouz, 2018).

The learning experience
Several studies of EMI students' perceptions in the SE context (Ackerley, 2017;Clark, 2017;Guarda, 2018) have explored EMI students' perceptions from different Italian HE institutions and have reported positive EMI experiences. Particularly, Ackerley (2017) conducted a survey study drawing from a range of EMI programmes/courses to explore how a total of 111 students felt about EMI. Despite drawing from a heterogeneous group (diverse English competences and varied EMI experiences), results pointed out that students preferred to enroll on EMI subjects because EMI is viewed as an opportunity to improve their English-language competence. In addition, they seemed to feel responsible for their own language development. Meanwhile, Clark (2017) focused on a single Master's degree, which had recently shifted from Italian-medium instruction to EMI. Drawing from survey and interview data, Clark looked at students' perceptions of their own language competence and that of their lecturers, as well as students' general evaluation of the course. Results showed students' satisfaction in regard to lecturers' level of English; they also self-reported improvement in their own level of English.
Likewise, students' satisfaction with EMI and their perceived language development has been found in Eastern contexts such as Qatar, China or Turkey (Ellili-Cherif & Alkhateeb, 2015;Kirkgöz, 2014;Muthanna & Miao, 2015;Yeh, 2011). Li and Ruan (2015) carried out a longitudinal study of how English for Academic Purposes (EAP) students' beliefs changed when they became EMI students in Mainland China, revealing that they valued EMI as a setting for English-language learning and reported vocabulary acquisition as a key aspect. Elsewhere, Galloway, Numajiri, and Rees (2020) employed multiple data collection instruments (questionnaires, interviews and focus groups) with national and international 265 BALBINA MONCADA-COMAS Vol. 10(2)(2022):  students to study how the EMI policy was conceptualised at the classroom level in Chinese and Japanese universities. The vast majority of students viewed EMI as a beneficial and positive experience, improving their general English proficiency and their specialised knowledge/vocabulary (see . In the Spanish context, Aguilar and Rodríguez (2012) reported similar findings based on their analysis of the opinions of lecturers and students. Their findings indicated that students perceived an improvement in the acquisition of specialised vocabulary as well as in speaking and listening skills. The majority of the students reported a positive experience, rating the acquisition of specialised technical vocabulary highly. Likewise, Aguilar and Arnó (2020) explored how an effective lecturer shifted to English and whether this shift from the L1 to EMI affected teaching quality. One of their particular goals was to analyse how EMI affected students in terms of comprehension challenges and overall course satisfaction. Their results, once again, indicated a positive, highly satisfactory, EMI experience. More recently, Arnó-Macià and Aguilar-Pérez (2021) unveiled the students' motivations for choosing either L1 instruction or EMI. Students who opted for EMI reported a positive experience and a "sense of achievement" (Arnó-Macià & Aguilar-Pérez, 2021: 36). In response to the challenges, students mentioned the need for scaffolding techniques to develop their disciplinary literacies.

Challenges
In fact, students' positive EMI experience is often accompanied by several challenges. In the Italian context, some studies (Ackerley, 2017;Clark, 2017;Guarda, 2018) noted that a minority of students reported that they would have learnt more in their L1, associating EMI with a reduction in content quality and the amount of subject matter. These findings suggest that some students perceive English as a barrier to disciplinary content learning, presuming a potential loss in content because of the language shift. In another study at a Japanese university, Aizawa et al. (2020) examined the impact that L2 proficiency level had on the perception of academic challenges. Drawing on questionnaire data, findings suggested that speaking and reading skills were the most challenging ones. In contrast, the study by Kamaşak, Sahan, and Rose (2021) at a Turkish university revealed that students found writing and speaking more difficult. Taking into account the challenges faced by students, Galloway and Ruegg (2020) examined to what extent support is provided to EMI students in Japan and China (e.g. support classes, self-access support and additional support from content faculty), concluding that students need support to overcome challenges such as language-related issues. The findings highlighted the importance of considering students' specific needs and their English proficiency when providing relevant support to study in English-medium programmes.

Learning in L1 and EMI
In relation to students' perception(s) of disciplinary knowledge learning, they report that successful comprehension and mastery/acquisition of concepts and new specific terminology demands higher levels of concentration in EMI, as shown by Guarda (2018) in the Italian context. Similar findings are found in the Spanish context where Dafouz et al. (2007) examined the reactions of lecturers and students towards CLIL in a university context by means of surveys. One of the conclusions from Dafouz et al. (2007) is that students are cautious about enrolling in Englishtaught courses, as they report that their L1 degree-courses are already challenging enough, hence EMI may prevent students from full content understanding. Elsewhere, Kirkgöz (2014) carried out a qualitative study of university students' perceptions of EMI and L1 instruction. Again, students reported learning more disciplinary content in L1 subjects than in EMI subjects, although they also mentioned that EMI improved their English language skills. In general, across studies and contexts, students seem to associate L1 instruction with a better and more detailed understanding/retention of disciplinary knowledge.

Multilingual practices
Focusing now on multilingual practices, such as the use (or not) of the L1, translation or translanguaging, students may actually rely on L1 translation of specialised technical vocabulary in order to follow the EMI class (Aguilar & Rodríguez, 2012). Thus, Maiz-Arévalo and Domínguez-Romero (2013) investigated the students' response to EMI and found that translation is one of the most common strategies employed by students to improve their learning in English. In addition, Aguilar and Arnó (2020) revealed that participants often resorted to the L1 equivalents of technical terminology and viewed instances of code-mixing, or spontaneous translanguaging (Cenoz & Gorter, 2017), as natural strategies when asking questions to the lecturer.
Overall, students engaged in EMI in HE institutions seem to report a positive attitude towards the implementation of EMI subjects; more importantly, it seems that the acquisition of specific and technical vocabulary of the discipline is a significant gain. Nevertheless, students seem to assign a higher difficulty to EMI in comparison to subjects taught in their first language.

THE STUDY: EXPLORING STUDENTS' PERCEPTIONS ON EMI
This paper presents a case study that investigates the experiences and concerns of eleven (out of twenty-six) EMI students enrolled in 'Swine Production', a STEM (Science, Technology, Engineering and Mathematics) content subject taught through 267 BALBINA MONCADA-COMAS Vol. 10(2)(2022):  English at the University of Lleida (UdL). In particular, the present study discusses four main foci that emerged from the data analysis: 1) students' perceptions of their own and their classmates' attitude(s)/performance(s); 2) students' perceived differences between learning through English and through L1, that is, the transmission/learning of disciplinary knowledge; 3) instances of English language teaching/learning; and 4) plurilingual strategies, the multilingual dynamics, adopted to make up for poor comprehension. Taking into account these foci of analysis, the aim of this study is to analyse how EMI students talk about their experience of changing the language of instruction from their L1 to English. From this, two specific questions arise: a) To what extent do EMI students report a positive learning experience? b) What are the concerns of EMI students when engaged in an EMI subject?

Setting
In Catalonia (Northern Spain), there was no top-down policy in the introduction of EMI courses. EMI implementation actually occurred with limited support (Mancho-Barés & Arnó-Macià, 2017), possibly due to the Englishisation requirements set by the Spanish Government (Dafouz, 2018). This study is contextualised in a multilingual HE institution where Catalan and Spanish (the local languages) are traditionally used for teaching and learning, and where English has been introduced recently through the implementation of EMI subjects (Llurda & Cots, 2020).
Catalan universities have been encouraged to promote (preferably) English or other third languages for management, education and research and, in turn, to improve students' knowledge of that language (CTLU, 2019). The teaching of disciplinary subjects in English was officially recognised at the UdL in 2004 as part of the university's internationalisation plan and since then EMI has increased dramatically at the UdL with 10% of bachelor's degrees being now taught in English (GDLP, 2018: 31).

Data collection
The data 1 collected belongs to the research project called ASSEMID, 2 particularly to the pilot-study period, which explored the impact of EMI on foreign language 1 The data were collected with informed consent and the collection followed all university ethics procedures and confidentiality and anonymity protocols. 2 The data discussed in this paper are from the project entitled Towards an empirical assessment of the impact of English medium instruction at university: Language learning, disciplinary knowledge and academic identities (ASSEMID). The project was funded by the Spanish Ministry of Economy, Industry and Competitiveness (El Ministerio de Economía, Industria y Competitividad MINECO), code FFI2016-76383-P (30 Dec 2016-29 Dec 2019).

WHAT STUDENTS HAVE TO SAY ABOUT EMI: EXPLORING UNIVERSITY STUDENTS' PERSPECTIVES ON CHANGING THE LEARNING/TEACHING LANGUAGE TO ENGLISH
Vol. 10(2)(2022):  teaching/learning, disciplinary knowledge and academic identities (Block & Khan, 2021). The particular data set (audio-logs) presented in this paper was chosen from the following sources: 1. A semi-structured pre-interview (53 minutes) with the lecturer. The interview was audio-recorded and conducted in the informant's language of choice (Spanish). It aimed at eliciting some general information about the lecturer's language biography. 2. Video-and audio-recorded observations in the classroom of this lecturer and his group of 26 students (20 Catalan/Spanish students and 6 international students). The pilot stage consisted of the recording of a total of three classes: a) a lecture following the flipped-learning methodology (lessons where students look at the theory/content at home and then practise it in class); b) a problembased seminar class; and c) students' oral presentations. These were useful for an in-depth understanding of what was going on in the EMI class and to analyse the lecturer and the students' performance in general. 3. Twenty audio-logs (diaries), which were gathered for the entire classroom observation period. Three logs were sent by the lecturer and 17 by students (10 from the first observation, 7 from the second observation, 0 from the third observation), all upon the completion of each class. These logs asked participants, in common-usage terms (logs were given in Catalan, Spanish and English and participants could choose any language to respond to them), to answer several questions (see Appendix 1), which were adapted from Block (1996). Students were given the log's questions when the research was presented to the class. These logs were voluntary and participants could record themselves using the Vocaroo tool. The recordings were sent to researchers for transcription. These logs were regarded as self-reflective practices (Farrell, 2020) as they aimed at collecting further information on the students' EMI experience and perspectives. However, participation in log-completion was much lower than expected. For the purpose of this particular study, the 17 students' logs were considered (and translated into English), providing explanatory meta-reflections on different issues.
The data set presented in this paper draws primarily on the last data set, i.e. the students' logs. These logs asked participants to reflect on and to discursively frame their EMI experience(s). As mentioned above, each student voluntarily answered this list of questions which encouraged them to think about both EMI in general and their own participation in the course as EMI students. 2 for transcription conventions) in the original language (Catalan/Spanish). The sections of the logs which are cited in the text (see Section 4) were translated into English by the author. First, the input of each individual log was examined and then a second analysis allowed us to come up with particular foci of discussion to organise the data. The logs were therefore subject to thematic analysis (Airey, 2011;Guest, 2012;Saldaña, 2013) in order to identify recurring comments. Although only the data from the students' logs (which have been anonymised through pseudonyms) will be analysed here, there were complementary instruments used to contribute to a bigger picture of this study: semi-structured pre-interview with the lecturer, video/audio-recorded observation of three classrooms and lecturer's logs. Therefore, the ethnographic nature of this research provided a comprehensive set of data, given that "ethnography privileges the direct observation of human behaviour within particular 'cultures' and settings and seeks to understand a social reality from the perspectives of those involved in the observed interactions" (Starfield, 2010: 50).

Method of analysis
A qualitative approach to data analysis entails "iterative cycles" in order to find key patterns, "[e]ach cycle results in loosely labelled categories that may then be split up, renamed or amalgamated in the next iteration" (Airey, 2011: 41). Therefore, the data analysis proceeded as follows: collection of logs, transcription of logs, and analysis of the accounts of each individual's experiences. Once all the logs were collected and transcribed, the input of each individual was examined and then a second analysis revealed particular foci of discussion, derived from the logs' questions, in order to organise the data. Each log was considered as a whole and then a comparative analysis was carried out throughout the 17 logs to identify similar comments. Initially, this involved the searching for words/phrases that (re)appeared throughout the data and finding connections between the different logs on the basis of the meanings conveyed. Thanks to the log collection process, participants were able to describe through reflection how their attitudes to learning had developed due to the change in the language of instruction.

FINDINGS
The process of data analysis resulted in four foci of analysis: 1. students' perceptions of their own and their classmates' attitude/performance, which is related to the English-level used in class and the students' own selfperceived linguistic competence; 2. the transmission/learning of disciplinary knowledge, which describes the differences that students perceive between learning in English and learning through their L1; 3. English-language teaching/learning, which targets students' perceptions about how EMI enhances their language skills in relation to both personal skills (confidence) and technical specialised vocabulary; and 4. multilingual dynamics, which addresses the plurilingual strategies used, such as spontaneous translanguaging and translation employed by students to ensure comprehension of disciplinary content.

Students' perceptions of their own and their classmates' attitude/performance
Students comment on their perceptions of the English level used in class and they often state that the linguistic standards of the class are "basic" or "understandable".
As one student puts it: (1) a basic language / … from my point of view quite understandable / without being / I think it could be understood perfectly (Mario, 2 Mar 2017).
In the same line, another student reports that content transmission in EMI is more schematic and dynamic: (2) it's an adapted class since he gives us much more schematized information for our comprehension / he provides us with knowledge based on videos and articles / not the way it would be in another class using PowerPoint and in a totally theoretical way (Carla, 7 Mar 2017) However, not all students agree with these accounts. Some of them suggest that the English level of the class requires much deeper concentration and some students report that it is more difficult to follow the lecturer's explanation without losing focus: (3) it's usually difficult to follow non-stop the English level of such a specific subject like it's swine production (Jorge, 2 Mar 2017) (4) doing this subject in English is much more difficult because the concentration you need to have in class is far superi-much more / you have to be much more focused (Ana, 2 It is interesting to note that the nature of the class also has an impact on the students' perceptions. Particularly, one student mentions that the first class was easy to follow: Vol. 10(2)(2022):  In contrast, he reports that the second class was difficult because he could not understand what was being expressed: (6) the teacher has increased the level of English and it was difficult to understand it (Adrian, 7 Mar 2017) This could be explained by the different nature of the lessons. While the first lesson was a flipped lecture where students looked at the theory and content at home and then asked questions and practised in class, the second was a problemsolving seminar. It is indeed striking that in only five days Adrian changes so much his opinion about the level of the class.
Further to this, the majority of students position themselves as efficient users of English and at the same time position their classmates as the ones who do not have a high enough level. They often report that they themselves do not have problems following the lesson but that other students might. For example: (7)  It is interesting to highlight that students seem to refer to 'other' students as the ones who have the insufficient level of English to express themselves or understand the class efficiently. By doing so, they seem to claim having an appropriate or good enough competence in English themselves. Taking into account these comments about fellow students' low level of English, it is perhaps not surprising that students also report being worried about not learning all the content that they would normally expect to learn in a course taught in Catalan/Spanish, which is the topic of the following focus.

The transmission/learning of disciplinary knowledge
The students in this study expressed the view that, when the lecture is in English, the content is addressed in less depth or detail so that they do not learn as much as if they were taught in their L1. (10) I don't think it's right to do the class in English / because we are neither learning English nor animal production / neither one thing nor the other / if we did the class Ivan considers his learning of disciplinary knowledge on swine production as flawed because of EMI. This student also adds: (11) we chose this subject because we are interested in the topic / we are interested in swine production / and by doing the subject in English the only thing that we have managed to do is to make the subject boring and a lot of things remain unclear … we'll finish this course with four general ideas about swine production / by doing the class in English we never go into any depth and we don't move forward (Ivan, 2 Mar 2017) There is an evident concern in Ivan's responses: EMI may lead to missing important disciplinary knowledge because the subject matter is taught in a foreign language. This particular student's comment shows that he feels he would learn more if the course was taught in his L1, hence he views English as a barrier to content development/learning. He is not the only student that reports this same experience: (12) we don't go in depth / and without going in depth / we are going to finish this course with very little knowledge / we won't have the basics / we'll know four words in English / but you can learn that on the internet / I don't think this class is being taught well and I'm pretty upset about that (Ana, 2 Mar 2017) (13) this subject in Catalan or Spanish would be three times harder than in English … we are taught basic stuff because it's a new language / it's a language less used by us / and they have tried to adapt a third year course so it can be done in English … there are technical words but not as many as there would be in Catalan or Spanish (Aitor, 7 Mar 2017) Therefore, these comments show that some students have a negative opinion about the EMI subject: they believe that they learn neither English nor content.

English-language teaching/learning
While some students view EMI as content loss, other students believe that it contributes to their language learning process. These students take 'a maintenance stance' in the sense that they can practise their English:

Multilingual dynamics
Turning now to plurilingual strategies, students expressed that resorting to L1s occurs when there are doubts or in cases in which concepts need to be clarified: Another reason for employing this plurilingual practice is because students themselves sometimes do not know how to express themselves in English: (21) all the class was in English / except for the questions asked by the students who don't know how to ask in English / or they ask the question in Catalan / or in Spanish / and the teacher tried to answer or explain again in English / and if the content was still not clear / then the teacher proceeded to explain briefly in Catalan or Spanish (David, 2 Mar 2017) (22) some classmates didn't understand a word or they asked in Spanish and then the teacher answered them in English or in Spanish (Aitor, 7 Mar 2017) Therefore, the use of the L1 is a strategy employed when students do not know how to ask clarification questions. Students' opinions about the presence of Catalan/Spanish show that it is a strategy that contributes to the negotiation of meaning within the ongoing flow of the class. Another multilingual strategy that students commonly experienced in their EMI class is the use of translation. In this case, translation is connected to vocabulary learning since, apart from doing their Students seem to have a positive attitude towards this translation strategy because it helps them to understand the terminology and it serves to clarify doubts about new vocabulary.

DISCUSSION
Several interesting discussion points emerge from the findings of this study. The discussion will make connections to the four foci of analysis and the research questions posed previously: a) To what extent do EMI students report a positive learning experience? (RQ1) b) What are the concerns of EMI students when engaged in an EMI subject?
In relation to the first focus of analysis (Students' perceptions of their own and their classmates' attitude/performance), there is a clear reference to content simplification and explicitness in EMI, suggesting that EMI content is taught and described in less depth. Indeed, EMI lectures seem to be more dynamic because of the teaching methodology employed: classes are less theoretical compared to those in L1 and more videos and/or readings are used to approach the content from different perspectives. Although some students report that the language used in class seems to be simple and pitched to a low level so that all students can understand it, this opinion is far from consistent. Some students have difficulties following the lecture because of the language of the discipline itself (highly specific and academic), which requires greater concentration so as not to lose track of content due to language. For this reason, one of the concerns (RQ2) of EMI students is their own English proficiency, as their inadequate competence may cause language-related problems, influencing the acquisition of content through English and the students' performance in the EMI course. Surprisingly, even students who reported following the lecture without problems (2 March 2017) also reported a negative perception of the second class (7   275 BALBINA MONCADA-COMAS Vol. 10(2)(2022):  March 2017), probably because of its problem-solving task nature. Indeed, the second class may have posed more of a challenge as students had to apply their theoretical knowledge to solve problems. Therefore, students may be indirectly referring to the linguistic challenges that EMI students face in the classroom setting. This may be due to an inadequate linguistic competence, which jeopardises their understanding of disciplinary knowledge. However, it could be also explained by a lack of (or insufficient) scaffolding by the lecturer, who may fail to accommodate and adapt to students' diverse proficiency levels.
In terms of students' performance, students reported that they had problems with regard to speaking so that they were not able to properly express themselves due to limited linguistic proficiency (Doiz et al., 2019). Students perceived that the student cohort as a whole did not have a sufficiently high level to ask questions in English. Furthermore, the findings throw light on a clear criticism of the introduction of EMI, which is the lack of in-depth disciplinary content learning because students are taught in English. In fact, this shows another concern (RQ2) about the transmission/learning of disciplinary knowledge: students perceive that the language shift to English is at the expense of content. These comments are in line with a minor group of Guarda's (2018) respondents who also expressed a similar concern in relation to loss in subject content because EMI does not guarantee full and profound access to disciplinary knowledge (see Ackerley, 2017 andClark, 2017). Nevertheless, there are clear contradictions in students' responses as they view the subject as a space for solely content learning, but then again they also criticise EMI because they do not really develop their linguistic competence. These responses seem to make reference to both content and language learning, hence they are actually CLILising EMI.
Students argue that if content was taught in Catalan/Spanish then everyone would learn more content. Students' opinions resonate with the findings in Huang's (2018) study, where students found the content easy and non-challenging in EMI, suggesting that L1-mediated courses are more cognitively demanding than EMI ones. It is interesting to note that it is not only students who have this feeling, but also lecturers share this perception. In fact, some of the lecturers from ASSEMID reported that language shift is at the expense of content consolidation as lecturers only focus on the essential knowledge obviating extra material (Moncada-Comas, 2020). Elsewhere, Airey (2011: 44) found that EMI lecturers often note that when teaching in English there is less detail, referring to "the level of disciplinary detail in the lecture".
Although the perception that there is less or simplified subject matter in EMI may be just a self-perception, other studies address the same issue with similar findings. Vinke, Snippe, and Jochems (1998: 383) found that teaching content subject in English "reduced the redundancy of lecturers' subject matter presentation". Likewise, Wilkinson (2013: 14) stated that in EMI "the narrowness of depth that one might expect in an L1 programme has been replaced by a shallower breadth". While lecturers can hold this view, as they have been present at their own teaching across a range of courses taught in both their L1 and English, the students' criteria for this comparison are not so clear. Therefore, students should be reassured that they are not missing crucial disciplinary knowledge because of the change in language of instruction, as some studies have shown (Thogersen & Airey, 2011).
Apart from expressing their concerns, students also reported on several positive learning experiences (RQ1), as they commented on the gains of enrolling in an EMI subject and referred to English-language teaching/learning episodes. For example, some students view EMI as an opportunity to be in contact with the language and so they have the chance to practise it and be immersed in it, but they do not make any particular reference to actually "learning" the language. EMI can be considered as something positive since it helps students with their confidence level given that they have the opportunity to practise their English by asking questions. In addition, they are also exposed to input in English, input which is relevant and authentic, since it is disciplinary knowledge of their interest.
Furthermore, students mention that in the EMI class they learn highly specific and specialised terminology. To do so, they often use glossaries as the lecturer encourages them to write down key words as a scaffolding technique for vocabulary learning. In fact, this attention to vocabulary is shared with the lecturer, who also emphasises the importance of vocabulary learning (Moncada-Comas & Block, 2019). This was also found in Aguilar and Rodríguez (2012), Ackerley (2017) and , where students valued EMI for the acquisition of disciplinary vocabulary, reporting that one of the perceived advantages of EMI was the chance to expand their subject-specific terminology in English. Therefore, the development of vocabulary seems to be regarded as a positive outcome in EMI, students view it as an opportunity to develop their lexicon, in particular, highly specific technical terminology.
As a result, the majority of students report positively on EMI, experiencing a boost in confidence in terms of language learning and improvement of their specialised technical vocabulary. In particular, students view vocabulary learning as crucial in EMI subjects because it is often associated to content learning. Although content is still the goal, some students view EMI as an opportunity to be in contact with the English language and so maintain their level. Additionally, those students with the most positive attitudes see in EMI subjects the chance to learn English as well, embracing the added benefit as they also expect some language learning.
However, there is a clear clash of opinions towards the implementation of EMI subjects. These contrasting attitudes may be related to differences in motivation among students enrolling in EMI subjects. According to Muthanna and Miao (2015: 61), those students who consider that EMI renders less detail to the disciplinary subject have extrinsic motivation because English for them has an "instrumental end" to their final objective, which is a good mark. When students exhibit extrinsic motivation, they may be less engaged in the learning process and, as a result, "they will feel less self-efficacy toward English (and maybe towards their courses), and Vol. 10(2)(2022):  will benefit much less from EMI courses" (Lee & Lee, 2018: 724). In contrast, those students who report that EMI resulted in more learning and confidence seem to exhibit an intrinsic motivation towards English as they have a positive attitude towards the language and the implementation of EMI, hence engaging in the learning of the language itself.
Finally, the EMI class also put an emphasis on multilingual dynamics. All students commented on the different plurilingual practices employed to make up for comprehension problems. Spontaneous tranlanguaging and translating were considered efficient and effective compensatory strategies, which are seen as normal exceptional non-uses of English (Sabaté-Dalmau, 2020) resulting in an EMI pedagogy that takes into account the language ecologies of the students. In fact, spontaneous translanguaging is considered by Cenoz and Gorter (2017: 904) as "the reality of bi/multilingual usage in naturally occurring contexts where boundaries between languages are fluid and constantly shifting".
Students regard the lecturer's use of spontaneous translanguaging as a technique accommodating to their English levels (see Zhang, 2018) so that everyone can follow the class. In addition, they do not criticise the lecturers' use of the L1, probably as a result of their possibly insufficient English proficiency. Students consider that the lecturer shifts to the L1 when communication in English is at risk. Therefore, this spontaneous translanguaging is seen as a way to reinforce and ensure content transmission.
Students position the use of L1 as "an additional meaning-making resource" (Jones & Martin-Jones, 2010: 61), because concepts are reiterated and clarified. This communicative strategy serves as a means of keeping the lesson content accessible to the students, something that students actually appreciate. As in Yeung and Lu (2018: 32), the L1 is used to "explain more difficult concepts and terms related to the subject being taught", hence L1 use is seen as an acceptable and common strategy. Similarly, in their research of students' experience in seminars, Evans and Morrison (2011: 155) concluded that as both students and lecturer are more concerned with the understanding and application of disciplinary knowledge, there was a considerable use of the L1 in an attempt to achieve the learning objectives that could have been inhibited due to the English language.
Likewise, the use of translation has a similar pedagogical function. While spontaneous translanguaging was linked with solving problems and re-explaining content knowledge that was not clearly understood, the pedagogic motivation for employing translation is to provide students with the L1-equivalents of particular English concepts. Vocabulary learning in English is seen as crucial and the strategy students employ to learn this highly specific vocabulary is translation. As Doiz and Lasagabaster (2018: 674) pointed out, students welcome and acknowledge when teachers are aware of students' gaps in English and actually try to deal with language issues. Therefore, according to students in the current research, the lecturer was willing to help students with their learning difficulties by using L1-mediated strategies: translation for vocabulary learning and spontaneous translanguaging for longer explanations. Indeed, one could argue that translation can be considered a language-teaching strategy, and as such, its implementation may embody the CLILisation of EMI.
The use of L1s is seen as a pedagogical tool or a scaffolding device and students resort to Spanish and Catalan on several occasions, so translation and spontaneous translanguaging secure the students' attention, maintain communication and make comprehension of the content successful. None of the students complained about the presence of L1s more generally. For this reason, the use of L1(s) in an EMI class is seen as a positive learning tool (RQ1) rather than as a concern or something to be avoided. Although the implementation of these plurilingual practices does not seem to be well-planned, L1-mediated strategies do seem to have a pedagogical value (see Aguilar & Rodríguez, 2012;Rose & Galloway, 2019) at least to a certain extent, as students acknowledge both strategies as profitable and accept them within the emergent culture of EMI classes.

CONCLUSIONS
Drawing on logs from EMI students, this paper has offered a picture of students' perceptions of their EMI experience at a Catalan university. Although some previous research has focused on whether or not EMI lecturers, on occasion, identify themselves as language teachers (Airey 2011;Block & Moncada-Comas, 2022;Kling 2015;Moncada-Comas, 2020), there is little research on whether students identify themselves as both content and language students when shifting to EMI. Their responses highlight that students, in line with lecturers, position themselves as content-students aiming to develop content knowledge. Nevertheless, students can also be said "to develop their academic biliteracy" (Curle et al., 2020: 34) by expanding their academic and subject-specific vocabulary or practising their receptive skills (listening and reading), which may be considered instances of language learning. When students refer to language learning instances, they seem to recognise a language asset in EMI. In turn, this may suggest that students indirectly CLILise EMI (see Moncada-Comas, 2020). Some students view EMI as a means for improving their English proficiency, thus considering EMI as a place for both content and language learning and as a context where acquiring disciplinary knowledge is primary but is accompanied by language learning in the development of their disciplinary literacies.
This study contributes to the gap in research by focusing on the voices of EMI students and what they have to say about an EMI subject, EMI practices, EMI lecturers and how all these factors affect their learning process. The study has allowed students to reflect on their learning experience, giving them the opportunity to express themselves openly about how they deal with learning disciplinary knowledge in English. In view of this, some pedagogical implications can be outlined, which are in line with recent findings (Galloway & Ruegg, 2020) Vol. 10(2)(2022):  concerning the support that students should receive in EMI. Depending on the nature of the class (lecture or problem-solving seminar), students may encounter distinct challenges that may lead to more difficult content acquisition processes, due to language-related issues or insufficient scaffolding. This study therefore highlights the need to not only take into account the specific needs of students but also the nature of the class in order to provide them with the relevant support in accordance with what they are expected to do (i.e. listen and take notes in a lecture or apply the theory and solve problems in a seminar). EMI research can benefit from this study by: (1) further considering EMI students' self-reported understanding, opinions and experiences of EMI courses; (2) designing methods of qualitative data-collection that gather information from the students' side to understand how they interpret their experience and the meaning attributed to it; and, finally, by (3) using students' EMI interpretation for EMI lecturers', policy-makers' and researchers' self-reflexive development concerning the potential of students' perspectives.
There are however a number of limitations to the study presented in this paper. Firstly, it focuses on a specific EMI context and so other contexts may report different results. Secondly, only the local students who share the L1 with the lecturer provided logs with their attitudes about the language of instruction. Therefore, we do not know what the international students, who did not share this linguistic profile with the other students and the lecturer, made of such plurilingual strategies such as translation and spontaneous translanguaging in Catalan/Spanish or what they had to say about the language level of the classroom. Finally, this paper is based on qualitative data dealing with students' expressed perceptions of changing their learning language from their L1 to English, and these attitudes can only be subjective.
Indeed, this study points to a future area of exploration: students' self-inhabited positionings as EMI students. That is to say, how EMI students undergo the process of identity construction as they are influenced by the shift in language of instruction from their L1 to English and the extent to which they inhabit both a content-student and a language-student identity. Indeed, students may implicitly view EMI subjects as a space for language learning, hence CLILising the subject and their own participation in it. This paper provides evidence of students' views of EMI implementation, which, in order to be successful, needs to take students into account as key and active participants of EMI. All in all, students' perceptions and opinions are illustrative of the complex and secret realities of the EMI classroom environment.