English in higher education in the Kyrgyz Republic, Tajikistan, and Uzbekistan

Correspondence AnastasiyaBezborodova,University ofWestminster, School ofHumanities,Collegeof LiberalArts andSciences, 309Regent Street, London, EnglandW1B2HW. Email: a.bezborodova@my.westminster.ac.uk Abstract In Central Asia, English is not only considered a language of intercultural communication, but it is also by far the most studied foreign language at schools and universities. The globalization and internationalization of education resulted in English becoming a medium of instruction in all levels of schooling. English functions as the language of research and academia. This article focuses on comparative research conducted on the role(s) of English in higher education institutions (HEIs) in three Central Asian republics – the Kyrgyz Republic, Tajikistan, and Uzbekistan. The main objective of this study was to explore the role(s) of English at HEIs to have a better evidence-based understanding of the benefits and drawbacks of English medium instruction (EMI) in these countries.

In recent years, English instruction has increased specifically in higher education because English medium instruction (EMI) is associated with prestige and high status (Altbach & Knight, 2007). Many students choose to study in English because they believe that it guarantees a brighter future and better employment (Shohamy, 2013). However, the way English is introduced and implemented in countries where it is a foreign language poses various challenges.
Lack of English support for teachers and students at EMI universities, shortages of materials, and poor language proficiency are among these challenges (Alhabahba et al., 2016;Ng, 2016). Besides these constraints, there are other struggles for students and teachers. In settings where English is a foreign language, English requires more time from teachers and students both in terms of understanding materials and in terms of preparation for the class (Macaro et al., 2018). Additionally, Airey (2011) and Probyn (2006) reported that studying in a foreign language slows down content learning. Relatedly, students studying in EMI contexts participate less in classroom discussions than those who study in their native language (Kiliçkaya, 2006). Lower language proficiency, as a result, lowers students' general performance in class and affects classroom dynamics. It also makes students struggle with their studies (Choi, 2013;Tan & Ong, 2011).
Attitudes towards EMI and its practical use in higher education have been widely investigated in various contexts.
This article draws from a broader study that explored the experience of English in HEIs across three post-Soviet Central Asian countries -the Kyrgyz Republic, Tajikistan, and Uzbekistan. Examining students' and lecturers' opinions, this article conceptualizes the role of English following the relevant dimension in the ROAD-MAPPING theoretical framework proposed by Dafouz and Smit (2016). To contextualize the experience of English in multilingual HEIs in three Central Asian countries, the article provides background for each country and compares the findings drawing on several benefits and drawbacks associated with EMI in the region. The research findings reported in this article directly influence English language education policy in the countries of the Expanding Circle. The research contributes to a broader scholarship around world Englishes by drawing attention to the lived reality of English in three Central Asian countries and initiating a debate about the EMI practices in HEIs.  Table 1 provides a brief overview of each country's general statistics (World Bank, 2020a, 2020b, 2020c.

BACKGROUND
Language-related reforms were happening in the region before and after the countries' independence. The Russian Empire and then the Soviet Union's expansion into Central Asia in the 18 th and 19 th centuries affected not only the socio-political situation but also the linguistic landscape in the region and largely shaped its current state. Since 1991, because Russia and the Russian language experienced a backlash, many Russian speakers left the countries in the early years of independence. Following the de-Russification movement, in 1993 Uzbekistan shifted from a Cyrillic alphabet to Latin, while the Kyrgyz Republic and Tajikistan still use the Cyrillic script. Despite this backlash, however, labor migration to Russia, the availability of an array of printed, online, and media resources in the Russian language, as well The Central Asian region has always had a rich language ecology. At present, apart from the use of titular languages and various dialects, many people still use Russian as a language of communication while beginning to implement English education. English is the most learned foreign language, and it is used as a medium of instruction at various institutions on all levels of education but especially in higher education. Not only are branches of foreign institutes opening their campuses in the region, but also local universities are implementing EMI to attract more international students. International students' motivation to study EMI in Central Asian countries is the relatively low school fees and the cost of living. To contextualize the research findings on the role of English in three countries, it is necessary to provide brief background information about each country's major educational developments. In the following section, the country descriptions are mostly related to English language development and particularly EMI HEIs.

Kyrgyz Republic
In the past, the education system in the Kyrgyz Republic was often criticized for lack of quality and clarity. Merrill AUCA accepts students from around the world and is a popular destination for students from Afghanistan.
Finally, the British Council and the US Embassy are two active organizations in the country that arrange teacher trainings, professional conferences, and projects where teachers of English and other subjects may participate. They provide libraries that both teachers and students may use free of charge to improve their English.

Tajikistan
Tajikistan developed its education system mostly through the reforms in consultation with international donors.

Uzbekistan
According to the World Bank (2018b), in 2017, the government of Uzbekistan spent more than 6 per cent of its GDP on education, with more than half of this money (56%) spent on the secondary education sector. Also, in 2018, IELTS and/or TOEFL test scores were accepted as an alternative for the English language section of the national standardized test that students take to enter the HE institutions.
Uzbek English language teachers are often involved in in-service training courses provided by the Uzbekistan Teachers of English Association, the British Council, the Regional English Language Office of the US Embassy, and various foreign universities. Additionally, to support students' English language skills development, the US Embassy opened several 'American Spaces' in main libraries in the capital and big cities where students may study English and use up-to-date materials. In terms of higher education, back in AY1988-1989, there were 43 HEIs in Uzbekistan (Ruziev & Burkhanov, 2018

METHODOLOGY
Since the goal of the broader study was to explore the experience of English in HEIs in the Kyrgyz Republic, Tajikistan, and Uzbekistan, the following research questions were explored: 1. What is the lived reality of English for staff and students at HE institutions in the Kyrgyz Republic, Tajikistan, and Uzbekistan?
2. What differences and similarities can be observed in the roles of English at universities in three countries?
This paper draws from a broader study and focuses on the roles of English in HEIs with EMI. Exploring the experience of staff and students with English, this study aims at getting a better understanding of the lived reality of English in academia, particularly in teaching and learning.
The data was collected at EMI international institutions as well as state institutions with separate programs taught in English. Study participants were from both universities -larger educational institutions that offer undergraduate and graduate courses in various disciplines -and institutes -educational institutions that focus on a specific area of study. There were two data collection instruments -a questionnaire and focus group discussions. Both were organized around three central themes -the context of English language use, the experiences of students, teachers, and administrators with English, and views and attitudes towards the use of English at HEIs, both within and beyond the classroom.
The questionnaire was designed in English. English was also the language of communication during the interviews. The data collection instruments were adapted from a study on language policy development at Westminster International University in Tashkent (Linn et al., 2020). The questionnaire also included several sections of the survey by Bolton and Kuteeva (2012) that was initially conducted at Stockholm University.
The questionnaires were digital, and the links were sent out to HEIs in all three countries through the administration, faculty members, and students. To get more responses, we have also prepared paper-based copies and distributed them during field visits. Even though there is Internet connectivity in all three countries, most HEIs did not have university-wide Wi-Fi and filling out the paper questionnaires was more convenient for the participants. This generated more responses and made data collection inclusive. Overall, 793 questionnaire responses were recorded, but after eliminating all the outliers, the final number of the analyzed responses was 782. In total, 583 students and 199 faculty members from 58 HEIs completed the questionnaire. The data was then analyzed using SPSS software.
The fieldwork visits took place in Bishkek (the Kyrgyz Republic), Dushanbe and Khujand region (Tajikistan), as well as in Tashkent and Syrdarya region (Uzbekistan). The main purpose of these visits was to carry out interviews and focus group discussions at the HEIs. The visits were held not only in the economically developed capitals with a number of universities but also in the other regions. This was done to be inclusive of different participant bodies. In total, 85 faculty members and 330 students from all three countries took part in the interviews and focus group discussions (Table 2).
During the fieldwork visits, focus group interviews included eight to 12 students or faculty members at a time.
Students were interviewed separately from their lecturers to hear unbiased opinions. The interviews with the administrative staff were held individually for the same reason. This allowed us to minimize power-relationship bias and made interviews as comfortable for everyone as possible. The interview questions and all collected responses were in English. Group interviews with lecturers and students were transcribed and later analyzed using NVivo software. The universities visited during the field trips are indicated in bold font in Table 3. Overall the questionnaire response rate and the interest towards this study was rather high in the Kyrgyz Republic and Uzbekistan, which might have been due to relatively easy access to HEIs, more developed Internet connection, and support from fellow teachers in the data collection process. However, access to the Tajik respondents was challenging largely due to Internet-use restrictions in the country.

Theoretical framework
To provide a theoretical foundation for this study, the article utilizes the ROAD-MAPPING framework proposed by Dafouz and Smit (2016). Originating from sociolinguistic, ecolinguistic, and language policy research, ROAD-MAPPING framework includes wider research agendas and particularly focuses on tertiary level education EMEMUS, the concept of EMI is substituted with a broader notion of EME that helps to explain the complex nature of EMI-based education (Dafouz & Smit, 2016). The main point of access to all dimensions is the discourse of social written and spoken origin that include policy documents, English language practices, discussions about English language use, and so on.
Even though English in EMI HEIs serves in various roles ranging from being a gatekeeper in students' and lecturers' intake to disseminator of scientific ideas, in all three countries, English neither has a recognized status nor does it function as a language of wider communication. Therefore, from a macro-level perspective, in the context of this study, English is considered a foreign language (EFL). It is also important to note that this article does not cover all dimensions of the framework. Since the aim of this article was to examine the lived reality of English and roles and functions it carries in EMI HEIs, only the first dimension of the framework, namely 'Roles of English,' is used to synthesize the findings. Reflecting the ecological perspective of Spolsky's framework (2004), this dimension allows studying the functional breadth of English on institutional and national levels. Therefore, considering the factors relevant for the dimension of Roles of English, Smit (2017, 2020) cover both macro-and micro-levels of language use and subdivide the dimension in the societal, institutional, pedagogical, and communicational sub-dimensions. The article follows these four functions in the analysis of the results and discussion of the findings. In the first sub-dimension (that is, societal), English functions outside the educational institution and as a 'foreign language, language of environment, and language in (inter)national relations' (Dafouz & Smit, 2017, p. 299). Explaining institutional sub-dimension, Dafouz and Smit listed teaching, research, and administration as the main areas of English language use. In this subdimension, English is viewed as a subject, medium of instruction, and means of internal and external communication.
The authors regarded the English language as an 'explicit or implicit learning aim or assessment criterion' in the pedagogical sub-dimension (p. 299). They also included online and printed teaching resources in this sub-category. Since the ROAD-MAPPING framework considers the use of languages other than English, the communicational sub-dimension covers the aspects of multilingualism and bilingualism. It also focuses on communicational purposes, language skills, and proficiency. Because in these four sub-dimensions the roles of English are identified through the collected data on instruction, delivery, teaching, and learning of academic subjects in different disciplines, the article uses the concept of EMI and not EME with its broader focus on English medium education. In this article, EMI is defined as 'the use of the English language to teach academic subjects (other than English itself) in countries or jurisdictions where the first language of the majority of the population is not English' (Macaro, 2019, p. 19). Dafouz and Smit (2020) defined the societal function of English as the use of languages in a broader environment outside the institution. The interview data showed that, on a societal level, knowing English is seen as trendy and prestigious. The main reason for this kind of thinking is connected with internationalization, which is often associated with the English language. In addition to this, Kyrgyz lecturers and the Uzbek teachers and students accentuated the parents' involvement in their children's lives. Parents were the decision-makers who encouraged, supported, and funded their children to learn English.

Societal roles
( As seen in excerpt 1 above, while the Tajik lecturers and students did not associate the English language with prestige, most of them emphasized the importance of knowing English to get an opportunity for better qualifications, more employment perspectives, and an affluent lifestyle. These beliefs were consistent across all three countries. Handling international business deals, travelling, studying, and growing professionally were the most popular associated benefits that English can bring about (excerpt 2 below).
(2) LKG: English learning is becoming popular because of the necessity of getting better jobs and improving lifestyle.
STJ: A person who knows English can work in the international finance sectors.
LTJ: Thanks to my English I could travel around the world easily. I have been to the US, UK, Scotland and many other countries.
LUZ: Knowing English gives an opportunity to work on international projects.
Although Kyrgyz and Uzbek participants in excerpt 2 highlighted the prestige of English in their countries, during the interviews most students confessed that they rarely used English outside the university. This was true among students in Tajikistan as well. The questionnaire results showed that in their spare time, however, students seemed to use English more than staff. The numbers for 'no English in spare time' and 'all in English in spare time' were rather low, ranging from 2 per cent to 10 per cent.
Students also stated that they felt uncomfortable using English outside the classroom and university because other peers accused them of boasting about their English skills. Thus, communication in English outside the universities was limited. However, as seen in excerpt 3, students could practice their English by participating in various extracurricular clubs, doing their assignments, socializing with tourists, surfing the Internet, and so on.∖ (3) SKG: There is not much opportunity to speak English; only if students attend American Corner or speaking clubs for free.
STJ: Outside the university I only read some scientific articles, books or chat with friends. SUZ: We have some assignments which require us to communicate with people outside of the class and include their ideas in the assignments. These activities helped us to improve our English.
Excerpt 4 shows that the English language was considered a useful resource that helped Central Asian students and lecturers to improve their language skills and knowledge of subject content. For these participants, English was a resource that provided an opportunity for professional and personal development.
(4) LKG: International organizations, where English is a working language, invite teachers to attend seminars, conferences that are very useful.
STJ: Most sources are in English, and if students want to learn about their field in-depth they have to know English.
SUZ: English gives more opportunity to study my major in more depth.
LUZ: English is promoted among Uzbek doctors, lawyers and engineers to update their knowledge of their speciality.
The respondents were also asked if they considered English to be a threat to their native languages (excerpt 5).
Since governments of all three countries have strengthened the role of titular languages -the languages of the majority ethnolinguistic group -the general response to this question was negative. However, most study participants did consider Russian to be a threat to the titular languages. This is striking since Russian is by far the most popular foreign language to study in Uzbekistan and Tajikistan, while in the Kyrgyz Republic it has official language status.
(5) LKG: In the Kyrgyz Republic, many people do not speak Kyrgyz but speak Russian, so English is not a threat.

STJ: I am very hesitant about that [English is a threat to titular languages] because there is Russian which is widely
spoken in this region.
SUZ: English will not influence Uzbek because Russian has its unofficial status of the language of communication.
During the interviews, it also became clear that Russian is used by the representatives of all age groups, while English is used mostly for educational and professional purposes among the younger generation.
To summarize, on a societal level, English is considered to provide opportunities to thrive both professionally and personally, despite its limited use outside universities. English, with its growing popularity as a foreign language, has also become the main competitor of Russian, which is viewed as a threat to the development of national languages.

Institutional roles
The questionnaire results showed that there was diverse language ecology at HEIs in the three Central Asian republics. On an institutional level, in the participant institutions, English generally coexisted with at least one other language. Over 37 per cent of the respondents signified that they used three languages at their universities. Russian was the most frequently used language after English. Since the Russian language has official status in the Kyrgyz Republic, it was used there more than in Tajikistan and Uzbekistan. Uzbek, English (13.6) English (10.2) developing materials. They also stated that more support was needed from the administration of the HEIs in conducting research and writing articles.
It is impossible to discuss the role of English on the institutional level without any mention of students' transition to the HEIs from schools and colleges in these three countries. The main issue identified during the focus group discussions with students and lecturers was insufficient English provision at secondary schools. As seen in excerpt 6, students blamed their secondary school education for their insufficient knowledge of English, specifically noting inadequate English teaching hours, a grammar-focused curriculum, teachers' low levels of English proficiency, and lack of teaching materials. To prepare for the university entrance exams, students had to take additional English classes that usually cost much money.
(6) STJ: English taught at schools was not enough to enter the university and I had to take additional English lessons.
But not everyone can afford it.
SKG: English at school only focuses on grammar rules, but it is not enough.
LUZ: Teachers are doing their best to teach English at schools, but gaps still exist because of bureaucracy and lack of resources.
However, even after students are admitted to the universities, they felt they were not provided with enough language support. Those students who studied in EMI programs at state HEIs did not have access to foundation level courses that could prepare them for further studies in their specialized subjects. Students, as well as lecturers, felt that they were not trained to function in EMI settings. As a result, different stakeholders described EMI as generally challenging in the Central Asian context. Although EMI has been widely 'implemented' throughout the Central Asian region, it has many shortcomings on an institutional level which should be addressed to improve the quality of the education in EMI programs.

Pedagogical roles
On a pedagogical level, English functions as a language of teaching. In the questionnaire, the status of English at HEIs in all countries was evaluated as either 'used for teaching' or 'studied as an academic subject.' Participants from business, medical, polytechnic, information technology, law, and English translation departments identified English -not only as a subject, but also a language of instruction -for specific disciplines.
The focus group discussions revealed students' views on their teaching and learning experience in EMI programs at state HEIs (excerpt 7). For example, Uzbek students highlighted the effectiveness of communicative teaching methods, group work, and interesting materials, which made classes interactive and motivating. While most Uzbek students were satisfied with the way they were taught, Kyrgyz and Tajik students were generally demotivated to study. Their main complaints were regarding traditional methods of teaching and old resources -which were considered boring, theoretical, and aimless. In contrast, in all three countries, students in the international branch campus HEIs were very satisfied with the quality of classes. Motivation to study through EMI also depends on the teachers' role.

(7) LKG: It [EMI] is not difficult, but it all depends on the teacher; if the lesson is interesting and the methodology is
proper, then students are motivated to study in English.
LKG: Some teachers make it very interesting to study in English.
SUZ: My teacher motivates me to work harder on my English through different class activities.
SKG: Teachers use old Soviet books and teaching methods which make the lessons so boring.
Faculty members also admitted that they experienced difficulties teaching through EMI (excerpt 8). Low language proficiency and mixed-ability classes often make teaching and learning challenging and create more confusion about language use. For the lecturers without any proper training on these matters, the teaching process was very frustrating and stressful, as seen in excerpt 8. To overcome these difficulties, lecturers and students tried to find ways of making teaching and learning more effective (excerpt 9). Allowing for code-switching was by far the most widespread practice. To make explanations clearer, lecturers were sometimes encouraged by students to switch from English into other languages. Such instances were normally accepted by Kyrgyz and Tajik students; however, Uzbek students expressed their dissatisfaction with the lecturer's code-switching. Therefore, some lecturers do this willingly, while some were very hesitant. Students, on the contrary, were discouraged when changing codes and many lecturers encouraged them to use English all the time when in class. However, during the interviews, students confessed that when doing pair or group work, they often switched to non-English languages to make the discussion more effective and less time-consuming.
(9) SKG: Teachers should sometimes switch into Russian or Kyrgyz to explain something that is not clear to students.
STJ: We usually translate terminologies into Tajik to understand them better.
SUZ: Speaking in English is challenging, especially when teaching Law to students with a low level of English. It is also hard for students to study in English, so they code-switch.
Another important finding was regarding dissatisfaction with the study resources in the HEIs' libraries. International branch campus universities with better funding opportunities did not experience such issues, while at the state institutions, this was a critical problem (excerpt 10). Students and faculty members confessed that they often breached copyright law and shared e-books illegally downloaded from the Internet. Therefore, at the state HEIs, students often used e-books which were provided either by lecturers or upper-level students. Although this issue was not pointed out by Tajik lecturers and students, this might be a reality for them as well.
(10) SKG: It is expensive to support students' learning, and the university does not have funds for that; there is a great need for books. No matter how challenging, exhausting and discouraging studying/working in an English-medium program may be in a foreign context, students and lecturers were working hard to find books and to pay for extra lessons to improve their language skills to make it a fruitful experience.

Communicational roles
There was often a mismatch between students' expectations of EMI at state universities and the realities of students' frustration. For example, lack of university-wide policies made it difficult for students to speak English not only inside but also outside the classroom (excerpt 11).
(11) LKG: English is not a medium of instruction university-wide, so students always switch into Russian or Kyrgyz outside classroom.
SKG: When students just entered the university they expected that all subjects will be in English and this motivated them to enter, but in reality, there is a lot more Russian and Kyrgyz (especially during the first two years).
Even though the interview data show that there are certain expectations that students carry regarding the use of languages other than English in EMI HEIs, their general frustration with the lived reality was not the only reason for codeswitching. The comparative survey data illustrate that English language proficiency, absence of language policies, and lack of English language support in HEIs were among the reasons for both complementary and conflicting communicative role of English HEIs. When respondents were also asked to evaluate how well their proficiency (language skills) in English meets their needs at their university, the responses were mostly positive with 40 per cent and over for 'well' and from 23 per cent to 30 per cent for 'very well' (Table 5). The exception was the respondents from Tajikistan, who rated their English proficiency as 'very poor' (6.42%). The other answers, however, were rather consistent across most categories.
Another finding showed that most respondents in the three countries generally evaluated their level of English to be rather high with advanced, upper-intermediate, and intermediate levels (Table 6). However, the percentage for lower levels (beginner and elementary) was still around 9 per cent to 10 per cent in Tajikistan and 3 per cent to TA B L E 7 English language proficiency tests (%) More able in non-English 28.7 13 11.9 9.4 34.6 11

Country Yes No
Much more able in non-English 8.1 6.5 8.2 3.8 6.7 1.9 Does not apply 2.9 5.4 3.7 5.7 0.4 0 4 per cent in the Kyrgyz Republic and Uzbekistan, which signified either low self-esteem or significantly poor command of English.
Another way of assessing English language access was to see how many participants had taken an international English language assessment like the IELTS or TOEFL exams (Table 7). As seen in  istan, 2018). The question about the respondents' language ability to discuss academic and professional interests revealed that most Kyrgyz, Tajik, and Uzbek participants thought that they were equally able to discuss their interests in both English and native languages, which requires a high perceived proficiency level in English. However, 28.7 per cent of students in the Kyrgyz Republic, 11.9 per cent in Tajikistan, and 11 per cent in Uzbekistan thought that they were more able to do so in their native languages (Table 8). Even though the questionnaire data showed that the participants consider their English to be generally high, as mentioned above, code-switching often took place. Since there was no specific 'English only' language policy at the universities, lecturer and students maintained flexibility in switching languages. During the focus group discussions, participants said that faculty members and students usually had a negotiated informal agreement to use only English in EMI classes unless there was a necessity to code-switch. For everyday conversations, students were also encouraged to use English, but that was rarely the case. The questionnaire results in all three countries showed that the respondents were interested in additional professional English language support if this support was offered at the university. Additionally, more than 60 per cent of respondents in each country stated that they had taken English language courses to develop their language skills since they have started to either study or work at the university. During the interviews, students confessed that they continued taking additional English classes and they would likely take part in free language courses if the university offered any. Overall, on a communication level, most lecturers and students expressed their satisfaction with their English language skills; however, in reality, they used their full linguistic repertoire whenever needed.

DISCUSSION
The findings raised key issues within the societal role of English in the HE system of these three countries, where English was used, not only at the international universities with EMI, but also in separate programs at state institutions.
First of all, since most students came to EMI HEIs from secondary schools where other languages were the medium of instruction, the transition to an EMI setting took time and effort. Even though the government of all three countries implemented English classes from grade one and three of primary schools, by the time students entered HEIs, they did not have enough English language skills. Therefore, many students admitted that they had to take additional tuition-based classes to enter the university. Further evidence for this interest was the growing popularity of English language proficiency tests like the IELTS and TOEFL exams. Because a certain level of language proficiency was often one of the entrance requirements to EMI HEIs, the demand for these tests continues to grow among students. The societal situation of English directly affected its role in various institutions. The findings of this study revealed that English was often viewed as a powerful tool for future development and improvement in every aspect of participants' lives. English was being promoted at and by HEIs as equipping students not only with subject knowledge but also English language competency, which was believed to be important to get a better job after graduation. While international branch campuses had more funds because they have other funding streams, state universities depended on government funding, which was often limited. Limited funds resulted in poor facilities and shortage of resources, which was particularly the case for Kyrgyz and Uzbek students. For all three countries, a lack of funding resulted in the absence of special EMI training that was needed for both students and teachers. In addition to this, at state universities, EMI was imposed on students and lecturers in a top-down manner, which was both challenging and frustrating. To what extent this objective is achieved, however, often depended on students and lecturers themselves.
The lived reality of English in three countries was that it functioned at HEIs without any overt policies that regulated language use. The state HEIs followed a covert EMI policy based on top-down policies, while international universities were EMI by default. From the faculty members' perspective, EMI classes took more time for preparation and explanation. Students stated that they often spend more time understanding the subject matter or refer to their native language sources for additional explanation and support. These findings were supported by Hellekjaer and Wilkinson (2001), who found that students studying EMI subjects spent 10 per cent to 25 per cent more free time on self-study than on subjects with their first language. There was also a demand for additional language support from the HEIs and clear evidence that both lecturers and students spent their funds on the development of English competencies. The interview data showed that both students and lecturers often initiated or organized additional extracurricular activities to improve their English, which showed great motivation to study.
On a pedagogical level, English also functioned but not without its drawbacks. A strong association of EMI with modern teaching methods and communicative approach to teaching was one of the primary motivations for students to study in EMI programs. The findings suggested that students at state universities were dissatisfied with the way they were taught. Claiming that teachers' main concern was attendance rather than learning, students were not motivated to study. Boring lectures, heavy focus on theory and absence of practice, and no articulated learning objectives were students' main complaints. The greatest levels of dissatisfaction were voiced by Kyrgyz and Tajik students, who claimed that traditional and teacher-centred pedagogy made classes boring and did not develop their professional and language skills. From teachers' perspectives, teaching through EMI at state universities was particularly difficult due to lack of materials and training. At international universities, the situation was the opposite. A wide variety of tasks and modes of delivery along with up-to-date materials resulted in general satisfaction with the quality of teaching.
In terms of preferences, since there was no explicit policy that would prioritize one variety of English over another, the preferences for British or American varieties of English were strictly personal or depended on available resources.
The interview results illustrated that the books that lecturers used usually define the use of a particular variety of English. Some lecturers and students could not explain why they preferred a certain variety or admitted that they are biased about the country of its origin (usually the United States or the UK). Whether English development requirement was explicitly included in the curriculum or not did not emerge in this study. However, it is assumed that this is a requirement at least for the students of English translation and English linguistics faculties in Uzbekistan since it was the only one of the three countries that had aligned all of its study objectives to the CEFR in English language-focused degrees. However, language development is an independent learning aim for most of the students and lecturers. As mentioned above, they invested in their self-improvement.
Finally, the role of English on a communicative level largely depended on the language proficiency of its users.
Across all three countries, the findings suggested similarity in this matter. Even though participants generally considered their English to be sufficient enough to operate in higher education, there was a great interest in English proficiency development. It is worth noting that around 20 per cent of students and faculty members in all three countries responded that they rated their proficiency in English as acceptable, which showed the awareness and justified the need for language improvement. For EMI lecturers in EFL settings, poor language proficiency may affect teaching and result in problems with rapport with their students. However, English was not the only language used at universities and other languages also played a significant role in different aspects of university life. The data showed that apart from informally or formally assigned EMI, most students and academic staff used Russian and their national language during classes, in training, and in daily communication. The main reason for this was the natural inclination for multilingual people to communicate in the language or languages in which they feel safe and comfortable. Moreover, students' and lecturers' admitted lower English proficiency levels might have resulted in a preference for native language use over the use of English. Although in the survey students and faculty members generally evaluated their English proficiency rather highly, during the interviews they stated that they code-switched during the lessons. Most research studies report that code-switching mainly happens when there is a misunderstanding or for the necessity to clarify certain points (Baker & Hüttner, 2018;Macaro et al., 2018;Tarnopolsky & Goodman, 2014). Both aspects were evidenced in the findings of this study as well.

CONCLUSION
This article investigated the experience of English and the societal, institutional, pedagogical, and communicational roles it plays in EMI HEIs in three Central Asian countries. Since English in such HEIs does not exist in isolation from the titular languages and Russian, the ROAD-MAPPING framework -and specifically the four aforementioned sub-dimensions -that focuses on English in multilingual settings was used for the analysis and interpretation of findings. The use of this framework allowed grounding the findings in a broader scholarship of EMI research in the context of three Central Asian countries. This study was a pioneering attempt to examine the experience of English, as well as the roles it plays in HEIs and in wider Kyrgyz, Tajik, and Uzbek societies. It also contributes to a broader scholarship around world Englishes because it highlights the lived reality of English in three countries of the Expanding Circle. The findings revealed that both academic staff and students viewed English as an opportunity for a brighter future. Study participants were motivated to improve their language proficiency in English, but they needed more support from the administration of HEIs and the government. Among the other shortcomings associated with English in all countries were limited funds, shortage of materials, and lack of professional EMI training.
What became apparent in this study was that the roles of English are interrelated and intricately connected one to another. Tackling the issues on the societal level is impossible without dealing with the challenges faced in teaching and learning in pedagogical and institutional domains. Thus, the experience of English in the three countries was uneven. Subsequently, EMI HEIs, even though positively welcomed, should address various issues to be both effective and enjoyable for the participants. Therefore, drawing on the research findings, the HEI administrations, Ministries of Education, and countries' governments are encouraged to consider the following: • English language teaching curriculum at schools and lyceums should align with the CEFR to ensure that students are equipped with the required level of English language proficiency by the time they enter the university. • More attention should be paid to the transition period from secondary schools to EMI HEIs through foundation level courses on core subjects or providing ongoing EAP or ESP support.
• Both students' and lecturers' lower language proficiency impedes learning and teaching and, thus, their language development should be better supported at institutional and state levels.
• HEIs should have sufficient funding to update their materials, invest in teacher training with the focus on EMI pedagogy, and support language development initiatives.
• The challenges that faculty members and students experience in EMI settings in and out of the classrooms should be freely discussed with the administration and addressed.
• HEIs with EMI should consider having an explicit language policy that would celebrate linguistic diversity on campus and clearly define the areas of English language use.
The scope of the findings suggests many avenues for further research in the region. Future research may focus on the transition between educational non-EMI and EMI schools, the impact of EMI on social equality in the region, inadequate English language proficiency at schools, EMI management and policy, as a few examples. These findings may initiate the debate and engage all EMI stakeholders to work on improving the experiences of all stakeholders.