Skip to content
BY 4.0 license Open Access Published by De Gruyter Mouton November 4, 2022

Language policy in the internationalisation of Higher Education in Anglophone countries: The interplay between language policy as ‘text’, ‘discourse’ and ‘practice’

  • Florence Bonacina-Pugh ORCID logo EMAIL logo , Elisabeth Barakos ORCID logo and Qi Chen


In order to better compete in an increasing neoliberalised education system, many Higher Education Institutions (HEIs) have developed an internationalisation strategy that aims at incorporating an intercultural and global dimension into curricula and learning environments for all. This internationalisation agenda raises important language policy issues that are often side-lined in the UK and other Anglophone countries where an English monolingual ethos prevails. Centrally, the question arises indeed as to whether internationalisation processes have an impact on HEIs’ language policies in Anglophone countries. This paper takes the case of a Russell Group University in the UK and focuses on two masters programmes that attract annually a ‘multilingual elite’ (Barakos and Selleck 2019). It examines the institution’s language policy adopted at the levels of ‘texts’, ‘discourses’ and ‘practices’ (Bonacina-Pugh 2012), using a critical discourse analysis of policy documents and a conversation analysis of classroom interactions. We argue that language policy is at the core of HEIs’ internationalisation processes even in Anglophone countries and that, methodologically, the articulation of findings from critical discourse and conversational analyses represents a step forward in the field of language policy.

1 Introduction

As a response to globalisation, that is, the increasing interconnectedness of transnational communications and movements (see also Block 2006: 3), the global marketisation of education and broader socio-economic processes in the knowledge economy, Higher Education Institutions (HEIs) have adopted ‘internationalisation’ strategies. These strategies refer to the integration of “an international, intercultural, or global dimension into the purpose, functions or delivery of post-secondary education” (Knight 2004: 11). Knight (2006) further distinguishes two streams of internationalisation: ‘internationalisation abroad’ which consists of students and staff mobility and supporting education across borders; and ‘internationalisation at home’ which consists of transforming the curriculum and more broadly the teaching and learning environments of all students to make higher education more responsive to the new demands of a globalised world. In non-Anglophone countries, ‘internationalisation at home’ strategies usually result in a change of language education policy, namely a switch to using English as a medium of instruction (EMI). This is the case for instance of Finland, where English has replaced Finnish and Swedish in many colleges and universities (e.g. Saarinen 2012). In Anglophone countries, however, ‘internationalisation at home’ strategies rarely include a reflection on language policy. This paper addresses this issue and examines the language policy of a HEI undergoing internationalisation in an Anglophone country. It takes the case of a Russell Group University in the UK that claims to have a strong international tradition and reputation, and focusses more specifically on two Masters programmes that attract annually a ‘multilingual elite’ (Barakos and Selleck 2019), and in particular Chinese students. It examines the interplay between the Institution’s language policy as ‘text’, ‘discourse’ and ‘practice’ (Bonacina-Pugh 2012) by conducting a Critical Discourse Analysis of policy documents and a Conversation Analysis (CA) of classroom interactions. This paper constitutes a step forward in the field as it articulates issues of language policy, multilingualism and internationalisation in the under-researched context of Anglophone countries. Methodologically, it represents a rare attempt at bringing together findings from Conversation Analytic and Critical Discourse Analytic approaches.

2 Internationalisation of Higher Education, multilingualism and language policy

To examine how internationalisation is shaping language policy, we articulate three notions that have not been often articulated in the context of HEI in Anglophone countries: internationalisation; multilingualism; and language policy. We now present our understanding of these three notions, central to our argument.

The number of students attending a HEI is constantly on the rise across the globe. UNESCO, together with the International Institute for Educational Planning and the Global Education Monitoring Report (2017) argue that between 2000 and 2014, the number of students in HEIs more than doubled to 207 million world-wide. The student population is becoming increasingly more mobile, with more and more students attending a HEI in a country different from the one they grew up in. In the UK for instance, in 2016, 89,318 students moved from China to study in a UK HEI (UNESCO Institute for Statistics 2016). In response to the globalisation of Higher Education, institutions, and especially universities, have adopted ‘internationalisation’ strategies. These strategies take on different forms depending on the context in which a particular institution is situated but they attempt on the whole to make HEIs more competitive on the global market. Many scholars have investigated the impact of globalisation on HEIs (e.g. Altbach 2009; Robertson and Kedzierski 2016) and the study of ‘internationalisation of Higher Education’ has become a research field in its own right. This paper focuses on one aspect of internationalisation of Higher Education, namely what Knight (2006) proposes to call ‘internationalisation at home’. This is understood as being “the purposeful integration of international and intercultural dimensions into the formal and informal curriculum for all students within domestic learning environments” (Beelen and Jones 2015: 76). Whilst the focus seems to be on internationalising the curriculum, very little attention is given to the issue of language in internationalisation processes by HEIs (e.g. Soler and Gallego-Balsa 2019). We argue here that, along with the formal and informal curriculum, language policy should also be at the core of internationalisation processes ‘at home’.

For the last decade, language policy issues in the internationalisation of HEIs have been investigated in non-Anglophone countries. In these contexts, internationalisation has often led to language policy changes such as the adoption of EMI (for a systematic review see Macaro et al. 2018). Some scholars analysed staff and students’ experiences of EMI (e.g. Haberland et al. 2008), or the policy challenges and realities of EMI (e.g. Bolton et al. 2017). Others examined staff and students’ multilingual language practices against monolingual EMI policies (e.g. Mazak and Carroll 2017; Moore 2016).

In Anglophone countries, however, there is a paucity of studies related to language policy in internationalised HEIs. This is perhaps because internationalisation of HEI in Anglophone countries has not often led to new language policy texts and English is assumed to be the one and only medium of instruction and interaction. As Martin noted, British HEIs have maintained a monolingual ethos, which resulted in “linguistic myopia” (Martin 2010: 110). For that reason, most existing applied linguistic studies on HEIs have focused so far on raising awareness of the multilingual ecology of international universities. Preece (e.g. 2011, 2018), for instance, has consistently argued that despite being sites of multilingualism, universities in the Anglophone centre are ignoring, marginalising or silencing staff and students’ multilingual repertoire (see also Preece and Martin 2009: 4). Similarly in Canada, Marshall (2009) observes that although international students perceive themselves as multilingual speakers, they are often confronted with a deficit ‘remedial ESL’ (English as a second language) identity. As a result, the institution concentrates on improving their English rather than on imagining ways in which their multilingualism could be used as a resource for teaching and learning.

With a view to recent political developments in the UK, the impact of Brexit on HEIs and language policy more generally still remains to be seen. Mac Giolla Chríost and Bonotti’s work (2018) shows that Brexit will certainly re-shape public policy norms and legal frameworks as they relate to linguistic diversity. In fact, they argue against the view that English might lose its importance in a post-Brexit EU. In this paper, we recognise along with others (e.g. Martin 2010; Preece and Martin 2009) the role of language policy in creating “multilingual spaces” in HEI. Preece (2011: 139) writes that: “universities in the Anglophone centre have not kept pace with the changing student demographic and need to devise institutional language policies that take pluricentric, rather than monocentric, perspectives to linguistic diversity”. We argue that internationalisation processes also affect language policy in Anglophone countries, albeit in a more subtle way than it does in non-Anglophone countries, where a clear change of language policy is easily observable.

The second key notion to our study is that of ‘multilingualism’. Multilingualism is a social phenomenon that can be investigated using a variety of lenses at the intersection of discourses and practices. We take the stance that multilingual classroom participants do not use language A or B but rather a “medium of classroom interaction” (Bonacina and Gafaranga 2011; see also Gafaranga 2007), that is, a code of interaction which can include a variety of linguistic and non-linguistic signs. In a ‘monolingual medium’, speakers use semiotic resources that would traditionally be seen as belonging to one language. In a ‘bilingual medium’, speakers use semiotic resources that would traditionally be seen as belonging to one or more languages (and not simply two as the prefix ‘bi’ would indicate). Following an emic perspective to interaction, language alternation is only seen as such by the analyst if, and only if, classroom participants themselves recognise that two languages are being used.

The third notion central to our study is that of ‘language policy’. We adopt Bonacina-Pugh (2012) tripartite conceptualisation of language policy as ‘text’, ‘discourse’ and ‘practice’. Building on Ball’s (1993) conceptualisation of policy as ‘text’ and ‘discourse’, Bonacina-Pugh (2012) proposed to conceptualise language policy as ‘text’ and ‘discourse’. Bonacina-Pugh further proposed a third conceptualisation, namely that of language policy as ‘practice’, following Spolsky’s (2004) idea that there is a language policy at the level of practices. This tripartite conceptualisation of language policy can be explained as follows: language policy as text is a written artefact that regulates language (use); texts are products of discourse, which is one form of social action. Language policy as discourse thus produces knowledge and meaning in context and stands in a dialectical relationship to the social. Language policy as practice (or ‘practiced’ language policy) refers to a set of implicit interactional norms that influence the production and interpretation of language choice acts. We thus conceptualise ‘language policy’ as being a multi-layered social and discursive process that involves interconnected texts, discourses and practices. Drawing on critical discursive (Barakos 2016) and practice-based (Bonacina-Pugh 2020) approaches to language policy, our aim here is to investigate the extent to which the language policy at the levels of texts, discourses and practices reflects the internationalisation agenda of the university under study.

3 Context

In the academic year 2016–2017 (which is the closest statistics to our year of study in 2017–2018), there were 2.32 million students studying at UK HE institutions, of which 442,375 were non-UK students (Universities UK). Of these non-UK students, the UK Council for International Student affairs reports that 307,540 were ‘international’ students, that is, from non-EU countries, and almost one third of these international students was from China (see also the UNESCO Institute for Statistics). This represents a 14% rise of students from China since 2012–2013. This paper takes the case of a prestigious Russell Group university in the UK where these trends are also noticeable. In the university under study (referred to as University X for the sake of anonymity), and at the time of the study in 2017–2018, there were 41,312 students, of which 13,280 were non-EU. We note that 9781 students attended taught post-graduate programmes, and a large number of these students (i.e. 4686) were non-EU.

Two masters programmes comprising a total of 164 students were studied. 136 of the 164 total were from China. However, it is important to note that the Chinese students themselves are a ‘superdiverse’ group (Wei and Zhu 2013), who brought with them a variety of linguistic resources. The twelve core teaching staff on these two programmes at the time of the study were all English bilinguals: six of them were British, three from the EU, and three from outside the EU. These two Masters programmes sit within one of the 21 Schools of University X.

4 Data sets and methods

Two data sets were collected: (1) a set of institutional documents; and (2) a set of audio-recorded classroom interactions. These two data sets were analysed using: (1) critical discourse studies (CDS); and (2) CA respectively. Whilst we are aware of the theoretical tensions between CDS and CA particularly around their respective understanding of ‘context’ (e.g. Billig and Schegloff 1999; Wetherell 1998), we do not feel it is necessary to engage in the debate over their potential (in)compatibility for the purpose of this paper, for the simple reason that we use CDS and CA to analyse very distinct data sets and phenomena. Ultimately, findings from CDS and CA will be jointly articulated to discuss the interplay between language policy as text, discourse and practice in relation to the internationalisation strategy of University X.

Our first set of data consists of three key institutional policy documents related to the internationalisation agenda of University X: The University’s Internationalisation Strategy “University Global”, the University’s Strategic Vision 2025, and a School-internal Internationalisation Strategy (2015). Unlike many universities in non-Anglophone contexts, which would have an explicit and written language policy, there is no ‘declared’ language policy (Shohamy 2006) formulated by University X. Notable exceptions are for minority languages, such as the University X’s Gaelic Language Plan, its policy regarding English language entry requirements for students, and its British Sign Language Plan. As a result, for our first data set, we selected texts on the basis of their salience to the topic of internationalisation at University X. The selected texts serve as surrounding documents to the debate over diversity, interculturality and a global education. We aim to trace processes of recontextualisation across these documents to show how certain themes, arguments and discourses circulate and change in different discursive contexts.

Our second data set consists of a corpus of classroom interactions audio-recorded in Semester two of the academic year 2017–2018. Data was collected during an eight-week option course that sits within the two Masters programmes described above. This particular option course was attended by 40 students. Two groups of six students each were audio-recorded during pre-workshop activities (without the tutor) and workshop activities (with the tutor) for a total of 12 h over a period of eight weeks. Group 1 consisted of three Chinese students, one Taiwanese student, and one Mexican student. Group 2 consisted of one Singaporean student, one Taiwanese student, and four Chinese students. Together with the French and English bilingual tutor, these groups presented a linguistic repertoire that included English, Mandarin-Chinese, varieties of Chinese, Spanish and French. In Group 1, students only had English as a shared language, whereas in Group 2, students shared English as well as Mandarin Chinese. Audio-recordings have been transcribed and translated into English when necessary.

To investigate language policy as text and discourse, CDS serves as an analytic lens. CDS is a context-sensitive pluralistic approach in critical social research (for overviews, see e.g. Hart and Cap 2014). It is both a theory and a method concerned with unpacking “what people say and do in their use of discourse in relation to their views of the world, themselves and their relationships with each other” (Paltridge 2012: 191). The discursive approach to language policy (motivated by Barakos 2016) employed here takes discourse as an entry point to examine language policy as a multi-layered phenomenon that manifests itself in regulations (declared policy texts), people’s ideologies and experiences about language as well as their language practices. It is shaped by the situated context and broader social structures at play. For the purpose of this paper, the policy analysis will gauge how ‘internationalisation’ is talked about in policy texts by paying attention to the text-internal analysis, including a focus on key discourse topics such as language, diversity and culture, lexico-syntactic features such as vocabulary and phrases as well as discourse elements such as genre and topoi (common-sense ideas as grounds for arguments), intertextual and interdiscursive connections between the strategies and the social variables that shape the creation and appropriation of the texts and their underlying ideological values. The analysis couples linguistic analysis (e.g. argumentation, how social actors and concepts are constructed, intertextuality) with social-theoretical analysis. It is thus paramount to investigate processes of recontextualisation, that is, how discourse topics and arguments are repeated, reformulated, and transformed in the particular context of Higher Education policy discourse on internationalisation. We are also interested in embedding circulating discourses within broader sociocultural, economic and political contexts and employing a critical gaze at taken-for-granted assumptions, circulating norms and problematic discursive practices by policy makers.

To investigate the ‘practiced’ language policy, we analysed the second data set through a CA lens (for a rationale for the use of CA for the study of ‘practiced’ language policy see Bonacina-Pugh 2012). A ‘practiced’ language policy is the set of norms of language choice that is being shared and negotiated in interaction to legitimise the use of particular languages at the local level (of a classroom in our study). As previously argued (Bonacina-Pugh 2012, 2020), this set of norms is considered to be a policy because it influences and informs the interpretation and production of language choice acts. We first approached our data set with a view to identify language choice ‘practices’, that is, regular patterns of language choice acts. We then took each of these language choice practices and tried to unravel the norms of language choice that speakers oriented to when engaging in these regular language choice acts. In other words, we tried to identify what classroom participants referred to when interpreting the use of a particular language(s) as being either legitimate or deviant. Following Shohamy’s and Spolsky’s advice that to best discover a language policy within language use one has to study its “nonobservance” (2000: 29), we conducted what Conversation Analysts call a ‘deviant cases analysis’ (Heritage 1984). Using a turn-by-turn analysis and adopting an emic approach to interaction, we focused on how classroom participants themselves reacted to the use of particular language acts. If, for instance, classroom participants interrupted, repaired, or translated talk in a particular language(s), we treated these interruptions, repairs and translations as evidence that classroom participants perceived that particular language(s) as ‘deviant’, or in other words, inappropriate. It is by looking at what classroom participants perceived to be deviant or repairable that we were able to deduce what they perceived to be normative, that is, in line with the agreed set of norms of language choice in the classroom, i.e. the ‘practiced’ language policy.

5 Language policy as text, discourse and practice

In what follows, we first focus on the critical discursive policy analysis of a range of institutional internationalisation strategies. We pay attention to the ways internationalisation is talked about and how it gets interlocked with language, culture and diversity by discussing selected passages of discursive data that are most relevant in constructing the link of internationalisation, diversity, culture and language. We then turn to our set of classroom talk to identify whether the ‘practiced’ language policy in these classrooms reflect a similar understanding of language, culture and diversity.

5.1 Language policy as text and discourse: Insights from institutional internationalisation strategies

Internationalisation and global outreach are central discursive features of the positioning, identity and branding strategies employed in the internationalisation documents of University X. The particularity of this institution is though that the international character gets discursively negotiated through claims over global outreach and a promotion and preservation of local heritage, culture and pride. What is of interest to this analysis is the ways internationalisation is discursively constituted and coupled with diversity discourses, and in what ways, if at all, language is made relevant across the various documents. The texts chosen for analysis pertain to the academic promotional genre. The nature and purpose of internationalisation strategies is similar to that of university mission/vision statements which fulfil a “telling” and “selling” function (Fairclough 2010: 184). That is, there is a dual focus on informing the customers (staff, students) of the internationalisation agenda and on promoting the institutions’ goodwill in doing so (see also Morrish and Sauntson 2013).

What these statements then share is the ways the university declares its public vision and how it conceives its strive for excellence and uniqueness before detailing its specific action plans on internationalisation. Key recurring discourse topics of the internationalisation agenda across all policy statements analysed encompass globalisation, diversity, investment, educational quality, culture, knowledge, resource and capital in terms of learning, teaching and research. It is also interesting to note how the same few broad and elusive terms of the discourse on internationalisation (e.g. world, worldviews, diverse; culture; investment; collaborate) are recycled and recontextualised across and within these various documents, from the top university level’s vision statement and its internationalisation strategy to the school level’s own internationalisation strategy.

University X positions and markets itself as a global and yet local institution, emphasising the connection between education for global citizenship and maintaining a local perspective. The following discourse fragments in Extracts 13 exemplify the localised dimension of globalisation as a marketable feature of the vision for internationalisation across the strategies:

Extract 1:

The University X: enlightenment for the twenty-first century world.

As a truly global university, rooted in Scotland, we seek to benefit society as a whole (University X Strategic Vision 2025, bold for emphasis)

Extract 2:

We are a distinctly Scottish University based in Scotland […], but our reach and aspirations are international and it is in that context we must be measured. […] This strategy sets out how we intend to deliver on our aspirations of becoming a place of first choice in the minds of the world” (University’s Internationalisation strategy, bold for emphasis; deletion for anonymisation purposes)

Extract 3:

To reach our overarching objective to become a campus of mutuality, collaboration and conversations between Scotland and the world” (Schools’ Internationalisation Statement, bold for emphasis)

As the extracts demonstrate, a range of nomination strategies (“a truly global university”; “a place of first choice”; “a campus of mutuality”) are at play to construct the social actor University X as a global player that can build a bridge between its distinct historical location in Scotland and its international outreach to the world. The use of personal deixis (“we”; “our”) to encode the university, foregrounds the institution’s values and beliefs as a personalised collective entity, similar to that of an individual person. This strategy serves to engage and connect the reader (Fairclough 1993). The fragments here, whilst eclectic and selective in nature, are also exemplary of the high density of positively connoted nouns (“aspirations, collaboration”), and adjectives and adverbs used as modifiers (here, “distinctly Scottish”; “first choice”; “truly global”) which characterise the entire texts and their promotional genre.

As Hultgren et al. (2014) work shows, in competitive climates, universities need to increasingly position and market themselves as international players in a global education market. Fairclough’s work in the context of UK Higher Education (e.g. 1993) and Urciuoli’s (e.g. 2003) work for US Higher Education diversity discourse have also critically traced the marketization, neoliberalisation and popularization of university discourse. For University X, then, part of its internationalisation agenda lies in branding itself as distinctly Scottish and truly global institution by blending global and local values in order to gain a competitive edge over other HEIs and attract foreign staff and student talent to this specific geographical part of the UK.

Another element of the University’s branding strategy is based on interlocking arguments about internationalisation with economic rationalities as part of marketisation processes as discursive actions, as illustrated in Extract 4.

Extract 4:

University’s Internationalisation strategy

We have succeeded in attracting increased numbers of international students from more than 130 countries on the back of our reputation, teaching quality and location. Having a good proportion of international students offers all of our students the opportunity to learn more about the world. In partnership with approximately 200 institutions worldwide, we also participate in exchange and mobility schemes that offer students and teachers a challenging and exciting international experience, providing them with new perspectives and an understanding of different cultures (bold for emphasis)

In this paragraph of the University’s internationalisation strategy, University X persuades its readers of its international appeal by drawing on the topos of numbers (“more than 130 countries”; “approximately 200 institutions”) as common-sense reasoning. The numerical evidence on international students and staff numbers adds officiality and factuality. It aids to make a case for the university’s global and diverse orientation, whilst the reference to ‘location’, as juxtaposed to reputation and teaching quality, recycles the key promotional argument about the distinctiveness of University X in terms of its geographical location and history (as discussed previously in Extracts 1 and 2). Furthermore, international students and staff / mobility scheme are promoted and mobilised as gateways to the world (“to learn more about the world”) and to access “different cultures”, with culture being treated here in a rather abstract way and based on difference.

The marketised Higher Education discourse on the global and local values of University X becomes further blended with a repetitive rhetoric on diversity and culture, which is especially salient in the School’s internationalisation strategy.

Extract 5:

School’s internationalisation strategy

School X is proud of the diversity of its student and staff population. Our key aim is to prepare 21st century graduates who will have the knowledge, understanding, skills and vision to live in and contribute responsibly to a globally interconnected society. School X is dedicated to developing an inclusive teaching, research and study culture that celebrates and upholds the principles of equality and diversity. We aim to foster a campus and online culture which stimulates collaborations and conversations across cultures, geographies and disciplines (bold in original)

As exemplified in Extract 5, this discourse fragment is characterised by a density of positively connoted nouns and predication strategies that position the social actor, here the School, as an ambitious, aspirational and active agent (“is proud of”; “is dedicated to”, “aim to foster”). In terms of nomination strategies, the School oscillates between adopting a more distancing, authoritative, third-person (“the School”) and a more inclusive, personalising first-person narrative (“our”; “we”). Here, the school prides itself on its diversity ethos that stretches to “its students and staff”. Yet, it remains vague about what type of diversity they mean (ethnic, religious, racial, linguistic, national, socio-economic etc.) and whether this diversity encompasses the local, regional, national and/or international level.

In terms of intertextuality, the term “21st century” is one that is recontextualised from the sub-title of the University’s Strategic Vision 2025 (“The University X: enlightenment for the twenty-first century world, see Extract 1) in its appeal for future-orientation and the identification of generic graduate attributes (“knowledge, understanding, skills, contribute responsibility”) needed for a global society. The diversity ethos in terms of staff and student cohorts then gets repeated in connection with the broader, more universal principles of equality and diversity and gets tied with inclusiveness in teaching, research and study practices. The term “culture” is used here varyingly to refer to “study culture” and “a campus and online culture”. As indicated in the last sentence, it is also employed with regards to the common-sense understanding of culture as a cross-border, intercultural phenomenon. From an argumentative perspective, then, the topos of culture as a common-sense way of reasoning is invoked in this extract as well as throughout the internationalisation statement. Other phrases such as “staff and students from a diverse range of cultures and countries” or “we will organise intercultural events to celebrate diversity” exemplify how culture is treated as part of an essentialist and narrow understanding of interculturality, centred on awareness raising of ‘difference’. This celebratory discourse on diversity and its interlocking with internationalisation intersects with broader socio-political and political-economic phenomena such as the growing hegemony of neoliberalism in education, which frames learning and teaching in competitive, efficient, productive and entrepreneurial terms. Internationalisation is hence one way for universities to participate as genuine competitors in an increasingly neoliberalised education system.

What is notable from the discussion of all data extracts thus far and the wider analysis of the sampled internationalisation strategies is that the discourse on linguistic diversity is invisible. Culture and diversity are abstracted from language, that is, connections to linguistic knowledge, language skills or the relevance of multilingualism for twenty-first century graduates are obscured. Furthermore, the linguistic capital of international and home students as well as that of existing and incoming staff is absent from the internationalisation discourse of University X. Language/language skills/multilingualism are obliterate topics. The only explicit references to language can be found in the School’s internationalisation statement, as exemplified in Extract 6.

Extract 6:

School’s Internationalisation Statement

Promoting worldviews

As priorities, we will:

  1. Organise student led contributions during the annual ILW week to enable discussions about different worldviews on a range of topics

  2. Organise events (including during ILW) that aim to alert home students of the value of learning languages and of learning in education settings in a second or third language (italics and bold in original)

Here, the School aims to prioritise awareness-raising events for home students about language learning and education in “a second or third language”. It discursively frames language under the broad banner of ‘worldviews’. The fragment is characterised by vagueness through generalised phrases (e.g. “a range of topics”; “value of learning languages”) and exclusion in that the organised event in point 6 are uni-directional and aimed at home students only, and not at international students. Similar to the above extracts, what is erased from this worldview promotion is the notion of communication skills and how the multilingual repertoire of students and staff can be used as a resource in teaching and learning.

Based on the analysis thus far, we can argue that the diverse and international education envisioned by University X seems to focus on the knowledge of ‘other’ cultures. The domestic Scottish culture, including the University’s own existing linguistic diversity, is sidelined. The internationalisation discourse does not embrace how the existing linguistic diversity of University X, and within the School, can be promoted, accommodated, planned for and linked up with its internationalisation agenda. Language policy is thus never discussed explicitly, which begs the question of the role of English in HEIs in Anglophone, ‘native’ English-speaking countries such as the UK. Liddicoat’s (2016) study on language planning in universities found that English serves as the academic lingua franca and the dominant language of public life in the UK, so universities have shown only little interest in managing the linguistic consequences of internationalisation. When for many universities in non-Anglophone countries, internationalisation means Englishization (e.g. Hultgren et al. 2014; Piller and Choo 2013), this aspiration does not necessarily need to be met by University X. Rather, there appears to linger a taking-for-granted assumption about the pervasiveness of English at University X as a ‘natural’ medium of instruction and medium of communication, based on the fact that said University is based in an Anglophone country.

5.2 Language policy as practice

Turning now to our corpus of classroom talk, we investigate the language policy as practice (also referred to as the ‘practiced’ language policy in Bonacina-Pugh 2012 and 2020) that the classroom participants under study orient to in interaction, that is, the set of norms of language choice that inform the interpretation and production of their language choice acts. In doing so, our aim is ultimately to show the resonances and dissonances between the institutional internationalisation discourse of University X and the ‘practiced’ language policies observed on the ground.

A CA of our corpus first reveals that English is adopted as a monolingual ‘medium of classroom interaction’ (Bonacina and Gafaranga 2011) in teacher-led talk (see Section 2 for an explanation of the notion of ‘medium of classroom interaction’). This choice of medium is never negotiated, talked about or challenged by either the teacher or the students. What is more, we note that there are no instances of alternation to languages other than English in teacher-led talk. Despite the fact that the teacher is bilingual in French and English and that students are also all bilingual in their respective languages and English, only English is used in workshop interactions between students and the teacher. Furthermore, English seems to be the only normative language choice in teacher-led talk, that is, the only language that is seen as legitimate and appropriate. In other words, it seems that only the shared language among classroom participants is appropriate.

Interestingly, however, in the corpus as a whole, English is not the only language that is used with a sense of normativeness. Mandarin is also seen as a legitimate medium of classroom interaction, both in pre-workshop groups (when the teacher is absent from the room) and workshop groups. In Extract 7 below, for instance, two students talk in Mandarin whilst the teacher is addressing the whole class in English. The teacher is going over the answers of a matching exercise, where students had to match a concept from the lecture with its definition. Here, the teacher acknowledges a students’ response about the definition of ‘coherence’ and ‘cohesion’.

Extract 7 (Group 1):

  • T: teacher, English and French bilingual speaker

  • S2 and S4: students, English and Mandarin bilingual speakers

In this extract, a problem arises as S4 does not understand which concept the teacher is discussing. S4 turns to her peer S2 to seek help and uses Mandarin (line 7). In line 8, S2 gives the clarification requested, also using Mandarin. This brief exchange between S4 and S2 forms a schisming (e.g., Egbert 1997) away from the teacher-led participation framework. Interestingly, the use of Mandarin here is not seen as a repairable matter. This indicates that Mandarin is seen as acceptable among S2 and S4, most likely because it is for them a shared language. In this sense, S2 and S4 seem to orient to the norm of language choice according to which it is appropriate to use a shared preferred language as a medium of classroom interaction, even if that language is not English.

Similarly, when students share more than one preferred language, these languages can be adopted as a “bilingual medium of classroom interaction”[1] (Bonacina and Gafaranga 2011). In our corpus, this is the case when Mandarin and English bilingual students choose to use both Mandarin and English as one medium, without seeing any of these two languages as needing repair or translation. English and Mandarin can be used within the same turn by the same speaker or, as illustrated in Extract 8 below, each speaker can choose to use a language each. Here, S1 and S2 are working in a linguistically homogeneous group where all speakers share English and Mandarin. They are presenting to each other the picture books they would like to use for their lesson plan.

Extract 8:

  • (Group 2): S1, S2: students, English and Mandarin bilingual speakers

This extract is a clear example of the “parallel mode” (Gafaranga 2007) of a bilingual medium of classroom interaction, as S1 uses Mandarin throughout the interactional episode, whilst S2 uses English. Since both English and Mandarin are shared languages among the speakers of this interactional episode, both languages are seen as appropriate and legitimate.The use of these two languages is not challenged, repaired or translated by any participants and everything indicates that the conversation is going smoothly and naturally. English and Mandarin are used as one code, or in other words, one ‘medium of classroom interaction’ (Bonacina and Gafaranga 2011).

We have shown so far that language choice is never explicitly discussed in our corpus but that, instead, the classroom participants seem to rely on an implicit understanding of what language(s) is appropriate or not in this classroom, that is, on a ‘practiced’ language policy that is continuously being negotiated in interaction. We have also shown that an English monolingual medium is used and taken for granted in teacher-led talk but that two other mediums also carry this sense of normativeness, namely a Mandarin monolingual medium and an English and Mandarin bilingual medium. Whilst linguistic diversity was ignored in language policy texts and discourses (as discussed earlier), it seems to be recognised and legitimised by the ‘practiced’ language policy at play in these two workshops (see also Bonacina-Pugh 2020 on the role of the ‘practiced’ language policy in legitimising multilingual classroom talk). However, for that linguistic diversity to be accepted and therefore drawn upon in interaction, it needs to be shared and understood by all interactional participants. This is what Extract 9 below indicates. Here, students are interacting among themselves in a linguistically heterogeneous group (Group 1), where S1 and S5 speak Mandarin and English whilst S3 speaks Spanish and English. The three students are working together to design a lesson plan based on a picture book of their choice. The picture book they chose is entitled ‘turn left turn right’.

Extract 9:

  • S1 and S5: English and Mandarin bilingual speaking students

  • S3: Spanish and English bilingual speaking student

S1, S5 and S3 have been conducting the task successfully in English until in line 6, S1 switches to Mandarin (“中文向左走向右走”). This switch is heard by S3 who does not understand Mandarin and who immediately reacts by asking ‘what’ and laughing (in line 8). Laughter here serves as an indicator of repair (a self-initiated other repair) and points to the fact that the choice of Mandarin is deviant and therefore inappropriate. By laughing, S3 conducts a “medium repair” (e.g. Gafaranga 2011) and requests that English be used as the medium of interaction. In doing so, she points to the implicit norm (i.e. policy) she is orienting to when interpreting S1’s language choice act. It appears that, according to that norm, only a shared preferred language(s) is appropriate in interaction. In the following turn, in line 9, S1 acknowledges orientation to that norm, apologises, and explains what she was saying in Mandarin back in English (line 11). S1 simultaneously switches back to English, providing a second position medium repair (Gafaranga 2011) and establishing mutual understanding of the practiced language policy with S3. S3 then acknowledges receipt of this information with a change-of-state token (Heritage 1984) ‘ohhh’ and the conversation resumes in English. In sum, this extract shows that languages other than English are allowed if, and only if, they are shared and understood by all interactional participants. In turn, this means, as a result, that languages that are not shared by many participants, such as French (the teacher’s first language) or Spanish (the first language of two students), rarely find an interactional space in classroom talk. Consider this last extract below, where a Spanish and English bilingual student (S3) is missing a word in English and therefore turns to her other language Spanish.

Extract 10:

(Group 1)

  • S3: Spanish and English bilingual speaking student

  • S1 and S5: Mandarin and English bilingual speaking students

S3 is searching for a word in English, as evidenced by her statement “I’m not sure how it is in English” and the multiple self-interruptions and pauses. She offers a paraphrase of the word she is missing in lines 5 and 6. Her peers S1 and S5 seem to understand what S3 is trying to say and offer possible candidates (“in-between”, line eight). As none of these candidates seem to be what S3 is looking for, she turns to her other language and offers to give the Spanish equivalent of the word she is missing in English (line 13). However, her attempt to switch to Spanish is interrupted by S1, who requests further clarification on the missing word (line 14). The fact that S3 is interrupted as she is about to switch to Spanish indicates that her peers do not see the use of a language they do not understand as being helpful for meaning making. S3’s peers are orienting to the implicit norm of language choice mentioned above according to which only a shared language(s) is appropriate in interaction. This extract clearly shows how a language that is not shared by all interactional participants is not given space to be drawn upon.

Although it appeared at first that the multiple languages available in these two classrooms are used in interaction, only English and Mandarin are in fact used. Furthermore, they are used exclusively in interactional episodes where all classroom interactants understand them. Languages that are not shared and mutually understood are not allowed. In a sense, this indicates that the so-called ‘linguistic diversity’ in these two classrooms is acknowledged and drawn upon only in interactional episodes where there is no diversity; that is, when there is no linguistic heterogeneity among participants and that everyone understands the same language(s). In cases like Extracts 8 and 9 above, where there is a linguistically heterogeneous group and therefore an actual linguistic diversity, that diversity is not mobilised. This may further reveal that the classroom participants only value and allow languages when these languages serve a shared purpose of constructing meaning jointly. When languages are used by one person to make meaning, they are not given space in joint talk.

6 Discussion and concluding remarks

The internationalisation of higher education and language policy are two issues that are frequently conjointly examined in non-Anglophone countries, where HEI undergoing internationalisation adopt new language policies (usually EMI). Our study contributes to the rapidly expanding discussions of internationalisation of Higher Education and language policy by offering a case study of a lesser observed context, namely a HEI in an Anglophone country. We have taken the case of University X, a Russell group university in the UK, with a view to analysing the extent to which language policy was an integral part of internationalisation ‘at home’. To do so, we have investigated University X’s language policy texts, discourses and practices. Findings show that there was indeed a variety of language policies such as the British Sign Language plan and the Gaelic plan but that there was no overt language policy stating what the medium of instruction and assessment should be. Yet, University X’s ethno-linguistic landscape is highly diverse with numerous multilingual staff and students. Due to the noticeable absence of an overt language policy, we examined a set of internationalisation texts and applied a CDS lens to unravel discourses of language, culture and diversity. We then took the case of two workshop groups at postgraduate level. A CA lens was used to analyse a set of audio-recorded classroom talk and understand the ‘practiced’ language policy classroom participants orient to. Given that University X has not formalised an explicit language policy, staff and students have come up with their own rules and negotiations of diversity in practice. That is, they have negotiated a ‘practiced’ language policy that they use as a common reference point to know what language(s) is/are appropriate or not.

Our CDS analysis has borne out a dominant promotional rhetoric about diversity and culture as forming part of a global education in the internationalization discourse. At the same time, the linguistic diversity of international staff and students (as well as home students and local staff) is ignored. The University communicates a clear message without articulating it discursively: it excludes any language matter from its internationalisation agenda and global vision, with an underlying ideology of English as the ‘normal’ and sufficient academic lingua franca and the language of internationalisation. This exclusion of language matters is quite surprising in view of the broader celebratory discourses currently circulating on linguistic diversity and the marketised and commodified view of languages as key skills for employment in future life. Similarly, our CA analysis of classroom talk has revealed that English is also, in practice, the normative choice of medium of interaction between the teacher and her students. It also became clear that among students, only shared languages were seen as acceptable. In the linguistically homogeneous group, students adopted either a Mandarin monolingual medium or an English and Mandarin bilingual medium but no other varieties of Chinese were heard. In the linguistically heterogeneous group, students adopted an English monolingual medium and no interactional space was given to students’ other languages that were perhaps not shared among many other students, like Spanish.

In sum, although it may appear that the classroom participants were acknowledging linguistic diversity, it became clear that their ‘practiced’ language policy only legitimises those languages that a majority of the classroom participants understands. In this sense, the full linguistic diversity of the students and the teacher was not drawn upon and languages that were not shared by all were silenced or altogether absent. Mandarin was therefore the only language other than English that was seen as legitimate and normative.

Interestingly then, it appears that language policy texts, discourses and practices are here clearly interconnected. The assumed ‘naturalness’ or ‘normativeness’ of English is visible in texts, discourses and practices and so is the silencing of languages that are not seen as being shared by all. The ‘practiced’ language policy observed in the two classrooms further revealed that the use of Mandarin was another language that was also seen as normative because it was shared among students in small group talk. As such, our CDS and CA analyses both indicate that linguistic diversity is not valued in language policy texts, discourses and practices. Only shared language(s) (namely, English at the level of texts and discourses; and English and Mandarin at the level of practices) are seen as legitimate.

Methodologically, this research presents a first and significant attempt at articulating findings from a CDS and CA lens in language policy research. Further work is needed to fully comprehend the extent to which CDS and CA can be articulated in language policy research. This paper further sheds light onto the multi-level nature of language policy and the dialogic relationship between texts, discourses and practices. The door has been opened on the language policy challenges that internationalised universities are facing in Anglophone contexts. Language policy issues should be at the heart of internationalisation processes of HEIs in Anglophone countries (as it is in non-Anglophone countries), in order to challenge a pervasive English monolingual ethos as well as the assumption that only shared language(s) can be used.

Corresponding author: Florence Bonacina-Pugh, The School of Education, The University of Edinburgh, Holyrood Road, EH8 8AQ, Edinburgh, UK, E-mail:

Transcription conventions


Altbach, Philip G., Liz Reisberg & Laura E. Rumbley. 2009. Trends in global higher education: tracking an academic revolution. Executive summary. A report prepared for the UNESCO 2009 world conference on higher education: United Nations Educational, Scientific and Cultural Organisation.10.1163/9789004406155Search in Google Scholar

Ball, Stephen J. 1993. What is policy? Texts, trajectories and toolboxes. Discourse 13(2). 10–17. in Google Scholar

Barakos, Elisabeth & Charlotte Selleck. 2019. Elite multilingualism. Discourses, practices, and debate. Journal of Multilingual and Multicultural Development 40/5. 367–374.10.1080/01434632.2018.1543691Search in Google Scholar

Barakos, Elisabeth. 2016. Language policy and critical discourse studies: Towards an integrated approach. In Elisabeth Barakos & Johann W. Unger (eds.), Discursive approaches to language policy, 23–49. London: Palgrave Macmillan.10.1057/978-1-137-53134-6_2Search in Google Scholar

Beelen, Jos & Elspeth Jones. 2015. Redefining internationalization at home. In Adrian Curaj, Liviu Matei, Remus Pricopie, Jamil Salmi & Peter Scott (eds.), The European higher education area. Dordrecht: Springer.10.1007/978-3-319-20877-0_5Search in Google Scholar

Bonacina, Florence & Joseph Gafaranga. 2011. ‘Medium of instruction’ versus ‘medium of classroom interaction’: Language choice in a French complementary school classroom in Scotland. The International Journal of Bilingual Education and Bilingualism 14(3). 319–334.10.1080/13670050.2010.502222Search in Google Scholar

Bonacina-Pugh, Florence. 2012. Researching ‘practiced language policies’: Insights from conversation analysis. Language Policy 11(3). 213–234.10.1007/s10993-012-9243-xSearch in Google Scholar

Bonacina-Pugh, Florence. 2020. “Legitimizing multilingual classroom practices: The role of the ‘practiced language policy’”. International Journal of Bilingual Education and Bilingualism 23(4). 434–448.10.1080/13670050.2017.1372359Search in Google Scholar

Billig Michael & Emanuel A. Schegloff. 1999. Critical discourse analysis and conversation analysis: An exchange between Michael Billig and Emmanuel A. Schegloff. Discourse and Society 10. 543. in Google Scholar

Block, David. 2006. Multilingual identities in a global city: London stories. London: Palgrave Macmillan.10.1057/9780230501393Search in Google Scholar

Kingsley Bolton, Werner Botha & John Bacon-Shone. 2017. English-medium instruction in Singapore higher education: policy, realities and challenges. Journal of Multilingual and Multicultural Development 38(10). 913–930. in Google Scholar

Chríost, Diarmait Mac Giolla & Matteo Bonotti. 2018. Brexit, language policy and linguistic diversity. London: Palgrave.Search in Google Scholar

Egbert, Maria M. 1997. Schisming: The collaborative transformation from a single conversation to multiple conversations. Research on Language and Social Interaction March 2014. 37–41. in Google Scholar

Fairclough, Norman. 1993. Critical discourse analysis and the marketization of public discourse: The universities. Discourse and Society 4(2). 133–168.10.1177/0957926593004002002Search in Google Scholar

Fairclough, Norman. 2010. Critical discourse analysis. The critical study of language, 2nd edn. Harlow: Longman.Search in Google Scholar

Gafaranga, Joseph. 2007. Talk in two languages. London: Palgrave Macmillan.10.1057/9780230593282Search in Google Scholar

Gafaranga, Joseph. 2011. Transition space medium repair: Language shift talked into being. Journal of Pragmatics 43(1). 118–135. in Google Scholar

Haberland, Hartmut, Janus Mortensen, Anne Fabricius, Bent Preisler, Karen Risager & Susanne Kjærbeck (eds.). 2008. Higher education in the global village. Roskilde: University of Roskilde.Search in Google Scholar

Hart, Christopher & Piotr Cap (eds.). 2014. Contemporary critical discourse studies. Bloomsbury: Academic.Search in Google Scholar

Heritage, John. 1984. Garfinkel and ethnomethodology. Cambridge: Polity Press.Search in Google Scholar

Hultgren, Anna Kristina, Frans Gregersen & Jacob Thøgersen. 2014. English in Nordic Universities: Ideologies and practices. Amsterdam: Benjamins.10.1075/wlp.5Search in Google Scholar

Knight, Jane. 2004. Internationalization remodeled: Definition, approaches, and rationales. Journal of Studies in International Education 8(1). 5–31. in Google Scholar

Knight, Jane. 2006. Internationalization of higher education: New directions, new challenges: IAU.Search in Google Scholar

Li, Wei & Hua Zhu. 2013. Translanguaging identities and ideologies: Creating transnational space through flexible multilingual practices amongst Chinese university students in the UK. Applied Linguistics 34/5. 516–535. in Google Scholar

Liddicoat, Anthony J. 2016. Language planning in universities: Teaching, research and administration. Current Issues in Language Planning 17(3–4). 231–241. in Google Scholar

Macaro, Ernesto, Samantha Curle, Jack Pun, Jiangshan An & Julie Dearden. 2018. A systematic review of English medium instruction in higher education. Language Teaching 511. 36–76. in Google Scholar

Marshall, Steve. 2009. Re-becoming ESL: Multilingual university students and a deficit identity. Language and Education 24(1). 41–56. in Google Scholar

Martin, Peter. 2010. ‘They have lost their identity but not gained a British one’: Non-traditional multilingual students in higher education in the United Kingdom. Language and Education 24. 9–20. in Google Scholar

Mazak, Catherine M. & Kevin S. Carroll (eds.). 2017. Translanguaging in higher education: Beyond monolingual ideologies. Bristol: Multilingual Matters.10.21832/9781783096657Search in Google Scholar

Moore, Emilee. 2016. Conceptualising multilingual higher education in policies, pedagogical designs and classroom practices. Language, Culture and Curriculum 29(1). 22–39. in Google Scholar

Morrish, Liz & Helen Sauntson. 2013. ‘Business-facing motors for economic development’: An appraisal analysis of visions and values in the marketised UK university. Critical Discourse Studies 10(1). 61–80. in Google Scholar

Paltridge, Brain. 2012. Discourse analysis. An introduction. London: Continuum.10.5040/9781350934290Search in Google Scholar

Piller, Ingrid & Jinhyun Cho. 2013. Neoliberalism as language policy. Language in Society 42(1). 23–44. in Google Scholar

Preece, Sian. 2018. Identity work in the academic writing classroom: Where gender meets social class. Journal of English for Academic Purposes 32. 9–20. in Google Scholar

Preece, Sian. 2011. Universities in the Anglophone centre: Sites of multilingualism. Applied Linguistics Review 2. 121–146. Retrieved 13 Nov. 2018, from in Google Scholar

Preece, Siân & Peter Martin. 2009. Imagining higher education as a multilingual space. Language and Education 24(1). 3–8. in Google Scholar

Robertson, Susan L. & Matt Kedzierski. 2016. On the move: Globalising higher education in Europe and beyond. The Language Learning Journal 44(3). 276–291. in Google Scholar

Saarinen, Taina. 2012. Internationalization of Finnish higher education—Is language an issue?. International Journal of the Sociology of Language 2012(216). 157–173. in Google Scholar

Shohamy, Elana Goldberg. 2006. Language policy. Hidden agendas and new approaches. New York: Routledge.10.4324/9780203387962Search in Google Scholar

Soler, Josep & Lídia Gallego-Balsà. 2019. The sociolinguistics of higher education. Language policy and internationalisation in Catalonia. London: Palgrave.10.1007/978-3-030-16677-9Search in Google Scholar

Spolsky, Bernard. 2004. Language policy. Cambridge: Cambridge University Press.Search in Google Scholar

Spolsky, Bernard & Elana Shohamy. 2000. Language practice, language ideology, and language policy. In Richard D. Lambert & Elana Shohamy (eds.), Language policy and pedagogy: Essays in honour of A. Ronald Walton. Amsterdam: John Benjamins Publishing.10.1075/z.96.02spoSearch in Google Scholar

UK Council for international Student affairs. (accessed 18 November 2018).Search in Google Scholar

UNESCO Institute for Statistics. 2016. Global flow of tertiary level students. (accessed 18 December 2018).Search in Google Scholar

UNESCO International. 2017. Institute for educational planning, and the global education monitoring report. ‘Policy Paper 30: Six ways to ensure higher education leaves no one behind’. (accessed 18 December 2018).Search in Google Scholar

Universities UK. Higher education in numbers. (accessed 1 November 2018).Search in Google Scholar

Urciuoli, Bonnie. 2003. Excellence, leadership, skills, diversity: Marketing liberal arts education. Language & Communication 23(2003). 385–408. in Google Scholar

Wetherell, Margaret. 1998. Positioning and interpretive repertoires: Conversation analysis and post-structuralism in dialogue. Discourse & Society 9(3). 387–412. in Google Scholar

Published Online: 2022-11-04
Published in Print: 2022-11-25

© 2020 Florence Bonacina-Pugh et al., published by De Gruyter, Berlin/Boston

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 28.5.2023 from
Scroll to top button