Abstract
At a time of increasing internationalization in tertiary education, English-Medium Education in Multilingual University Settings (EMEMUS) has become a common practice. While there is already ample research describing this phenomenon at a local level (Smit and Dafouz 2012a), the theoretical side needs to be elaborated. This article thus aims to develop a conceptual framework that considers the dynamic nature of EMEMUS. Drawing on recent sociolinguistic orientations and discursive approaches (e.g. Scollon and Scollon 2004; Shohamy 2006; Blommaert 2010; Hult 2010), our framework regards EMEMUS as a social phenomenon and views discourse as the access point to six relevant dimensions. These dimensions are considered as inherently complex, contextually bound, and intersecting dynamically with one another. Focusing on an example from a higher education institution, the article argues for the utility of the proposed framework.
INTRODUCTION: ARGUING FOR THE RELEVANCE OF A CONCEPTUAL FRAMEWORK
While the inherently global nature of universities has been recognized throughout history ( de Ridder-Symoens 1996 ), the last 20 years have seen an unprecedented surge in the number of students (both national and international) attending institutions of higher education. Although complex and varied, the main reasons for this are broadly linked to the rising forces of globalization and internationalization, 1 migration, and the emergence of a knowledge economy and information society ( Knight 2008 ; Cancino et al. 2011 ). As a result, the student population of universities in the 21st century is greater and more heterogeneous regarding its academic, linguistic, and cultural background than ever before ( Doiz et al. 2013 ; van der Walt 2013 ). In Europe, the establishment of a European Higher Education Area (EHEA) has greatly strengthened this view of universities as global institutions, whose main aims include student and staff mobility, curricular harmonization, and international research collaboration.
As a result, an increasing number of universities are adopting the use of an additional language (very often English) as a means of instruction. Though this has educational and linguistic consequences, such consequences have been generally neglected by higher education institutions and research. A clear indication of the secondary importance ascribed to language issues is the fact that explicit institutional language policies are usually only invoked when language ‘problems’ arise ( van der Walt 2013 : 13). These problems generally derive from the decreasing use of home languages (e.g. Denmark, Finland, Sweden, or Norway) or from a need for languages to be equally distributed in multilingual contexts (e.g. Luxembourg, Bolzano, or Catalonia). For the most part, however, neither universities nor researchers seem to be immediately concerned with focusing explicitly on the educational and linguistic outcomes of such internationalization strategies. In many cases, ‘programmes [are] being introduced with scant underpinning of research findings into the relationships between language and content, either generally or specifically’ ( Wilkinson and Zegers 2007 : 12).
To counteract this ‘invisibility’ ( Saarinen 2012 ), a number of events have taken place since 2000. For instance, between 2003 and 2010, Fribourg, Helsinki, Bolzano, and Luxembourg hosted conferences on multilingual institutions, while two congresses ( Wilkinson 2004 ; Wilkinson and Zegers 2007 ) in Maastricht served as a springboard for the founding of an association under the acronym of ICLHE (Integrating Content and Language in Higher Education). 2 A decade later, an overview of the research publications stemming from these ICLHE conferences reveals a variety of research foci that span a wide range of topics, from studies on (foreign) language issues, genres or educational theories, to research drawing on new literacy and cultural studies or language management policies ( Jacobs 2013 ). While these works, on the whole, reflect a growing concern for applied linguistic research, they also display a lack of consensus in their theoretical orientations. In other words, they rest on major constructs (language, content, and the integration of these) that are not sufficiently developed. As a result of these highly fragmented perspectives, there is a pressing need for a conceptualization that encompasses the diversity and complexity of the specific settings analysed ( Smit and Dafouz 2012a ) as well as the range of contextual needs and the variety of policy aims. The specific objective of our article is therefore to provide a theoretically grounded framework based on core dimensions that operate dynamically across higher education institutions that use an additional language as the means of instruction. The framework also aims to combine various perspectives and serve as a reference when analysing specific contexts while still acknowledging the importance and constant interaction of global and local forces. Most importantly, it will permit researchers to go beyond specific cases and engage in analyses that contrast and draw conclusions across different settings.
To embrace the multiplicity of higher education sites and the different labels used to depict the variety of present-day multilingual tertiary education (e.g. Smit and Dafouz 2012b : 3–4), our work proposes the use of an alternative notion: English-Medium Education in Multilingual University Settings (EMEMUS). This label is semantically wider, as it does not specify any particular pedagogical approach or research agenda. Our conceptualization of multilingual universities departs from the narrow view of balanced multilingual speakers, official multilingual programmes and explicitly multilingual pedagogical methods. Instead, it encourages an organic understanding of the university setting as ‘a multilingual situation where students, using the languages they know and those they are getting to know, are enabled to succeed’ ( van der Walt 2013 : 12). Multilingual universities are seen as sites where bilingual or multilingual education, whether official or unofficial, partial or comprehensive, pedagogically explicit or implicit, may be represented. Furthermore, EMEMUS underlines the growing multilingual nature of higher education worldwide ( García 2009 ; Doiz et al. 2012 ; Fortanet-Gómez 2013 ). At the same time, it focuses on English-medium education because of the particular role that English plays both as an academic language of teaching and learning as well as a means of international communication across professions and trade.
In the following, we will discuss the theoretical orientations that inform our conceptual framework and the dimensions that we argue are relevant for EMEMUS. After justifying the centrality of these specific dimensions, we will provide a brief account of each, and finally present an example from a higher education institution in Spain to show that the proposed framework provides a basis for theoretically sound analysis.
THEORETICAL ANCHORING
Our theoretical argumentation starts with the recent conceptual developments in sociolinguistics in relation to present-day societal processes. These considerations on language and society are complemented by equally dynamic, but more established, ecolinguistic ideas, which, in turn, have informed multilingual language policy (LP) research. The theoretical overview further argues for using discourses as the point of access.
Sociolinguistics in the 21st century
While the focus of interest in sociolinguistics is still largely on ‘the role of language in the construction of social relations and social organization’ ( Heller 2008 : 504), its central tenets have changed remarkably ( Blommaert 2010 ; Pennycook 2010 ). Twenty-first century societies are increasingly permeable, changeable, and in flux, constructing themselves by language and other semiotic means in fittingly transient, dynamic, and fluid ways. In an attempt to come to grips with these realities, scholarship in this field has undergone some major ideological shifts. By leaving behind well-established, largely static, and bounded concepts (such as ‘the language of nation X’ or the ‘speech community’), new conceptualizations centre on the ‘transnational flows’ ( Pennycook 2007 ) and ‘super-diversity’ of present-day migration and their influence on communication and language ( Blommaert and Rampton 2012 ). Although these latest developments are too recent to have reached a level of theoretical maturity, the readiness with which a new ‘sociolinguistics of globalisation’ ( Blommaert 2010 ) has been taken up pays tribute to how relevant such a re-conceptualization is perceived to be ( Blommaert et al. 2012 ).
This is also true of our undertaking. While we do not wish to claim that universities showcase societies, they can be viewed as prime sites for super-diversity in terms of the increasing mobility of staff and students, and their varying constellations of individual and group-specific linguistic repertoires. Within the power hierarchies of academia, the use of such complex and multilingual communicative resources by differing groups of social actors can lead to tensions or even conflict ( Gnutzmann 2008 ). At a conceptual level, it is thus important to foreground the dynamic complexity inherent in the main ‘ingredients’, that is, social agents and language, recognizing that neither comes in clearly measurable or bounded units. When dealing with ‘English-medium education’, for instance, the conceptualization of English needs to go beyond the standardized, abstracted norm that the label seems to reduce it to. Thus, apart from the supposedly fixed language code that functions as a tool for academic publishing, and for teaching and learning disciplinary knowledge, English-medium education encapsulates discursive and other social practices. Typically, such practices involve multilingual teachers and students engaging in knowledge construction by drawing on their diverse language resources (e.g. Creese and Blackledge 2010 ).
In sum, postmodern sociolinguistics offers important considerations for EMEMUS in our globalized age as regards the fluidity, complexity, and multifunctionality of English and other languages that academics, students, and administrators draw on for their institutionalized practices. To outline the contours of the relevant dimensions for EMEMUS, we cast our theoretical net wider, integrating basic conceptualizations of ecolinguistics and of LP.
The ecology of language and LP research
Following Haugen’s (1972) initial suggestion to study language as an integral part of its environment, the ecology metaphor has been developed further to aid in ‘explorations of the relationship of languages to each other and the society in which these languages exist’ ( Creese and Martin 2003 : 161). Such an integrated view acknowledges the interdependence of the language varieties used by a particular social group, while also recognizing the need to sustain diversity in a linguistic ecosystem ( Fill and Mühlhäusler 2001 : 3). If, as in our case, the metaphorical ecology includes multilingual university settings, an ecolinguistic perspective needs to recognize the dynamic interrelatedness between the relevant languages (functions and forms) and their academic habitats, not least because these contain, and are constructed by, academics, students, and administrators in their actual and virtual university spaces. Besides English and other languages, important ecolinguistic factors for multilingual university settings are thus the agents themselves, as well as communicative practices and academic cultures in their global and local realizations.
In line with its activist concerns, ecolinguistic considerations have been taken up in discussions of LP options. For instance, Phillipson and Skutnabb-Kangas (1996) argue for the ‘ecology-of-language paradigm’ as recognizing people’s language rights and promoting language diversity in support of language maintenance. This is an interventionist call that reverberates in Hornberger’s (2003) discussion of micro-level instances of teachers and learners ‘counteracting language endangerment’ (p. 296) by drawing on their multilingual practices in the classroom. In addition, the ecology-of-language approach resonates well with LP theorizing, where the basic premises of the ecolinguistic paradigm find their reflection in the recent ‘ideological underpinnings’ ( Hornberger 2002 : 35) of LP research. Rather than viewing languages as distinct and bounded entities, language ecologists perceive human communication and the language varieties it draws on as a resource that members of a society use in situated dynamic ways ‘within the[ir] vast cultural, educational, historical, demographic, political, social structure’ ( Kaplan and Baldauf 1997 : 13).
Furthermore, such a view holds implications for the notion of ‘language policy’ itself. In contrast to its widespread, but arguably reductionist, interpretation of primarily referring to policy declarations or statements, the dynamic and evolving nature of a language ecology presupposes a wider, multilayered, and socially aware modelling of LP. For instance, when dealing with the LP of a particular higher educational institution, it is paramount to also consider the actual language practices that teachers and students are engaging in as well as the potentially different and conflicting communicative and academic aims agents might be pursuing.
Spolsky (2004) offers a valuable clarification in this context in the form of a tripartite model. From the vantage point of the social agents of an ecology, LP is modelled as encompassing the three components of ‘language management’, ‘language practices’, and ‘language beliefs’ ( Spolsky 2004 : 5–14). Language management deals with the ‘direct efforts [taken] to manipulate the language situation’ ( Spolsky 2004 : 8), that is, with LP statements and documents. Language practices, on the other hand, focus on an individual’s sociolinguistic choices and how they reflect ‘habitual pattern[s] of selecting among the varieties that make up [a community’s] linguistic repertoire’ (ibid.). Finally, language beliefs refer to the social players’ ideologies as regards to their language practices. Furthermore, each component is multilayered and complex in itself in relation to the various political levels, from supra-national to institutional as well as other linguistic and extralinguistic environmental factors.
This ‘expanded’ model foregrounds the contested nature of the societal mechanisms behind ‘organizing, managing and manipulating language behaviours’ ( Shohamy 2006 : 45). Since different societal groups pursue different interests, it is the norm rather than the exception that language practices, managerial decisions, and language beliefs stand in partial contradiction or even conflict (see also Hornberger 2002 ; Tollefson 2006 ). As a result, there is a need for considering LP concerns in EMEMUS according to three separable dimensions: language management, practices, and the agents’ beliefs. In extension to the expanded model, we further identify LP more generally as being centrally constituted and co-constructed in and through discourses. Hence, the conceptual framework for researching EMEMUS integrates the expanded model from the point of view of social practices as discourses.
Social practices as discourses
The notion of discourse has been approached from a wide variety of perspectives that range from the more classic and linguistically oriented definition of ‘discourse as the study of language in context’ ( McCarthy et al. 2002 : 57) to a broader societal understanding, whereby discourse(s) are regarded as ‘meaning systems which have historical, social and institutional implications’ ( Foucault 2002 : 131).
In line with this stance, our work views discourse as ‘a locus of co-construction’ ( Hüttner et al. 2013 : 4) and not merely as a way to make the situationally relevant factors visible and explicit through social practices. In other words, we see this relation as a reciprocal process. Discourses are viewed as a form of social action, or ‘discourse as mediated action’ taking place in an interactional context or ‘site of engagement’ ( Scollon 1998 ). We also believe that many social practices (e.g. LP documents, teacher interviews, or student exams) are shaped and built through discourses.
From this perspective, discourse suggests itself as a viable access point to the analysis of social practices since ‘discourse-oriented work (…) is well-suited to understanding relationships between language policies and the social actions of individuals’ ( Hult 2010 : 8). In this line, Hult’s (2010) analysis of LP discourses and Halonen et al. 's (2014) examination of multisited LP seem to be especially relevant. Both perspectives take discourse as the object of analysis while also acknowledging the dynamic interplay of factors. Consequently, these approaches strive to go beyond well-established sociolinguistic categorizations, such as time and space, community members, or macro vs. micro layers. 3
The latter distinction in particular ‘seems to imply a certain hierarchy in which macro-level phenomena somehow take place on a different plane of existence from micro-level phenomena’ ( Hult 2010 : 18). On the contrary, social phenomena are constantly interrelated and layers are seen as contingencies of the methodological take pursued. With the help of the microscope analogy, Hult (2010 : 14) explicates that we may focus on our object of study and its features depending on how we set the lens. This means that the kinds of details observed depend on the power of magnification (i.e. on the degree of zooming in or out). What is therefore being examined can no longer be described as layers. Instead, Hult draws on Blommaert's (2007) construct of scale to refer ‘to the fluid and dynamic nature of relationships among discourse processes across dimensions of social organization’ ( Hult 2010 : 14). A similar idea is encompassed in the construct of multisitedness, which, in turn, conceptualizes social practices as ‘taking place, being constructed, contested and reproduced on different horizontally and vertically linked levels simultaneously, and in different times and places’ ( Halonen et al. 2014 ).
Summing up the theoretical foundations, we identify three main postulates for developing an EMEMUS framework. First, the core dimensions constituting the model are developed from the preceding sociolinguistic, ecolinguistic, and LP considerations (see the next section for a description). Secondly, we adopt the view of discourse as point of access to analysing EMEMUS; thirdly, we acknowledge the general interplay of the dimensions.
THE FRAMEWORK
The framework that we are proposing identifies six relevant components, namely, Roles of English (in relation to other languages), Academic Disciplines , (language) Management, Agents, Practices and Processes , and Internationalization and Glocalization . For ease of reference and memory, the model will be referred to as ROAD-MAPPING.
The dimension of Roles of English (in relation to other languages) and its intersection with Agents , Processes and Practices and Academic Disciplines springs directly from the aforementioned ecolinguistic considerations, while the importance of Internationalization and Glocalization for sociolinguistic phenomena and their impact on Agents and Processes and Practices is put into relief by the recent developments in sociolinguistics. Expanded LP, on the other hand, underlines the relevance and interplay of the components of Management , Agents ’ ideology and, again, Practices and Processes . Finally, Discourses , as explained in detail before, is seen as the intersecting access point through which all six dimensions can be examined. Figure 1 displays visually these components and their constant interplay.
The ROAD-MAPPING framework for EMEMUS
The ROAD-MAPPING framework for EMEMUS
As Figure 1 shows, Discourse(s) are placed at the centre, functioning as the point of access to the different components. Having already explained the fundamentally discursive nature of EMEMUS, the following brief descriptions will focus on the six dimensions. For the sake of clarity, we will deal with each dimension separately following the sequence given in the resulting acronym, ROAD-MAPPING. In this approach, we follow an applied linguistic perspective, while still acknowledging that each of these components is multifaceted and potentially interdisciplinary.
Roles of English
Reflecting our ecological perspective, the functional breadth of English must be considered in relation to the complete linguistic repertoire of a specific higher education site. Generally, expanded language policies of universities range across a broad dimension of mono- vs. multilingualism with regards to language management, language practices, and educational aims ( van der Walt 2013 : 76–8). Whatever the LP constellation may be, however, English can be expected to hold a central position in higher education language planning, reflected in suggestions to focus on English vis-à-vis other languages, be they foreign, national, regional, minority, or migrant languages. The resulting ‘English-plus’ multilingual policies pay tribute to the presently unrivalled position of English as the main language of dissemination of scientific ideas, and an increasingly relevant language of education ( Jenkins 2014 ).
The latter role is particularly relevant in this context and encompasses a range of functions, each of which deserves attention for its own sake as well as in combination with each other. As proof of English proficiency is one of the entry requirements for many English-medium study programmes worldwide, it is used as a gatekeeper regulating (non-native) student intake ( Shohamy 2013 ). In some cases, this function has also been extended to staff, stipulating a certain proficiency level as a necessity for teaching in English-medium programmes. At the exit level, English can function as an outcomes criterion. This may take the form of one of the educational aims linked to future professional language requirements, or be linked to English for Specific Purposes classes.
With regard to coursework itself, English is used in a range of ways, as a subject in English for Academic Purposes classes, as means of teaching and learning, or in English-Medium Instruction (EMI) ( Hellekjaer 2010 ). Additionally, English can be drawn on in relation to some or all communicative skills. In some classes, for instance, English will be used only for reading scientific publications, while in others it will also be used when producing written texts or interacting.
While English remains mainly an academic language amongst teachers and students sharing another language, the steady increase in staff and student mobility also positions it as the only shared language or lingua franca of many higher educational settings, thereby adding the role of joint language of communication to the well-established academic functions of English ( Smit 2010 ; Hynninen 2012 ). In view of this multifunctionality, it can be expected that English language practices form a contested terrain in need of socially aware and critical analysis ( Tollefson 2006 ; Jenkins 2014 ).
Academic Disciplines
One of the most demanding aspects of university education is understanding and complying with the social practices that rule it. Within those social practices, acquiring academic literacy stands as a first stage towards academic acculturation, and academic disciplinary practices gradually follow. These two stages are tightly connected to their contextual settings and are subject to an ideological stance that can neither be isolated nor treated as neutral ( Street 2011 ). Social theories of learning ( Gee 1990 ; Lave and Wenger 1991 ) suggest that as academic literacy knowledge usually operates at an unconscious level, the rather tacit and implicit principles that govern it are acquired through socialization into academic communities of practice ( Wenger 1998 ).
A number of different classifications have been widely used in an attempt to make these explicit and to show how, for instance, teaching and learning practices, curricular design or assessment methods may vary depending on the inherent epistemological characteristics of each academic discipline or sets of disciplines ( Becher 1989 ; Neumann et al. 2002 ). Becher’s (1989) widely used proposal offers a two-dimensional space for classifying disciplines according to a hard vs. soft and pure vs. applied continuum. This categorization is contingent with different teaching methods and assessment. Within business studies, for instance, a subject such as Economic History (which would be classified as soft/pure ) is believed to favour essay-based assessment formats rather than multiple choice exams or problem-solving tests ( Dafouz et al. 2014 ).
A more discursive perspective addresses academic literacy using a genre-based approach. Here, genre is seen as a means to ‘map’ the types of discourse and specialized language that are used to communicate knowledge in different disciplinary areas ( O’Halloran 2004 ; Coffin 2006 ). More specifically in EMEMUS, courses in English for specific or academic purposes pay close attention to academic disciplinary practices and their resulting subject literacies ( Swales 1990 ; Bhatia 2004 ).
Given the contingency of disciplinary knowledge and academic literacies, multilingual settings pose two immediate challenges. First, we need to recognize the different discourses operating in different disciplines and to encourage their constructive integration (see Practices and Processes below). Secondly, it is vital to remain critically aware of the risk of homogenizing disciplines and following an (Anglocentric) monocultural model potentially triggered by the use of English as the language of instruction ( Smit and Dafouz 2012b : 8–9).
(Language) Management
As explained above, language management is concerned with LP statements and declarations. These ‘direct efforts to manipulate the language situation’ ( Spolsky 2004 : 8) come in different channels, that is, written, spoken, or Internet based, and are issued by social agents representing collectives at various socio-political and hierarchical levels (see next section). Reflecting their authors, language management statements vary in terms of range of application and legal status. For instance, a language management measure can be binding, but applicable to a specific classroom only, while others, such as European White Papers (e.g. European Commission 1995 ), are recommendations designed for a supranational region like the EU.
Additionally, managerial decisions might turn out to be largely ignored or replaced by what relevant agents believe to be appropriate, thus creating ‘envisaged’ or ‘de facto’ policy statements ( Shohamy 2006 ). Furthermore, regulations can contradict each other, as, for example, in the case of the University of Kebangsaan (Malaysia), which decided to promote the use of Bahasa Melayu for education in its strategic plan 2000–2020 ( Gill 2004 ). This decision was in line with the original mission of the university, but stood in contrast to the novel and surprizing LP decision taken at that time by the Malaysian minister of education to teach science and mathematics in English at all levels of education.
Finally, language management also needs to consider a factor that is conspicuous by its absence, namely, a lack of explicit regulations in higher education institutions ( Fortanet-Gómez 2013 : 78–9). Yet even in multilingual university settings that do come with explicit statements, certain aspects of LP remain unmentioned, which creates sociolinguistic relevance in itself. A case in point is the internationalization development plan for Finnish higher education, which, in addition to the official languages (Finnish and Swedish), argues for the use of ‘other’ languages without explicit reference to English. By thus keeping English ‘invisible’ ( Saarinen 2012 ), it could be argued that its unique status is camouflaged behind a call for multilingualism.
Agents
A significant number and variety of agents take part in the planning, implementation, and assessment of language policies in higher education institutions worldwide. Research on the articulation of the notion of ‘agency’ in LP reveals that it comes in different and multiple shapes ( Saarinen and Ursin 2012 ). Structural approaches, for example, acknowledge the existence of actors (e.g. the EHEA, the University of London), but regard these as secondary. Actor approaches, on the other hand, move along a scale that ranges from individual actors (e.g. teachers, students, and administrative staff) to collective or institutional actors (e.g. faculties, rectorates, and student unions). Agency approaches ( Marginson and Rhoades 2002 ) highlight the interactive nature of policy change as well as the simultaneous significance of global, national, and local forces, while Actor-network theory (e.g. Latour 2005 ) proposes that ‘no organization or no agent is ever complete or autonomous’ ( Saarinen and Ursin 2012 : 151).
Irrespective of the specific approach adopted in the conceptualization of agency, it is essential to acknowledge the multifaceted and dynamic nature of the roles that stakeholders engaged in EMEMUS may adopt. Furthermore, it is important to examine how these roles and the individuals and/or institutions that implement them may also display different hierarchical status and, consequently, pose distinct views and interests that sometimes conflict. 4 A clear example of this is observed in the different agendas that content teachers (disciplinary experts) and language specialists put on the table when using English as a means of instruction for subject learning. Thus, while content lecturers ‘have initially engaged in this new scenario by embracing a top-down internationalisation plan alongside a chance for professional and academic development’ ( Smit and Dafouz 2012b : 8), for language teachers, pedagogical concerns about language learning are, inevitably, of maximum importance. Reconciling these conflicting interests is no easy matter (but see Winberg et al. 2013 ).
Practices and Processes
In keeping with Leung and Street’s ( 2012 : 9) description of social practices as a ‘cultural conception of particular ways of thinking about and doing’, this component focuses on the teaching and learning activities that construct and are constructed by specific EMEMUS realities. It thus takes a process rather than a product view. From this vantage point, it is possible to analyse praxis and developments that might otherwise go undetected. To illustrate the range of relevant practices, we will briefly sketch three types, the first of which relates to ‘ways of doing’, the second to ‘ways of thinking’, and the third combining both.
Within a social constructivist understanding of learning, the discursive practices in the classroom function as an analytical window onto the joint development and construction of knowledge ( Mercer 2000 ; Vygotsky 2012 ). In the case of multilingual groups, these practices reveal the localized process of developing a shared repertoire appropriate to the academic and social communicative purposes which arise from drawing on English in its roles as a disciplinary and educational language and also as a lingua franca ( Hynninen 2012 ). The second type of practices focuses on the views and beliefs teachers have regarding the learning process and how their teaching can best support students (e.g. Doiz et al. 2012 ). Such views influence the actual practices themselves and how policies are implemented in the classroom ( Borg 2011 ). Finally, we turn to the development of academic literacy skills, which is a contested area of teaching between language specialists and subject specialists. While language development has traditionally been the terrain of language specialists, the New Literacy School (e.g. Gee 1990 ; Street 1999 ) has argued that disciplinary specialists are best placed to induct students into the discourses of their disciplines. In view of the complementary strengths of both groups, Jacobs (2007) proposes a model for the process of integrating academic literacies into disciplines based on collaborative partnerships between language experts and content specialists, thus developing joint ways of ‘doing’ and ‘thinking about’ literacy development.
Internationalization and Glocalization
Although often used interchangeably, internationalization and globalization address different issues. Drawing on Scott (2011 : 60–1), internationalization
describes a process of intensifying exchange between nations (or other securely internationalized organizations and agencies), most of which occurs within the public domain (…), [while] globalisation describes the progressive integration of economic structures within global (but also volatile) arrangements and the homogenization (but also hybridization) of distinctive national cultures, both of which occur largely in the private domain.
Against this backdrop, higher education institutions display ‘the tensions but also the synergies’ ( Scott 2011 : 61) between their multifaceted roles in society. As such, universities need to address international, global, national, and local forces and interests to succeed. For instance, while the global or horizontal mobility process provides access to international students, the local or vertical mobility process aims to widen participation of minoritized communities ( van der Walt 2013 : 38–44). This ‘glocalization’ process ( Robertson 1995 ) compels universities to keep a delicate balance between the two conflicting but also complementing forces that interact and struggle for ground.
In EMEMUS, the complex interplay of local and global drives includes a number of diverse questions and decisions which should go beyond the most noticeable (and sometimes only) criterion applied with regards to internationalization, namely, student and teacher mobility. Higher education institutions need to design systematic and institutionally supported implementation measures and guidelines that will deal with increasingly multicultural and multilingual university scenarios. This would allow for international research collaboration and professional networks and, most importantly, would assist stakeholders in taking informed decisions such as internationalizing curricula ( Svensson and Wihlborg 2010 ). By doing this, higher education would probably be seen as a much more global institution than it is now.
Having explained briefly the six dimensions that configure our framework, that is Roles of English (in relation to other languages) (RO), Academic Disciplines (AD), (language) Management (M), Agents (A), Practices and Processes (PP), and Internationalization and Glocalization (ING), we will refer to it by the acronym made up of the initial letters of the dimensions: ROAD-MAPPING. In the next section, we will exemplify how the framework operates with the help of an extract from an instance of EMEMUS.
EXEMPLIFICATION OF THE FRAMEWORK
By using an example from a data set collected at the School of Economics and Business Administration, Universidad Complutense de Madrid (UCM), we will now illustrate how the ROAD-MAPPING framework may be used in a specific context. It is not our intention to provide evidence for its uniform applicability across settings; neither can we give detailed or in-depth analyses of individual cases. Rather, we intend to exemplify the structure of the framework and illustrate its functioning and potential. The example, which is taken from a data set collected for an earlier study and is reanalysed here for our present purposes, will display the relevance of all six dimensions and their interplay.
In the UCM context, teachers and students are largely Spanish nationals who have English as a foreign language and mostly use it in instructional settings (‘internationalization at home’, see Nilsson 1999 ). The English-medium strands follow the same curricula as Spanish-medium programmes, are often taught by the same teachers and implement similar assessment procedures. Teachers and students enrol in these degrees on a voluntary basis. Students are required to either certify a B2 level of English (Common European Framework of Reference for Languages) in the form of official certificates or take an entry test designed by the university. In contrast, teachers do not need to verify their foreign language competence but in practice either hold English language certificates and/or have experience abroad in English-speaking countries. Interestingly, while internationalization is advertised as a key component of the UCM profile, most of the initiatives to implement English-medium programmes have operated in an experimental fashion, with only a relatively small percentage of the total university population taking part.
Extract 1 is taken from a semi-structured interview with a teacher from the School of Economics and Business Administration. The data come from a larger corpus of teacher interviews and classroom discourse gathered by the CLUE project (Content and Learning in University Education). This project investigates the impact of internationalization on higher education in the Madrid context and its implications in other university settings ( Dafouz and Núñez 2009 ). This specific extract is chosen because it is representative of the views expressed regarding english-medium instruction (EMI) in other interviews conducted in the same faculty and in other disciplinary divisions (e.g. Aeronautical Engineering, Physics). Moreover, it illustrates how the dimensions are discursively drawn up within an interview session, and how these components are interconnected. The interviewee, a female teacher not directly involved in these English-medium programmes (known as ‘bilingual’ in the Spanish setting), was responding to the interviewer’s question: ‘what do you think of the faculty’s decision of implementing “bilingual” degrees?’
Extract 1. Teacher interview (translated from Spanish)
My view of the bilingual groups comes from a wider perspective—as a bridge towards
internationalization . I understand that English is the language of business and trade
but if it were French, we would teach in French. This is not a group in EFL, because in
that sense it would probably be better that we set up a group taught through Chinese!
[…] University authorities need to hire teachers with a good command of English and
with research stays abroad so that internationalization can be speeded up and
becomes natural . It’s also important that society sees this need. […] I am a world
citizen . With internationalization you realize that things can be done differently, that
the customs or traditions that one has don’t have to be better or worse than what
other people have.
The analysis of the extract reveals the interviewee’s view of Internationalization as the driving force behind EMI programmes, followed by her opinion in lines 6–7 that this process should be accelerated by the university authorities so that it eventually ‘becomes natural’. In other words, this teacher expresses her wish to see student numbers in EMI groups grow and eventually include all beginners in business studies.
In this line, she argues that internationalization needs to expand and engage well-prepared actors to become a reality, that is, a normal and ‘natural’ condition for the university. This connects with another dimension, Agents , who adopt a collective appearance (‘university authorities’, ‘society’ in lines 5 and 7), and are also regarded as the instruments in charge of implementing (language) Management statements. As the interviewee states, it is the university’s responsibility to hire adequate staff with good language skills and international experiences (‘stays abroad’, line 6). This illustrates explicitly a long-standing teacher complaint for more institutional involvement in the form of concrete organizational and structural measures (i.e. Management ) that go beyond the vague policy principles of internationalization that many institutions seem to have followed in the absence of precise management guidelines.
Returning to the Agents dimension, the individual agent perspective phrased in lines 7–8 (‘I’m a world citizen’) views the speaker herself as an actor operating glocally and, at the same time, showing awareness of the need to do ‘things differently’ (line 8). This line links both with the Internationalization and Glocalization dimension and also intersects with Processes and Practices , as the speaker implicitly suggests that classroom practices need to vary so that an understanding of intercultural differences may also be acknowledged. Academic Disciplines is not so visibly voiced in this interview, nor in the other sets of interviews gathered, which suggests that language, and more specifically disciplinary language issues, initially do not seem to play a central role in the content teachers’ view of EMI and if so, as line 5 shows (‘hire teachers with a good command of English’), these are mostly related to language proficiency issues.
Finally, as regards to Roles of English , the interviewee sees English as ‘the language of business and trade’ (line 2) but also admits that the current role of English as a language of trade is not a static one but dynamic as she immediately adds that other languages (i.e. French or Chinese) could very well adopt that function in the future.
Overall, this example highlights the applicability of the ROAD-MAPPING framework. It displays the six core dimensions in EMEMUS and how these dynamically intersect within the respective ecology. Moreover, the interpretation of the extract suggests that the multiperspectivity offered by the framework permits a more in-depth analysis of data within and across instances of EMEMUS.
CONCLUSIONS AND IMPLICATIONS
Inspired by recent theoretical orientations of sociolinguistics, ecology of language, expanded LP, and social practices as discourses, this article has argued for a conceptual framework for analysing EMEMUS. Reflecting its six components, Roles of English (RO), Academic Disciplines (AD), (language) Management (M), Agents (A), Practices and Processes (PP), Internationalization and Glocalization (ING), the framework has been given the acronym ROAD-MAPPING.
As EMEMUS practices have outpaced conceptualization, we have discussed the need to provide such a theoretically grounded and holistic framework, with which research may describe, analyse and compare EMEMUS within and across contexts. As advocated, the dynamic nature of the framework is conceptualized in the interplay of the six dimensions, in the internal complexity that each of them displays, and in the positioning of discourse as central point of access. Simultaneously, the fact that the dimensions are seen as intersecting with each other supports its arguably holistic nature in the sense that ‘the whole is more than the sum of its parts’. This concurs with an ecolinguistic orientation, as it takes all environmental factors into account, thereby making an activist or interventionist stance possible.
Of course, this inclusive and dynamic understanding inevitably implies grey areas, which may be interpreted as rendering the framework imprecise. For example, (language) Management and Agents can be seen as contingent dimensions in so far as the one requires the other. Managerial decisions are taken by higher educational agents, who, in turn, enact management. In this line, similar dependencies can be identified between all other dimensions. While this shows that there is overlap between them (as illustrated in Figure 1 ), our exemplification has supported our theoretical take in that the dimensions also function individually in shaping EMEMUS and, at the same time, retain their independent explanatory relevance. In other words, instead of cancelling each other out, the dimensions, such as Management and Agents , complement each other by intersecting. It is exactly these intersections that pay tribute to the inherent complexity of such varied contexts. Thus, these should not be sidelined but ‘zoomed in on’ and discussed in their intricacy, depending on the researcher’s ‘lens’ or specific research focus. As integral to the lens metaphor, researchers may foreground one or more components without losing sight of the overall picture or object of analysis. This is precisely what the ROAD-MAPPING framework intends to do: it offers a blueprint for outlining an ‘object of analysis’ that is intrinsically dynamic and potentially elusive.
Given the rationale of the framework, there is obviously a clear need to test its applicability across contexts. This can be done in a variety of ways, for instance, by describing and comparing different cases of EMEMUS. Such a comparative study is underway, focusing on university agents and their views on internationalization across four different contexts in Austria, Finland, Spain, and the UK ( Dafouz et al. in preparation ). Alternatively, it would be interesting and necessary to try out the viability of the framework through interdisciplinary collaboration, engaging stakeholders, such as university management and content specialists. Conceptually, the framework may also benefit from this interdisciplinary stance by expanding the applied linguistic perspective adopted here. This would support the aim of a truly holistic framework. In this sense, we invite further research from scholars in diverse contexts and disciplines.
ACKNOWLEDGEMENTS
Our thanks go to the anonymous reviewers who offered constructive comments and to Miya Komori–Glatz for her meticulous proofreading. We would like to extend our thanks to the Academy of Finland for their financial support and especially to Tarja Nikula as director of the ConCLIL project (2011–14).
