This book is about a bilingual educational approach in which the study of academic content is combined with the use and learning of a foreign language (referred to as FL hereafter). In Europe, this approach is usually known as Content and Language Integrated Learning (CLIL). While the term CLIL was developed in Europe, it can be seen as part of a global trend, especially as regards the use of English as a medium of instruction (Graddol, 2006). The rise in popularity of CLIL in Europe can be seen in the context of the European Commission’s white paper (1995) (Teaching and Learning: Towards the Learning Society) in which a stated objective was the ‘1+2 policy’, that is, for EU citizens to have competence in their mother tongue plus two Community foreign languages. Even at this early stage, the importance of CLIL or CLIL-like initiatives was envisaged, with the document pointing out that ‘it could even be argued that secondary school pupils should study certain subjects in the first foreign language learned, as is the case in the European schools’ (p. 47). By 2006, according to a Eurydice Report, CLIL had become ‘a fast developing phenomenon across Europe’ (Eurydice, 2006: 2). On the whole, though, European CLIL has been mostly a bottom-up movement, with many local small-scale initiatives in different parts of the continent. However, as Björklund (2006) suggests, it may be time to move beyond personal experiences, intuition and individual adaptations to ‘scientifically justified’ and generalisable principles for CLIL. As she puts it, ‘the search for common, effective core features must be more intensive than the ambitions for local uniqueness’ (p. 194). This book is intended as a step in the direction of establishing generalisable principles about the roles of the first L in CLIL: language.
Many writers on CLIL use a wide definition of the phenomenon, which includes the combination of academic content learning and the learning of heritage and community languages. In CLIL, an additional language ‘is often a learner’s “foreign language”, but it may also be a second language or some form of heritage or community language’ (Coyle et al., 2010: 1). One way of defining the core features of European CLIL is to distinguish it from other bilingual educational approaches such as Canadian immersion programmes, content-based instruction in the United States or other programmes that involve the use of regional minority or heritage languages as medium of instruction. As Lasagabaster and Sierra (2010) argue, the terms CLIL and immersion have often been used interchangeably, and this has led to some confusion and the blurring of important differences between them. These authors describe differences between European CLIL and immersion programmes in the following areas: language of instruction, teachers, starting age, teaching materials, language objectives, inclusion of immigrant students and research. With the aim of clearly identifying the scope of the term CLIL as it is used in this book, in what follows we comment briefly on each of these areas.
In terms of language of instruction, Lasagabaster and Sierra point out that, in CLIL programmes, the language of instruction is a foreign language which, unlike in immersion contexts, is not present in the students’ local communities. Nevertheless, even in immersion contexts, exposure to the L2 is largely confined to the classroom (Swain and Johnson, 1997: 7–8). As for CLIL teachers, unlike most immersion teachers, they are non-native speakers of the language used as a medium of instruction. In immersion contexts teachers are usually bilingual or native speakers of the language of instruction. In terms of starting age, CLIL learners often start studying content in the new language later than their immersion counterparts, with the result that there are large differences in the amounts of exposure between CLIL and immersion students. However, this situation is changing as many CLIL students now start learning in English at primary school, or even earlier, and some contexts provide the students with more possibilities of exposure to the target language outside the classroom than others (for example, Northern European countries like Sweden have more exposure than Mediterranean countries like Spain).
Turning to teaching materials, in immersion programmes these are normally the same as those used by native speakers, while in CLIL materials may be adapted or written specifically for a CLIL programme. Regarding the degree of language competence as an objective, Lasagabaster and Sierra claim that immersion programmes aim at native-speaker competence, while in CLIL the expectations are significantly lower. However, this is open to debate, as in most immersion contexts as well as in CLIL the main aim is functional competence. If we take the examples of the roles of French in Quebec and English in Europe, we could argue that, in both cases, the aim for the non-natives would be similar: non-native speakers in Quebec need French in order to participate in the local community and students in European countries need English to participate in the European community, when travelling or doing business, for example. In terms of the role of immigrant students in CLIL programmes, Lasagabaster and Sierra point out that these students might be at risk of exclusion, especially in those contexts where they already have to deal with two additional languages (for example, Spanish and Basque in addition to English in the case of the Basque Country). Finally, these authors acknowledge that there has been a long-standing research effort in immersion programmes, but CLIL is still relatively underresearched. This book will contribute to filling this gap, with a new approach to the roles of language in CLIL and how language is integrated into the teaching and learning of content.
While recognising the very varied realities of CLIL, our focus is on contexts in which a foreign language, English, is used as a medium of instruction. In this book, we draw on a corpus of CLIL classroom data from four European countries (Spain, Austria, Finland and the Netherlands). However, the fact that nearly everywhere in the world the target language in content-based language instruction is usually English makes the contents and suggestions presented in this book easily transferable to non-European contexts like China or Latin America.
The core corpus used in this book consists of 500,000 words of secondary CLIL classroom interaction recorded in the aforementioned contexts. This corpus includes data collected by the Universidad Autónoma in Madrid as part of the UAM-CLIL corpus, and data collected by Christiane Dalton-Puffer in Austria, Tarja Nikula in Finland and Liz Dale in the Netherlands. The secondary school corpus is supplemented by a further 200,000 words of preschool and primary CLIL classroom data compiled by the Universidad Autónoma in Madrid, again as part of the UAM-CLIL corpus. These corpora have been used in a wide range of published research studies carried out by the authors and the other contributors. The findings from these studies have played a fundamental role in the development of the ideas about the roles of language in CLIL presented in this book. Throughout the book we use extracts from this database to illustrate these ideas. Because the data extracts are genuine examples of CLIL practice, readers can build a picture of CLIL as it is actually implemented in four European contexts, and can compare what happens in these classrooms with what happens, or may happen, in their own contexts.
The use of this corpus allows us to identify core features within a limited range of variation, something which would be more difficult to do if we included contexts in which community or heritage languages are used as medium of instruction. We can thus identify the broad sociolinguistic and educational parameters of foreign language CLIL initiatives. A very useful instrument for doing this are Cenoz’s continua of multilingual education (Cenoz, 2009). By using the continua (see Figure 0.1) to more precisely delimit the features of the type of European CLIL that is the focus of this book, readers from other contexts will more readily be able to compare and adapt the ideas we present.
Cenoz’s continua of multilingual education show how linguistic, sociolinguistic and school factors combine in different ways in different bi- and multilingual education contexts, thus making it possible to compare different situations by seeing them as lying at different points on a range of continua. Although Cenoz uses the term multilingual, she points out that ‘bilingual schools can also be considered a type of multilingual school because the term “multilingual” refers to multiple languages and this can be understood as two or more languages’ (2009: 33).
Starting at the top of the diagram with school-based factors, schools can be more or less bi- or multilingual depending on how many languages are taught as subjects (school subject), how well they are integrated in the curriculum, the age at which they are introduced and the time devoted to them. In most European CLIL contexts, at least two languages (the majority language and one or more regional or foreign languages) will be taught as subjects. There will be differences in the extent to which these languages are integrated into the curriculum. For example, in some primary schools foreign languages can be used in theme-based teaching across different subject areas. In other schools, the second or foreign languages will not be integrated with the rest of the curriculum.
| Image not available in HTML version |
Schools are more multilingual the more languages are used as the language of instruction. In all of the CLIL contexts which we draw on in our European CLIL corpus, at least two languages are used as the medium of instruction: the national or majority language, and English. This would place these schools in the category of what Baker (2006: 216) describes as strong bilingual education, that is, mainstream bilingual education in two majority languages. However, the extent of content instruction through the medium of the foreign language also has to be measured. In some contexts, there is a sampling approach to CLIL, in which a relatively small part of a subject is taught in the FL, often in collaboration with a language teacher. This weak version can be seen as closer to the vision of CLIL that has emerged in foreign language teaching circles. A strong version of CLIL would see a whole subject from the curriculum taught and assessed in the FL. Most of the data from our European corpus are drawn from these strong CLIL experiences, in which the entire subject is taught and assessed in the FL, English. However, in the same way as research on Canadian immersion programmes based on French has influenced many studies and applications to other foreign / second languages throughout the world, the ideas in this book are likewise applicable to other contexts.
The teacher continuum refers to whether the teachers in a school have multilingual competences themselves, and to whether they have received training in multilingual education. One of the defining features of CLIL outlined by Lasagabaster and Sierra (2010) is that the teachers are generally non-native speakers of the language of instruction, usually English. Added to this is the fact that many have received little training in CLIL pedagogy. This is the case in the settings in our corpus. Although there are some native speakers of English, most of the teachers are non-native, and many have had little formal training in CLIL methodology. Some of the teachers have a background in language teaching, and all are certified as teachers of their subjects. In some of the contexts in the study, the teachers have the benefit of curricular guidelines which specify content and language objectives, but most do not have this support in integrating content and language.
School context, in Cenoz’s model, refers to the use of different languages inside the school for different types of formal and informal communication outside classroom lessons. As the corpus only contains classroom discourse, we do not have transcribed data showing the extent of FL use outside classrooms. This contextual information, where necessary, is provided by informants working in the different contexts. In any case, our focus on the roles of language in CLIL in this book is on the FL in representing subject knowledge and organising the social world of the classroom. This, in many ways, is seen as the main justification for CLIL. Extending FL use outside the classroom, both in terms of oral communication and the linguistic landscape (posters, examples of students’ work etc.) is, where it is achieved, an extremely useful benefit of CLIL as a context for language use.
In terms of linguistic distance, the languages involved in multilingual or CLIL programmes may be closer or more distanced from each other, regarding both the language typology and the amount of contact there has been between the languages in any setting. Two languages may not be related to each other in terms of the language families they belong to, but may have been in close contact, so that one has an impact on the other, for example with the use of ‘loan words’ (Baker 2006: 51–2). In the CLIL contexts in the corpus, there is of course great variation in the degrees of similarity and difference between the language of instruction, English, and the main languages spoken in the four countries (Dutch, German, Finnish and Spanish). However, English’s role as an international language or lingua franca (Graddol, 2006) means that there is, to a greater or lesser extent, an impact of English in these settings, with some people concerned about a possible impoverishing of the local languages, not only due to English loan words permeating them, but also to the influence of English on much wider domains of use, such as business or education.
Sociolinguistic variables can be described at the macro and the microlevels. At the macrolevel, it is important to consider the number of speakers of the target languages, the status of the languages in society, their use in the media and their general presence in the local linguistic ecologies, or the way different languages interact with each other in a specific setting (Cenoz, 2009: 37). The more use there is of different languages, the more multilingual the environment will be. In the contexts of the corpus used in this book, English can be seen as having a high status in that it has been chosen as a medium of instruction in the CLIL programmes. This is explained by the shared sociolinguistic status of English, which is not like any other foreign language, as it is the international lingua franca. However, there are differences in the presence of English in everyday life among the different contexts in the corpus. The Dutch and Finnish contexts can be seen as occupying the high-presence end of the continuum, with Austria somewhere in the middle and Spain at the low-presence end of the continuum. In the Spanish context, the only contact with English for most learners is the classroom. It is true, of course, that technology means that there is easy access to English (through the Internet, easily available DVDs in English etc.) but this does not mean that learners are generally willing to use these resources (Lasagabaster and Sierra, 2010). These features of the Spanish context may be of interest to readers of this book who are planning to implement CLIL in other contexts where there is a low presence of the FL outside the classroom.
The micro-sociolinguistic level refers to the students and their local communities of families, friends and neighbours. The context will be more multilingual if more languages are used for everyday communication. In many European countries, there are significant numbers of speakers of co-official regional languages, for example Finland with an important minority of Swedish speakers. Also, in most European countries, and this includes those in our corpus, there have been considerable increases in immigration, with the result that there is much more linguistic diversity in many, mainly urban, local communities. However, as Cenoz says, even if a context is multilingual in terms of the languages used in the local community, it is not so in educational terms if the aim of the school is not to promote multilingualism. In the case of European CLIL, there are clearly issues in introducing a foreign language such as English as a medium of instruction in situations where there are already two languages used in education. And there are also issues in changing the language of instruction in contexts where there are large numbers of students in the classroom who are still learning the majority national language. As we saw in the discussion of Lasagabaster and Sierra’s paper (2010), there is a risk of elitism if students who are still learning either the national or regional languages are excluded from CLIL programmes. Thus, it is important that bilingual programmes such as CLIL are inclusive, and that they do no harm to the educational chances of learners who do not speak either of the languages of instruction at home.
By describing how the different CLIL contexts included in the corpus used in this book are located on the continua of multilingual education, we hope to help readers to more precisely describe their own CLIL contexts. This should help CLIL practitioners in situations in which, for example, there are differences along the sociolinguistic, school or teacher dimensions, to make more principled assessments of how relevant the ideas presented in this book are to them. For example, some of the data in the book represent relatively mature CLIL situations, in that practices are quite well developed: they do not, then, represent beginning and/or experimental experiences. As will be seen in the extracts of oral and written language presented throughout the book, both the teachers and the learners in the different classrooms are able to produce at times fairly extended stretches of the L2 for a range of purposes. However, this should not be taken as an indication of homogeneity. The examples used in the book show learners’ production of language from a fairly wide range of stages of development, and often the extracts show how learners struggle to make meaning with limited resources. Wherever readers are located along the continua of multilingual education, the framework presented in this book will better equip them to make principled decisions about the roles of language in their local CLIL contexts.
Why do we need a book about the roles of language in CLIL? After all, as CLIL stands for content and language integrated learning, we could assume that language is already included in the package. However, things are more complicated than that. In spite of the increasing popularity of CLIL across Europe and around the world, there has not always been clarity about key issues such as how language is involved in doing CLIL, what aspects of language should be targeted, how learners’ language develops through CLIL, and whether and how language should be assessed along with content. Indeed, much research on CLIL has largely seen language and content as separate issues. As Leung points out, there is a need to bring the two dimensions together:
[c]urriculum content learning and language learning, which are still generally seen as two separate pedagogic issues, should be consciously taken into account in an integrated way in classroom-based bilingual research.
(Leung, 2005: 240)
Leung’s point is that it does not make much sense to argue for bilingual education initiatives such as CLIL without a greater understanding of the ways in which languages are actually used in classroom interaction and activities. Without this understanding, justifications for CLIL tend to be vague, a kind of ‘language bath’ (Dalton-Puffer, 2007: 3), in which learners, simply by participating in lessons in which they study subject matter in a foreign language, will somehow pick up the foreign language by osmosis. In this view, CLIL classrooms, unlike language classrooms, are a ‘natural’ environment for language learning in which students can come into contact with the language as it is used in everyday life. In this way, as Dalton-Puffer (ibid.) observes, the CLIL classroom becomes a kind of replacement for the street, especially when the foreign language is not used in the real streets surrounding the classroom.
This book sees the role of language in CLIL rather differently from the language-bath approach. In working with content, students will encounter and have to use a whole range of the language which shapes educational knowledge. This kind of language can be broadly referred to as ‘the language of schooling’ (Schleppegrell, 2004). It is this language that we focus on in this book. In this sense, the language that can be learned in CLIL classrooms is, in very important ways, unlike the language of ‘the street’ and the kind of language that is often the focus of communicative language teaching. As Byrnes puts it,
(…) educational knowledge is shaped through language that fundamentally differs from language used to transact life’s tasks in, for example, social encounters or to seek or provide information – areas of language use that have dominated communicatively oriented educational practice.
(Byrnes, 2008: 48)
Thus, the focus of CLIL is not to equip learners with the language they need to transact everyday life tasks, such as ordering a meal or buying a train ticket. However, as we will see in this book, in CLIL classrooms learners can and do transact everyday tasks and talk about things which are personally meaningful to them and, therefore, they will find opportunities to develop ‘everyday language’. These everyday tasks relate to the organisation of the social world of the classroom, and talk involving personal responses or opinions is generally related to the content-based learning objectives for the subject being studied. In fact, the ability to communicate one’s personal experiences and attitudes in a foreign language is fundamental to achieving understanding of complex subject matter taught through that language. The exposure to and practice of the foreign language in different classroom tasks and activities is likely to be transferable to other non-academic contexts.
By giving the book the title The Roles of Language in CLIL, we are referring to language in two main ways: that which is involved in representing the meanings which are crucial to any academic subject, and that which is used in organising and orienting the social world of the classroom. In focusing on these two broad roles of language in CLIL, we build on important research carried out in European CLIL contexts. In terms of the social organisation of the CLIL classroom, we draw mainly on work that looks at the pragmatic aspects of CLIL lessons, for example at how students are given opportunities to use language for such functions as issuing requests (Dalton-Puffer, 2007; Dalton-Puffer and Nikula, 2006). As for the language through which educational knowledge is shaped, we build on the large body of work in systemic functional educational linguistics as well as the influential work on CLIL by Do Coyle and colleagues (Coyle et al., 2010), which has already offered frameworks for describing language use in CLIL classrooms. In our own description of the roles of language in CLIL, we acknowledge their distinction between the language of learning (language needed to express key aspects of content), language for learning (language needed to participate in tasks and activities) and language through learning (language which emerges when CLIL students are being stretched to think about and express meanings related to content). Throughout the book, where relevant, we refer to these distinctions, but our own approach draws on a different range of theoretical perspectives, particularly those of systemic functional linguistics (SFL), sociocultural theory and second language acquisition (SLA). In the next section, we provide a brief overview of these three approaches.
The perspective on the roles of language in CLIL presented here needs a theory of language to sustain it. This theory needs to show in a principled way how, at the same time, social activities such as education shape language use and how language itself constructs knowledge. Theories of language in which language is seen as an abstract system removed from contexts of use will not be adequate to this task. In this book we adopt the systemic functional linguistics framework originated by Michael Halliday (Halliday and Matthiessen, 2004). From its inception, researchers have used this framework to develop understanding of how language and learning are related (Halliday, 1993). Put briefly, systemic functional linguistics or SFL is a meaning-based theory of language, in which all choices speakers or writers make from the lexical or grammatical systems of a language are shaped by the social activities, such as education, in which they are involved.
In making these choices, speakers and writers draw on three types of meaning or metafunctions of language: ideational, interpersonal and textual (see Box 0.1 for an explanation).
Box 0.1: The three metafunctions of language
According to the systemic functional model of language, there are three basic functions of language:
the ideational, through which we construe or make sense of our experience;
the interpersonal, through which we enact our social relationships;
the textual, which facilitates the first two by enabling us to construct sequences of discourse which flow and have cohesion and continuity (Halliday and Matthiessen, 2004: 29–30).
These are labelled ‘metafunctions’ as they are much more general and intrinsic to language use than the ‘functions’ of individual examples of language use (ibid.: 31).
| Image not available in HTML version |