CLIL in Higher Education: Towards a Multilingual Language Policy 9781847699374

Citation preview

CLIL in Higher Education

BILINGUAL EDUCATION & BILINGUALISM Series Editors: Nancy H. Hornberger, University of Pennsylvania, PA, USA and Colin Baker, Bangor University, Wales Bilingual Education and Bilingualism is an international, multidisciplinary series publishing research on the philosophy, politics, policy, provision and practice of language planning, global English, indigenous and minority language education, multilingualism, multiculturalism, biliteracy, bilingualism and bilingual education. The series aims to mirror current debates and discussions. Full details of all the books in this series and of all our other publications can be found on http://www.multilingual-matters.com, or by writing to Multilingual Matters, St Nicholas House, 31–34 High Street, Bristol BS1 2AW, UK.

CLIL in Higher Education Towards a Multilingual Language Policy

Inmaculada Fortanet-Gómez

MULTILINGUAL MATTERS Bristol • Buffalo • Toronto

Library of Congress Cataloging in Publication Data Fortanet, Inmaculada. CLIL in Higher Education: Towards a Multilingual Language Policy/Inmaculada Fortanet-Gómez. Bilingual Education & Bilingualism: 92 Includes bibliographical references and index. 1. Multilingual education—Government policy—Spain. 2. Language arts (Higher)— Correlation with content subjects—Spain. 3. Education, Higher—Spain. I. Title. II. Title: Content and language integrated learning in higher education. LC3727.F67 2013 370.117'50946–dc23 2013001048 British Library Cataloguing in Publication Data A catalogue entry for this book is available from the British Library. ISBN-13: 978-1-84769-936-7 (hbk) ISBN-13: 978-1-84769-935-0 (pbk) Multilingual Matters UK: St Nicholas House, 31–34 High Street, Bristol BS1 2AW, UK. USA: UTP, 2250 Military Road, Tonawanda, NY 14150, USA. Canada: UTP, 5201 Dufferin Street, North York, Ontario M3H 5T8, Canada. Copyright © 2013 Inmaculada Fortanet-Gómez. All rights reserved. No part of this work may be reproduced in any form or by any means without permission in writing from the publisher. The policy of Multilingual Matters/Channel View Publications is to use papers that are natural, renewable and recyclable products, made from wood grown in sustainable forests. In the manufacturing process of our books, and to further support our policy, preference is given to printers that have FSC and PEFC Chain of Custody certification. The FSC and/or PEFC logos will appear on those books where full certification has been granted to the printer concerned. Typeset by Techset Composition Ltd., Salisbury, UK. Printed and bound in Great Britain by Short Run Press Ltd.

To the loving memory of my father, Antonio Fortanet (1934–2012), a model of perseverance and optimism

Contents

Acknowledgements Introduction

ix xi

Part 1: Multilingualism and Multilingual Education

1

1

Multilingualism 1.1 Definition of Multilingualism 1.2 Types of Multilingualism 1.3 Linguistic Characteristics of Multilinguals 1.4 Social and Individual Multilingualism at Universitat Jaume I

3 3 6 17 18

2

Multilingual Education 2.1 Definition of Multilingual Education 2.2 Factors Conditioning Multilingual Education 2.3 What can be Achieved in Multilingual Education Programmes? 2.4 Multilingual Education Approaches 2.5 Content and Language Integrated Learning 2.6 Multilingual Education at Universitat Jaume I

22 23 25 29 31 40 50

Part 2: Multilingualism in Higher Education

53

The Sociopolitical Context 3.1 Societal Multilingualism and Language Policy 3.2 Research into Language Policies 3.3 Economics and Language Policy 3.4 Factors for a Multilingual Language Policy in Education 3.5 Supranational Policies 3.6 National and Regional Language Policies 3.7 University Policies

55 55 56 59 61 68 74 77

3

vii

viii CL IL in Higher Educat ion

4

The Language Component 4.1 Languages as Target Products 4.2 Languages as Media of Instruction 4.3 Languages of Research 4.4 Languages of Administration

97 99 109 115 125

5

The Pedagogical Component 5.1 Teaching and Learning in Higher Education 5.2 Foreign Language Teaching Approaches 5.3 Multilingual Education: CLIL Pedagogy

127 130 143 147

6

The Human Factor 6.1 Students 6.2 Academic Staff 6.3 Administrative Staff

153 153 158 170

Part 3: Towards a Multilingual Higher Education Institution: The Case of Universitat Jaume I

173

7

Description of the Study 7.1 Method 7.2 Objectives and Research Questions

181 181 183

8

Background and Context of Universitat Jaume I 8.1 Profiles of Community Members 8.2 External Institutional Context 8.3 The Internal Factors

185 186 209 213

9

Proposals for a Multilingual Language Policy 9.1 The Multilingual Language Policy 9.2 The Multilingual Language Policy and University Community Members 9.3 Multilingual Education 9.4 Follow-up, Assessment and Evaluation

218 218 225 229 235

Conclusions Appendix 1: Questionnaires Appendix 2: Complete Tables References Index

236 247 257 261 280

Acknowledgements

This book is the result of much effort and work shared with a number of persons, each of whom has had a different and specific role in its making. Chronologically, I would like to thank Eva Alcón for inviting me to get involved in the first project for the introduction of English as a third language of tuition at Universitat Jaume I in 2003. I am also grateful to the colleagues who initially took part in in the pilot project and those who later participated in the teacher training courses on ‘Teaching in English’ during the five years I taught them. These colleagues not only gave me insights into the problems and difficulties of multilingual education I would have never reflected on without their help, but they also showed me the advantages and benefits of the efforts they were making, encouraging me to carry on exploring its possibilities. During the several years I have devoted to the book, other colleagues have also supported my work and have provided good ideas about the development of some of the chapters. I would like to mention Bob Wilkinson (University of Maastricht), Ute Smit (University of Vienna), ChristianneDalton Puffer (University of Vienna), Yolanda Ruiz de Zarobe and David Lasagabaster (Universidad del País Vasco), Emma Dafouz (Universidad Complutense de Madrid), Belinda Crawford-Camicciottoli (Università de Firenze), Cecilia Jacobs (Stellenbosch University, South Africa), my colleagues at the research group GRAPE (Group for Research on Academic and Professional English) and, in particular, Christine Räisänen (Chalmers University of Technology, Gothemburg, Sweden). I would also like to express my gratitude to Mercedes Querol, who has helped me through the difficult task of collecting data for the research and analysing it, as well as Pilar Jara and Francisco Herrero for their advice on the statistical treatment of the data. Jasone Cenoz has been one of the most significant people in the creation of this book; she read the first draft and gave me very insightful feedback, helping me to produce a manuscript which could be interesting for its readership. ix

x

CL IL in Higher Educat ion

I should not forget the anonymous reviewers from the publishing house, who also contributed their opinions and comments to improve previous versions of this book. I would also like to thank my Rector, Vicent Climent, for giving me the chance to design and implement a multilingual language policy for Universitat Jaume I, based on the ideas presented in this book. He and his team have always backed my ideas and have supported them in every context. Finally, I would like to thank my family, my parents, my husband Hector and my daughter Marina for their patience and understanding when the work absorbed my time and attention more than they deserved.

Introduction

Multilingual education has been studied from many perspectives, although few studies have focused on higher education. This book aims to offer a different view of multilingualism by presenting a contextualised case of a multilingual language policy taking the Content and Language Integrated Learning (CLIL) approach. CLIL seems to have been gaining support in recent years as a new approach to multilingualism. There have been some studies on its implementation, although most of them have presented experiences or partial perspectives on this topic without any deep theoretical foundation. The aim of this book is to deal with multilingualism in higher education from a global perspective in a coherent single-authored volume. The book starts with a general overview on multilingualism and multilingual education. In Part 2 several aspects of multilingual education are discussed, considering the previous literature. Finally, a case for an application of multilingualism to higher education is presented in the form of a proposal for a multilingual language policy for Universitat Jaume I. Universitat Jaume I is situated on the eastern coast of Spain, in the Valencian Community autonomous region. It was created in 1991 and has been bilingual in Spanish and Valencian from the start. In bilingual regions like the Valencian Community, higher education institutions (HEIs) play an important role in the preservation and promotion of the local language. Most universities in bilingual areas have traditionally maintained language policies which describe the relevance of the local language and the role the university must have in promoting and particularly developing it as a language of the academy in tuition, research and also in administration. English, as is stated in its statutes, has always been considered a working language at Universitat Jaume I, since it is essential in many fields of research as well as in the workplace for many graduates. However, up to now the use of this language has not been officially monitored. A new language policy is needed in order to include English as a third language of instruction and to xi

xii

CL IL in Higher Educat ion

redefine its uses alongside those of the two official languages, Valencian and Spanish. Valencian is the language of cultural identity of the university and has to be promoted in all areas. Spanish is the state language, also still the majority language. It does not need to be promoted, but its correct use is a duty and constitutional right of all Spanish citizens. English, as stated before, is a working language. Its use has to be encouraged in the areas and situations where it is needed, particularly for international relations in teaching, research and administration. Like many universities in Europe, Universitat Jaume I has been considering the possibility of introducing English as a language of instruction for several years. In the last five years the university has designed new study plans following the Bologna Agreement; 2010 was established as the deadline to harmonize European university studies. During this process, a university directive established that 12 of the 240 credits for all new study plans should be taught in English. This was a turning point for the introduction of English as the third language of instruction in undergraduate studies. Moreover, multilingual language policies are being implemented or are in the process of development in several universities in the Spanish higher education system where English is being introduced as the second or third language of instruction. There seem to be many factors suggesting that this is the right moment to design a new language policy for Universitat Jaume I. In the past few years I have found myself playing a significant role in the development of a multilingual language policy for the university. In order to guarantee its success I started a thorough review of previous research on multilingualism, multilingual education and language policies, especially at the higher education level. A research project was eventually designed with the aim of producing an analysis of the current situation at Universitat Jaume I and the possibility of accepting and supporting a multilingual language policy there. The final aim was to provide some solid recommendations for this policy. The first chapter in this book is devoted to multilingualism. The reason why it has been included is because it is necessary to understand the background and the reality of multilingualism before tackling the question of how to implement a multilingual language policy at a university. I have started with definitions of the two main types of multilingualism: societal and individual. The characteristics of societal multilingualism refer to the social status of the languages, the association of multilingualism with certain geographical locations, the policies of language use and language planning, and the associations the members of a social group make between languages and cultural identity.

Introduc t ion

xiii

Individual multilingualism refers to an individual choice to know and learn several languages. The dimensions reviewed here relate to the individual characteristics of the multilingual, such as relative competence in both or each language, the cognitive organisation of languages, the age and order of acquisition, the context when and where each language was acquired, the social and cultural status given to each language by the individual, and the cultural identity assigned to each language. Most dimensions of both societal and individual multilingualism can be measured, and the research techniques most often used are also presented in this chapter. The next section is devoted to the linguistic characteristics of multilinguals, the aspects that differentiate them from monolinguals. These include plurilingualism, transfer and interference, the accommodation or adaptation of one’s speech to that of the interlocutor’s, the possibility of choosing the most appropriate language for each communicative event, or the use of codeswitching, loans, borrowings or mixing several languages in the same communication event. The contribution of this chapter to the research part of the book is to provide a basis for the identification of the societal and individual characteristics of multilingualism that can be found at Universitat Jaume I. This identification will help to establish the characteristics of the sample of the study and hence of the population. Chapter 2 reviews multilingual education. It starts with a comprehensive definition, and continues with the factors that condition multilingual education. Three types of factors are described: sociopolitical factors, such as those of former colonies; individual factors which may hinder success in multilingual education, such as the socioeconomic situation of families, the academic ability of the learner, age, the level of command of the languages before entering school; and pedagogical factors which affect the ways of teaching and learning languages, such as programmes that focus on the meaning and other programmes that focus on the form of the language, time of exposure to the several languages within and outside school contexts, and teaching by means of a specific subject or in immersion programmes. The next section takes account of the main achievements that can be expected from multilingual education programmes: language achievements, literacy achievements and content achievements. This chapter also includes a review of the main models of multilingual education institutions and programmes that have been implemented in recent years in different parts of the world, and which have contributed to new forms of multilingual education. The models reviewed are: the language across the curriculum approach, originally developed in the UK in order to improve the English-language proficiency of learners at all levels of education, and which has had a strong influence on other later approaches; languages for specific purposes, also

xiv

CL IL in Higher Educat ion

developed originally in Britain and later in the United States, and eventually applied to teaching English as a foreign language; content-based instruction, linked initially to students with limited English proficiency in the United States, mainly immigrants, and which later has had an impact on other approaches such as immersion programmes or CLIL; and immersion programmes, developed in Canada to teach French to English-speaking children, and which is considered to be the approach which has had the greatest influence on CLIL. The final section of Chapter 2 introduces CLIL as yet another variety of multilingual education, one which draws from the experience of the approaches described so far. This section looks at CLIL from several perspectives – focusing on higher education, but also taking in ideas, research and experiences related to other approaches. This chapter will be the basis for establishing the type of multilingual education that will be presented in the study in the third part of the book. Chapter 3 deals with the sociopolitical context and focuses on several factors that condition language policies, ranging from economics to situational, operational and outcome factors. In this chapter, the reader can also find a review of national and international policies, as well as the internal provisions of universities regarding multilingualism, and how they may have an influence on the implementation and success of multilingual education. In the last section of this chapter, several language policies of universities from all over the world are analysed, especially those developed in the Catalan universities which have a social and cultural environment which is closer to that of Universitat Jaume I. The next three chapters examine what I have considered the three key components of any multilingual education approach: the language component, the pedagogical component and the human factor. Chapter 4 deals with the language component as a target product and medium of instruction. The chapter starts by focusing on university languages at three levels: local, national and international languages. Looking at cases where English is one of those languages, there is a dilemma as to whether the native language or the lingua franca should be used as a teaching medium. The next section discusses whether multilingual education should focus on the language of communication (BICS) or on the language of academic cognition (CALPS). There have been many studies about academic discourse in university classrooms and about the language of research. These are important contributions to the success of multilingual education, since language is the means used to learn and to disseminate research, and it is in language use where some of the major difficulties may be found. A review of these studies and their main findings can provide interesting insights into

Introduc t ion

xv

the major characteristics of the classroom and research genres; these will be of interest to teacher trainers, who can find in this research some materials they need for their courses. Chapter 5 deals with pedagogy. Higher education teaching pedagogies deserve special consideration. In this chapter I start by describing the basic learning principles in higher education which explain how learners build on previous knowledge in order to achieve transformative learning. Secondly, I deal with discipline conventions and practices. After this, some language learning approaches are reviewed, and I finally describe how all these pedagogies and approaches contribute to the specific pedagogy of CLIL. The teachers’ profile will be a focus of interest in Chapter 6, where we explore the characteristics and roles of academic staff, both content and language teachers’ profiles, as well as team teaching and multiple language inputs, all related to the CLIL approach. Several options for needs analysis and the resulting teacher training programmes are also presented. In addition, the roles of the administrative staff are considered. These three chapters have significantly contributed to the research put forward in Part 3, especially regarding the design of the questions on language in the questionnaire and the determination of the profiles of students, lecturers and administrative staff. Chapter 5 has also been very insightful for the interest group discussions on the implementation of multilingual education, particularly CLIL, in the context of the pedagogies applied today in higher education. Part 3 introduces the research on Universitat Jaume I. Chapter 6 begins with a description of the institutional and social context of Universitat Jaume I and a justification for the study. The section on method describes how questionnaires and interviews have provided us with information in order to make a ‘linguistic mapping’ of the population of Universitat Jaume I, including students at undergraduate and graduate level, lecturers and administrative staff. The factors that have been analysed include: the degree of multilingualism among university members and the languages involved; the languages mainly used and the relationships established with identity; how and when individuals have learned these languages; and the perceptions and attitudes they have towards the languages they know and use. These factors provide a profile of the university community members which is used as a point of departure for the proposal of a multilingual language policy. The other two sections in this chapter describe on the one hand the external institutional context, and on the other hand the internal factors that condition a language policy in this university. All this information is used to propose guidelines for a multilingual language policy for Universitat Jaume I. It is recommended that the policy takes

xvi

CL IL in Higher Educat ion

the form of a plan; the languages selected for this policy should be Valencian and English, in addition to Spanish. This selection is justified in the first part of Chapter 9. Then I explain the objectives of the plan and the expected outcomes, taking into account the present circumstances. Furthermore, I also provide some guidelines about the communication plan to disseminate the advantages of the multilingual language policy and how multilingual education could be implemented. I hope the readers of this book – researchers, students or policymakers – find it interesting and useful, and can easily adapt it to their universityspecific contexts.

Part 1 Multilingualism and Multilingual Education The two chapters that form the first part of this book are devoted to multilingualism and multilingual education. The concept of multilingualism and the reality of multilingualism all over the world are used as a point of departure to identify what is known as multilingual education. The second part builds on the first part in an attempt to join multilingualism and multiliteracy by means of formal instruction. Multilingual education has been adapted to fit the context where it has been implemented. The resulting types of multilingual education will be discussed and one of them, CLIL, will be highlighted as the approach which is nowadays promoted in European education systems.

1

1 Multilingualism

1.1 Definition of Multilingualism Although bilingualism and multilingualism refer to the knowledge and use of two and of more than two languages, respectively, very often the term ‘bilingualism’ has been used to refer to both concepts, as in the case of Baetens Beardsmore’s (1982) definition: Bilingualism (. . .) must be able to account for the presence of at least two languages within one and the same speaker, remembering that ability in these two languages may or may not be equal, and the way the two or more languages are used plays a highly significant role. (Baetens Beardsmore, 1982: 3) Moore and Gajo (2009) distinguish between multilingualism (the study of social contact) and plurilingualism (the study of individuals’ repertoires and agency in several languages). In this book I have chosen the term multilingualism to refer to both the social and the individual situation in which two or more languages are known and used by speakers. Plurilingualism is, in turn, considered as one of the characteristics of multilingual individuals, as explained in Section 1.3. Multilingualism is not something new. It has happened all over the world and at all times in the history of humankind. However, according to Aronin and Singleton (2008: 4), current multilingualism is ‘assimilable to politicoeconomic aspects of globalisation, global mobility and “postmodern” modes of thinking’, and as such has different characteristics from historical multilingualism. Multilingualism is related to today’s ‘new linguistic dispensation’, that is, to the modern concepts of time, space and identity as a background to language. In relation to time, linguistic communication has 3

4 Par t 1: Mult ilingualism and Mult ilingual Educat ion

been affected by the use of technologies that have made it instantaneous both in its written and oral forms (mobile phones, audio and video communication, computer software, etc.). Space is no longer fixed and its relationship with languages has varied. Mobility culture and the accessibility to modern means of transport have also altered the concept of identity and the idea of belonging to a certain culture and place. In addition, virtual existence in cyberspace has created new identities and new social groups to be part of. Although it is true that today’s way of life, particularly in the West, favours multilingualism, throughout history a large number of people have always been multilingual, that is, they utilised and many still use two or more languages in their everyday lives (Cenoz & Genesee, 1998). Research on multilingualism has unveiled the complexity of linguistic and cultural variation in the multilinguals’ use of languages, which may vary over time. The competence of multilinguals in each of the languages they know is no longer the only important aspect of multilingualism. It is also important to learn the status of the languages each person possesses, which may differ according to the situation and place where these languages are used. They can be official or non-official languages, majority or minority languages, international, national, regional or local languages. The use the people make of the language can be individual or social. Social multilingualism can be explained in relation to historical, political and educational dimensions. The concern for multilingual education in different parts of the world relates to the status of the language in any of the possibilities described above. In countries that went through a long colonial period in their history, such as India or South Africa, one of the major concerns is still the attitude towards the colonial versus the local or regional languages in educational contexts (Bekker, 2003; Ferguson, 2003). Although most of these countries have approved constitutions that declare as official languages several of their own languages along with the colonial language, it is the latter that is normally used in the classroom. The reasons usually given by the education authorities often relate to the great variety of languages used among the pupils (as many as 11 are official in South Africa out of the 25 that are spoken by a significant number of people). However, the underlying reason is often the power relationship between the speakers of those languages (Baetens Beardsmore, 2003). Even though European countries have been colonisers rather than colonies, a similar situation can be found in European regions which had their own language, such as Friesland in the Netherlands, Brittany in France or the Ruthenian region in Slovakia. Nowadays these languages have become mostly secluded within the family and home domain, and the state language has been

Mult ilingualism

5

adopted in the education system (Hornsby & Agarin, 2012). Some other languages went through a similar situation but social, political and historical events swung in their favour so that a few years ago they began to be incorporated in some way or other into the classroom and the public and administrative domains. This is the case for Welsh in Wales, and Galician, Basque and Catalan in Spain. In addition to the minority local languages, some countries have received a great number of immigrants who may keep their own language through one or several generations. Sometimes these languages are considered to be problematic for their integration and immigrants are encouraged to use only the national language of the host country with their children, forcing them to become monolingual. In cases where they decide to preserve their mother tongues, their multilingualism is most often reduced to the private domain only. Most countries in Europe, as in many other parts of the world, have received immigrants at some time in their history, especially in the large cities, so this situation needs to be considered when dealing with multilingualism. Regarding individual multilingualism, the situation is rather different depending on whether the bilingual or multilingual speakers are immersed in a society that shares this multilingualism (social multilingualism) or, on the contrary, they are surrounded by monolinguals or people who speak languages different from the ones spoken by them. A case thoroughly studied, especially from the point of view of psycholinguistics and second language learning, is that of a child with a home language which is different from the social or school language, such as in the case of the immigrants described above (Cots & Nussbaum, 2008; Kasanga, 2008; Lotherington & Eamer, 2008). On the other hand, looking through the social perspective, a country may be multilingual even when most of the citizens have competence in just one language, such as for example in Switzerland or Canada, where several languages are official, but depending on the area of the country only one is spoken. On the contrary, there may be countries where most of the population is multilingual whereas the official languages consist of only one or two, as in many countries in Africa (Edwards, 2006). The social consideration of languages may also have a major influence on their use; majority national languages may not be useful when they are used in another country where those languages do not have a high status or are perceived with a negative attitude by the population (Nunan & Lam, 1998). This may happen, for example, in China, where all the population is compelled to learn Mandarin, a language very few people understand outside China and which is not very popular in other countries (Verma, 2002). In addition, the social status of a language may change over time, as is the case

6 Par t 1: Mult ilingualism and Mult ilingual Educat ion

in the Philippines, where Spanish was the official language until the end of the 19th century, and the language of instruction until the 1920s when it was replaced by English, and nowadays has completely disappeared from the educational context, where local languages co-exist with English (Gonzalez, 1998). Moreover, there is also the case of borderline areas where, for pragmatic reasons, several languages co-exist, some of them as official languages, others being the official language of a neighbouring country, and sometimes a third language being spoken in both states when they share a linguistic area (for example in some parts of the Pyrenees where Spanish, French and Catalan co-exist). A different question that needs to be addressed here is the ‘lingua franca’.1 These are the international languages used in extensive parts of the world to communicate with people of different origins and mother tongues (Hamers & Blanc, 2000). The most important languages used as lingua francas nowadays are English, Spanish, Russian, French and Chinese. Even in those countries where these languages are not official, some of them have been adopted as the language of instruction in the education system, on their own or coexisting with one or several other languages. In recent years a new idea has been developed about the linguistic competence of multilinguals. Some researchers (Cook, 1992; Herdina & Jessner, 2002; Moore & Gajo, 2009) assert that the knowledge they have of their languages differs from that of monolingual speakers, since they can develop language learning, language management and language maintenance skills that are seen to contribute to the kind of metalinguistic awareness associated with multilingualism – the M-factor (multilingualism factor) and CLIN or cross-linguistic interaction.

1.2 Types of Multilingualism There have been several attempts to provide a typology of multilingual phenomena. As Baetens Beardsmore (1982: 4) stated, ‘typologies are only descriptive labels that help to establish certain limits of a complex field of investigation into smaller areas with a clear frame for reference’. Two main types of multilingualism have been established by most researchers, which serve as a framework for further classification: (1) Social or societal multilingualism. A group of individuals share the knowledge of two or more languages. (2) Individual multilingualism. An individual has certain knowledge of at least two languages.

Mult ilingualism

7

These two main types are not exclusive and mainly differ in standpoint. Societal multilingualism tries to explain the tensions and relative powers of the languages in the community, their interrelationships, and the connections between language and political, economic, social, cultural and educative forces (Baetens Beardsmore, 1982: 4). For the purpose of this book it will be interesting to learn more about societal multilingualism and the role of bilingual or multilingual education, especially in the context of higher education upon which this book is focused. As for individual multilingualism, it can logically follow the societal multilingualism of a certain community or it can be an exception, as in some of the cases described above for minority languages or immigrants, or for individuals who live in a monolingual society and decide to become multilingual. In any of these cases, it is interesting to look at how individuals acquire their second and foreign languages and the psychological processes involved in this acquisition (see Section 2.2). Nowadays the opinion that is most often supported is that multilingualism is a matter of degree. Somebody who knows a few words in another language is multilingual but to a very low degree. At the other end of the continuum, a person who has a perfect domain of several languages possesses a high degree of multilingualism (Edwards, 2006: 7). Multilingualism can be measured in order to ascertain the current situation in a particular community, as in the case presented in Part 3 of this book, Universitat Jaume I. Picturing the current situation can help to establish the best language policy in order to achieve certain goals. The success of the multilingual language policy can then be determined by measuring the same factors of multilingualism again several years later.

1.2.1 Societal multilingualism Unlike what was stated in the first definitions, one of the admitted characteristics of multilingualism nowadays is the different use speakers make of the languages they know. In societal multilingualism, very often the use of the languages is determined by certain dimensions: (1) (2) (3) (4)

social status; geographic bilingualism; language use and language planning; cultural identity.

One of the most common ways to measure societal multilingualism is censuses or a survey; that is, the whole population, or just a sample of it, are

8 Par t 1: Mult ilingualism and Mult ilingual Educat ion

asked to respond to a questionnaire or to answer some questions in a structured interview (Baetens Beardsmore, 1998; Baker, 2001; Hamers & Blanc, 2000). Most censuses are mainly interested in the geographical boundaries of multilingualism, although others try to investigate further by asking questions about the mother tongue, preferred language or dominant language. However, these are rather subjective and ambiguous concepts. It is difficult to determine the mother tongue of an individual whose parents have each spoken to him or her in a different language since the day she was born. The preferred language may change depending on context and is often strongly influenced by the social status conferred by each individual to a language. On the other hand, the dominant language can be measured in terms of frequency of use, though again subjective perceptions are the only way to measure this variable. Much more interesting may be questions regarding attitudes towards a language. These can provide information about the social perception of a language as elite multilingualism, about the cultural identity the multilingual subject relates to the knowledge and use of a language, and the motivation to maintain that language, for example, instrumental motivation, when the language is perceived as having only utilitarian purposes, or integrative motivation, when the language is perceived as an element of integration in a linguistic community (Baetens Beardsmore, 1998). Regarding the units that can be used to measure the four dimensions determined for societal multilingualism, all of these can follow a continuum that goes from ‘elite’ or the highest social status of a language to ‘folk’ or the lowest status, or from a language most related to the geographic and historical factors of the place where it is spoken, to the least geographic relation. For example, English in Romania or in Spain cannot be associated with geographical multilingualism, whereas it can be in other parts of the world such as Sri Lanka or India. In relation to language use and language planning, the continuum will have at one end the languages with the widest use, and at the other end those with the most restricted use, i.e. instrumental languages. Finally, in order to measure social identity, the continuum will show the degree of identification of a certain language with the cultural values of a social group.

1.2.1.1 Social status Throughout history, a certain type of societal bilingualism has been associated with the upper classes, in what has been coined as elite bilingualism, in contrast to folk bilingualism. Educated people throughout history voluntarily learned a prestigious language such as Greek, Latin or French, in order to be recognised as belonging to an elite group. On the other hand, folk

Mult ilingualism

9

bilingualism has usually been driven by necessity and frequently learned in the street without any formal instruction. This is the case for most immigrants (Edwards, 2003). However, nowadays it is difficult to distinguish one from the other in many social contexts, since everybody is supposed to have access to education in developed countries, where infant, primary and secondary education are public and obligatory, including learning one or several official languages of the country and some foreign languages. Therefore, in many cases multilingualism has become obligatory (a characteristic of folk multilingualism) and learned through formal instruction (as described for elite multilingualism). A phenomenon associated with societal multilingualism, which has its origin in the unbalanced status of languages in contact, is that of diglossia, which was described by Ferguson as early as 1959 as: a relatively stable language situation in which in addition to the primary dialect of the language, which may include a standard or regional standard, there is a very divergent, highly codified, often grammatically more complex, super-posed variety, the vehicle of a large and respected body of literature, heir of an earlier period or another speech community, which is learned largely by formal education and is used for most written and formal purposes, but is not used by any sector of the community for ordinary conversation. (Ferguson, 1959) This definition accounts for extreme situations where one language or a variation of a language is imposed for political, economic or religious reasons. For example, in Arab-speaking countries a high form of Arabic is used for education, religion and formal communication, whereas a low form is used for informal and local communication (Baetens Beardsmore, 1982; Ferguson, 1959). Other researchers have developed this concept (Fasold, 1984; Gumperz, 1971; Hudson, 1994; Williams, 1992) in order to relate it to the functional distribution of languages and the power relations between their speakers, where languages with a relatively low status are used for at-home and informal interactions, and those with a relatively high status are used for formal communication. The status and use given to a language may vary geographically; that is, it may be conceived as the low variety, dialect or language in precedence to a dominant language in a certain area, whereas in an area only a short distance away the situation may be the opposite. Additionally, a diglossic situation may also vary chronologically, when one variety or language takes over the functions formerly reserved for the other language as power relations change in the society (Hamers & Blanc, 2000). Diglossic situations have been identified in some areas in which multilingual education

10 Par t 1: Mult ilingualism and Mult ilingual Educat ion

was introduced, including English where this language is considered a language of power in contrast to the local languages.

1.2.1.2 Geographic multilingualism Societal multilingualism can also be associated with geographic areas. The criteria adopted to relate languages to a territory have been to look at this relationship in the regions of certain countries since, as we know, a common language has not been one of the main aspects taken into account when forming a new country. Some countries have two or more official languages, whereas others have just one. Among the former, several situations can be found. Some countries such as Canada or Switzerland claim to be multilingual although very few of their inhabitants can be described as such. In these countries there is territorial monolingualism in two or more different languages, which are spoken in different regions (Baetens Beardsmore, 1982; Edwards, 2003). In order to overcome problems with mobility, an individual right has been recognised for those who move from one region to another one with a different language to have formal instruction in that language (e.g. in Canada, where this was the original reason for the start of the ‘immersion programmes’ described in Chapter 2). Multilinguals in these countries can be considered horizontal multilinguals, since the languages they use have equal status and either language can be used in similar circumstances (Baetens Beardsmore, 1982). A different situation is that of countries where a large number of languages are spoken. It has been reported that over 70% of all the languages in the world are found in just 20 nation states, such as Papua New Guinea (860), Indonesia (670), Nigeria (427), India (380), and so on. Most of these countries are in southeast Asia and Africa, and are among the poorest in the world. The reason for the survival of this wide variety of languages is often based on the isolation of the groups who speak them. In addition to the above-described cases of multilingualism associated to geography, there are the cases of border areas. As mentioned earlier, political boundaries very rarely take into account linguistic communities. As a result, the inhabitants of areas of several neighbouring countries often share a common language or dialect. This is particularly common in Europe where, for example, Catalan is spoken in the northeast of Spain, the southeast of France, in Sardinia in Italy and in Andorra, a small country in the Pyrenees. Additionally, border areas have traditionally been bilingual due to the need to communicate both with the people in their country and those in the neighbouring country. So, in the border between Austria and Hungary, people speak both German and Hungarian. In many cases there is an additional reason for this bilingualism: due to frequent wars and treaties, borders

Mult ilingualism

11

move and some bordering areas have historically alternated between belonging to one country or the other (Romaine, 2004: 390–391). Although geographic multilingualism is a fact today, a new tendency, as will be seen below, is to consider language as ‘not centered on personal spaces or territories’, but as a means ‘to maximize efficiency and equality among speakers’ (Garcia 2009: 79). Calvet (1998) has proposed what he calls a ‘gravitational model’, where powerful languages can co-exist with official and national languages, even local vernaculars, without threatening them in any way. The relationship between language and territory has been blurred for many languages, as well as their identification with the inhabitants of that territory, especially for largely expanded majority languages.

1.2.1.3 Language use and language planning As noted above, frontiers between countries have not been drawn in accordance to the ethnic, cultural, social or linguistic groups of the population. As a consequence, many countries are confronted with a wide range of different unintelligible languages within their boundaries. Throughout history several solutions have been adopted for this problem. Some powerful countries have imposed their language on the nations they have conquered by war; this has happened in Central and South America since the end of the 15th century, when Spanish was imposed, ignoring and frequently eliminating the vernacular languages of the native population. In more recent times, states have used what has been coined as ‘language planning’ (Hamers & Blanc, 2000), a particular form of political, economic and social planning. There are two aspects to language planning: internal and external. The former, also called ‘language engineering’, ‘constitutes a systematic interference with the internal dynamic processes to which languages are subject’ (Hamers & Blanc, 2000: 311). This interference can be restrictive, by means of standardisation and the ‘purification’ of the language from geographical variations and foreign influences, or it can be creative, by means of resources such as the introduction of a written system for an unwritten language, standardisation of alphabet or orthography, or the expansion and modernisation of its lexicon (Baker, 2001). External language planning is ‘concerned with artificially interfering with existing status relations between languages in contact’ (Hamers & Blanc, 2000: 312). Status planning is a political development (Baker, 2001) and usually aims at gaining more recognition and functions for a language. Two main approaches have been distinguished in external planning: nationalism and nationism (Fishman, 1968b). In the former, a language is chosen as the official national language for reasons of ethnic identification and national consciousness. In the nationism approach, on the other hand, a language is selected for its efficiency, such as it being the one spoken

12 Par t 1: Mult ilingualism and Mult ilingual Educat ion

by most of the population. These two approaches are not exclusive and on many occasions multinational states adopt a combination of both. Among the different possibilities are the following: •

•

•

The use of one or more indigenous languages for all circumstances. This is the solution adopted, for example, in China, where the authorities in the 1950s decided to raise the status of the Mandarin dialect to become the official national language after standardising its speech and phonetic alphabet. China thus created a new standard language that could be used as the language of tuition all over the country, in an effort to make most of the population literate, but also as a tool for political unity and ideological propaganda. The adoption of one or more regional languages as official languages together with an international language (many former colonies have adopted this solution). This was the option selected by South Africa, where language policies have been central to the historical development of the country. The Constitution of 1993, later ratified in 1996, recognises 11 official languages: English, Afrikaans, isiNdebele, Northern Sotho, Sesotho, siSwati, Xitsonga, Setswana, Tshivenda, isiXhosa and isiZulu. Nowadays the government has issued several documents supporting multilingualism in education, including tertiary education, and the right of students to decide in which language they wish to be taught (Bekker, 2003). Only the international language is given the status of the official language. This decision was the one most usually found in colonies in the past, although it is not so common to find it today. In southern Africa, former British colonies such as Malawi, Botswana or Zimbabwe maintained English as the only official language after becoming independent. In past Portuguese colonies, the colonial language has remained the sole official language, such as in Mozambique (Kamwangamalu, 2004). Even in those countries where an African language has become co-official, the situation is that of diglossia, with African languages playing a minor role in lower primary education and family-and-friends communication, whereas English is the elite language used in higher levels of education, administration, law, etc.

1.2.1.4 Cultural identity Language has also been categorised as one of the salient characteristics of cultural, ethnic and social identity. It helps groups of individuals perceive themselves as sharing certain values and gives cohesion to the group (Hamers & Blanc, 2000). A willingness to preserve one’s identity may be a good reason to retain the language of origin or the mother tongue in situations

Mult ilingualism

13

of immigration or where languages or dialects are not officially supported. For example, even though Romani has undergone stages of near extinction, it is still preserved as the language of Roma gypsies all over the world, to accompany their characteristic lifestyle and the closeness of their groups. Another example is Hebrew in the Jewish culture, as much a sign of identity as their religious practices. Language is not the only pillar of ethnic and cultural identity, but it is an important one, and it becomes the main one when the language is threatened. This was the case of Basque, Catalan and Galician in Spain during the dictatorship of General Franco (1939–1975). These languages were prohibited as signs of banned identities and cultures. During the 1960s and 1970s there were demonstrations and riots in defence of freedom; the use of these languages was also part of the protest. Nowadays, the Spanish regions where these three languages are spoken are officially bilingual. However, there are still groups who associate the language and the culture with extreme nationalism and claim monolingualism of the vernacular language and political independence. The spread of societal multilingualism can be the basis for multilingual education, when the aim is the preservation of native languages, the dissemination of a new version of a national language, the necessity of a language for group religious, social or political integrity, or the learning of a lingua franca for international communication. The situations described above regarding societal multilingualism are not permanent but are often in constant flux. Multilingual education can be used in some of these changes, such as turning elite multilingualism into an accessible situation for most people, as we will see in Chapter 2.

1.2.2 Individual multilingualism The second main type of multilingualism is individual multilingualism, which analyses the choice of languages and how these languages are learned by an individual who has access to more than one linguistic code as a means of social communication (Hamers & Blanc, 2000). The types of individual multilingualism can be analysed according to certain dimensions: (1) (2) (3) (4) (5) (6)

relative competence in both languages; cognitive organisation; age of acquisition; context of acquisition; social cultural status; cultural identity.

14 Par t 1: Mult ilingualism and Mult ilingual Educat ion

1.2.2.1 Relative competence in the several languages As described above, the first definitions of multilingualism considered an equal competence in the two languages used, which has been coined as balanced bilinguals (Cummins & Swain, 1986) or ambilinguals, i.e. individuals who can function in all contexts in all languages (Baetens Beardsmore, 1982). This should not be confused with a high competence in two or more languages, since most bilinguals and multilinguals show a higher degree of competence in the use of one or another language for certain functions, leading to the idea of functional multilingualism.2 Ambilingualism is a very rare phenomenon. The degree of competence of a multilingual is very difficult to establish, since the parameters to measure it should be specific and different from those used for monolinguals. In some cases, there is a marked variation between the competence attained in different skills. Some multilinguals are very competent in receptive skills (reading and oral comprehension), but not in productive skills (writing and speaking). Edwards (2006) calls the first case receptive multilingualism or semibilingualism, as compared to productive multilingualism (including both receptive and productive skills). Competence has been measured over a long period, especially by means of second language or foreign language testing (Bachman & Palmer, 1996; Hughes, 1989). Although these tests can be very useful for measuring the progress in certain aspects of communicative competence, they are not so useful for measuring multilingualism because their results cannot be compared with those obtained for the mother tongue and they also neglect aspects such as the pragmatic use of the language or non-verbal behaviour associated with the language. On the other hand, most of these tests only compare the competence of a learner of a second or foreign language with the ‘ideal’ competence of a native speaker of a language. However, as stated by Hamers and Blanc (2000: 34), ‘native competence is not necessarily synonymous with a high level of competence’. There are many factors affecting the multilingual which differentiate him or her from the monolingual. For example, when comparing the extent of vocabulary known by a monolingual and a multilingual, the multilingual may know a reduced number of words in one of the languages, but very frequently he or she will know a wider range of vocabulary by adding the other languages, since not all words are learned in all languages at the same time.

1.2.2.2 Cognitive organisation In relation to cognitive organisation, a distinction has been made between compound and coordinate language systems (first pointed out by Weinrich, 1953; then developed by other researchers such as Baetens Beardsmore, 1982 or Ervin & Osgood, 1954). In the case of compound multilinguals, certain

Mult ilingualism

15

linguistic units in two or more languages are associated with the same meaning or representation of reality. On the other hand, coordinate multilinguals relate translation equivalents in the languages they know to different sets of representations of reality. This is associated with other dimensions such as the age and context of acquisition. This distinction has nothing to do with competence, and very often multilinguals can be more compound for certain concepts and more coordinate for others, in such a way that their situation can be located on a continuum on which the extremes are coordinate and compound multilingualism. For example, a child who has learned the Arabic word for ‘family’, and learns the English word in the same social context, understands the reality of a family in the Arab world, with two different linguistic forms. However, a child who has learned Arabic as a mother tongue, and learns the word ‘family’ after emigrating to Britain may associate the words in the two languages with different realities of family. According to this dimension, additional languages can be associated with the culture in which they are spoken as primary languages (for English: the USA, Britain, Australia, etc.), or they can be associated with the culture of the mother tongue without making differentiations, or they may be associated with the culture of the person who is using this language. The latter should be the case when using a language as a lingua franca, when the purpose is just communication between people with mother tongues and cultures not related to the language being used. Some psychology researchers have applied several techniques to find out the degree of coordinate–compound cognitive organisation of multilinguals, such as semantic satiation and semantic generalisation tasks or wordassociation techniques, which try to compare the identification of concepts with words in different languages. The results of some early studies (Lambert et al., 1958) seem to have proved that: • •

The semantic difference between translation equivalents is larger for coordinate multilinguals. Compound multilinguals have two interdependent association networks; it is easier for them to identify a common concept and to make more frequent use of translation resources.

However, later studies do not always corroborate these conclusions. Language biographies and observation also seem to be effective methods of research to evaluate the cognitive behaviour of multilinguals, since they can provide information about age, context of acquisition and the circumstances of use of a language, such as the combination of several languages when interacting with other multilinguals (Hamers & Blanc, 2000).

16 Par t 1: Mult ilingualism and Mult ilingual Educat ion

1.2.2.3 Age of acquisition Much has been said about the ideal age for the acquisition of a second or a third language. Research evidence seems to suggest there is a critical period (Critical Period Hypothesis; CPH) for language development, when children can acquire one or more languages simultaneously. This starts at approximately the age of 2 years. However, different times of closure have been given by researchers: 5 years old (Krashen, 1973); 6 years old (Pinker, 1994); 12 years old or ‘puberty’; and 15 years old (Johnson & Newport, 1989). After the critical period, additional languages can also be learned, although there are deviations from native-like performance, especially with regard to pronunciation and intonation (Butler & Hakuta, 2004). Although no consensus has been reached, it is now generally accepted that a child should first acquire a mother tongue and, once he has some fluency in that language, other languages can be introduced, in what has been coined as consecutive multilingualism.

1.2.2.4 Context of acquisition The acquisition of a language can certainly follow different processes depending on the social use of the language in the immediate environment of the learner. A language can be ‘endogenous’ if it is used by the community for daily communication, or ‘exogenous’ if it is restricted to certain situations such as administration and public careers. However, again this can be viewed as the extremes on a continuum, since a language may not be used for daily street communication, but may be the medium of instruction in the educational system, such as English or French in many western African countries (Hamers & Blanc, 2000). In this case it can be considered an exogenous language, but it will be still more exogenous learning Chinese in this context where a learner cannot use it at all. As in previous dimensions, interviews, language biographies and a good knowledge of the context can be suitable ways to measure this dimension.

1.2.2.5 Sociocultural status The learner’s environment may be different from the social environment where he lives, or it may change over a short period of time. The individual environment may determine the status given to a language through the family, group of friends, teachers at school, and even transmission by means of political propaganda. A line showing the continuum for this dimension will indicate the sociocultural status given by an individual to each of the languages he uses, which may also be rather different from that given by the social group. For example, French may have a very high status for somebody living in Algeria, especially if he feels there are possibilities for him to immigrate to France in order to start a new life there. However, another individual

Mult ilingualism

17

who is obliged to learn French at school without seeing any possibility of practising this language may not give it the same status.

1.2.2.6 Cultural identity This is another concept dealt with in societal multilingualism. As in the case of sociocultural status, the cultural identity given to a language can also vary at an individual level, as part of the subject’s linguistic biography. Multilingual persons can associate the languages they know with one or several cultures, and identify themselves more or less with those cultures. Some multilinguals, even though they keep a good command of their mother tongue, may become identified with only the culture of their second language. This has been coined as L2 acculturated multilinguals (Sachdev & Giles, 2004). Some techniques, such as multidimensional scaling, role playing, psychometric tests, as well as ethnographic and sociolinguistic approaches, have been applied to the measurement of the multilingual’s cultural identity. Language can be a very important factor in cultural identity, which can also be measured on a continuum. As extreme cases, some people like the Flemings in Belgium or the Quebecois in Canada have built their cultural identity exclusively on the defence of their language (Hamers & Blanc, 2000: 203). However, the identification of language and culture may vary according to individuals in the same linguistic community. For example, the cultural identity felt by many individuals who learn and use English as a lingua franca is almost nonexistent, whereas the learning of that same language by some other people can be closely related to a profound admiration for the culture of its native speakers.

1.3 Linguistic Characteristics of Multilinguals Multilinguals have the ability to communicate in different ways. One of these is what Garcia (2009: 45) calls ‘translanguaging’, that is, ‘multiple discursive practices in which bilinguals engage in order to make sense of their bilingual worlds’. Multilinguals need translanguaging in order not only to facilitate communication with others, but also to make sense of their multilingual worlds. Translanguaging involves accommodation, language or code choice, code-switching, loans and borrowings, and code or language mixing. Multilinguals can negotiate meaning in different ways, depending on the situation and their interlocutors, and the choices made by the speaker. Similarly, the Council of Europe (2002) refers to ‘plurilingualism’, a term widely used by this institution to emphasise the fact that the multilingual

18 Par t 1: Mult ilingualism and Mult ilingual Educat ion

individual does not ‘keep languages and cultures in strictly separated mental compartments, but rather builds up a communicative competence to which all knowledge and experience of language contributes and in which languages interrelate and interact’ (Council of Europe, 2002: 4). The characteristic that defines plurilinguals is their capacity to choose among the many options and possibilities they can use as communication strategies, which monolinguals lack. Canagarajah (2009: 6) says of plurilingual competence: (1) Proficiency in languages is not conceptualised individually, with separate competencies developed for each language. What is emphasised is the repertoire – the way the different languages constitute an integrated competence. (2) Equal or advanced proficiency is not expected in all the languages. (3) Using different languages for distinct purposes qualifies as competence. One does not have to use all the languages involved as all-purpose languages. (4) Language competence is not treated in isolation but as a form of social practice and intercultural competence. Moreover, Herdina and Jessner (2002) identify the specific characteristics of multilinguals as the M-Factor or multilingual factor. The multilingual learner develops new skills which are not developed by the monolingual, such as language learning skills, language management skills and language maintenance skills. All these skills contribute to what has been regarded as metalinguistic awareness, that is, activities of reflection on each and every language and its use and the ability to consciously monitor and plan methods of linguistic processing in both comprehension and production. Some examples of this ability are the capacity to reduce the number of performance errors by correcting him or herself, developing and applying conversational strategies, drawing on common resources for the learning of more than one language, and – with regard to language management and maintenance – keeping the systems apart by trying to avoid transfer phenomena or relating them when it is convenient for the communication process (Herdina & Jessner, 2002: 29).

1.4 Social and Individual Multilingualism at Universitat Jaume I Multilingualism can be regarded from two points of view: societal multilingualism and individual multilingualism. These two points of view help

Mult ilingualism

19

to determine the factors that need to be measured. To start with, the aim at Universitat Jaume I is to introduce multilingualism into mainstream education in order to have an influence on the knowledge and status of the three priority languages in society (Valencian, Spanish and English), in this way affecting societal multilingualism. Something to be borne in mind is that several of the factors described in Chapter 1 have ideological, historical and political connotations, especially with regard to Valencian, the local language. This part of Spain was conquered by King James I in the 13th century and most of its Moorish inhabitants were expelled. James I decided to bring settlers from Catalonia, who brought with them the Catalan language. However, Valencia was always considered a separate kingdom and its language evolved in parallel to the northeastern Catalan. This language was commonly used until the 18th century, when it was banned by the king of Spain, who wished for a homogeneous state. After the succession war, Valencia’s special Law and Rights were abolished and the king and his court, and later several state governments, tried to take strict control over the region, since it has always been rich and politically strategic (Fuster, 1980). Since the 18th century Valencian has survived as an oral language, only written by some minority authors and avoided by nobles and famous people, especially in large cities like Valencia or Alicante, except for short periods of time when it recovered some prominence. It was not until the 1970s that it gained the status of an official language, after a long dictatorship which was particularly tough on local languages. For people who had been punished and mocked for speaking the local language it was difficult to assume this change of status. The generation who had lived most of their lives under the dictatorship decided not to use their mother tongue in many situations, to avoid using it in public, and most importantly to avoid teaching it to their children. This happened mainly in the cities; the language was best preserved in rural areas. As a consequence, a large percentage of the next generation, now in their fifties or sixties, did not have Valencian as a mother tongue but heard their parents and grandparents speak it naturally, and some of them recovered the language and started using it again as a family language. Fortunately, a group of people in this generation claimed their language rights in the 1960s and 1970s and, thanks to them, Valencian was given official status, at the same level as Spanish. The younger generations have lived through a completely different situation. Valencian has been introduced as a language of instruction in schools. It is no longer just a family language but also a school language, and all children learn how to read and write it correctly, as well as Spanish, thus becoming horizontal multilinguals as they have two languages with the

20 Par t 1: Mult ilingualism and Mult ilingual Educat ion

same status. Contrary to what has happened in other countries where language planning and standardisation were made top-down, in this area it has responded to a bottom-up movement. In 1932 a group of cultural associations wrote and subscribed to the so-called Normes de Castelló or Normes d’ortografia valenciana (1932). Nowadays, the Acadèmia Valenciana de la Llengua (the institution that sets the rules and preserves the language) still considers these rules as a guide to the standards of the language. Taking all this into account, one can easily understand that it is difficult to establish a single societal interpretation of the Valencian language. There are people who still claim more presence and importance for the local language, the only one they relate to their cultural identity, even claiming their right to monolingualism in that language, which according to them would give the language the status it deserves. This position is often associated with nationalist and left-wing political positions. On the other hand, there is another group of people who consider Spanish their only language and whose cultural identity is associated with the Spanish state. For them, Valencian should just be kept as a family language, especially in rural areas. This position is often linked to right-wing and national parties, who aim to preserve the centrality of the state, as well as groups of immigrants, both nationals and foreigners. Regarding societal multilingualism in relation to English, this language was first introduced as a subject in primary and secondary education in Spain in the 1970s, without much success. French was at that time the preferred foreign language for schools. The consequence is that, in general, many people who are today in their fifties or over have never taken any classes in English at school. However, in the late 1970s and 1980s English became the preferred foreign language, and eventually the only one offered to students in most schools. For many years speaking English has been considered a privilege of wealthy families, who could afford to pay for extra classes so that their children would learn this language ‘properly’, especially regarding oral skills. English had therefore become elite multilingualism. The reason for this need for extra classes is that the traditional methodologies followed in schools do not promote the use of the spoken language, as well as the generally perceived consideration of foreign language subjects as secondary both by parents and by the school system. After an overview of the perceived societal situation in the context of Universitat Jaume I, it would be interesting to learn how this applies individually. In Part 3 of this book, I present the results of a survey which tries to find out the individual characteristics of the university community regarding multilingualism, following the recommendations of Baetens Beardsmore (1998), Hamers and Blanc (2000) and Cenoz (2009a), who consider that

Mult ilingualism

21

questionnaires and structured interviews given to a sample of the population are the best ways to measure multilingualism. In this survey, I have tried to find answers to questions such as the relative competence in several languages of the members of the university community, their age of acquisition, how they acquired the various languages they know, and their cultural identification with these languages. As stated at the beginning of this section, the degree of multilingualism has to be measured as a point of departure and, after the implementation of certain policies, also as an indicator of success. Multilingual education is one of the most important policies a university can implement in order to improve individual and, in turn, societal multilingualism. Before deciding on the best policy and implementation of a multilingual programme in an institution, it is essential to establish the characteristics of its population. The degree and the type of multilingualism are of special concern, as well as the goal the institution wants to achieve and its attitude and motivation towards this goal. In the scenario I will develop in the third part of the book, the goal is multilingual education in an already bilingual Spanish-Catalan university. Research by means of questionnaires and interviews will provide the relevant information about the starting-point situation of multilingualism, along with the attitude and motivation. The findings of this research will inform decisions about the goal or target multilingual situation to be achieved. Having introduced in a general way the concepts and characteristics of multilingualism, in Chapter 2 multilingual education is investigated as a means to create or to maintain multilingualism.

Notes (1) Some authors do not like this term, because it may also refer to those languages that mix the characteristics of several languages, simplifying their grammar and lexis: pidgins or Creole languages. For example, Crystal prefers the term ‘global language’ to indicate an international language which can serve as a lingua franca all over the world, referring to English (Crystal, 2003). (2) The original term here and in subsequent classifications is ‘bilingualism’, although for consistency we use both ‘multilingualism’ wherever possible, and ‘bilingualism’ as a synonym for it.

2 Multilingual Education

Multilingual education develops in response to a conscious policy within the language planning of a state. By adopting this kind of education, the authorities interfere in the social status and power relations of languages. As the education we refer to is not private or individual, the aim of multilingual education is to build or maintain societal multilingualism. In this chapter we will deal with multilingual education, its definition, conditioning factors, and types and models of multilingual education programmes. Most of the research carried out up to now has focused on primary and secondary school experiences. Not much has been said or done regarding multilingual education at university level, which is the main objective of this book. However, a successful programme for tertiary education presupposes previous success at the lower levels of education. In addition, some of the problems encountered and the good practices developed in primary and secondary schools can also be applied in some ways at university level. For these reasons, multilingual education will be discussed globally, and specifications for the introduction of more than two languages will occasionally be pointed out. There are many reasons to support multilingual education in the 21st century. Globalisation is reaching all corners of the world, with a particular emphasis on business and commerce and electronic communications. Business and commerce have been the main inducers of multilingualism throughout history, and today this is even more true, as multinational companies become more numerous, together with trade relationships between distant countries in both a geographic and a linguistic sense. International relations have been enhanced by advances in electronic communication which have created a need for proficiency in a lingua franca of communication, generally English. On the other hand, migrations are still very common, and these can be motivated by political, economic, educational, cultural, medical or even personal reasons (Genesee, 2004). These migrations are also encouraged by 22

Mult ilingual Educat ion

23

better means of transport and communication, allowing people to have an idea of the place where they aim to live, and at the same time, once they have migrated, to maintain contact with their families and friends, which is a way to keep their mother tongues alive. Furthermore, the importance of local and regional languages has been upgraded as a result of the organisation of indigenous peoples all over the world in order to preserve and promote their own languages in addition to other national or international languages. A second reason for the increasing relevance of local languages is the emergence of new states, which have conferred on previously regional languages the status of national official languages, thus becoming the medium of instruction in formal education. This chapter starts with a definition of multilingual education and the factors that condition this type of education system. Section 2.3 describes the language, literacy and content achievements that can be expected in multilingual education. The final sections include the most well-known models of multilingual programmes.

2.1 Definition of Multilingual Education All levels of an education system play an important role in providing the necessary skills to communicate effectively in several languages. Multilingual education can be defined as ‘education that aims to promote bilingual (or multilingual) competence by using both (or all)1 languages as media of instruction for significant proportions of the academic curriculum’ (Genesee, 2004). The term multilingual education was introduced by UNESCO in its Resolution of 1999, although it is in the declaration of 2003 that it is defined as ‘the use of at least three languages, the mother tongue, a regional or national language and an international language in education’ (Genesee, 2004: 17). Multilingual education programmes teach content through a target language, i.e. the language the programme intends to teach. Although some researchers define this target language as other than the student’s home language or mother tongue (García, 2009; Hamers & Blanc, 2000), there are some situations in which the aim of a multilingual programme may be to preserve or reinforce the home language or mother tongue of the student, especially in the early stages of literacy, as will be explained below. Although multilingual education has been implemented for centuries, since the second half of the 20th century there has been a development of a different kind of multilingual education. It is no longer conceived as the privilege of a few, but as a resource most people should be able to make use of. Article 2 of the Universal Declaration of Human Rights (1948) establishes

24

Part 1: Mult ilingualism and Mult ilingual Educat ion

that ‘everyone is entitled to all the rights and freedoms set forth in this Declaration, without distinction of any kind, such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status’,2 giving a mother tongue the status of a basic human right. In 1953 UNESCO issued a resolution declaring that children should begin their education through the medium of their mother tongue, following the many problems children had in becoming literate in a foreign language they did not know in new countries which had become independent from colonies. Although this problem occurred mainly during the second half of the 20th century, this situation has not changed in many developing countries which were former colonies, especially in Africa and South America. In many places, although the mother tongue is used in the first stages of education, the colonial or state language is introduced progressively, so that it becomes the only language of instruction in the upper levels such as university studies. This has been called ‘transitional bilingual education’ (Garcia, 2009). During the 1970s and 1980s transitional bilingual education was criticised, and some language minorities who felt threatened started claiming their language rights as a sign of identity, as was the case of Basque and Catalan in Spain and Irish in Ireland. These movements have been maintained in the 21st century, when new groups of minorities have joined. Associations focusing on the defence of minority languages have been created and have made an impact, such as SIL International (an organisation devoted to building the capacity for sustainable language development through research, training and materials development for over 2590 languages, based in Dallas, TX) or the Association Internationale pour la Défense des Langues et des Cultures Menacée (based in Liege, Belgium), in addition to associations for the defence of a single language, like the Kadazandusun Language Foundation in Malaysia, the Fenno-Ugria Foundation in Estonia or the Instituto de Lengua y Cultura Aymara in Bolivia, to name just a few. As described above, the situation we have now in the 21st century favours the awareness of and the need to know several languages, and though governments still play a relevant role in promoting some of them, especially through schooling and the public education system, there are other initiatives to incentivise multilingualism. In recent years UNESCO’s declarations have gone further in matters of language learning and multilingual education. The resolution of 2003 states that: UNESCO supports bilingual and/or multilingual education at all levels of education as a means of promoting both social and gender equality and as a key element of linguistically diverse societies. (UNESCO, 2003: 32)

Mult ilingual Educat ion

25

This multilingual education includes mother tongue, the national language and an international language. According to this document, when children leave secondary school they should be able to work with three languages ‘which should represent the normal range of practical linguistic skills in the twenty-first century’ (UNESCO, 2003: 32).

2.2 Factors Conditioning Multilingual Education Multilingual programmes are implemented in schools and higher education institutions (HEIs), embedded in a social and political context which cannot be ignored. Three main factors can be distinguished which may condition the characteristics and the success of multilingual programmes: (1) sociopolitical and economic factors; (2) individual factors; (3) pedagogical factors.

2.2.1 Sociopolitical and economic factors Political history may be a key factor in some communities. Some countries which were former colonies are multilingual as a result of arbitrary divisions of the territory by the colonial powers. Very often the peoples who share a state are enemies, especially in Africa, where there are in addition serious problems of economic and social development. Even though they may wish to become linguistically independent, they cannot, since it is the colonial language that allows them to work with more powerful countries which may provide external economic aid. On the other hand, multilingual programmes have generally been assumed to be positive for students from discriminated and low socio-economic backgrounds, especially in the case of immigrants. Although several studies have demonstrated that there are no major problems in the integration of these students (Caldas & Boudreaux, 1999), other factors may have an effect on the maintenance of their mother tongue, such as the perception that the language they speak is of a lower status, and therefore is not suitable as a medium of instruction. In addition, the reduced number of speakers in a certain area, either because of few immigrants or because of the low number of people speaking that language in a country, may impede its inclusion in the education system. Situations of economic difficulty such as that persisting in the first decades of the 21st century are not of much help in solving these problems, since political authorities are not prone to investing in the preservation of the languages of small groups of people.

26

Part 1: Mult ilingualism and Mult ilingual Educat ion

As an opposing tendency, and usually as a bottom-up movement, in recent decades there has been an ‘ethnic revolution’ (Fishman, 1977) of minority groups who claim the right to preserve their own culture and their language in particular as a symbol of identity. This has been backed up by the expansion and democratisation of education throughout the world (Hamers & Blanc, 2000). Even so, the success of these movements has been heterogeneous, because the status of minority languages varies dramatically from one state to another. For example, in Spain there are three official languages recognised apart from Spanish: Galician, Catalan and Basque; all of them are co-official in the specific territory where they are spoken, and have been introduced as media of instruction in the education system together with Spanish. The situation is different when the minority languages co-exist with the official language without a marked territory, as may happen with immigrant groups. Even more complex is the situation of multilingual countries such as India, where almost 400 mother tongues have been reported (Srivastava, 1984). As recommended by UNESCO, there is general agreement that education should begin all over the world with the mother tongue. However, this seems to put a very heavy burden on developing countries with such high numbers of minority languages. Almost without exception these are economically disadvantaged countries, and the groups of people still speaking some of these minority languages are among the poorest.

2.2.2 Individual factors Some individual factors have been studied by researchers as factors that may favour success in multilingual education. Research conducted in Canada (Holobow et al., 1991) attempted to find out if socioeconomically disadvantaged situations had a negative effect on the acquisition of first (L1) and second (L2) languages in immersion programmes. It was found that these children achieved as well as comparable students in conventional programmes of language acquisition, and in mathematics and science which were taught in L2. These results were similar to those obtained by Caldas and Boudreaux (1999) for English-speaking American students attending French immersion programmes in Louisiana. This seems to demonstrate that socioeconomic factors are not decisive in determining the success of a multilingual education programme at an individual level. Another personal factor that has been focused on as problematic for multilingual education is academic ability. Genesee (1987) carried out some studies to find out if below-average students had serious problems when taking part in immersion programmes. When comparing these students in

Mult ilingual Educat ion

27

multilingual and regular primary education programmes, results showed no differences in academic achievement in L1; moreover, those in multilingual programmes performed better than the ones in regular programmes in L2, especially as regards speaking and reading skills. According to Genesee (2004: 563) ‘academic ability alone does not distinguish students who can benefit from multilingual/immersion education and those who cannot’. Age has also been included among the factors that may affect the acquisition of more than one language. Although some studies have pointed out that an early age is better for acquiring a second language, clear empirical evidence to sustain this argument has not been found; although some results support a critical period when a second language is acquired more easily, others argue against it (Genesee, 2004). In all the studies other aspects appear to condition the results obtained, such as their attitude and motivation, the acquisition of the language outside school, and pedagogical factors in the school setting, which may have a great influence on the successful acquisition of a language by a child or group of children. According to the results obtained by research into immersion programmes in Canada, early immersion programme students perform better in the L2 than late immersion students until they are at the end of secondary school, when their level of performance becomes very similar (Genesee, 2004). This might presuppose that both early and late immersion students should be well prepared for multilingual education at the tertiary level. There are also a number of additional factors in the planning of languages in education that may affect its success, including the level of acquisition of both languages the child has reached before entering school and the cognitive level of the languages as representing reality. In later acquisition, it is even more common to find an imbalance between competence in L1 and L2 of the individual; often the order and circumstances of the acquisition of the languages lead to functional multilingualism, in which each language is used for a certain function and not for others. Therefore, the sequence and level of previous acquisition of L1 and L2 are also very relevant in the first stages of multilingual education. There is no clear answer as to what is the optimal age to start multilingual education. It depends on the objectives, the needs and the resources available. In multilingual communities it may be better to introduce at least two languages earlier, so children become accustomed to and start using both in everyday life and in school. In communities which aim at trilingualism the introduction of the third language at a later stage seems to be a better option since it allows a settlement of the other two languages before introducing the third (Genesee, 2004). In any case, the introduction of bilingual or trilingual programmes requires qualified personnel, appropriate instructional

28

Part 1: Mult ilingualism and Mult ilingual Educat ion

materials and administrative resources in order to start the learning and to provide continuation along the primary, secondary and tertiary levels of education.

2.2.3 Pedagogical factors The third group of factors affecting multilingual education success relates to the way of teaching and the circumstances of the programme. Genesee (2004) identifies two approaches in the use of the target languages to teach academic subjects. One is ‘focus on meaning’ programmes, in which ‘students are expected to learn the target language by using it for academic or authentic communicative purposes’. The second approach is given the name of ‘focus on form’, in which the learning of formal structures and the rules of the language is a prerequisite to the functional use of the language. Some studies have revealed that students who have followed programmes based exclusively on a focus on meaning approach often fail to master important aspects of the language such as verb tense or the use of pronouns or prepositions (Harley & Swain, 1984). Moreover, more classroom exposure to the language has not produced an increased linguistic competence in the target languages (Genesee, 1987). The reason is that students develop the ability to use the language only in a number of school settings which are limited to the requirements of the teachers. In contrast, when language instruction is linked to the students’ communicative needs in the classroom and there is a more explicit focus on form, the result is a greater advance in target language competence (Lyster, 1990; Norris & Ortega, 2000). According to Garcia (2009), there are also operational factors within the educational system affecting the pedagogy that have an effect on the success of the multilingual programme. One of them is the timing of the introduction of the L2: some programmes introduce the language as a subject before using it as a medium of instruction (Mäsch, 1993); others prefer total immersion in the target language (Swain & Lapkin, 1982); some others keep the L2 subject along with the subjects taught in that language; and in some programmes the L2 subject disappears gradually from the curriculum as it becomes the language of instruction (Baetens Beardsmore, 1995). In addition, the selection of subjects to be taught in English in primary and secondary education may also have an influence on the success of the programme. Many programmes tend to choose social sciences, history, art and physical education. In some countries where the indigenous language has not developed scientific discourse it is often the sciences that are chosen. In Europe there are a great variety of subjects taught in an L2 (Fortanet-Gómez & Räisänen, 2008). Some programmes also involve several languages in the same

Mult ilingual Educat ion

29

subjects, with parts of the subject taught in one language and parts of it taught in another language; sometimes the parts taught in the L2 follow a project-based approach and may be related to L2 subjects. At tertiary level there is also a wide range of possibilities, from complete programmes taught in a second or foreign language, to special groups or a selection of subjects. A situation that is becoming widely accepted today is the co-existence of several languages of instruction. All in all, there do not seem to be any comparative studies of these situations in order to prove which of them may be more effective. However, unsuccessful multilingual programme results are often related to lack of continuity or of progressive or graded introduction of the languages.

2.3 What can be Achieved in Multilingual Education Programmes? 2.3.1 Language achievements Linguistic achievement is closely related to the linguistic goals of the programme. Research comparing immersion programmes and traditional language teaching in language subjects carried out in several regions of the world (Christian & Genesee, 2001) has demonstrated that students in multilingual programmes which involve dominant languages acquire a more advanced level of functional proficiency in a L2 than those taking conventional instruction through only language subjects. Assessment in the two programmes included the four basic skills. Even so, generally speaking, it can be stated that students in immersion programmes develop comprehension skills (reading and listening) further than production skills (writing and speaking), perhaps due to the pedagogical factors described above, and to the priority given to comprehension skills in these programmes (Garcia, 2009). Regarding relative competence, balanced multilingualism is often the goal promoted by programmes in Canada and Europe. The same can be said when the aim is the maintenance or the development of a minoritised language, which is the case in many areas of the world with a situation of geographical multilingualism, as Catalonia or the Basque Country in Spain (Cenoz, 2009a). With regard to threatened languages, although balanced multilingualism could be the ideal goal, this is not always possible, and some time is needed to reverse language shift (Baker, 2001). In these situations, the education system may help by introducing the use of the language where it is possible to do so with a continuation along several levels. In some countries the aim is not only multilingualism, but plurilingual proficiency for the

30

Part 1: Mult ilingualism and Mult ilingual Educat ion

whole population, as encouraged by the European Union (EU); the results may be functional interrelationship and complementarity in the use of several languages, though nowadays much still remains to be done to accomplish this ambitious goal.

2.3.2 Literacy achievements It is generally assumed that all multilingual education programmes promote multiliteracy, that is, ‘instances in which communication occurs in two or more languages in and around writing’ (Hornberger, 1990: 213). Although multiliteracy is related to the goals of language achievement described above, these do not always coincide. For example, programmes for the revitalisation of a threatened language may aim at full multilingualism but monoliteracy; that is, successful students will be multilingual, but literate only in one of the languages. This is also commonly found in some transitional programmes for immigrant communities. Receptive or partial multiliteracy may also be the objective of programmes which favour some skills rather than others; in most cases receptive skills are prioritised and productive skills are considered as secondary. The ideal multilingual programme is that in which the goal is the best possible command of all skills in all languages, although this ideal is almost non-existent. It is often assumed that literacy acquisition is a cognitive process, which is universal and culturally neutral. However, reality shows that educational systems, usually at a national level, follow standardised procedures to teach children to read and write in a certain language. When a second or third language is introduced into the system, there is a tendency to use the same procedure as with the first language, which is not always the most appropriate. For example, the way to learn how to read in Spanish is syllabic (pa-pepi-po-pu leads in a further step to read pa-pa, pe-pa, pi-pa, which are already meaningful words in Spanish). However, the first step in learning to read in English is phonetic (the combination of /m/ /æ/ and /n/ gives /mæn/) (Denton et al., 2000). Starting from these first stages of literacy acquisition, sociocultural contexts play a more and more important role in the nature of the literacy practices needed by multilinguals in their environment. Lüdi (1997) proposes a general definition of biliteracy: ‘the possession of, or access to, the competences and information required to accomplish literacy practices which an individual wishes – or is compelled to – engage in everyday situations in two languages, with their corresponding functions’. The student does not only have to learn to read and write in the appropriate way in the various languages, but also to choose the most suitable language for each situation (or even a combination of both, or several of them, in certain

Mult ilingual Educat ion

31

situations), the most suitable style, and the most suitable genre. In addition, the provenance of the text should also give the reader enough clues to make the right interpretation of what the writer wrote, and also of his or her intentions when writing it.

2.3.3 Content achievements It is difficult to measure the content achievements of students in multilingual education programmes due to several circumstances: many of these programmes are only implemented with selected groups of students, who may have been pulled out of the programmes when their academic results fell below expectations (Garcia, 2009), and there are very few studies of academic results comparing students in multilingual programmes with control groups with similar characteristics. Additionally, we could also presume here a political or ideological bias, since most researchers are in favour of multilingual education; negative results, especially regarding content achievements, would not benefit the future support of those programmes, which often find some resistance by parts of the population. In general, the expected outcome of a multilingual education programme regarding content achievement is to have the same results as monolingual standard programmes. The benefits of the programmes are not expected to come from content achievement, but from language achievements; that is why the former has not been given so much attention, and it is often assumed that if the linguistic achievements are satisfactory, the content achievements should be as well. As an exception, some studies by Cenoz (2009b) compare L2 acquisition with achievements in the acquisition of content in subjects taught in the L2 and even in L1 competence. Results indicate that using a minority language as the language of instruction in multilingual education programmes leads not only to better proficiency in the minority language, but also to similar levels in other subjects such as mathematics or the majority language. However, as stated above, additional factors such as motivation or the background of the students should be carefully accounted for in this kind of study.

2.4 Multilingual Education Approaches In this section a number of models of multilingual education programmes will be reviewed. As will be seen, there are frequent points in common between them, although each has been designed to fulfil specific objectives to solve the problems or the ambitions of a larger or smaller group of people in a particular part of the world.

32

Part 1: Mult ilingualism and Mult ilingual Educat ion

2.4.1 Language across the curriculum Although the language across the curriculum approach is not related to the development of multilingualism, I have decided to mention it here briefly due to the influence it has had on subsequent approaches first on second or foreign language learning, and then on CLIL. This approach started after the report of a committee on a study of all aspects of English teaching in Britain in 1975 (A Language for Life, 1975)3. The most important recommendation of the report was that first language instruction in the education system in Britain should involve all subjects; that is, language should also be part of the instruction in other curricular areas. These recommendations have been very influential in the last 30 years not only in British but also in North American educational systems, affecting primary and secondary schools, as well as in higher education where more attention has been paid to the relationship between content and language learning. Students are encouraged not only to learn to write and to read, but also to write and read to learn. It is obvious that establishing this aim implies a close cooperation between language and subject-matter teachers (Brinton et al., 1989). As a result of this movement, there was a development in research and materials design focused on the language skills required for advanced schooling, such as understanding and expressing complex decontextualised content in reading and writing. Such a research trend has continued today, dealing both with first and second (foreign) language teaching, as can be seen in publications such as Carson et al. (1990), Cumming (1998), Dudley-Evans (1995), Herrington and Moran (1992), Hyland (2002), Johns (1997), Schunk and Zimmerman (1997) and Silva and Brice (2004).

2.4.2 Languages (English) for specific purposes This approach also saw its major development during the 1970s and 1980s, and though it was originally developed in Britain and for the teaching of English, it was soon exported to other countries, first by English teachers working for the British Council in Commonwealth countries, and then to other mainly European countries, for the teaching of English as a foreign language and for other languages. English for specific purposes (ESP) is seen as an approach rather than a product, and is based on the needs of the learners (Hutchinson & Waters, 1987). Strevens (1988) establishes some basic characteristics of ESP: • •

It is designed to meet specific needs of the learner. It is related in content to particular disciplines, occupations and activities.

Mult ilingual Educat ion

• •

33

It is centred on language appropriate to those activities in syntax, lexis, discourse, semantics, etc. One of the main research methodologies to find the specific characteristics of these activities is discourse analysis.

ESP courses develop from a needs analysis, which ‘aims to specify as closely as possible what exactly it is that students have to do through the medium of English’ (Robinson, 1991: 3). In addition, ESP makes use of the underlying methodology and activities of the disciplines it serves, while being centred on the language skills, discourse and genres appropriate to these activities (Dudley-Evans & St. John, 1998). This approach has been followed in many countries around the world, especially at universities and in occupational settings, and has produced a great deal of research, as reported in Robinson (1991) for early research, and in more recent publications such as Arnó-Macià et al. (2006), Fortanet-Gómez (2010a), Fortanet-Gómez and Räisänen (2008) and Gotti (2003). One of the branches of ESP that deserves special attention is English for academic purposes (EAP), which has undergone further development as related to the specific needs of students involved in multilingual programmes. EAP has as an aim assisting learners in their study and research through English (Flowerdew & Peacock, 2001; Hyland, 2006). EAP received a strong influence from the language across the curriculum approach, which focused teachers’ and researchers’ attention on language proficiency, especially that of students in higher education. EAP was fostered by the internationalisation of universities in the 1980s and 1990s, especially of those in English-speaking countries, and eventually it was also implemented in many other countries where English is not the official nor the dominant language, but which incorporate it as the medium of instruction. Although it is not limited to higher education, most studies on EAP focus on the university academic context. In this respect, Hyland (2006: 1) lists the areas of communicative practice EAP analyses: • • • • •

pre-tertiary, undergraduate and postgraduate teaching (from the design of materials to lectures and classroom tasks); classroom interactions (from teacher feedback to tutorials and seminar discussions); research genres (from journal articles to conference papers and grant proposals); students’ writing (from essays to exam papers and graduate theses); administrative practice (from course documents to doctoral oral defences).

34

Part 1: Mult ilingualism and Mult ilingual Educat ion

These new academic contexts have evidenced a growing need for academic discourse tuition, especially at university level. English-speaking countries have found it necessary to teach this type of discourse to their students even when English is their mother tongue. Today, it is also essential in non-English speaking countries where a growing number of courses are being offered in English as part of multilingual programmes. Even when the medium of instruction chosen is not English, many materials, especially those related to research, can only be found in this language. For this reason there is also a need to learn EAP for non-native English-speaking academics who must gain confidence in order to be able to perform academic tasks in that language. As Hyland reports: The professional and institutional expectations of these academics are closely aligned with those in the ‘metropolitan’ English-languagespeaking countries and whether the academic is a native or non-native user of English is seen as immaterial to the roles they play and the jobs they perform. (Hyland, 2006: 3) EAP is today an area of expanding research, as can be seen in publications such as Burgess and Martín-Martín (2008), Del Lungo Camiciotti and Tognini Bonelli (2004) and Hyland and Bondi (2006).

2.4.3 Content-based instruction Content-based instruction (CBI) started in the USA in the 1970s, though it was not widely applied until the late 1980s. Grabe and Stoller (1997) discuss early versions of CBI with ESP programmes, L2 immersion programmes for K-12 students, university-level foreign language instruction, and some multilingual contexts in Europe. According to Brinton et al. (1989): in a content-based approach, the activities of the language class are specific to the subject matter being taught, and are geared to stimulate students to think and learn through the use of the target language. (Brinton et al., 1989: 2) CBI has been linked from its beginning to limited English-proficient (LEP) students, i.e. immigrant children and adolescents who found education in English too difficult to follow and frequently abandoned their schooling. This became an important problem, especially in areas where immigrant children were a majority in the schools.

Mult ilingual Educat ion

35

CBI draws from and develops in parallel with other approaches such as language across the curriculum, language for specific purposes (LSP) and immersion programmes, also explained in this chapter. Brinton et al. (1989) describe three content-based models which have been applied to high secondary school and university levels: • • •

theme-based language instruction; sheltered content instruction; adjunct language instruction.

These three models share a number of features: content is the main focus of the course; only authentic tasks and materials are used; and, in order to help students comprehend the content materials, there should be some degree of accommodation to the learners’ needs, especially in teacher classroom discourse, type of assignments and assessment procedures. Theme-based language instruction is closely related to LSP. It includes language courses based on a certain topic; this can be a major topic such as world economics, and then be subdivided in more specialised topics, or be restricted to a specific topic. The duration of these courses depends on the topic and their relevance in the curriculum. Short courses are often restricted to a single activity or skill, such as reading comprehension or assignment writing, whereas long courses tend to include all skills. One of their distinctive characteristics is that materials are usually generated by the teacher or adapted from original sources. These courses are generally rather demanding in terms of language processing, since students have to deal with a number of spoken and written genres used in their disciplines. The primary purpose of theme-based language instruction is to develop competence in the second language within specific topic areas, and it is usually a language teacher who is in charge of it. Sheltered content instruction consists of content courses taught in a second language to a particular group of students by a content area specialist, usually a native speaker of the target language. A group of L2 learners is segregated from the native-speaking students as a ‘sheltered’ group, in which the teacher makes some adjustments and modifications especially regarding class discourse, making it more organised and giving more emphasis to receptive skills and less to speaking and writing. The focus of these courses is on content mastery; language learning is only incidental, though the content teacher needs to be aware of L2 acquisition processes. In adjunct language instruction, students are enrolled in two related courses – a language course and a content course. Both courses share the same content as a base and are complementary, often sharing the same assignments. Second language learners are encouraged to take both, while

36

Part 1: Mult ilingualism and Mult ilingual Educat ion

native-speakers are only registered in the content course. This type of instruction requires a great deal of coordination between language and content teachers. The objective of the adjunct model is to assist L2 students in mastering content material as well as to introduce students to L2 academic discourse and academic skills. In order to achieve this objective, language and content teachers need training in curriculum coordination and syllabus design, materials development and team teaching. These three models have been implemented successfully in the USA and in other parts of the world since the 1980s (Brinton et al., 1989), and other developments have followed that stem from or combine these basic models, such as CLIL (Section 2.5).

2.4.4 Immersion programmes Drawing from Swain and Johnson (1997), Lyster (2007: 9) defines an immersion programme as that in which ‘students’ exposure to the second language tends to be restricted to the classroom where it serves as a medium of subjectmatter instruction’. Immersion teachers are usually bilingual, and students share as their main language of communication a majority language which is the one used outside the classroom, so there is a home–school language switch. What have been called immersion programmes started in Quebec, a French-speaking province in Canada, which is a country with geographical bilingualism, i.e. composed of two monolingual areas speaking two different languages, French and English. Quebec became a monolingual province during the 1960s, when the Charte de la Langue Française was approved. This was the end of a period of struggle by the majority French-speaking population to gain control of economic and political power in the province. At this time, the English-speaking community realised they should become proficient in French in order to use it as a working language and provide their children with teaching methods that could guarantee quick and efficient learning of the French language, which they could not attain through traditional L2 teaching. The first immersion programme in French for Englishspeaking children was started by parents in Saint-Lambert, a middle-class neighbourhood of Montreal, in collaboration with a group of psychologists from McGill University. The experience was described by Lambert and Tucker (1972). At that time, experiments on the effect of early bilingual experiences were starting to demonstrate the advantages of bilingual education. Hamers and Blanc (2000: 332) classify immersion programmes into three main types: (1) Early total immersion. Children receive all their tuition in the L2 (French in the Saint-Lambert project), from nursery to the second year of

Mult ilingual Educat ion

37

elementary school. From then on, the time devoted to instruction in French diminishes and is only about 50% at the end of elementary school, when some subjects are taught in English and others in French. In the final years of elementary school, they can choose the subjects they prefer in each language. No particular methodology is followed to introduce both languages. (2) Early partial immersion. Both languages are used as the medium of instruction from the beginning of schooling. The percentage of time devoted to each language varies depending on the programme. (3) Late immersion. Programmes designed for high school students who have only had the established L2 subjects in primary education. In the Canadian case, during the first year 85% of the teaching time is in French, and only 15% in English. During the next years the student can attend 40% of the classes in French, but only as an option (Swain & Lapkin, 1982). Since immersion programmes started as a joint project between schools and university research teams, they have been studied and assessed from the beginning. The results of these studies have shown that immersion programmes help students to learn a second language without any negative effect on the learning of the mother tongue or on content subjects (Swain, 1982). There also seem to be indications that early total immersion gets better results than partial or late immersion (Morrison, 1981), as well as the fact that this type of programme may favour the child’s cognitive development (Barik & Swain, 1978; Scott, 1973). The most relevant negative result of the studies carried out on immersion programmes seems to have been the use of the language learned – French – outside the school context. After developing a competence that allows them to interact in that language, learners lack the necessary social motivation to use it as a native speaker (Blanc, 1980). The teaching methods which do not encourage children to be participative and the lack of opportunities to use the language with native speakers outside school were held responsible for these drawbacks (Swain & Lapkin, 1991). In Spain, immersion programmes have been used for the maintenance and spread of regional languages. The most extended language apart from Spanish, the state language, is Catalan. This is a Romance language, autochthonous of a geographical area on the eastern part of Spain covering Catalonia, Valencia and the Balearic Islands. Catalan is also spoken in Andorra and, to a lesser extent, in the South of France and in the Sardinian city of Alghero in Italy. The situation of the Catalan language has historically depended on the general politics of the central government of Spain. Since the 18th century until the second part of the 20th century, after the

38

Part 1: Mult ilingualism and Mult ilingual Educat ion

end of the military dictatorship, successive governments promoted the linguistic homogenisation of the country. The situation changed with the introduction of a constitutional, parliamentary monarchy in 1975, and the adoption of a territorial structure which divides Spain in autonomous regions. These regions are allowed to implement their own language policies, which has led to the development of Catalan language education policies in the three Spanish autonomous regions where this language is spoken (Vila i Moreno, 2009). In 1994 Catalonia adopted what was known as the ‘Model of Conjunction in Catalan’, which makes Catalan the usual medium of instruction. Initial instruction in Spanish is given only when applied for by parents, and this is rarely done. Teaching in Catalan accounts for over 90% of primary schools, 50% of secondary schools and about 60% of tertiary education4 in Catalonia. In the Balearic Islands the introduction of Catalan as the language of instruction has only recently been implemented. In 2003 over 50% of infant and primary schools were teaching in Catalan, and in secondary education it was the language of instruction in about 70% of the subjects. Nowadays there are some movements to introduce other languages of instruction, such as English. There is discussion as to whether this could lead to a decrease in the number of hours taught in Catalan (Vila i Moreno, 2008). Valencia is an autonomous region divided in two linguistic areas; in the western area Spanish is spoken, whereas in the eastern part, Valencian, the local variety of Catalan, is most commonly used. The school system is organised into several programmes5: •

•

•

Teaching in Valencian programme (Programa d’Ensenyament en Valencià [PEV]). This is addressed to pupils whose mother tongue is Valencian. The language of tuition is Valencian. Spanish is progressively introduced, and eventually a third language, usually English, is taught. This programme is also implemented in secondary education. Linguistic immersion programme (Programa d’Immersió Lingüística [PIL]). This is addressed to L1 speakers of Spanish. Valencian is the language of tuition and Spanish is incorporated progressively when pupils are 6–8 years old. This programme has continuation into secondary education. Progressive incorporation programme (Programa d’Incorporació Progressiva [PIP]). Tuition is basically in Spanish, with the progressive introduction of Valencian from the third year in primary education through the whole of compulsory secondary education.

The results shown by the autonomous government indicate an important increase in the number of centres offering tuition in Valencian. However, it

Mult ilingual Educat ion

39

is difficult to find reports and statistics about the process of implementation and the outcomes. The introduction of Catalan in schools is nowadays accepted and well considered, though there are still a minority who reject it. Most students in the Catalan-speaking areas can use both Catalan and Spanish without any problem, at least in academic contexts. However, it has been reported that the use of Catalan out of school when there is not a majority of Catalan speakers is not usual. There is a tendency to switch to the state language, a strategy known as ‘automatic convergence language norm’ (Vila i Moreno, 2009). Schools and authorities expect that this situation will change in the future, by extending the use of Catalan to leisure activities, such as sports, cinema, music, etc. Galician is a Romance language with Celtic origins closely related to Portuguese, used in northwestern Spain. The consideration of Galician as a language of education in Spain is rather recent. Although it had been included as a language subject for over a decade, it was not until 2007 that a decree6 established some order into the use of the language in the public education system. This decree establishes that in infant education the medium of instruction should be mainly the mother tongue of most of the students, although Galician should be used at least as much as Spanish. In primary education, it is obligatory to teach mathematics, natural environment, social and cultural education, and education for citizenship and human rights in Galician, and the rest in Spanish. In obligatory secondary education (until 16 years of age), it is natural science, social science, geography and history, mathematics and education for citizenship that must be taught in Galician. In secondary education leading to the Baccalaureate, at least 50% of the teaching time should be in Galician; this also applies to occupational education and adult education. There is no norm regulating language use in tertiary education. Basque is the only non-Indoeuropean language in western Europe, and its origin is unknown. It is used along the Bay of Biscay, north and south of the Pyrenees in France, and in Spain, respectively. Multilingual education in Basque and Spanish or even including French had already occurred in the Basque Country in the 19th century. However, during Francisco Franco’s dictatorship it was completely banned from education. During the 1960s groups of parents and teachers started some ‘illegal’ Basque-medium schools (Cenoz, 2009b). Some years after the dictatorship ended, in 1982, Basque was recognised as an official language in the Basque country, and three models of language schooling were implemented: Model (A). Intended for native speakers of Spanish. Basque is a second language subject taught during the education system for 4–5 hours a week.

40

Part 1: Mult ilingualism and Mult ilingual Educat ion

Model (B). Also intended for native speakers of Spanish who wish to be multilingual in Basque and Spanish. Both Basque and Spanish are used as a medium of instruction, sharing 50% of the teaching time, though some variations can be found when comparing schools. Model (D7 ). Intended for Basque speakers. Basque is the language of instruction and Spanish a second language subject taught for 4–5 hours a week. At the moment, this model has also attracted a large number of Spanishspeaking students, and it is the model with most students in the Spanish Basque country. The three models are available both in public and in private schools all over the territory. In many bilingual schools, especially in the Basque country, English has been introduced from an early age with no adverse effect on the learning of the other two languages (Baker, 2001).

2.5 Content and Language Integrated Learning 2.5.1 Definition and origins Content and Language Integrated Learning (CLIL) has been defined as ‘a dual-focused approach in which an additional language is used for the learning and teaching of both content and language’ (Maljers et al., 2007: 8). Following this definition, what is characteristic of CLIL is the use of an ‘additional language’ in the classroom, and the combination of teaching/ learning content and language at the same time. First of all, the meaning of an ‘additional language’ should be given some thought. The CLIL approach has two main purposes, to support minority languages and to reinforce the teaching and learning of foreign languages, as stated in the definition of CLIL in the Eurydice Report, Content and Language Integrated Learning (CLIL) at School in Europe (2006)8: ‘CLIL is a generic term to describe all types of provision in which a second language (foreign, regional, minority, official language) is used to teach non-language subjects’. However, for many practitioners and researchers, CLIL is associated exclusively with foreign languages (especially English), as suggested by the expression used in Marsh (2002: 58) for whom the term CLIL involves activities in which ‘a foreign language is used as a tool in the learning of a non-language subject’. Taking an external perspective from outside the EU, Garcia (2009: 208) considers CLIL ‘an umbrella term that embraces any type of program where a second language is used to teach non-linguistic content-matter’, but which excludes ‘minority or lesser-used languages, where the entire curriculum is given in these

Mult ilingual Educat ion

41

languages for their speakers’. This definition follows the tendency described above, by excluding some languages considered ‘minority and lesser-used’. From my point of view, this definition could be controversial in Europe, where even some national languages, such as Icelandic, Estonian or Maltese, with less than one million speakers, could be included in the group of ‘lesserused languages’. Nevertheless, García’s definition tackles a relevant issue: it is not just a question of which languages are used in the classroom, but the approach chosen for the teaching and learning of those languages. On the other hand, CLIL has most often been associated with pre-school, primary, secondary and vocational education. Notwithstanding, there is no reason why CLIL should be relegated to only the first stages of the education system; quite the reverse, it has been proved that one of the secrets of success for CLIL is continuity along the education process. On that account, during the conference held in Maastricht (The Netherlands) in 2003 about CLIL in Higher Education, a new term was coined: integrating content and language in higher education (ICLHE). According to some researchers, there was a reason behind having a different term for tertiary education: at this stage language learning is not so relevant and very often the balance is more on content (Smit, 2009). Another unexpressed argument may have been the need to include more international views on the matter, even from scholars who could have related the term CLIL only to European experiences. Be that as it may, CLIL seems to have already become a familiar term at an international level, as is proved by the existence of a Latin American Journal of Content and Language Integrated Learning.9 On the other hand, for the purposes of this book, we will follow Wilkinson’s (2004b: 10) premise that learning content through a foreign language ‘involves the conscious design of programmes that integrate both content and language goals’, referring to all languages adopted as media of instruction provided they are ‘additional languages’ which are used along with the majority language. As a point of departure, CLIL responds to a European need for multilingual education, and one of its main characteristics is that its major concern is communication rather than producing balanced multilinguals, in contrast with other approaches. The general premise is that ‘multiple language proficiency is an “added value” which can be obtained at no cost to other skills and knowledge, if properly designed’ (Baetens Beardsmore, 2002). According to Coyle (2002), every student should be given the opportunity to learn content effectively through the medium of a language not considered to be the usual language for subject instruction in the regular curriculum, so that all learners can benefit from this approach during compulsory education. This characteristic differentiates CLIL from other approaches where the aim is to scaffold students who have special needs in order that they can successfully

42

Part 1: Mult ilingualism and Mult ilingual Educat ion

join mainstream monolingual curricula. Four key principles (4Cs) have been established as essential in CLIL approaches (Coyle, 2002): (1) Successful content learning and the acquisition of knowledge, skills and understanding through another language. The close relationship between language and subject matter demands a focus on how content is taught through another language. This means reflection on the use of language, and on methodologies of teaching, which should include interaction and output as essential (De Bot, 2002). (2) Language use leads to communication and to learning. Language is learned by using it in authentic situations, though sometimes with the need of ‘scaffolding’. The aim is to enable students to operate successfully in multilingual settings. (3) CLIL should challenge learners cognitively to develop thinking skills in conjunction with language. (4) CLIL can support learning about other cultures through studies in an alternative language. Another specific characteristic of CLIL is that only a certain percentage of the curriculum is taught in the target language. Teaching a subject in a foreign language is not the same as the integration of language and content. CLIL requires differentiated objectives for language and content learning. Moreover, full immersion with the target language is very unusual. In CLIL classrooms, translanguaging and other characteristics found in multilingual speakers (see Chapter 1) are not only allowed but encouraged, since they are skills that also need to be acquired along with the target language and content. On the other hand, language and content teachers need to work together in order to coordinate the language and content aims, as well as the teaching methodology and the evaluation criteria. This is not easy, since it involves good personal and professional relationships, as explained in Chapter 6. Moreover, at the institutional level, it becomes necessary to incorporate multilingual tracks in the educational systems, which are offered from primary or even pre-school level up to tertiary education, since the benefits of CLIL obtained in one stage of education may vanish if there is no continuation in higher stages. Finally, the social and political context also needs to be taken into account. As Takala (2002) points out, ‘CLIL needs to be tailor-made to fit the national/local circumstances’. In order to fulfil this requirement, cooperation between all stakeholders: policymakers, educational authorities, and the academic community including teachers and students is essential. In addition, there is a need for social support, which will confer motivation for learning through this approach. This implies involving relevant members of

Mult ilingual Educat ion

43

society such as representatives from commerce and industry who may provide information about the needs of professionals for related knowledge of language and content, or teacher unions and parent organisations who might organise debates on multilingual education in order to circulate the main ideas about the CLIL approach (Baetens Beardsmore, 2002; Ruiz-Garrido & Fortanet-Gómez, 2009). Summing up, CLIL proposes a twofold pedagogical objective: subject matter and languages are given equal status. This implies the cultivation of a ‘global didactic concept’ or integrated instruction model, which involves the conscious design of programmes that integrate both content and language goals (Maillat & Serra, 2009; Wilkinson, 2004a). Content and language are placed in a continuum from a clearly content-based perspective to a language-based perspective, as can be observed in Figure 2.1 (p. 46). According to Räsänen (2008), adopting a CLIL approach presupposes that there are separate goals for content learning and language learning. Regarding language, as will be explained in Chapter 4, the perspective can be very different depending on the eye of the beholder. While it is a tool for interaction and strategic communication from the point of view of its use, for a language teacher and learner it may be a subject to be taught and learnt. For linguists, then, it is their discipline and object for research. Finally, for a CLIL academic professional, language should be seen as a tool and mediator for constructing knowledge and sharing expertise.

2.5.2 Differences between CLIL and other approaches CLIL draws from other approaches, such as immersion programmes (Artigal, 1993), bilingual education (Crawford, 1997), CBI (Brinton et al., 1989), DLBE (Lindhold-Leary, 2001) or sheltered content instruction (Echevarria & Graves, 2007). However, there are substantial differences between what is understood by these approaches. The question is: what makes CLIL different from the other approaches? In definitions like the following by Brinton et al. (1989) CLIL and CBI seem to be very close: More specifically, it [the integration of content learning with language teaching aims] refers to the concurrent study of language and subject matter, with the form and sequence of language presentation dictated by content material. The language curriculum is centred around the academic needs and interests of the students, crossing over the barrier between language and subject matter courses which exists in most secondary and post-secondary institutions. (Brinton et al., 1989: vii)

44

Part 1: Mult ilingualism and Mult ilingual Educat ion

However, as explained in Section 2.4.3, CBI is linked to LEP students in order to assist them to assimilate into mainstream education in English. This is one of the main differences between CBI and CLIL because, as was said above, CLIL is addressed to all students as part of mainstream education. However, together with immersion education, CBI is the approach that has contributed the most to CLIL. For example, the three CBI teaching models (Brinton et al. 1989: 14–23) have had a strong influence on the development of CLIL experiences. Immersion education has also traditionally been related to CLIL. One similarity between the two is that in immersion education at least 50% of the instruction is conducted through a non-native language, and another similarity is that every teacher communicates with his or her pupils through the target language. However, in immersion education, translanguaging is not allowed and the focus is on content learning through the target language. In contrast, in CLIL both language and content teaching and learning are regarded as equally important. The specific blend of content and language will be made according to the specific objectives of the educational context and both the language-learning and the content-learning objectives will be clearly identified. In addition, CLIL is comprised of some situations where all instruction is conducted through a foreign language and others when foreign-language enhanced education is only applied to certain parts of the curriculum, as for example in situations where the small-scale effort is to encourage students to participate in exchange programmes (Nikula & Marsh, 1998). In this sense, CLIL could be related to DLBE. The difference between these two approaches is that DLBE classrooms contain a mixture (which aims to be 50:50) of language majority and language minority students. Additionally, in DLBE the central idea is language separation and compartmentalisation, since in each period of instruction (days, subjects, classes, etc.) only one language is used (Baker, 2001). On the contrary, in CLIL programmes code-switching and translanguaging are not only allowed but also encouraged if they facilitate communication. The term multilingual or bilingual education is still in general use by many people, even when they refer to CLIL. The reason why bilingual education was not adopted as the accurate term for this approach is that it refers to just two languages, and has been widely used to refer to children who are brought up in bilingual family environments (Marsh, 2002). The objective of bilingual education is most often balanced bilingualism, although, as was observed in Chapter 1, this is an ideal situation, generally too difficult to achieve, and is never the aim of CLIL programmes. Another term widely used nowadays is trilingual education, especially with regard to bilinguals who learn a third language (Alcón, 2010; Cenoz, 2003; Safont, 2012). However,

Mult ilingual Educat ion

45

this term has the same problem as bilingual education, since it limits the number of languages involved. Thirdly, some scholars see many points in common between CLIL and EAP or ESP. As early as 1992, Ann Johns (1992) answered the question ‘what is the relationship between Content-Based Instruction and English for Specific Purposes?’ Although she did not refer directly to CLIL, the characteristics of CBI she referred to are shared by both approaches. For her, the first similarity is that both emerge as a response to the need to combine language teaching and the contexts and demands of real language use. On the other hand, the main difference is that CBI was conceived as a solution for foreign students who arrived in the USA with low proficiency in the English language. ESP, however, has always been related to foreign language learning, i.e. it has a more international perspective, something that is shared by CLIL. Additionally, as well as EAP, ESP is language centred, i.e. the main purpose is to learn the English language appropriate for certain situations. Räsänen (2008) considers CLIL as a continuum, where LSP finds a place towards the end of the continuum which is more language based as opposed to more content-based programmes (see Figure 2.1). This is the reason why many researchers focusing on ESP or EAP go a bit further and consider their research as part of CLIL (Fernández, 2009; Ruiz-Garrido & Fortanet-Gómez, 2009). After justifying the differences between CLIL and other existing multilingual approaches, there is a need to know how the term ‘content and language integrated learning’ came into being.

2.5.3 Origin and history After a number of years during which the different organisations within the EU had proved their interest in the teaching and learning of EU languages, in 1996 the concept of CLIL was first mentioned as a topic for a workshop by the Council of Europe. The topic of the workshop was phrased as ‘Bilingual education (both in situations where children from regional and migrant/refugee communities study alongside others from the majority community and in “mainstream” education where a foreign language is the medium of instruction in teaching other subjects)’.10 This was the first time regional and migrant/refugee communities had been cited along with foreign languages. CLIL was soon considered as an approach, targeted mainly at the promotion of the teaching and learning of the main languages of the EU, and only occasionally regional languages; migrant and refugee community languages were not considered in further developments. Through the late 1990s and the early years of the 21st century, CLIL was one of the topics under discussion

PRE-CLIL (content)

-language specialists providing discipline-specific LT to support learning -no (systematic) collaboration -with subject specialists, FL teacher chooses materials -possible power plays, role formation

-courses/programmes provided systematically by subject specialists to mixed, multicultural and multilingual groups (>25% exposure) -language learning expected due to exposure, but outcomes not specified; implicit aims and criteria -collaboration possible, but rare

PRE-CLIL (language)

Figure 2.1 Steps from non-CLIL to CLIL in higher education

-no concern for language learning, other agendas -no (pedagogical) collaboration -e.g. visiting experts, individual lectures -incidental, unsystematic, limited exposure (