Colin baker foundations of bilingual education

BILINGUAL EDUCATION & BILINGUALISM

Series Editors: Nancy H. Hornberger (University of Pennsylvania, USA) and Colin Baker (Bangor University, Wales, UK)

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.

BILINGUAL EDUCATION & BILINGUALISM
Series Editors: Nancy H. Hornberger (University of Pennsylvania, USA) and Colin Baker (Bangor University, Wales, UK)

Foundations of Bilingual
Education and Bilingualism

Fifth Edition
Colin Baker

MULTILINGUAL MATTERS
Bristol • Buffalo • Toronto

Library of Congress Cataloging in Publication Data
Baker, Colin
Foundations of Bilingual Education and Bilingualism/Colin Baker. 5th ed.
Bilingual Education and Bilingualism: 79
Includes bibliographical references and index.
Previous ed.: 2001.
1. Education, Bilingual. 2. Education, Bilingual--Great Britain.
3. Bilingualism. 4. Bilingualism--Great Britain. I. Title.
LC3715.B35 2011
370.117–dc 22 2011000603

British Library Cataloguing in Publication Data
A catalogue entry for this book is available from the British Library.

ISBN-13: 978-1-84769-508-6

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 © 2011 Colin Baker.

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 Bookcraft Ltd, Stroud, Gloucestershire.

Printed and bound in the United States of America by McNaughton & Gunn Ltd

Contents

Acknowledgements

Introduction

CHAPTER 1 Bilingualism: Definitions and Distinctions

CHAPTER 2 The Measurement of Bilingualism

CHAPTER 3 Endangered Languages: Planning and Revitalization

CHAPTER 4 Languages in Society

CHAPTER 5 The Early Development of Bilingualism

CHAPTER 6 The Later Development of Bilingualism

CHAPTER 7 Bilingualism and Cognition (updated by Panos Athanasopoulos)

CHAPTER 8 Cognitive Theories of Bilingualism and the Curriculum

CHAPTER 9 Historical Introduction to Bilingual Education: The United States (updated by Wayne Wright)

CHAPTER 10 Types of Bilingual Education

CHAPTER 11 Education for Bilingualism and Biliteracy

CHAPTER 12 The Effectiveness of Bilingual Education

CHAPTER 13 Effective Schools and Classrooms for Bilingual Students

CHAPTER 14 Literacy, Biliteracy and Multiliteracies for Bilinguals

CHAPTER 15 The Special Educational Needs, Assessment and Testing of Bilinguals

CHAPTER 16 Deaf People, Bilingualism and Bilingual Education

CHAPTER 17 Bilingualism and Bilingual Education as a Problem, Right and Resource

CHAPTER 18 Bilingualism and Bilingual Education: Ideology, Identity and Empowerment

CHAPTER 19 Bilingualism in the Modern World

Bibliography

Index

Acknowledgements

The idea of this introductory text derived from a fellow Essex gentleman, Mike Grover. He wrote one simple sentence in the early 1990s that has affected my academic life ever since: ‘Consider writing THE textbook on Bilingual Education’. My gratitude goes to Mike for not only trusting me with this responsibility, but also for his continual encouragement and friendship. Ken Hall of Multilingual Matters meticulously worked with me in the production of previous editions. Tommi Grover and Anna Roderick are two of the new generation in MLM constantly giving wise, astute, sensitive, tactful and efficient advice. Anna has worked with me in the production of this edition, and completion could not occur without her dedication and considerable expertise.

Multilingual Matters perceptively appointed Ofelia García as Academic Consul tant when this project began. I have always received detailed, sensitive, wise and judicious advice from her. Much gratitude is owed to Ofelia who helped to shape the book, including chapters in this edition. Muchas gracias.

In the previous editions, I recorded my grateful thanks to those who helped in the construction of those editions. I wish to repeat my sincere gratitude to: Hugo Baetens Beardsmore, Claudine Brohy, Jasone Cenoz, Tony Cline, Jim Crawford, Jim Cummins, Margaret Deuchar, Nancy Dorian, Viv Edwards, Peter Garrett, Tamar Gollan, Nancy Hornberger, Annick De Houwer, Sharon Lapkin, Hilaire Lemoine, Christer Laurén, Karita Mard, Stephen May, Aneta Pavlenko, Bernard Spolsky, Merrill Swain, Terry Wiley and Iolo Wyn Williams.

For the fifth edition, various scholars constructively gave expert advice on improvements and needed developments: Panos Athanasopoulos, Jasone Cenoz, Viv Edwards, Annick De Houwer, Ofelia García, Nancy Hornberger and Wayne Wright. Panos Athanasopoulos and Wayne Wright took over responsibility for the revision of two chapters (seven and nine). Jeanne Clark kindly provided the support needed with checks and corrections, bibliography and graphics, indexes and insertions. Crede quod habes, et habes. Noah Cameron, my PA, helped enthusiastically with computers, calculations and crises. Sylvia Prys Jones wrote the Encyclopedia of Bilingualism and Bilingual Education (1998) with me, and during three years of cooperation, I learnt much from such a brilliant intellect. The chapter on ‘Bilingualism in the Modern World’, and the discussion on one-parent bilingual families owes much to Sylvia’s superlative contribution to the Encyclopedia. Continuous dialogue with Meirion Prys Jones, Chief Executive of the Welsh Language Board, Dr Cen Williams, and Dr Gwyn Lewis, both colleagues at Bangor University, has been a constant source of encouragement and enlightenment. Diolch i chi.

The context for writing and researching on bilingualism has been both Bangor University and home. I have worked with a succession of colleagues (e.g. W.R. Jones, Iolo Wyn Williams, S.P. Jones, Gwyn Lewis), and lately in the ESRC Bilingualism Research Centre at Bangor University, that has stimulated and challenged my thinking. Working on bilingualism has always been inspired by living in a bilingual family. My Welsh wife, Anwen, and our three bilingual offspring (Sara, Rhodri and Arwel) have each taught me the gifts of bilingualism that go beyond language. Two grandsons (Ioan Tomos and Joseff Rhys) have allowed me to observe closely without the responsibility of parenthood. Their loving support is always appreciated. Diolch yn fawr iawn am bopeth.

The help and support given me by all those mentioned above has been extremely generous and far more than is deserved. However, the responsibility for all that is not perfect is totally mine.

Colin Baker

Permissions: Every attempt has been made to contact copyright holders and gain permissions where needed. If there are any omissions, we will be pleased to correct them in a future edition.

Introduction

The fifth edition of this book is intended as a comprehensive and modern introduction to bilingual education, bilingualism and multilingualism (as bilingualism often includes multilingualism). Written from a multidisciplinary perspective, the book covers a wide range of topics: individual and societal concepts in minority and majority languages; childhood developmental perspectives; general bilingual education issues; bilingual classrooms, and political and ideological perspectives. Bilingualism and multilingualism relates to, for example, the use of two or more communication systems, identity and personality, globalization and assimilation, thinking and reading, education and employment, politics and culture. All of these, and more, are encapsulated in this book.

In compiling five editions, increasingly tough decisions have been made as to what to include and exclude, what to present in detail and what to summarize, what assumptions to explore and what to take ‘as read’. I have often been asked: ‘Why don’t you put Chapter X earlier?’ I agree, everything should be earlier. Other frequently asked questions are ‘Why don’t you expand on Y?’, ‘Why don’t you leave out Z because it is irrelevant to me?’, ‘Why can’t we have an edition just for our region?’, and ‘Why isn’t a chapter in an earlier edition still included?’ What follows are some explanations. An attempt is made to balance the psychological and the sociological; macro and micro education issues; the linguistic and the sociopolitical with discussion at individual and societal levels; and to be inclusive of major international concerns in bilingualism, multilingualism and bilingual education, based on research evidence and theory. Faced with the social and political challenges that surround bilinguals, students will find in this book an attempt to analyze constructively those problems and recognize the positive values and virtues of a future multilingual world.

This book starts with definitional, sociological and psychological issues that are essential to understanding bilingual/multilingual children, and bilingual education. Later discussions of bilingual education and bilingual classrooms are built on that foundation. However, the book is more than a cross-disciplinary foundation with a series of education layers built on top. Within the boundaries of clarity in writing style and structuring, explicit inter-connections are made between chapters.

In writing the book, a constant challenge has been ‘From whose perspective?’ There are majority language mainstream viewpoints, relatively advantaged minority language viewpoints and various disadvantaged minority language viewpoints. There are left-wing and right-wing politics, activist and constructivist ideas, debates about globalization, Europeanization, regionalism and preserving grass-roots. The book attempts to represent a variety of viewpoints and beliefs. Where possible, multiple perspectives are shared. Readers and reviewers (including on YouTube) have kindly pointed out some of the hidden and implicit assumptions made, and kindly provided alternative viewpoints that I have tried to represent faithfully in the text. Where there are conclusions and dominating perspectives, I alone stand responsible.

Another issue has concerned generalization and contextualization. The book was written for an international audience to reflect ideas that transcend national boundaries. The book attempts to locate issues of international generalizability. Unfortunately, space limits discussion of a variety of regional and national language situations. There are many other writings mentioned in the chapters that will provide necessary contextualization. Where particular situations have been discussed (e.g. US debates), it is often because of the thoroughness of documentation, plenty of research evidence, and the depth of analysis in the surrounding literature.

In an attempt to make the contents of this book relevant to a variety of contexts and regions, various chapters focus on integrating theories. From one individual research study, it is usually impossible to generalize. A study from Europe may say little about North America. Results on six-year-olds may say nothing about 16 or 60-year-olds. Research on middle-class children speaking French and English in a bilingual school may say little or nothing about children from a lower social class in a bilingual environment where the second language is likely to replace the first language. From a mass of research, but occasionally despite a paucity of research, a theoretical framework will attempt to outline the crucial parameters and processes. Thus a theoretical framework on a particular area of bilingualism may attempt to do one or more of the following: attempt to explain phenomena; integrate a diversity of (apparently contradictory) findings; locate the key parameters and interactions operating; be able to predict outcomes and patterns of bilingual behavior; be capable of testing for falsification or refinement; express the various conditions that will allow the theory to be appropriate in a variety of contexts.

However, in providing a relatively comprehensive synthesis of bilingualism, multilingualism and bilingual education, the danger lies in suggesting that there is a systematic coherence to the subject. While some teachers and many students want ‘recipes’ and clear assertions, the current state of our knowledge and understanding rarely provides that clarity. The book therefore attempts to represent contested positions, varied viewpoints and the limitations of research and theory.

Instructors, in particular, will wish to know what are the specific changes in the fifth edition. First, there are many minor changes. For example, text and references have been updated (e.g. there are over 400 new references), research findings have been judiciously added, and modifications made based on the most recent evidence. All websites were accessed and checked in November 2010. Second, and importantly, there are new or more thoroughly covered topics: endangered languages, language planning, early childhood bilingualism, cognition and ‘neurobilingualism’ (brain and bilingualism) with the help of Panos Athanasopoulos, recent changes in bilingual education in the United States (with the help of Wayne Wright), the assessment of bilinguals, bilinguals with additional learning needs, CLIL (Content and Language Integrated Learning), effective practices in bilingual schools and classrooms, and globalization.

Wayne Wright (University of Texas at San Antonio) and Panos Athanasopoulos (Newcastle University) very kindly joined me in this venture. Wayne took over responsibility for Chapter 9 – the historical introduction to bilingual education in the United States – with an excellent revision. Panos similarly took over responsibility for Chapter 7 on bilingualism and cognition and created a much more modern and wide-ranging text.

There is a Reader to accompany this book entitled Bilingual Education: An Introductory Reader (2006) published by Multilingual Matters. This Reader contains a selection of the most important and influential contributions on bilingualism and bilingual education and is edited by Ofelia García (City University of New York) and myself. Each is a ‘classic’ or a pivotal paper.

Organization of the Book
The starting point of the book is an introduction to the language used in discussing bilingual education and bilingualism. Not only are important terms introduced, but also key concepts, distinctions and debates which underpin later chapters are presented. There are important dualisms and paradoxes throughout the study of bilingualism and multilingualism: for example, the individual bilingual person as different from groups and societies where bilinguals live; the linguistic view compared with the sociocultural and sociopolitical view; language skills and language competences; subtractive and additive forms of bilingualism. The opening chapters (1 to 8) present foundational issues that precede and influence discussions about bilingual education. Before we can sensibly talk about bilingual education we need to tackle questions such as:

Who are bilinguals and multilinguals?
How does bilingual education fit into minority language maintenance, language decay and language revival?
How does a child become bilingual or trilingual?
What effect does the home and the neighborhood play in developing bilingualism and multilingualism?
Does bilingualism have a positive or negative effect on thinking?
What do we know about bilingualism in the brain?
Chapters 9 to 16 focus on the many aspects of bilingual and multilingual education. They commence with a broad discussion of different types of bilingual education, followed by an examination of the effectiveness of those types. After a focus on systems of bilingual education, the book proceeds to examine bilingual classrooms, multiliteracies and biliteracy and key bilingual education strategies. The underlying questions are:

Which forms of bilingual education are more successful?
What are the aims and outcomes of different types of bilingual education?
What are the essential features and approaches of a classroom fostering bilingualism?
What are the key problems and issues of bilingual classrooms?
Why are Deaf people an important group to study both as bilinguals and as recipients of bilingual education?
Chapters 17 and 18 are central to understanding bilingualism, multilingualism and bilingual education. They consider the political and cultural dimensions that surround bilingualism in society (and bilingual education in particular). Different views of the overall value and purpose of bilingualism and multilingualism join together many of the threads of the book. The finale of the book (Chapter 19) takes a look at the present and future with themes of multilingualism and the internet, employment, mass media, economy and tourism.

Thus the concluding issues of the book include:

Why are there different viewpoints about language minorities and bilingual education?
Why do some people prefer the assimilation of language minorities and others prefer linguistic diversity?
Study Activities
Study activities are placed at the end of each chapter. These are designed for students wishing to extend their learning by engaging in various practical activities. Such activities are flexible and adaptable. Instructors and students will be able to vary them according to local circumstances. More study activities are found in the Reader that accompanies this text entitled Bilingual Education: An Introductory Reader (2006) published by Multilingual Matters and also in Bilingualism: An Advanced Resource Book (2007) by Chin and Wigglesworth.

Finale
To end the beginning. The motivating force behind this book is to introduce students to the modern and ever-increasing globalization of bilingualism, multilingualism and bilingual education. The book has been written for minority language students as well as multilingual, bilingual and monolingual majority language students. The book is an attempt to contribute to the preservation and celebration of a linguistically and culturally diverse world.

CHAPTER 1
Bilingualism: Definitions and Distinctions

Introduction
Terminology
Some Dimensions of Bilingualism
An Individual’s Use of Bilingualism
Language Choice
Bilingual and Multilingual Ability
The Four Language Abilities
Minimal and Maximal Bilingualism
Balanced Bilinguals
The Monolingual View of Bilingualism
‘Semilingualism’/’Double Semilingualism’
The Holistic View of Bilingualism
Conversational Fluency and Academic Language Competence
The Structure of Language Competence
Bachman’s Model of Language Competence
Conclusion

CHAPTER 1
Bilingualism: Definitions and Distinctions

Introduction
Since a bicycle has two wheels and binoculars are for two eyes, it would seem that bilingualism is simply about two languages. Multilingualism is then about three or more languages. The aim of this chapter is to show that the ownership of two or more languages is not so simple as having two wheels or two eyes. Is someone bilingual if they are fluent in one language but less than fluent in their other language? Is someone bilingual if they rarely or never use one of their languages? Such questions need addressing before other topics in this book can be discussed.

To understand the answers to these questions, it is valuable to make an initial distinction between bilingualism and multilingualism as an individual characteristic, and bilingualism and multilingualism in a social group, community, region or country. Bilingualism and multilingualism can be examined as the possession of the individual. Various themes in this book start with bilingualism as experienced by individual people. For example, a discussion of whether or not bilingualism affects thinking requires research on individual monolinguals, bilinguals and multilinguals.

From sociology, sociolinguistics, politics, geography, education and social psychology comes a ‘group’ perspective. Bilinguals and multilinguals are usually found in groups. Thus linguists study how the vocabulary of bilingual groups changes across time. Geographers plot the density of bilinguals and minority language speakers in a country. Educationalists examine bilingual educational policy and provision for particular language groups. Such groups may be located in a particular region (e.g. Basques in Spain), or may be scattered across communities (e.g. the Chinese in the US).

The initial distinction is therefore between bilingualism (and multilingualism) as an individual possession and as a group possession. This is usually termed individual bilingualism and societal bilingualism. Like most distinctions, there are important links between the two parts. For example, the attitudes of individuals towards a particular minority language may affect language maintenance, language restoration, downward language shift or language death in society. In order to understand the term ‘bilingualism’, some important further distinctions at the individual level are discussed in this chapter. (While bilingualism and multilingualism are different, where there is similarity multilingualism is [for the sake of brevity] combined under bilingualism.) An introduction to bilingualism and multilingualism as a group possession (societal bilingualism) is provided in Chapters 3 and 4.

If a person is asked whether he or she speaks two or more languages, the question is ambiguous. A person may be able to speak two languages, but tends to speak only one language in practice. Alternatively, the individual may regularly speak two languages, but competence in one language may be limited. Another person will use one language for conversation and another for writing and reading. An essential distinction is therefore between language ability and language use. This is sometimes referred to as the difference between degree and function.

Terminology
Before discussing the nature of language use and abilities, a note about terminology. To understand bilingualism and bilingual education, an awareness of often-used terms and distinctions is needed. For example, apart from language ability there is language achievement, language competence, language performance, language proficiency and language skills. Do they all refer to the same entity, or are there subtle distinctions between the terms? To add to the problem, different authors and researchers sometimes tend to adopt their own specific meanings and distinctions.

Some Dimensions of Bilingualism
Bilinguals and multilinguals can be analyzed along the following over-lapping and interacting dimensions:

A BILITY : some bilinguals actively speak and write in both languages (productive competence), others are more passive bilinguals and have receptive ability (understanding or reading). For some, an ability in two or more languages is well developed. Others may be moving through the early stages of acquiring a second language and are emerging bilinguals (O. García, 2009a). Ability is thus on a dimension or continuum (Valdés et al., 2003) with dominance and development varied across Speakers.
U SE : the domains (contexts) where each language is acquired and used are varied (e.g. home, school, street, phone, email). An individual’s different languages are often used for different purposes. For example, one language is used at home and another in school.
B ALANCE OF TWO LANGUAGES : rarely are bilinguals and multilinguals equal in their ability or use of their two or more languages. Often one language is dominant. This can change over time.
A GE : when children learn two languages from birth, this is often called simultaneous or infant bilingualism or ‘bilingual first language acquisition’ (De Houwer, 2009a). If a child learns a second language after about three years of age, the terms consecutive or sequential bilingualism tend to be used. Chapters 5 and 6 consider age issues in detail.
D EVELOPMENT : Incipient bilinguals have one well-developed language, and the other is in the early stages of development. When a second language is developing, this is ascendant bilingualism, compared with recessive bilingualism when one language is decreasing, resulting in temporary or permanent language attrition.
C ULTURE : Bilinguals become more or less bicultural or multicultural. It is possible for someone (e.g. a foreign language graduate) to have high proficiency in two languages but be relatively monocultural. A process of acculturation accompanies language learning when immigrants, for example, learn the majority language of the host country. Bicultural competence tends to relate to knowledge of language cultures; feelings and attitudes towards those two cultures; behaving in culturally appropriate ways; awareness and empathy; and having the confidence to express biculturalism.
C ONTEXTS : Some bilinguals live in bilingual and multilingual endogenous communities that use more than one language on an everyday basis. Other bilinguals live in more monolingual and monocultural regions and network with other bilinguals by vacations, phone, text messaging and email, for example. Where there is an absence of a second language community, the context is exogenous (e.g. Russian bilinguals in the US). Some contexts may be subtractive, where the politics of a country favors the replacement of the home language by the majority language (e.g. Spanish being replaced by English in the US). This particularly occurs among immigrant bilinguals (e.g. in the US and UK). Other contexts are additive such that a person learns a second language at no cost to their first language, as occurs in elite or prestigious bilinguals.
E LECTIVE BILINGUALISM is a characteristic of individuals who choose to learn a language, for example in the classroom (Valdés, 2003). Elective bilinguals typically come from majority language groups (e.g. English-speaking North Americans who learn French or Arabic). They add a second language without losing their first language. Circumstantial bilinguals learn another language to function effectively because of their circumstances (e.g. as immigrants). Their first language is insufficient to meet their educational, political and employment requirements, and the communicative needs of the society in which they are placed. Circumstantial bilinguals are groups of individuals who must become bilingual to operate in the majority language society that surrounds them. Consequently, their first language is in danger of being replaced by the second language – a subtractive context. The difference between elective and circumstantial bilingualism is important because it immediately locates differences of prestige and status, politics and power among bilinguals.
An Individual’s Use of Bilingualism
Grosjean (2010: 4) proposes a definition of bilingualism that places emphasis on the regular use of languages rather than fluency (as well as including multilinguals and those who speak a dialect): “bilinguals are those who use two or more languages (or dialects) in their everyday lives”

Language use cannot be divorced from the context in which it is used, nor from the effects of the interactions of different combinations of people in a conversation. Language is not produced in a vacuum; it is enacted in changing dramas. As props and scenery, audience, co-actors, the play and roles change, so does language. As the theatre and stage where we act changes, so does our use of two or more languages. Communication includes not only the structure of language (e.g. grammar, vocabulary) but also who is saying what, to whom, in which circumstances. One person may have limited linguistic skills but, in certain situations, be successful in communication. Another person may have relative linguistic mastery, but through undeveloped social interaction skills or in a strange circumstance, be relatively unsuccessful in communication. The social environment where the two languages function is crucial to understanding bilingual usage. Therefore, this section now considers the use and function of an individual’s two languages.

An individual’s use of their bilingual ability (functional bilingualism) moves into language production across a wide range of everyday contexts and events. Functional bilingualism concerns when, where, and with whom people use their two languages. The table below provides examples of the different targets (people) and contexts (often called domains) where functional bilingualism is enacted in different role relationships.

EXAMPLES OF LANGUAGE TARGETS EXAMPLES OF LANGUAGE CONTEXTS (DOMAINS)
1 Nuclear family 1 Shopping
2 Extended family 2 Visual and auditory media (e.g. TV, radio, DVD)
3 Work colleagues 3 Printed media (e.g. newspapers, books)
4 Friends 4 Cinema/discos/theater/concerts
5 Neighbors 5 Work
6 Religious leaders 6 Correspondence/email/telephone/official communication
7 Teachers 7 Clubs, societies, organizations, sporting activities
8 Presidents, Principals, other leaders 8 Leisure and hobbies
9 Bureaucrats 9 Religious meetings
10 Local community 10 Information and communications technology (e.g. internet, phones)
Language Choice
Not all bilinguals have the opportunity to use both their languages on a regular basis. Where a bilingual lives in a largely monolingual community, there may be little choice about language use from day-to-day. However, in communities where two or more languages are widely spoken, bilinguals may use both their languages on a daily or frequent basis. When bilinguals use both their languages, there is often language choice. If the other person is already known to the bilingual, as a family member, friend or colleague, a relationship has usually been established through one language. If both are bilingual they have the option of changing to the other language (e.g. to include others in the conversation), although old habits die hard.

If the other person is not known, a bilingual may quickly pick up clues as to which language to use. Clues such as dress, appearance, age, accent and command of a language may suggest to the bilingual which language it would be appropriate to use. In bilingual areas of Canada and the United States for example, employees dealing with the general public may glance at a person’s name on their records to help them decide which language to use. A person called Pierre Rouleau or Maria García might be addressed first in French or Spanish, rather than English.

An individual’s own attitudes and preferences will influence their choice of language. In a minority/majority language situation, older people may prefer to speak the minority language. Teenagers (e.g. second-generation immigrants) may reject the minority language in favor of the majority language because of its higher status and more fashionable image. Heller (1982: 108) shows how in a conversation, perceptions about language and identity affect language choice, as in the following example from Québec:

I stopped in a garage … and struggled to explain … that my windshield wipers were congelé and I wanted to make them fonctionner. He listened in mild amusement and then said: ‘You don’t have to speak French to me, abilities fit into two dimensions: receptive and productive madame. I am not a separatist’.

In situations where the native language is perceived to be under threat, some bilinguals may seek to avoid speaking the majority or dominant language to assert and reinforce the status of the other language. For example, French-Canadians in Québec sometimes refuse to speak English in shops and offices to emphasize the status of French.

Li Wei et al. (1992), in a study of a Chinese community in northern England, indicate that the degree of contact with the majority language community is a factor in language choice. Their research shows that Chinese speakers who were employed outside the Chinese community were more likely to choose to speak English with other Chinese speakers. In contrast, those Chinese immigrants who worked in family businesses, mainly catering, and had less daily contact with English speakers, were more likely to use Chinese with other Chinese–English bilinguals.

Some minority languages are mostly confined to a private and domestic role. This happens when a minority language has historically been disparaged and deprived of status. In western Brittany in France, for example, many Breton speakers only use their Breton in the family and with close friends. They can be offended if addressed by a stranger in Breton, believing that such a stranger is implying they are uneducated and cannot speak French.

An individual may also switch languages, either deliberately or subconsciously, to accommodate the perceived preference of the other participant in the conversation. A language switch may be made as one language is regarded as the more prestigious or as more appropriate for the other person. To gain acceptance or status, a person may deliberately and consciously use the majority language. Alternatively, a person may use a minority language as a form of affiliation or belonging to a group. (Codeswitching is discussed in Chapter 5.)

Bilingual and Multilingual Ability
The Four Language Abilities
If we confine the question ‘Are you bilingual?’ to ability in two (or more) languages, the issue becomes ‘what particular ability? ’ There are four basic language abilities: listening, speaking, reading and writing. These four abilities fit into two dimensions: receptive and productive skills; oracy and literacy. The following table illustrates:

ORACY LITERACY
Receptive skills Listening Reading
Productive skills Speaking Writing

The table suggests avoiding a simple classification of who is, or is not, bilingual. Some speak a language, but do not read or write in a language. Some listen with understanding and read a language (passive bilingualism) but do not speak or write that language. Some understand a spoken language but do not themselves speak that language. To classify people as either bilinguals or monolinguals is thus too simplistic. Or, to return to the opening analogy, the two wheels of bilingualism exist in different sizes and styles. The four basic language abilities do not exist in black and white terms. Between black and white are not only many shades of gray; there also exist a wide variety of colors. Each language ability can be more or less developed. Reading ability can range from simple and basic to fluent and accomplished. Someone may listen with understanding in one context (e.g. shops) but not in another context (e.g. an academic lecture).

These examples show that the four basic abilities can be further refined into sub-scales and dimensions. There are skills within skills, traditionally listed as: pronunciation, extent of vocabulary, correctness of grammar, the ability to convey exact meanings in different situations and variations in style. However, these skills tend to be viewed from an academic or classroom perspective. Using a language on the street and in a shop require a greater accent on social competence with language (e.g. the idioms and ‘lingo’ of the street).

The range and type of sub-skills that can be measured is large and debated. Language abilities such as speaking or reading can be divided into increasingly microscopic parts. What in practice is tested and measured to portray an individual’s bilingual performance is considered later in the book. What has emerged so far is that a person’s ability in two languages is multidimensional and will tend to evade simple categorization.

Minimal and Maximal Bilingualism
So far, it has been suggested that deciding who is or is not bilingual or multilingual is difficult. Simple categorization is arbitrary and requires a value judgment about the minimal competence needed to achieve a label of ‘bilingual’. Therefore, a classic definition of bilingualism such as ‘the native-like control of two or more languages’ (Bloomfield, 1933) appears too extreme and maximalist (‘native-like’). The definition is also ambiguous (what is meant by ‘control’ and who forms the ‘native’ reference group?).

At the other end is a minimalist definition, as in Diebold’s (1964) concept of incipient bilingualism. The term incipient bilingualism allows people with minimal competence in a second language to squeeze into the bilingual category. Tourists with a few phrases and business people with a few greetings in a second language could be incipient bilinguals. Almost every adult in the world knows a few words in another language. The danger of being too exclusive is not overcome by being too inclusive. Trawling with too wide a fishing net will catch too much variety and therefore make discussion about bilinguals ambiguous and imprecise. Trawling with narrow criteria may be too insensitive and restrictive.

Valdés (2003) pictures bilinguals as existing on a continuum, where A and B are the two languages. The first letter is the stronger language, and font sizes and case suggest different proficiencies, e.g.

A
Monolingual Ab Ab Ab Ab Ab AB BA Ba Ba Ba Ba Ba B
Monolingual

Who is or is not categorized as a bilingual will depend on the purpose of the categorization. At different times governments, for example, may wish to include or exclude language minorities. Where an indigenous language exists (e.g. Irish in Ireland), a government may wish to maximize its count of bilinguals. A high count may indicate government success in language planning. In comparison, in a suppressive, assimilationist approach, immigrant minority languages and bilinguals may be minimized (e.g. Asian languages in the UK in the Census – see Chapter 2).

Is there a middle ground in-between maximal and minimal definitions? The danger is in making arbitrary cut-off points about who is bilingual or not along the competence dimensions. Differences in classification will continue to exist among different authors. One alternative is to move away from the multi-colored canvas of proficiency levels to a portrait of the everyday use of the two languages by individuals (see earlier).

Balanced Bilinguals
The literature on bilingualism frequently spotlights one particular group of bilinguals whose competences in both languages are well developed. Someone who is approximately equally fluent in two languages across various contexts has been termed an equilingual or ambilingual or, more commonly, a balanced bilingual. As will be considered in Chapter 7, balanced bilinguals are important when discussing the possible thinking advantages of bilingualism.

Balanced bilingualism is sometimes used as an idealized concept. Rarely is anyone equally competent in two or more languages across all situations. Most bilinguals will use their two languages for different purposes and with different people. For example, a person may use one language at work and the other language at home and in the local community.

Balanced bilingualism is also a problematic concept for other reasons. The balance may exist at a low level of competence in the two languages. Someone may have two relatively undeveloped languages that are nevertheless approximately equal in proficiency. While this is within the literal interpretation of ‘balanced’ bilingual, it is not the sense employed by many researchers on bilingualism. The implicit idea of balanced bilingualism has often been of ‘appropriate’ competence in both languages. A child who can understand the delivery of the curriculum in school in either language, and operate in classroom activity in either language would be an example of a balanced bilingual.

Is ‘balanced bilingualism’ of use as a term? While it has limitations of definition and measurement, it has proved to be of value in research and theory (see Chapter 7). However, categorizing individuals into such groups raises the issue of comparisons. Who is judged ‘normal’, proficient, skilled, fluent or competent? Who judges? The danger is in using monolinguals as the point of reference, as will now be considered.

An argument advanced by Grosjean (1985, 2008), Cook (1992, 2002a) and Jessner (2008b) is that there are two contrasting views of individual bilinguals. First, there is a monolingual or fractional view of bilinguals, which evaluates the bilingual as ‘two monolinguals in one person’. There is a second, holistic view which argues that the bilingual is not the sum of two complete or incomplete monolinguals, but that he or she has a unique linguistic profile. A monolingual view can be that it is ‘normal’ or ‘pure’ to have just one language, and therefore bilinguals are studied from that perspective. The multicompetence view is that bilingualism and multilingualism are also ‘normal’, and this has consequences for how bilingualism is studied (e.g. acquisition, use, storage, thinking, integration/interconnection/separation, - see Grosjean, 2008; Cook, 2002a; Jessner, 2008b).

The Monolingual View of Bilingualism
Some teachers, administrators and politicians look at the bilingual as two monolinguals in one person. For example, if English is a bilingual’s second language, scores on an English reading or English attainment test will often be compared against monolingual or native English speakers’ scores and averages. In the US and the UK, a bilingual’s English language competence is often measured against that of a native monolingual English speaker. This derives from a monolingual viewpoint – and with political overtones. It is unfair because bilinguals will typically use their two languages in different situations and with different people. Thus bilinguals may be stronger in each language in different domains.

One expectation from this fractional viewpoint is for bilinguals to show a proficiency comparable to that of a monolingual in both their two languages. If that proficiency does not exist in both languages, especially in the majority language, then bilinguals may be denigrated and classified as inferior. In the United States, for example, children of immigrant families, or of other language minority families, have been officially federally categorized as LEP (Limited English Proficient). In northern Europe, bilinguals who appear to exhibit a lack of proficiency in both languages may be described as ‘semilingual’.

In Africa, India, Scandinavia and parts of Asia, for example, bilingualism is often seen as the norm, while in countries such as the United States and England, the dominant view of the world is monolingual (Brutt-Griffler & Varghese, 2004). Although between a half and two-thirds of the world’s population is bilingual to some degree, the monolingual is often seen as ‘normal’ in these two countries, and the bilingual as an oddity or as inferior (see Ellis (2006) for a discussion of monolingualism). This ‘inferior’ viewpoint, for example that bilinguals have two half-developed languages, is encapsulated in the debate about ‘semilingualism’.

‘Semilingualism’/‘Double Semilingualism’
Bilinguals tend to be dominant in one of their languages in all or some of their language abilities. This may vary with context and may change over time with geographical or social mobility. For others, the dominance may be relatively stable across time and place. The topic of dominance will be considered in Chapter 2 when tests are discussed. For the present, a group has been proposed, one that is distinct from balanced and dominant bilinguals. Sometimes referred to negatively as semilinguals or double semilinguals, the group is regarded as not having ‘sufficient’ competence in either language. This section will suggest that such a label is more politically motivated than accurate or commonplace.

Hansegård (1975; see Skutnabb-Kangas, 2000) described semilingualism in terms of deficiencies in bilinguals when compared with monolinguals on the following dimensions: displaying a small vocabulary and incorrect grammar, consciously thinking about language production, stilted and uncreative with each language, and finding it difficult to think and express emotions in either language.

The notion of semilingualism, or double semilingualism, has received much criticism (e.g. Skutnabb-Kangas, 2000; Wiley, 1996a, 2005c; MacSwan, 2000). There are major problems with the term. First, the term took on disparaging and belittling overtones, particularly in Scandinavia and with immigrant groups in the US. Semilingualism may be used as a negative label that invokes expectations of underachievement and failure. It is most frequently applied to immigrant groups and ‘blames the victim’.

Second, if languages are relatively undeveloped, the origins may not be in bilingualism per se, but in the economic, political and social conditions that create such under-development. This is a theme considered in detail in later chapters. The danger of the term semilingualism is that it locates the origins of underdevelopment in the internal, individual possession of bilingualism, rather than in external, societal factors that co-exist with bilingualism. Thus the term may be used as a political rather than a linguistic concept.

Third, most bilinguals use their two languages for different purposes and events. Language may be specific to a context. A person may be competent in some contexts but not in others. Some children are competent in the school context, but are less competent in the vernacular of the street. Others the opposite. Some are competent in a language for religious purposes, but less so in the home.

Fourth, the educational tests that are most often used to measure language proficiencies and differentiate between people may be insensitive to the qualitative aspects of languages and to the great range of language competences. Language tests may measure a small, unrepresentative sample of a person’s daily language behavior (see Chapter 2). Thus ‘deficiencies’ are often an artefact of narrow academic tests. Standardized tests of language proficiency fail to measure the varied conversation patterns that children from different cultures use with considerable competence.

Test scores [are] based on specific language and literacy tests of the school. These tests, in turn, reflect particular literacy practices and social expectations favoring groups that control institutions. Also, because school tests are based on ‘standard’ academic language, there is an implicit bias against language variation within L1 (i.e., there is a bias against speakers of nonstandard and creolized varieties of L1). In interpreting results based on standardized tests, practitioners sometimes claim that students have ‘no language’, meaning that they have no standard academic language. (Wiley, 1996a: 167–168)

Fifth, the comparison with monolinguals may not be fair. Being different is not the same as having a deficiency. An apparent deficiency may be due to unfair ‘fractional’ comparisons with monolinguals.

These criticisms raise serious doubts about the value of the term ‘semilingualism’. However, this does not detract from the fact that there are language abilities on which people do differ, with some people being at the earlier stages of development, others where there is rapid language loss (Davies, 2003). Being at an early stage or undergoing language loss may not be the result of being bilingual. Economic and social factors or educational provision may, for example, be the cause.

Rather than highlight an apparent ‘deficit’ in language development, the more equitable and positive approach is to emphasize that, given suitable conditions, competence in language is capable of development to high levels. Instead of concentrating negatively on a ‘language deficit’, a more proper approach is to locate the causes in, for example, the type of language tests used, material deprivation, in the type and quality of schooling, and not in language itself (see Chapters 9, 15 and 17).

Second language users also need to be seen as legitimate speakers of a language in their own right. Second language speakers are not deficient communicators. Rather they are embryonic or developing bilinguals. This leads to a more holistic view of bilinguals.

The Holistic View of Bilingualism
Grosjean (1985, 2008) and Cook (1992, 2002a, 2002b) present a more positive alternative view of bilinguals, as those with multicompetences. Grosjean uses an analogy from the world of athletics, and asks whether we can fairly judge a sprinter or a high jumper against a hurdler. The sprinter and high jumper concentrate on one event and may excel in it. The hurdler concentrates on two different skills, trying to combine a high standard in both. With only a few exceptions, the hurdler will be unable to sprint as fast as the sprinter, or jump as high as the high jumper. This is not to say that the hurdler is a worse athlete than the other two. Any comparison of who is the best athlete makes little sense. This analogy suggests that comparing the language proficiency of a monolingual with a bilingual’s dual language or multilingual proficiency is similarly unjust.

However, this raises the question, should bilinguals only be measured and compared by reference to other bilinguals? When for example, someone learns English as a second language, should that competency in English only be measured against English ability in other bilinguals? Any assessment of a bilingual’s language proficiency should ideally move away from traditional language tests (with their emphasis on form and correctness) to an evaluation of the bilingual’s general communicative competence. This appraisal would be based on a totality of the bilingual’s language usage in all domains, whether this involves the choice of one language in a particular domain, or a mixing of the two languages.

There is sometimes a political reality that deters the blossoming of a holistic view of the bilingual. In Australia, much of Canada, the United States and the United Kingdom, the dominant English-speaking monolingual politicians and administrators tend not to accept a different approach or standard of assessment (one for monolinguals, another for bilinguals). There is also the issue of preparation for the employment market. In countries like Wales, where first-language Welsh-speaking children compete in a largely English-language job market against monolingual English speakers, the dominant view is that they should be given the same English assessments at school.

Yet a bilingual is a complete linguistic entity, an integrated whole. Bilinguals use their two languages with different people, in different contexts and for different purposes. Levels of proficiency in a language may depend on which contexts (e.g. street and home) and how often that language is used. Communicative competence in one of a bilingual’s two languages may be stronger in some domains than in others. This is natural and to be expected. Any assessment of a bilingual’s competence in two languages needs to be sensitive to such differences of when, where and with whom bilinguals use either of their languages. Such an assessment should reveal the linguistic multicompetences of bilinguals (Cook, 2002a), even though there is one underlying conceptual system (Kecskes, 2010, see also Chapter 8).

Conversational Fluency and Academic Language Competence
So far, this chapter has centered on the variety of language abilities and the danger of categorization using a small or biased selection of language sub-skills. One issue has been whether the variety of sub-skills can be reduced to a small number of important dimensions. Hernández-Chávez et al. (1978), for example, originally suggested there are 64 separate components to language proficiency. In comparison, many reading tests tacitly assume that reading can be reduced to one dimension. Oller (1982) claimed that there was one overall, global language dimension. An overlap between different academic language tests was sufficient for Oller and Perkins (1980) to suggest that there exists a single factor of global language proficiency.

The idea of a single language factor is contentious as the evidence indicates that there are both global and specific aspects of language proficiencies. Oller’s (1982) idea of a global language factor is based on quantitative testing. As will be considered later in the book, such tests leave qualitative differences between people unexplored. There is also an emphasis on language in an academic context. This leaves the out-of-school communicative profile of people relatively ignored.

Oller’s (1982) much-disputed claim for one global language factor provides a starting point for a distinction between two different language abilities. Oller’s (1982) single language proficiency factor has been allied to the language abilities needed to cope in the classroom. Most (but not all) language tests are closely linked to the cognitive, academic language skills of the classroom. Reading and writing tests are obvious examples. The notion of a curriculum-based language competence led various authors to make an important distinction.

Apart from academically related language competence, it has been proposed that there is a conceptually distinct category of conversational competence (Cummins, 2000b). Skutnabb-Kangas and Toukomaa (1976) proposed a difference between surface fluency and academically related aspects of language competence. Surface fluency would include the ability to hold a simple conversation in the shop or street and may be acquired fairly quickly (e.g. in two or three years) by second language learning. To cope in the curriculum, conversational language competence may not be enough. Academically related language competence in a second language may take from five to eight years or longer to acquire. This theme is considered in detail later in the book when a contentious distinction is made between ‘basic interpersonal communicative skills’ and ‘cognitive/academic language proficiency’ (BICS and CALP – see Chapter 8). Such a distinction between two levels of language competence is important as it involves disputing Oller’s (1982) ‘single factor’ language skill.

If language abilities are multicolored, and if bilinguals have a range of colors in both languages, then positive terms are needed to portray the variety. Calling bilinguals LEP (Limited English Proficiency) students in the US seems negative and pejorative. For example, in the US, Title III of the Elementary and Secondary Education Act of 1965, reauthorized by the ‘No Child Left Behind Act’ of 2001, was entitled ‘Language Instruction of Limited English Proficient and Immigrant Students’. Such a label can accentuate children’s perceived deficiency rather than their proficiencies, children’s perceived ‘deprivation’ rather than their accomplishments, their lower, marginalized, minority status through majority eyes rather than their bilingual potentiality. Such a label highlights past and present performance rather than potentialities and the possibility of functioning well in two or more languages.

The chapter now continues by considering various language structure theories. Theories about the structure of language competence provide an integrating consideration of the themes of the definition of bilingualism.

The Structure of Language Competence
The language theories of the 1960s (e.g. Lado, 1961; J.B. Carroll, 1968) tended to center on language skills and components. The skills comprise listening, speaking, reading and writing, and the components of knowledge comprise grammar, vocabulary, phonology and graphology. These earlier models did not indicate how skills and knowledge were integrated. For example, how does listening differ from speaking? How does reading differ from writing? Earlier models fail to probe the competence of ‘other’ people in a conversation. In a conversation, there is negotiation of meaning between two or more people. Real communication involves anticipating a listener’s response, understandings and misunderstandings, sometimes clarifying one’s own language to ensure joint understanding, plus the influence of different status and power between people.

It has also been suggested that such skill and knowledge models ignored the sociocultural and sociolinguistic context of language. Earlier models tended to be purely linguistic and ignore the social contexts where language is used. A more sociolinguistic approach examines actual content and context of communication called ‘speech acts’ or the ‘ethnography of communication’. This approach includes looking at the rules of dual language usage among bilinguals, their shared knowledge in conversation, and the culturally, socially and politically determined language norms and values of bilingual speech events.

Various holistic models of language competence have been developed. One example will be briefly outlined.

Bachman’s Model of Language Competence
A major model of language competence was proposed by Bachman (1990), refined by Bachman and Palmer (1996). Bachman’s model is valuable in that it considers both language competence and language performance. The model includes not only grammatical knowledge but also knowledge of how to use language in a particular communicative context. To fully define, refine and enable the testing of communicative competence, Bachman (1990) proposed a model that is summarized in the following table.

Language Competence

Organizational Competence
(i) Grammatical (e.g. Syntax, Vocabulary)
(ii) Textual (e.g. Written and oral cohesion).
Pragmatic Competence
(i) Illocutionary Competence (e.g. speech strategies, language functions)
(ii) Sociolinguistic Competence (e.g. sensitivity to register, dialect, cultural figures of speech).

To explain the table: for Bachman (1990), communicative competence is composed of two major components: organizational competence and pragmatic competence. Organizational competence is broken down into two parts, grammatical competence and textual competence. Grammatical competence comprises knowledge of vocabulary, syntax, morphology and phonology/graphology. For example, a person needs to arrange words in a correct order in a sentence with appropriate endings (e.g. high, higher, highest). Textual competence involves ‘the knowledge of the conventions for joining utterances together to form a text, which is essentially a unit of language – spoken or written – consisting of two or more utterances or sentences’ (Bachman, 1990: 88).

Pragmatic competence is composed of two sub-parts: illocutionary competence and sociolinguistic competence. Following Halliday (1973), Bachman (1990) lists four language functions as part of illocutionary competence: ideational (the way we convey meanings and experiences), manipulative (using language in an instrumental way to achieve ends), heuristic (the use of language to discover new things about our world and solving problems), and the imaginative function (using language beyond the ‘here and now’ (e.g. for humor or fantasy).

The second part of pragmatic competence is sociolinguistic competence. Sociolinguistic competence is sensitivity to the context where language is used, ensuring that language is appropriate to the person or the situation. This may entail sensitivity to differences in local geographical dialect, sensitivity to differences in register (e.g. the register of boardroom, baseball, bar and bedroom). Sociolinguistic competence also refers to sensitivity to speaking in a native-like or natural way. This will include cultural variations in grammar and vocabulary (e.g. Black English). Another part of socio-linguistic competence is the ability to interpret cultural references and figures of speech. Sometimes, to understand a particular conversation, one needs inner cultural understanding of a specific language. A Welsh figure of speech such as ‘to go round the Orme’ (meaning ‘to be long-winded’) is only fully understandable within local northern Welsh cultural idioms.

In order to represent language as a dynamic process, the listed components given in the table must be regarded as interactive with each other. Therefore, the notion of strategic competence is important, where individuals constantly plan, execute and assess their communication strategies and delivery. Bachman and Palmer (1996) see such strategic competence as cognitive executive processes that govern language behavior. In a revision of the model, Bachman and Palmer (1996) added the personal characteristics of the individual language user to the model (topical knowledge and affective schema). Critiques of this model are provided by McNamara (1996, 2003), Skehan (1998) and Purpura (2008).

Since competence in a language is viewed as an integral part of language performance and not abstracted from it, measuring language competence cannot just use pencil and paper tests, but also needs to investigate the language of genuine communication. Instead of tests that are artificial and stilted (e.g. language dictation tests), communicative performance testing involves creative, unpredictable, contextualized conversation. However, predicting ‘real world’ performance from such tests, and the ‘one sidedness’ that ignores the reality that conversations are jointly constructed and negotiated, remains problematic. This suggests that it will be difficult to measure communicative proficiency in an unbiased, comprehensive, valid and reliable way. Simple classroom tests are likely to be but a partial measure of the bilingual’s everyday performance.

Celce-Murcia et al. (1995) have extended this model to include communicative competence in speech acts. This emphasises elements such as the kind of knowledge needed in greeting people and leave-taking, in expressing feelings and persuading. They also accentuate the importance of socio-cultural competence including matters such as politeness in a particular context, and the style and formality of language. This includes body language (e.g. smiles, eye contact), even the use of silence.

The emphasis has therefore moved over time from the linguistic to the communicative, to interactional competence and the adaptivity of a person in using two or more languages (Fulcher & Davidson, 2007).

Discussions of language competence often move to questions about measurement (e.g. of students). To what extent can we measure someone’s performance in their two languages? How can we portray when, where and with whom people use their two languages? What are the problems and dangers in measuring bilinguals? These questions provide the themes for the next chapter.

Conclusion
Defining exactly who is or is not bilingual is essentially elusive and ultimately impossible. Some categorization, however, is often necessary and helpful to make sense of the world. Therefore categorizations and approximations may be required. Simple definitions (e.g. Bloomfield’s (1933) ‘native-like control of two languages’) offer little help as they are intrinsically arbitrary and ambiguous.

A more helpful approach may be to locate important distinctions and dimensions surrounding the term ‘bilingualism’ that help to refine our thinking. The fundamental distinction is between bilingual ability and bilingual usage. Some bilinguals may be fluent in two languages but rarely use both. Others may be much less fluent but use their two languages regularly in different contexts. Many other patterns are possible. Languages are not static, as with ‘emerging bilinguals’ (O. García, 2009a).

Profiling a person’s use of their two languages raises questions about when, where and with whom? This highlights the importance of considering domain or context. As a bilingual moves from one situation to another, so may the language being used in terms of type (e.g. Spanish or English), content (e.g. vocabulary) and style. Over time and place, an individual’s two languages are never static but ever changing and evolving.

In terms of ability in two languages, the four basic dimensions are listening, speaking, reading and writing. With each of these proficiency dimensions, it is possible to fragment into more detailed dimensions (e.g. pronunciation, vocabulary, grammar, meaning and style). Those sub-dimensions can subsequently be further dissected and divided.

Creating a multidimensional, elaborate structure of bilingual proficiency may make for sensitivity and precision. However, ease of conceptualization and brevity require simplicity rather than complexity. Therefore simple categorization is the paradoxical partner of complex amplification. This chapter has focused on the categories of balanced bilingualism, semilingualism, and one-factor ideas of language ability. These categories have received some depth of discussion and critical response in the research literature. As will be revealed in later chapters, these categories also relate to central research on bilingualism and bilingual education.

The chapter considered theories of the structure of language competence. In particular, the focus has been on linking a linguistic view of language competence with a social communicative view. Language can be decomposed into its linguistic constituents (e.g. grammar, vocabulary). It is also important to consider language as a means of making relationships and communicating information. This important dualism will follow us through the book: ability and use; the linguistic and the social; competence and communication.

Key Points in this Chapter
There is a difference between bilingualism as an individual possession and two or more languages operating within a group, community, region or country.
At an individual level, there is a distinction between a person’s ability in two languages and their use of those languages.
Bilinguals typically use their two languages with different people, in different contexts and for different purposes.
Language abilities are listening, speaking, reading and writing.
Balanced bilinguals with equal and strong competence in their two languages are rare.
There is a difference between a monolingual or fractional view of bilinguals and a holistic view. The fractional view sees bilinguals as two monolinguals inside one person. The holistic view sees bilinguals as a complete linguistic entity, an integrated whole.
The terms ‘semilingual’ and ‘double semilingualism’ have been used to describe a bilingual whose languages are both under-developed. However, the label has tended to take on negative, political and personally negative overtones.
A distinction is made between the kind of language required for conversational fluency and the type of language required for academic, classroom purposes.
Language competence includes not only linguistic competence (e.g. vocabulary, grammar) but also competence in different social and cultural situations with different people.

Suggested Further Reading
BAKER, C. & JONES, S.P., 1998, Encyclopedia of Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.

BHATIA, T.K. & RITCHIE, W.C. (eds), 2004, The Handbook of Bilingualism. Malden, MA: Blackwell.

EDWARDS, V., 2004, Multilingualism in the English-speaking World. Oxford: Blackwell.

GROSJEAN, F., 2008, Studying Bilinguals. Oxford: Oxford University Press.

GROSJEAN, F., 2010, Bilingual: Life and Reality. Cambridge, Mass.: Harvard University Press.

MYERS-SCOTTON, C., 2006, Multiple Voices: an Introduction to Bilingualism. Oxford: Blackwell

CHIN, N.B. & WIGGLESWORTH, G., 2007, Bilingualism: an advanced resource book. New York: Routledge

SKUTNABB-KANGAS, T., 2000, Linguistic Genocide in Education – Or Worldwide Diversity and Human Rights? Mahwah, NJ: Erlbaum.

See also the International Journal of Bilingual Education and Bilingualism for current viewpoints on bilinguals and bilingualism: http://www.tandf.co.uk/journals/1367-0050

Study Activities
Do you consider yourself and/or people known to you as bilingual or multilingual? Would you describe yourself, or someone known to you, as ‘balanced’ in ability and use of two or more languages? Which language or languages do you think in? Does this change in different contexts? In which language or languages do you dream, count numbers, pray and think aloud?
This activity can be based on self-reflection or you may wish to interview someone who is bilingual or multilingual. Make a table or diagram to illustrate how one person’s dual or multilingual ability and language usage has changed and developed since birth. Write down how different contexts have affected that change and development. The diagram or table should illustrate the life history of someone’s languages indicating changes over time and over different contexts.
In a school with which you are most familiar, find out how students are labeled and categorized in terms of their languages. Who applies what labels? Which students are seen positively and negatively? Are there consequences of labels? Is there interest in the language competences and language use of multilingual students?
CHAPTER 2
The Measurement of Bilingualism

Introduction
The Purposes of the Measurement of Bilinguals
The Assessment of Bilinguals in School
Limitations in Measurement
Communicative Language Testing
A Political View of Language Testing
Examples of the Measurement of Bilinguals in Research
Language Background Scales
Language Balance and Dominance Measures
Language Censuses
Limitations of Language Censuses
Conclusion

CHAPTER 2
The Measurement of Bilingualism

Introduction
The topic of measuring bilinguals and multilinguals both elaborates and illuminates the previous discussion about definitions, dimensions and distinctions. Problems of defining bilinguals are illustrated by the measurement and categorization of such individuals and groups. This chapter, therefore, begins by clarifying the different reasons for measuring bilinguals and multilinguals. This occurs not only in education but also in society (e.g. censuses). Illustrations are then given of such measurement as well as ways of profiling the language ability and use of individuals and language groups. It is important to develop a critical awareness of language measurement, both in the internal limitations of measurement and the politics surrounding language testing.

It is customary to try to categorize the complexity of individual differences in bilingualism. We make sense of our world by continual classification. People are constantly compared and contrasted. Yet the simplification of categorization often hides the complexity of individuality. Sorting often simplifies unsympathetically. Individual differences are reduced to similarities. Yet over-complexity can be unwelcome and confusing. Complications can confound those needing order and pattern. The measurement of bilinguals attempts to locate similarities, order and pattern.

The Purposes of the Measurement of Bilinguals
The measurement of bilinguals can take place for a variety of purposes, and it is valuable to differentiate between some of these overlapping aims.

Distribution
An example of the measurement of bilinguals is census questions requesting information about ability or usage in two or more languages (e.g. in US, Canada, Ireland). Such census data estimates the size and distribution of bilinguals in a particular area. For example, a survey of 850,000 children in London schools revealed the presence of over 300 different languages (P. Baker & Eversley, 2000). Geographers map the proportion and location of regional and minority language groups (C.H. Williams, 2004). For an example, see the Ethnologue web site: http://www.ethnologue.com/

Selection
Bilinguals can be distinguished as a ‘separate’ group for selection purposes. For example, a school may wish to allocate children to classes, sets, streams, groups or tracks based on their degree of bilingual proficiency or language background. Bilinguals may be assessed for placement in special education classes (see Chapter 15). A different example is measuring bilinguals at the outset of research. An investigation may require the initial formation of two or more groups (e.g. ‘balanced’ bilinguals, ‘partial’ bilinguals and monolinguals).

Summative
Summative means ‘totaling up’ to indicate the destination a person has reached in their language learning journey. Summing up someone’s language proficiency may occur, for example, at the end of a semester or a school year. When measuring the current performance level of a person, a wide variety of language proficiency and achievement tests are available (e.g. reading comprehension, reading vocabulary). Such tests may be used in schools to measure the four basic language abilities. In a minority language context, emphasis is often on measuring proficiency in both the minority language and the majority language.

In the United States, emphasis on minority language groups becoming proficient in English has been a dominant issue in the testing of bilinguals (see Chapter 9). Thus in most US schools, language testing is solely in English, with second language competence virtually ignored (Carrisquillo & Rodríguez, 2002).

With language proficiency testing, the measurement of bilinguals becomes fused with second language testing and the general area of language testing. Apart from general language tests, there are measures of the relative dominance of a person’s two languages and the mixing of a person’s two languages (sometimes pejoratively termed ‘interference’ – see Chapter 5). Such tests spotlight the particular characteristics of bilinguals as different from second language learners and first language development. Examples are provided later in this chapter.

Formative
Language proficiency tests are often used for summative judgments. In comparison, a test or assessment device that gives feedback (and feed-forward) during learning, and to aid further development, is formative assessment. A student may be profiled on a precise breakdown of language skills to provide facilitative feedback to the teacher that directly leads to action. If the test reveals areas where a child’s language needs developing, there can be immediate intervention and remedial help. A diagnosis of a problem in language may lead to the formation of a plan to effect a remedy. The assessment of bilinguals is considered in Chapter 15.

The Assessment of Bilinguals in School
Criterion Referenced Language Tests: A Curriculum Approach to Language Testing
A recent shift in testing has been away from norm-referenced tests to criterion-referenced tests. This is partly due to a movement in language education towards communicative skills, curriculum objectives, assessing learning outcomes and mastery learning. What is the distinction between norm-referenced and criterion-referenced tests?

Language proficiency tests are often classified into norm-referenced and criterion-referenced tests; the former usually being summative tests, the latter mostly being formative tests. Standardized norm-referenced tests essentially compare one individual with others, as in an IQ test. A norm-referenced test of reading ability, for example, may enable the teacher to compare one student with a national or regional average (norm). The student can then be exactly placed in an ordered list (e.g. in the top 16%).

Language Test Terminology

Language skills tend to refer to highly specific, observable, measurable, clearly definable components such as handwriting.
Language competence is a broad and general term, used particularly to describe an inner, mental representation of language, something latent rather than overt. Such competence refers usually to an underlying system inferred from language performance.
Language performance hence becomes the outward evidence for language competence. By observing general language comprehension and production, language competence may be inferred.
Language ability and language proficiency tend to be used more as ‘umbrella’ terms and therefore used somewhat ambiguously. For some, language ability is a general, latent disposition, a determinant of eventual language success. For others, it tends to be used as an outcome, similar to but less specific than language skills, providing an indication of current language level. Similarly, language proficiency is sometimes used synonymously with language competence; at other times as a specific, measurable outcome from language testing. However, both language proficiency and language ability are distinct from language achievement.
Language achievement (attainment) is usually seen as the outcome of formal instruction. Language proficiency and language ability are, in contrast, viewed as the product of a variety of mechanisms: formal learning, informal non-contrived language acquisition (e.g. on the street) and individual characteristics such as ‘intelligence’.

A criterion-referenced test moves away from comparing one person with another. Instead it profiles an individual child on a particular language skill. The profile will test what a child can and cannot do on a precise breakdown of language skills. The parallel is with a car driving test. There are a variety of components to car driving (e.g. backing around a corner, three-point turn, starting on a steep hill). Proficiency in driving often requires being able to satisfy an examiner on these sub-skills. Comparisons with other drivers are unimportant. An individual’s mastery of specific tasks is the criterion for passing or failing.

Specifying driving criteria is easier than identifying language criteria; hence the analogy is not exact. The sub-components of language proficiency will be contested, and may not be easily definable or measurable. Apart from language skills, there are the qualitative aspects of language that are not simply reducible for testing (e.g. the emotive, status and poetic functions of languages).

One advantage for bilinguals of criterion-referenced testing over norm-referenced testing is the point of comparison. Norm-referenced testing may compare bilinguals with monolinguals. The norm becomes the native speaker (Davies, 2003). Various authors (e.g. Frederickson & Cline, 1996; Cook, 2002a; Grosjean, 2008) regard such comparison as unfair and invalid (see Chapter 1).

In criterion-referenced testing, the bilingual can be profiled on specific language skills. In theory, unfair comparisons between bilingual and monolingual may then be avoided. In practice, however, criterion-referenced tests can be used to create comparisons between children (e.g. monolingual native speakers and bilinguals), between groups of children and between schools. Behind every criterion lurks a norm. The norm is usually the point of departure for setting a criterion (Davies, 2003).

An advantage of criterion-referenced language tests is that they may facilitate feedback to the teacher that directly leads to intervention. This is the notion of formative testing. If the test reveals areas where a child’s language requires developing, further action can be taken. One illustration of testing language proficiency by criterion-referenced tests is given by Harris and Murtagh (1999). Focusing on the attainment of Irish listening and speaking objectives, the table below specifies the test items. The number of items to measure each objective ranged from three (control of syntax of questions) to 25 items (general comprehension of speech). Fourth grade children were assessed for their mastery of 140 separate items.

LISTENING OBJECTIVES SPEAKING OBJECTIVES
1.Sound discrimination 1.Pronunciation
2.Listening vocabulary 2.Speaking vocabulary
3.General comprehension of speech 3.Fluency of oral description
4.Understanding morphology of verbs 4.Control of morphology of verbs
5.Understanding morphology of prepositions 5.Control of morphology of prepositions
6.Understanding morphology of qualifiers 6.Control of morphology of qualifiers
7.Understanding morphology of nouns 7.Control of morphology of nouns
8.Understanding syntax of statements 8.Control of syntax of statements
9.Understanding syntax of questions 9.Control of syntax of questions

These Irish language tests are based on phonology (pronunciation), morphology (knowledge of the functions of words), syntax (grammar structures) and lexicon (vocabulary). In contrast, the Council of Europe’s Common European Framework of Reference for language assessment (Council of Europe, 2001) particularly concerns language communication and use. The Common European Framework ‘provides a common basis for the elaboration of language syllabuses, curriculum guidelines, examinations, textbooks, etc. across Europe’ (Council of Europe, 2001: 1). The Framework defines levels of proficiency that allow assessment of learners’ progress irrespective of age, language or region.

COMMON EUROPEAN FRAMEWORK OF REFERENCE FOR LANGUAGES

Proficient User: C2
Can understand with ease virtually everything heard or read. Can summarise information from different spoken and written sources, reconstructing arguments and accounts in a coherent presentation. Can express him/herself spontaneously, very fluently and precisely, differentiating finer shades of meaning even in more complex situations.

Proficient User: C1
Can understand a wide range of demanding, longer texts, and recognise implicit meaning. Can express him/herself fluently and spontaneously without much obvious searching for expressions. Can use language flexibly and effectively for social, academic and professional purposes. Can produce clear, well-structured, detailed text on complex subjects, showing controlled use of organisational patterns, connectors and cohesive devices.

Independent User B2
Can understand the main ideas of complex text on both concrete and abstract topics, including technical discussions in his/her field of specialisation. Can interact with a degree of fluency and spontaneity that makes regular interaction with native speakers quite possible without strain for either party. Can produce clear, detailed text on a wide range of subjects and explain a viewpoint on a topical issue giving the advantages and disadvantages of various options.

Independent User B1
Can understand the main points of clear standard input on familiar matters regularly encountered in work, school, leisure, etc. Can deal with most situations likely to arise whilst travelling in an area where the language is spoken. Can produce simple connected text on topics which are familiar or of personal interest. Can describe experiences and events, dreams, hopes and ambitions and briefly give reasons and explanations for opinions and plans.

Basic User A2
Can understand sentences and frequently used expressions related to areas of most immediate relevance (e.g. very basic personal and family information, shopping, local geography, employment). Can communicate in simple and routine tasks requiring a simple and direct exchange of information on familiar and routine matters. Can describe in simple terms aspects of his/her background, immediate environment and matters in areas of immediate need.

Basic User A1
Can understand and use familiar everyday expressions and very basic phrases aimed at the satisfaction of needs of a concrete type. Can introduce him/herself and others and can ask and answer questions about personal details such as where he/she lives, people he/she knows and things he/she has. Can interact in a simple way provided the other person talks slowly and clearly and is prepared to help.

www.coe.int/t/dg4/linguistic/Source/Framework_EN.pdf

Such assessment tasks are important because they tend to accent what a child can do, rather than typical classroom tests which focus on what cannot be done. Communicative skills, knowledge and understanding are profiled rather than marks, percentages or grades. Competence is highlighted rather than errors and deficiencies. Such tasks compare the learner with a scheme of language development, rather than comparing the child with other children. Such tests are formative, giving direct feedback and ‘feed-forward’ to enable curriculum decisions about individual learners.

Criterion-referenced language tests should provide direct feedback into the following areas:

teaching decisions (e.g. diagnosis of curriculum areas not mastered by an individual student); reporting to, and discussing achievement with, parents;
locating children needing special support and the type of curriculum support they need;
identifying children for accelerated learning;
informing about standards in the class in terms of curriculum development through a subject.

In the US, single tests to measure bilinguals’ proficiency are becoming less common than multiple criteria (Carrasquillo & Rodríguez, 2002). Structured approaches (e.g. published standardized tests, checklists) are often combined with inspection of portfolios of students’ work (e.g. assignments, journals) and teachers’ own insightful observations.

Self-rating on Proficiency
Students may be asked to assess their language strengths and weaknesses. An example to illustrate self-rating on language proficiency follows:

CAN YOU UNDERSTAND: ENGLISH? SPANISH? CAN YOU SPEAK: ENGLISH? SPANISH?
Yes – fluently ❒ ❒ Yes – fluently ❒ ❒
Yes – fairly well ❒ ❒ Yes – fairly well ❒ ❒
Yes – some ❒ ❒ Yes – some ❒ ❒
Yes – just a little ❒ ❒ Yes – just a little ❒ ❒
No – not now ❒ ❒ No – not now ❒ ❒
CAN YOU READ: ENGLISH? SPANISH? CAN YOU WRITE: ENGLISH? SPANISH?
Yes – fluently ❒ ❒ Yes – fluently ❒ ❒
Yes – fairly well ❒ ❒ Yes – fairly well ❒ ❒
Yes – some ❒ ❒ Yes – some ❒ ❒
Yes – just a little ❒ ❒ Yes – just a little ❒ ❒
No – not now ❒ ❒ No – not now ❒ ❒
Limitations in Measurement
Self-rating covers the basic four language abilities across two languages (e.g. Spanish and English). The answers are possibly too broad (e.g. there are many gradations possible in between each of the answers). Apart from this problem of scaling, there are other problems frequently encountered when measuring language competence. These may be listed as:

A MBIGUITY . Words such as ‘speak’, ‘understand’, ‘read’ and ‘write’ include a wide variety of levels of proficiency. The range is from an absolute beginner to Bloomfield’s (1933) maximum notion of ‘native-like control of two languages’. Tests also often contain only a small and unrepresentative sample of the totality of language proficiencies (e.g. reading in an academic context).
C ONTEXT . A bilingual may be able to understand a language in one context (e.g. a shop) and not in another context (e.g. an academic lecture). Another bilingual may be able to read newspapers but not textbooks. Proficiency and usage will vary with changing environments. A response summated across contexts is required. This may not be sensitive to different levels of proficiency across different contexts.
S OCIAL DESIRABILITY /S UBJECTIVITY . Respondents may consciously or sub-consciously give a ‘halo’ version of themselves. Self-ratings are vulnerable to exaggeration or understatement. People may say they are fluent in a second language (when in reality they are not) for identity, self-esteem or status reasons. Others may indicate they do not speak a language when they actually can. This may occur, for example, in a low prestige, ‘subtractive’ (see Chapter 4) minority language environment where the introduction of the second language may replace the first language. Questions about proficiency can be interpreted as about personal attitudes and identity and not just language.
A CQUIESCENT RESPONSE . There is a slight tendency for respondents to answer ‘Yes’ rather than ‘No’ in self-rating questions. It appears preferable to be positive rather than negative. This also tends to hold with a slight preference for ‘Agree’ rather than ‘Disagree’ and ‘Like Me’ rather than ‘Not Like Me’.
S ELF - AWARENESS . A self-rating depends on accuracy of knowledge about oneself. For one person, the frame of reference may be other neighborhood children who are not so fluent. When compared to children in another community, apparent fluency may be less. What is competent in one environment may seem less competent in another. The age, nature and location of the reference group may cause self-assessment not to be strictly comparable across a representative sample of people. A child may also self-rate on surface fluency and not be aware of much less fluency in cognitively demanding language tasks (Skutnabb-Kangas, 2000).
P OINT OF REFERENCE . There is a danger in using monolingual proficiency and performance as the point of comparison (see Chapter 1).
T EST AURA . Another danger is raising language measurement to the level of scientific measurement with an accompanying exaggerated mystique. More ‘natural’ forms of language sampling may be given lower status (e.g. recording natural conversation) as they rarely carry the mystique of educational and psychological (psychometric) measurement, but they may provide more authentic samples of language.
N ARROW SAMPLING OF DIMENSIONS OF LANGUAGE . Language measurement may unwittingly be perceived as something tangible and concrete (as when measuring height and weight). Rather, language tests mostly contain a specification of language skills that is debatable.
I NSENSITIVITY TO CHANGE . It is customary and seen as good practice to produce measurement which is reliable over time and across occasions (give consistent scores for the same individual over weeks or months). However, the paradox is that such measurement may be insensitive to change within individuals. Test scores need an expiry date (e.g. not valid after one year).