Content uploaded by Angel Lin
Author content
All content in this area was uploaded by Angel Lin on Jul 13, 2016
Content may be subject to copyright.
Bilingual Education: Southeast Asian Perspectives
Hong Kong University Press thanks Xu Bing for writing the Press’s name in his Square Word
Calligraphy for the covers of its books. For further information, see p. iv.
Bilingual Education:
Southeast Asian
Perspectives
Angel M. Y. Lin
Evelyn Y. F. Man
Foreword by Jim Cummins
Hong Kong University Press
14/F Hing Wai Centre
7 Tin Wan Praya Road
Aberdeen
Hong Kong
© Hong Kong University Press 2009
Hardback ISBN 978-962-209-958-6
Paperback ISBN 978-962-209-959-3
All rights reserved. No part of this publication may be reproduced or
transmitted, in any form or by any means, electronic or mechanical,
including photocopy, recording, or any information storage or
retrieval system, without prior permission in writing from the publisher.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
Secure On-line Ordering
http://www.hkupress.org
Printed and bound by xxxxxxxxxxx Hong Kong, China
Foreword vii
Introduction xi
Chapter 1
Language Policy and Planning in Southeast Asian Contexts 1
Chapter 2
Bilingual Education in Different Contexts: Principles and Practice 11
Chapter 3
Key Issues in Immersion Education: Implications for Hong Kong 41
Chapter 4
History and Development of Bilingual Education in Hong Kong 73
Chapter 5
Research on Bilingual Education in Hong Kong 87
Chapter 6
Negotiating between Nationalist and Globalization Agendas: 105
Lessons from the Divergent Paths of Singapore and Malaysia
Chapter 7
Contexts of Language Policy and Planning in Southeast Asian Societies: 119
Need for Innovative Approaches
Bibliography 143
Index 159
Content
Blank page
Every society invests a signicant proportion of its economic resources in educating its
youth. Despite the fact that there will usually be no economic return on this investment
for at least twelve years after students enter school, there is consensus in most societies
among business interests, policymakers, and the wider community regarding the
importance of education for ensuring future economic productivity and social stability.
The way any society organizes its education system reects its current social priorities
and the implicit images it continuously constructs of its own future identity.
Education is also the gateway to social and economic rewards for individual students
and for the social groups they were born into and represent. There is a very substantial
(and rapidly increasing) disparity in income between those who graduate from university
and those who have obtained only a basic secondary school qualication.
In view of its importance for the future of society and the economic rewards for
individuals and social groups associated with education, it is hardly surprising that debates
about the organization of education have become extremely volatile in many countries.
Elite groups in society — those with wealth, power, and privilege — invariably ensure
that their children receive a quality education either within the public system or by opting
out of the public system into the private school sector. Perhaps the most universal nding
in comparative education is that schools tend to reproduce the broader social structure;
those with cultural capital of various kinds (economic, linguistic, etc.) choose schools for
their children that will reinforce the cultural capital of the home. The result is that these
children enjoy major advantages over those whose home and school experiences provide
them with much less access to these forms of cultural capital.
Despite the fact that schools in most societies very clearly reproduce social
inequalities, policymakers tend to adopt the rhetoric of egalitarianism and meritocracy
in discussing education. Schools should provide all students with equal opportunities
to succeed according to their abilities. Tests and examinations ostensibly serve this
Foreword
viii Bilingual Education: Southeast Asian Perspectives
purpose; all students, regardless of social class or economic background, are judged by
the same criteria. It seems that students who come from more privileged socio-economic
backgrounds just happen to perform better according to these criteria.
Some of these differences in educational outcomes can be attributed to the
intellectually enriching experiences that money can buy: access to better schools, private
tutoring, books in the home, good nutrition, etc. However, the reproduction of inequality
can also be attributed directly to the failure of education policies in many countries.
Policymakers have frequently opted for quick-x sound-bite “solutions” that play better
in the political arena than they do in the classroom. Rarely have they been willing to apply
imagination and funding to implementing policies and classroom practices that build on
the cultural resources that less privileged children bring to the school, and extend these
cultural resources in linguistically and intellectually powerful ways. Individual projects
that have systematically adopted this approach have proved highly successful (e.g., the
Kamehameha Early Education Project in Hawaii, the Molteno Project in South Africa),
but inertia continues to reign when it comes to applying these insights on a larger scale.
Hopefully, the preceding statement will be refuted by the educational initiatives
that Hong Kong will undertake during the next decade. There is every indication from
the research it has initiated and the reports it has commissioned that the Hong Kong
administration is committed to nding solutions to the disparities of linguistic access
and educational achievement that have characterized its education system under British
colonial rule. In Hong Kong, linguistic access and educational achievement have been,
in the past, almost synonymous insofar as command of English signicantly inuenced
the extent to which students understood instruction conducted through that language.
English achievement also determined the extent to which students could enter the more
prestigious schools and universities.
Because of the unique history of English, command of the English language has
long served as a marker of privilege and access to power within Hong Kong. The global
dominance of English is creating similar alignments between linguistic (English) access
and socio-economic status in other countries around the world. In the global marketplace,
ability to function in what is rapidly becoming the common language of commerce is seen
by parents and policymakers alike as essential for social and economic advancement. Thus,
there is widespread consensus in Hong Kong (and many other economies) that the school
system should provide students with strong functional command of both conversational
and literate aspects of English. However, there is also considerable evidence that the policy
of immersing students in an English-medium environment at the secondary level has not
worked well for much of the school population. For students who already have access
to socio-economic and cultural capital in the home (e.g., highly educated parents who
speak English uently), immersion in English has not been overly problematic, although
their level of Chinese literacy has not developed to the extent that it might. However, for
the majority of students, secondary education has been a matter of developing classroom
survival strategies to meet the demands of learning through a language still inadequately
understood. In many cases, a supercial command of English has been gained at the
expense of both mother-tongue literacy development and overall educational attainment.
Foreword ix
The current policy dilemma for Hong Kong educators and policymakers is how
to ensure that standards of English achievement are maintained, or improved, while
minimizing the social and educational costs of previous policies. Parental demand for
access to English-medium schools is extremely strong, and the imposition of restrictions
of access through streaming is perceived as inequitable and as reproductive of the
structures of privilege that existed in colonial times.
Are there alternatives to present and past policies that could ensure widespread and
equitable access to English prociency without sacricing the overall quality of education
that students receive? Is there the potential at this auspicious period in the history of Hong
Kong to organize the education system so that all students will have opportunities to develop
the ability to function effectively in English, Cantonese, and Putonghua? Can we conceive
of an education system that would simultaneously nurture among students a strong sense of
cultural identity together with condence in their linguistic and intellectual achievements?
Angel Lin and Evelyn Man provide clear, afrmative answers to all of these
questions in their analysis of policy options regarding medium of instruction in Hong
Kong schools. Drawing on the international literature on bilingual/immersion education,
together with recent research in the Hong Kong context, they demonstrate that there
are potentially different routes to bilingual and trilingual prociency. The either/or split
between English-medium and Chinese-medium options serves neither the interests of
students nor the society as a whole. It is possible to conceive of alternatives that provide
greater support for Chinese literacy development within predominantly English-medium
schools while at the same time providing bilingual or partial immersion options within
predominantly Chinese-medium schools. Rather than a sudden immersion into English-
medium instruction, a gradual transition is likely to be much less traumatic for both
students and teachers, and more effective in supporting linguistic and intellectual
development. While there will always be variation in outcomes among students, much
more can be done to align instruction with the cultural and linguistic knowledge that
students bring to school (as illustrated in the Molteno Project in South Africa) so that the
potential of all students is nurtured more effectively than is currently the case.
It is a privilege to write a Foreword to a volume that so lucidly and succinctly
articulates the policy options available to the Hong Kong administration in its search
to develop strong Chinese and English abilities in an equitable way among its school
population. During the past twenty-ve years, many countries and regions have been
faced with crucial language planning decisions as they have gained autonomy from
oppression or external control (e.g., many African countries, newly independent countries
in the Pacic Basin, the Basque Country and Catalonia in Spain, etc.). Clearly, in all
these contexts broader socio-political considerations inuence policy decisions at least
as much as do specically educational considerations. Sometimes, unfortunately, what is
in the best interests of children has played little role in language policy decisions, partly
because there has been little awareness of the research information regarding bilingual/
immersion education or the policy alternatives implemented elsewhere.
This report ensures that policymakers in Hong Kong will not be in this situation. I
know of no other language planning context in which the international experience regarding
x Bilingual Education: Southeast Asian Perspectives
bilingual/immersion education has been synthesized in such a useful and powerful way
and its implications for policy highlighted so insightfully. Fortunately, there appears to be
considerable consistency between socio-political and educational imperatives in the Hong
Kong context. Maintenance of Hong Kong’s status as an international city at the crossroads
of global commerce with direct access to the two most powerful languages in the world
(Chinese and English) requires that the educational system succeed in developing strong
Chinese and English language skills among a signicant proportion of its population.
This necessarily involves changing the predominant orientation with respect to medium
of instruction from either/or to both/and — it is feasible to develop both Chinese and
English literacy skills to a high level within the educational system rather than privileging
one over the other for different sectors of the population. It is not necessary to create an
elite group that has access to English while the bulk of the population remains excluded.
There is strong evidence from the international research and experience that the teaching
of English can be dramatically improved within predominantly Chinese-medium schools.
There is also strong evidence that a more rigorous approach to Chinese literacy within
predominantly English-medium schools might strengthen students’ linguistic foundation
at no cost to their English literacy or conversational skills development.
The challenge of implementing these directions should not be underestimated. It is
no easy task to change educational structures or the mindsets that support them. Change
is likely to be regarded with concern, even suspicion, by stakeholders in the system —
parents, teachers, and administrators. Yet there is also the potential to mobilize enthusiasm
and support for a vision of education and the future of Hong Kong society that refuses
to reproduce the structures of privilege that existed in colonial times. The educational
options described in this volume will potentially enable a much greater proportion of
students to aspire to bilingual and trilingual prociency than was previously the case. As
a result of their education, these students will have access to powerful linguistic resources
to enable their voices to be heard within the international arena. Part of the language
planning process, and the challenge for policymakers, is to communicate this vision of
what is possible to the population in such a way that they are willing to invest their
children’s future in these possibilities.
The rigorous analysis of educational options articulated in this volume confers
credibility on the claim that much more can be achieved educationally in the Hong
Kong context. The authors have made an immense contribution not only to Hong Kong’s
future as an international city but also to the development of language policies in many
other contexts in which confusion still reigns regarding the most appropriate educational
options for developing the language abilities necessary to communicate and thrive in our
global village.
Jim Cummins
Questions regarding whether a rst or a second/foreign language should be used as a
medium of instruction (MOI) in schools (and if yes, for whom, and when) have been
enthusiastically debated in recent years in Hong Kong. The public debates, however,
have largely not been able to benet from the existing international body of research in
bilingual and immersion education or the educational experiences of other regions. The
reason is that such knowledge is often either couched in specialized, technical language
or scattered over diverse journals and books, which are often off-putting to teachers,
parents, school principals, policymakers and the general public. There is an urgent need
to critically integrate and review the international research literature with a view to
informing public debates and policymaking regarding the medium of instruction in Hong
Kong schools. In January 1999, we obtained a research grant from the Hong Kong Special
Administrative Region Government’s Standing Committee on Language Education and
Research (SCOLAR) to embark on such a study. The objectives of the study were:
• To identify and critically examine the theories, concepts, and various options and
models relevant to an understanding of using L1/L2 as a medium of instruction in
schools;
• To identify and critically examine practical studies and empirical research on the
use of L1/L2 as a medium of instruction, and various options in using different
languages as instructional media at different stages of education which may be
relevant and applicable to the Hong Kong context;
• To identify and critically examine the best current practices in the world on using
students’ L1/L2 as a medium of instruction, giving special consideration to the
experiences of countries or areas reverting from using a second language to using
the mother tongue as an instructional medium; and
• To identify the conditions under which successful practices elsewhere might be
applicable in the local context to assist language planners, policymakers and school
Introduction
xii Bilingual Education: Southeast Asian Perspectives
practitioners to make informed decisions about the language(s) of instruction in
schools to raise language standards in Hong Kong.
The critical literature report we wrote in response to the task set for us by SCOLAR has
become only the starting point of the present book. Over the years, we have witnessed
signicant changes in both the policies of the Hong Kong government and those of other
Southeast Asian societies. The present book has come a long way from our starting point,
and now includes coverage of the recent developments of bilingual education policies in
Hong Kong, Singapore and Malaysia.
The book can be divided into three main parts. Part I consists of three chapters that
give an overview of the basic principles and prototypical models of bilingual education
originating in North America and Europe. It provides the reader with a general background
for understanding key issues in immersion education, a category within bilingual
education, and their implications for other Southeast Asian societies. Part II consists of
two chapters. Chapter 4 focuses on discussing the bilingual education policy changes
that happened after Hong Kong was returned to China by Britain. Chapter 5 summarizes
signicant research studies on Hong Kong’s bilingual education policy and practice, to
provide a picture of what we already know and what we still do not know about the
bilingual education situation in Hong Kong. Part III consists of two chapters. Chapter 6
compares the divergent paths in language education policy taken up by Singapore and
Malaysia since independence. While Singapore has taken a consistently pragmatic path
from day one, Malaysia’s nationalist MOI policy since independence and its recent policy
to bring back English as MOI for science and math subjects in the schools reects a
difcult, tension-lled negotiation between nationalist and globalization imperatives.
Chapter 7 outlines different theoretical frameworks and models for language-in-education
(LIE) planning and proposes and discusses the pros and cons of a range of policy options
based on these frameworks.
In concluding this introduction, we want to thank Professor Jim Cummins for his
encouragement and support for our work all through these years. We hope that this book
will serve as a bridge between international research and local research on bilingual
education and inform the debates in policymaking regarding MOI issues in Hong Kong
as well as in other Southeast Asian societies.
Angel Lin and Evelyn Man
Hong Kong, January 2008
The Global Spread of English
English has become an everyday presence in many cosmopolitan cities in Southeast Asia.
For instance, in international airports in Seoul, Singapore, Hong Kong and Kuala Lumpur,
bilingual or multilingual signs are everywhere, and among them are always English signs.
The global spread of English has arisen from a host of historical, political and socio-
economic factors. In many Southeast Asian contexts such as Singapore, Hong Kong and
Malaysia, where English was historically a colonial language imposed by former British
colonial governments, English has carried with it the baggage of colonial histories and
exploitations. However, today English has also become a predominant medium of global
trade, nance and commerce, science, technology and the Internet. It serves as a chief
medium of communication for different peoples coming from both within and beyond Asia.
It is a common scene in Southeast Asian cities that people of diverse ethnic backgrounds are
communicating in some variety of English. So, has English shaken off (or merely masked?)
its colonial history and become a widely used “lingua franca” (or common language)
1Language Policy and Planning in
Southeast Asian Contexts
In this chapter, debates revolving around the global spread of English, linguistic
imperialism, World Englishes and the theories of postcolonial performativity
are introduced. Is English a cultural imperialistic tool of the West, or is English
being increasingly hybridized and used for their own daily purposes by many
Southeast Asian people? How is language policy and planning (LPP) related
to the creation of social and educational (in)equalities? The chapter discusses
these issues and concludes with the proposal that Southeast Asian postcolonial
societies need to develop their own LPP frameworks.
2 Bilingual Education: Southeast Asian Perspectives
for intercultural communication among peoples from Southeast Asia and beyond? In the
following sections we review different arguments and perspectives on this question.
Linguistic Imperialism
The theory of linguistic imperialism has been put forward to describe and explain,
among other phenomena, the global spread and domination of English (Phillipson, 1992,
1994, 1997, 1998). Imperialism is typically chacterized by exploitation, penetration,
fragmentation, and marginalization of native peoples, their labours, cultures and resources.
Imperialism has taken many forms, including economic, political, military, cultural and
social penetrations and exploitations. Linguistic imperialism, which is a form of cultural
imperialism, “permeates all other types of imperialism, since language is the means used
to mediate and express them” (Phillipson, 1992, p. 65). English linguistic imperialism is
one example of linguicism, a notion dened by Phillipson (1992, p. 47) as:
ideologies, structures, and practices which are used to legitimate, effectuate, and
reproduce an unequal division of power and resources (both material and immaterial)
between groups which are dened on the basis of language.
Phillipson (1992) argues that in the postcolonial era — in the last phase of English linguistic
imperialism — the ex-colonizers need not be physically present in the “Periphery”
countries, for there exists an indigenous English-educated elite who identify with the
ex-colonizers’ Anglocentric interests and values, typically through having studied in a
“Centre” country, and it is in their own interests to perpetuate the domination of English in
their home countries at the expense of the natural use and development of the indigenous
language(s). The Centre countries, especially the UK and the US, exercise inuence
through hegemonic language ideologies (or ideas about language) by dictating the norms
of “standard” English to learners and speakers of English in Periphery countries.
Why do people in former colonies seem to willingly accept the continued domination
of English in their societies? To explain this, Phillipson (1992) draws on the Gramscian
notion of “hegemony”, which prevails in the third and last stage of imperialism called
“neo-neo-colonialism”:
The sophistication of the arguments grows on a scale advancing from the use of force to
the use of carrots to the use of ideas. At one stage, the colonial power could use coercion
when selling one of its products, English. When the counterpart became slightly more
equal, and brute force could no longer be applied or was no longer an ethically acceptable
alternative, carrots were more suitable. But the ideal way to make people do what you
want is of course to make them want it themselves, and to make them believe that it is
good for them. This simplies the role of the ‘seller’, who then can appear as ‘helping’ or
‘giving aid’, rather than ‘forcing’ or ‘bargaining with’ the victim. (p. 286)
The notion of “hegemony” attempts to explain why some ex-colonial peoples seem
to embrace their former colonizers’ cultures and languages as superior to their native
Language Policy and Planning in Southeast Asian Contexts 3
cultures and languages. English, for instance, can be seen to be standing in a hegemonic
relationship to many former British-colonized peoples when they have internalized
(e.g., through education and socialization) the belief that English is intrinsically a better
language for science and technology, for arts and cultures, is superior to their own native
language, or is the marker of civilized, modern citizenship. We shall return to a discussion
of different perspectives on this issue in the section on multlilingual and multicultural
identities in Southeast Asia. Now, let us look at another related set of questions revolving
around which variety of English should serve as a standard for learners in non-Anglo
countries, for instance, in the former colonies of Britain in Southeast Asia.
World Englishes
In many Southeast Asian cities such as Singapore and Hong Kong, it is common to nd
people conversing in different varieties of English. Which variety (or varieties) should be
put forward by language planners and educators as the pedagogical model(s) to teach and
learn in schools? In many ex-British colonies, British English norms have been used as
the target norms. American English has also become important because of its increasing
trade and political presence in Southeast Asia in the post-war era.
Researchers of World Englishes (Kachru, 1985, 1992, 1997) have differentiated
among different kinds of English, chiey based on geographic locations and national
boundaries. Those English varieties spoken in Anglo countries (e.g., the UK, the US,
Australia, Canada) are called “core” or “inner circle” varieties, while those spoken as
second languages (ESL) (e.g., India) are called “outer circle” varieties. Those spoken
in places as foreign languages (EFL) are called “expanding circle” varieties. An image
of three concentric circles (inner circle, outer circle, expanding circle) is used to build a
model of a hierarchy of Englishes, each having different status and authority. The inner
circle varieties are norm-giving; the outer circle norm-developing; and the expanding
circle varieties norm-dependent. This means that many learners in Southeast Asian
societies will nd themselves trying to learn the target varieties in the inner circle (e.g.,
British English or American English).
The World English (WE) paradigm (or theoretical framework) has changed our
concept of “English” from a monolithic notion to a pluralistic notion of “Englishes”.
That is, there is not one single legitimate English in the world; there are many legitimate
Englishes. It has also highlighted the notion of ownership of English by people in ex-
colonies of Anglo-speaking countries. That means English no longer belongs only to the
former colonial masters. Different Englishes are now being developed and appropriated
(i.e., taken as their own) in their own right by peoples who use them as their rst or
second languages and very often as a marker of their own identities (e.g., in Singapore
and India).
While the WE paradigm has been seen as progressive in liberalizing the ownership
of English and in pluralizing English, some researchers have observed that it misleadingly
presupposes that all people in a nation necessarily speak the same national variety of
4 Bilingual Education: Southeast Asian Perspectives
English. It has also failed to question the native speaker/non-native speaker (NS/NNS)
dichotomy in any profound fashion (Graddol, 1997; Pennycook, 2003). It continues to
privilege native speakers in the inner circle (as norm-giving) over non-native speakers,
and then ESL speakers in the outer circle (as norm-developing) over EFL speakers in
the expanding circles (as norm-dependent). We shall return to a discussion of this issue
later when we discuss developing frameworks for language policy and planning that are
appropriate in Southeast Asian contexts. In the next section, let us return to the questions
we raised at the beginning of the section on linguistic imperialism.
Emergence of Hybridized Multilingual and Multicultural Identities in
Southeast Asia
Is English an imposed language, a vehicle of linguistic and cultural imperialism, and a
killer language that threatens the continued existence (e.g., learning and use) of other
natural languages and cultures in ex-colonies (e.g., Phillipson, 1997)? Or is it merely
a medium for international communication that exists side by side with other local
languages which different peoples keep for expressing their local identities (e.g., Crystal,
1997)? It seems that both positions seem to be a simplication of what usually is a much
more complex situation. Instead of trying to argue for one or the other position in the
abstract, perhaps we should go beyond such a totalizing, dichotomous way of thinking
and actually look at each specic sociocultural context in all its concrete complexities.
For instance, in a study (Lai, 2003) of young people’s cultural identication patterns
and language attitudes, it was found that young people who identify themselves as Hong
Kongers are also affectively inclined towards both Cantonese and English. To them,
Cantonese and English are not mutually exclusive and they nd it natural (or almost
impossible; see Li & Tse, 2002) to mix English words into their everyday Cantonese.
Also, given the special socio-political, historical context of Hong Kong, it seems that
many Hong Kong people did not entirely accept British colonial rule in the pre-1997
era and yet are equally ambivalent about Socialist Chinese domination in the post-1997
era. Such mixed, ambivalent feelings in national and sociocultural identication seem to
correlate with the freely intertwining of Cantonese and English words in the everyday
public life of Hong Kong people, and these “non-pure” linguistic practices seem to be
playing an important role in marking out the Hong Kong identity — they seem to serve
as distinctive linguistic and cultural markers of “Hong Kong-ness” and seem to constitute
some deant acts of identity. It is almost like saying: We’re Hong Kong-ese and I don’t
care whether I’m speaking “pure Chinese/English” or not!
In this sense, then, if “Singlish” is a linguistic marker of the distinctive local
Singaporean identity (Chua, 2003), then the so-called “mixed code” of Hong Kong is
its counterpart in Hong Kong. Like Singlish, the so-called “Hong Kong mixed code” is
not a monolithic, stable entity. In practice, it consists of a whole continuum of different
styles of speaking and writing, from the use of here and there a few English lexical items
in otherwise Cantonese utterances/sentences to the intertwining of extended English and
Cantonese utterances/sentences (Lin, 2000). From the perspectives of performativity
Language Policy and Planning in Southeast Asian Contexts 5
theory on languages and communication resources (Pennycook, 2004), it is a better idea
not to view languages as separate stable systems with solid boundaries. As Pennycook
(2004) argues, the idea of languages as discrete, stable, monolithic entities with solid
boundaries is actually the product of colonial knowledge production. In practice, people
draw on a whole range of linguistic resources which cannot be easily pigeonholed as
“separate languages” in their everyday linguistic practices. Parallel to these hybridized
linguistic practices are their similarly hybridized sociocultural identities. At least among
many Hong Kong people as we witness it today, there do not seem to be any clear-
cut “pure” sociocultural identities: Hong Kong people’s identity seems to be always a
“hyphenated” one, indicating its “in-between-ness” (Abbas, 1997). We would want to
argue that, as a result of the rise of international cosmopolitan cities in Southeast Asia and
the rise of a whole new generation of bilingual speakers in these cosmopolitan cities, we
shall witness the rise of cosmopolitan varieties of Asian Englishes (Lin & Shim, 2004)
which will not t comfortably into the hierarchical WE paradigm of core, inner, outer
or expanding circles. We predict that these cosmopolitan speakers of Asian Englishes
will increasingly seek to assert the legitimacy and status of their speaking styles on an
equal footing with Anglo-American English speaking styles, ultimately bringing about a
paradigm shift in institutionalizing what target models to teach, learn and test in schools
in these societies (Luk & Lin, 2005), although as things stand now we are still a long way
from reaching these goals.
Recent research has actually found that, in East Asian cosmopolitan cities, there are
increasing transnational popular cultural ows and linguistic hybridization taking place.
For instance, Rip Slyme, a popular rap group in Japan, has used English in their lyrics to
fashion a kind of “double” identity (Pennycook, 2003):
… Rip Slyme locate their Japaneseness explicitly, yet at the same time they use the
English word for Japanese, seeming in the same instant to refashion their identity from
the outside. This Japanese identity is then both ‘freaky’ and ‘double’, the latter a recently
coined term to describe people of mixed origin. (p. 527)
Admitting research in this direction is still scarce, Pennycook (2003) concludes his article
by hinting at the development of bilingual and bicultural identities through transnational
popular cultural practices mediated by both English and local languages:
How does the use of English work as it locates its users both as part of the global imagined
community of English users and as participants in the global music industry, creating
links through the ‘international language’ and yet relocating through its juxtaposition
with Japanese? How do these new global raplishes work as tools for the performance of
identities? (p. 529)
What the theory of linguistic imperialism fails to show is perhaps how English can be
actively taken up, how people can actually appropriate (i.e., claim ownership of) English
and why people strategically choose to use English (Lin, Wang, Akamatsu & Riazi,
2002). Pennycook (2003) observes that the linguistic imperialism theory cannot account
for a sense of agency, resistance, or appropriation on the part of ex-colonized peoples. It
6 Bilingual Education: Southeast Asian Perspectives
tends to construct ex-colonized peoples as passive victims (Li, 2002). Somehow between
the dichotomous positions of uncritically celebrating the global spread of English as
an innocuous tool for communication, science and technology (Crystal, 1997), and
constructing English as a monolithic universal killer language colonizing relentlessly
the linguistic and cultural habitats of ex-colonial societies, we have to steer a level-
headed, middle way by taking a socioculturally situated perspective; i.e., we need to
look at each sociocultural context in all its complexities before jumping to a conclusion.
Going beyond the debate between the “imperialism-resistance” theories (e.g., Phillipson,
1992) and the “postcolonial performativity” theories (e.g., Pennycook, 2003, 2004), we
have to nd a way of understanding and exposing new forms of inequalities in education
and society and new productions of subaltern subjectivities (i.e., marginalized identities
and an underclass sense of self; see Ashcroft, Grifths & Tifn, 1998) under forces of
globalization. While doing critical education analysis we must also be wary of falling
into the trap of doing merely essentialist identity politics (e.g., arguing that one’s L1 must
be more important than one’s L2). Rather, we must struggle to study the new material
and institutional conditions that might lead to social and educational inequalities, and to
explore practical alternatives in LPP policy and practice.
The New Cosmopolitan Bilingual Elites and the Newly Ghetto-ized Locals
under Globalization-driven Bilingual Education Policies
As much as we would want to celebrate the new opportunities that globalization has
seemed to offer us in reworking and refashioning our identities as new transnational, global
Cosmopolitans, unbound by old forms of essentialist nationalism and culturalism and
binary frameworks of identity politics, we also see the anxieties created by globalization
forces. Zygmunt Bauman (1998) points out this economic underside of globalization in
his book, Globalization: The Human Consequences:
In the words of John Kavanagh of the Washington Institute of Policy Research:
Globalization has given more opportunities for the extremely wealthy to make money
more quickly. These individuals have utilized the latest technology to move large sums
of money around the globe extremely quickly and speculate ever more efciently.
Unfortunately, the technology makes no impact on the lives of the world poor. In fact,
globalization is a paradox: while it is very benecial to a very few, it leaves out or
marginalizes two-thirds of the world’s population.
As the folklore of the new generation of ‘enlightened classes’, gestated in the new, brave
and monetarist world of nomadic capital, would have it, opening up sluices and dynamiting
all state-maintained dams will make the world a free place for everybody. According to such
folkloristic beliefs, freedom (of trade and capital mobility, rst and foremost) is the hothouse
in which wealth would grow faster than ever before; and once the wealth is multiplied, there
will be more of it for everybody. The poor of the world — whether old or new, hereditary
or computer-made — would hardly recognize their plight in this folkloristic ction. … New
Language Policy and Planning in Southeast Asian Contexts 7
fortunes are born, sprout and ourish in the virtual reality, tightly isolated from the old-
fashioned rough-and-ready realities of the poor. The creation of wealth is on the way to
nally emancipating itself from its perennial — constraining and vexing — connections
with making things, processing materials, creating jobs and managing people. The old rich
needed the poor to make and keep them rich. That dependency at all times mitigated the
conict of interest and prompted some effort, however tenuous, to care. The new rich do not
need the poor any more. (Bauman, 1998, pp. 71–2).
Thus, increasingly under the forces of economic globalization, entire factories and
jobs can disappear overnight from one locality as fast, nomadic global capital holds no
allegiance to communities in any locality and roams from one locality to another across
the Globe searching for ever-cheaper land and labour (Bauman, 1998). Also, while
the Cosmopolitan multilingual elite well-versed in global English and new knowledge
technologies (often mediated through global English) can nd jobs anywhere across
the Globe (i.e., gaining transnational mobility), those monolingual locals who never
catch on to the new skills and new global languages (often due to lack of class-based
capital and habitus; see discussion below) are ever more locked up in non-mobility both
geographically and socio-economically.
In his plenary paper given at the Crossroads Conference of the International
Association of Cultural Studies on 25 June 2004, Larry Grossberg urged cultural studies
scholars and critical theorists to go beyond the mere analysis of expressive culture (e.g.,
popular culture and media), but to also pay attention to the policies of the state and the
global ows of capital. He urged critical theorists, educators and cultural researchers to
do what he called ‘conjunctual analysis’ — to analyze the historical conjunctures formed
by both cultural and material, economic and political forces. We also see a parallel in
critical education analysis in postcolonial studies of LPP contexts. For instance, Rani
Rubdy, in her paper on Singapore’s bilingual education policy (Rubdy, 2005), gives us
a sharp analysis of how the state’s English-dominant LPP framework has been driven
by both agendas of economic globalization and political management of different ethnic
groups. This has resulted in the wiping out of Chinese dialects (including Hokkien, which
used to be the mother tongue of the majority of Chinese in Singapore), creating cross-
generational linguistic and cultural discontinuities (e.g., English-speaking grandchildren
cannot communicate with Hokkien-speaking grandparents), and indirectly fostering the
development of “Singlish”, an ofcially denigrated but popularly spoken hybridized
Hokkien-sounding variety of English (Rubdy, 2005). Singlish, as a hybridized linguistic
variety, is certainly not a language and trademark of the high-ying Cosmopolitan
Singaporean identity but instead a marker of local Singaporean identity and a medium for
parodying ofcial discourses (Chua, 2003). While the Cosmopolitan, global Singaporean
can sometimes switch to Singlish for a joke or for showing “authentic” Singaporeanness,
what socially straties the Cosmopolitan multilingual high-yer from the monolingual
Singlish/Hokkien-speaking ghetto-ized local is their differential access to and differential
degrees of mastery of global or “standard” English. In the words of Singaporean
sociologist and cultural theorist Chua Beng-huat:
8 Bilingual Education: Southeast Asian Perspectives
The consequence, after 20 years [of the state’s language policies], is that Hokkien,
along with all other Chinese languages, has become a language spoken by those who
have never received a formal education and/or those who did not make the grade in the
highly competitive bilingual education system. It is thus reduced to a language of the
lowest-educated section of the working class and the illiterate. The linguistic hierarchy,
in order of economic and political advantages, is thus English, Mandarin and Hokkien,
as depicted in the lm, Money (Chua, 2003, p. 169).
… Ah Beng and his female counterpart, Ah Lian [in a popular TV sitcom] are
two caricatures of the Singlish-speaking Singaporeans who are ‘adoringly’ laughable to
the middle-class English-educated writers and audience, for whom switching code from
standard English to Singlish is a marker of ‘authentic’ Singaporean identity.
… Not surprisingly the show’s [here Chua is referring to another popular TV
sitcom] popularity is based on the ‘silliness’ of the Phua and his ‘Lian’ wife, even as
they triumph repeatedly over their commonsense-decient, university-educated architect
brother and his Westernised pretentious wife. (Chua, 2003, p. 162)
We must point out that our concern here lies not so much in a nostalgic mourning of
the loss of linguistic diversity (though this is a legitimate concern to many) as in the
production of socio-economic disadvantage and new subaltern identities: consequences
of Singapore’s Cosmopolitan-oriented ruling elite’s emptying out of the “local” (e.g.,
local languages) in their thorough-going pursuit of the “global” (e.g., standard English,
Standard Mandarin Chinese) under their linguistic engineering policies driven by
globalization desires. The emergence of Singlish as a surrogate for the “linguistic local”,
and in some sense as a reincarnation of Hokkien, is certainly not anticipated by the state’s
linguistic engineers and might represent the poor’s linguistic resistant “weapon”, a local
linguistic spectre that lingers on to continue to embarrass and haunt the authorities (see
Rubdy, 2005). However, as acquisition of standard English correlates with family-based
capital and “habitus” (see Lin, 1999, 2005), we can expect new forms of socio-economic
stratication along the lines of social class mediated by the (un)availability of family
capital predisposing the use and acquisition of global, standard forms of English versus
local hybridized forms of English such as Singlish in Singapore or “mixed codes” in
Hong Kong (see Lin, 2000). Here, we nd Bourdieu’s critical analytic tools very useful,
as delineated by British sociologist Nick Crossley (2003, p. 43):
Class-based cultural advantages are passed from parents to children through the habitus,
but as pre-reective and habitual acquisitions they are generally misrecognized within
the school system as ‘natural talents’ and are rewarded ‘appropriately’. The school thus
launders cultural advantages, albeit unwittingly, transforming them into the hard and
clean currency of qualications.
We see parallel social and linguistic processes taking shape in Iran (Lin, Wang, Akamatsu,
& Riazi, 2005) when parents who can afford it send their children to private schools
to acquire globally marketable communicative competences in English that the public
schools cannot offer, and in Turkey when private institutions that teach in the English
Language Policy and Planning in Southeast Asian Contexts 9
medium multiply and compete with state universities which struggle to offer a balance
of local languages and English. The economic drive (both from the local communities
and the postcolonial governments) towards the emptying out of the “linguistic local”
and the one-sided pursuit of the “linguistic global” must be problematized in our study
of the consequences of globalization. While taking care not to embrace linguistic and
cultural essentialism as a simple, reductionist reaction to these globalizing forces, we
have to both critically examine these processes in their consequences (e.g., new forms
of social stratication and new productions of subaltern subjectivities; see Lin, 2005),
and explore new alternatives in policy, pedagogy and curriculum drawing on theoretical
frameworks that go beyond traditional discrete models of languages, cultures and
literacies (Canagarajah, 2005).
Thus, in the nal chapter of this book, we shall look at the political roles played
by state LPP in shaping the linguistic, educational and cultural habitats of society. We
shall outline major approaches to LPP and propose that we need to develop our own
approaches which are appropriate for use in postcolonial Southeast Asian contexts and
that languages should not be seen and planned as discrete, separate entities but rather as
continua (Hornberger, 2003; Canagarajah, 2005).
Questions for Discussion
1. Pennycook (2003) uses Rip Slyme as an example of transnational popular cultural
ows and linguistic hybridization. Find more such examples from your own
sociocultural contexts and analyze them.
2. As an English user (if your mother tongue is not English), do you demonstrate
different sociocultural identities, either implicitly or explicitly, in different contexts?
Conduct a self-analysis by reecting on your own everyday linguistic practices.
3. Reect on your own English teaching or learning experiences to discuss whether
globalization-driven bilingual education policies in your area/country really lead
to social stratication and subaltern subjectivities. If they do, what is (are) the new
alternative(s) in language, pedagogy and curriculum that can improve the present
situation?
Blank page
Bilingual Education: Sociohistorical Contexts, Aims and Types
Bilingual education is a simple label for a complex phenomenon (Cazden & Snow, 1990).
It relates to a society’s “debates about the fundamental purposes and aims of education
in general: for individuals, communities, regions and nations” and must be considered as
situated in complex sociohistorical, economic, cultural and political contexts (Baker, 2001,
pp. 183–4). To understand the great variety of bilingual education programmes in the world,
one useful perspective is to consider the different sociohistorical contexts that have given
rise to the different aims of bilingual education and its diverse programme types.
Diversity of Aims and Contexts
Baker (2001, p. 193) cited Ferguson, Houghton and Wells’s (1977) examples of the
diverse aims of bilingual education as follows:
2Bilingual Education in Different
Contexts: Principles and Practice
This chapter presents an overview of the theoretical and empirical literature
of bilingual education which is relevant to an understanding of how a second
or foreign language (L2) can be used as a medium of instruction in schools
for the dual goal of achieving bilingualism/biliteracy and cognitive/academic
learning. Major theories, principles, concepts and programme options/models
in bilingual education are delineated. The factors and conditions promoting
or inhibiting success in bilingual education are critically reviewed. Their
implications for Hong Kong and other Southeast Asian contexts are discussed.
12 Bilingual Education: Southeast Asian Perspectives
1. To assimilate individuals or groups into the mainstream society; to socialize people
for full participation in the community;
2. To unify a multilingual society; to bring unity to a multi-ethnic, multi-tribal, or
multinational linguistically diverse state;
3. To enable people to communicate with the outside world;
4. To provide language skills which are marketable, aiding employment and status;
5. To preserve ethnic and religious identity;
6. To reconcile and mediate between different linguistic and political communities;
7. To spread the use of a colonizing language, socializing an entire population to a
colonial existence;
8. To strengthen elite groups and preserve their position in society;
9. To give equal status in law to languages of unequal status in daily life; and
10. To deepen understanding of language and culture.
Different aims seem to be important in different societal/political contexts. For instance,
aim (1) is important in the United States under the American “melting-pot” ideology to
assimilate non-English-speaking linguistic minorities/immigrants into the mainstream
unilingual English-speaking society, whereas aims (2) and (6) are far more important
in multi-ethnic societies like Singapore and Malaysia, for the political importance of
promoting national unity, social cohesion and inter-ethnic harmony. In pre-1997 Hong
Kong, aims (7) and (8) might be relevant. However, in post-1997 Hong Kong, it seems
that aims (3) and (4) should have become the more important goals. Also apparently
important in the post-1997 context is the not-often-mentioned sociocultural goal to
foster in the younger generation a sense of roots and pride in the Chinese language
and culture as well as a sense of belonging to the mother country of China, while at the
same time developing in them an international outlook and adequate English skills to
participate and compete in the global economy as well as scientic and technological
endeavours.
The above diverse list of aims also shows that bilingual education can be used by
language education planners as a means to a range of ends, which often goes beyond
merely linguistic and educational considerations. As Baker (2001, p. 193) pointed out:
... bilingual education does not necessarily concern the balanced use and development of
two languages in the classroom. Behind bilingual education are varying and conicting
philosophies of what education is for. Sociocultural, political and economic issues are
ever present in the debate over the provision of bilingual education. (Emphasis added)
It seems that, in designing language education policies in Hong Kong, the important
rst step is to explicate and prioritize the range of goals that are widely considered to be
important in and for Hong Kong. Policymakers, however, can anticipate animated public
debates and discussions on what constitute the most important goals and what appear
to be the optimal programmes in language education. It seems that, to achieve greater
success in implementing language education policies, these policies need to be rst
legitimized or supported by some public consensus. In the process of policy legitimation
Bilingual Education in Different Contexts: Principles and Practice 13
or public consensus-building, it is important to make accessible to the public research and
scientic information regarding different aspects of bilingual education. For instance, it
is important:
1. To inform the public of the different, often conicting goals and priorities in language
education;
2. To inform the public of the necessary conditions for success (often requiring certain
school, family and community resources), as well as the costs and benets of
different programme options that are available and feasible to achieve the goals,
under current government resource constraints and
3. To generate and promote informed public discussion on how to prioritize the different
goals and select from different feasible programme options to achieve the goals with
a view to resource implications.
Three Main Types of Programme: Maintenance, Transitional and
Enrichment
Another way of classifying bilingual education programmes would be to consider the
following set of variables proposed by Baker (1996):
1. Typical language(s) used by the child in daily life
2. Typical language(s) used in the classroom
3. The educational/societal aim(s) of the programme
4. The probable outcomes of the programme
Based on the above variables, ten types of bilingual education programmes can be
distinguished (summarized from Baker, 2001, pp. 193–201):
1. Submersion education/the “sink or swim” method, e.g., in the US
2. Submersion with withdrawal/pull-out/sheltered English classes, e.g., in the US
3. Segregationist education, e.g., in South Africa before Nelson Mandela was elected
president.
4. Transitional bilingual education (early exit or late exit) employing bilingual teachers
(as a result of advocacy by minorities), e.g., in the US
5. Mainstream education (with a foreign language taught as a subject), e.g., “core
French” and “drip-feed” French programmes in Canada.
6. Separatist education, e.g., small isolationist religious schools.
7. Immersion bilingual education, e.g., early total, early partial, delayed immersion,
late immersion programmes in Canada.
8. Maintenance and heritage language bilingual education, e.g., Navajo and Spanish in
the US, Ukrainian in Canada, Maori in New Zealand.
9. Two Way/Dual Language Immersion, e.g., in certain elementary schools in the
US
10. Bilingual education in majority languages, e.g., certain schools in Luxembourg, the
European Schools Movement.
14 Bilingual Education: Southeast Asian Perspectives
These ten types of bilingual education programme are summarized and contrasted in Table
1 (from Baker, 2001, p. 194). To facilitate discussion, these diverse types of bilingual
education programmes can be classied into the following three broad categories based
on their educational/societal aims:
Weak Forms of Education for Bilingualism
Type of Typical Type Language of Societal and Aim in Language
Program of Child the Classroom Educational Aim Outcome
Submersion Language Majority Assimilation Monolingualism
(Structured Minority Language
Immersion)
Submersion with Language Majority Assimilation Monolingualism
Withdrawal Minority Language with
Classes/Sheltered “Pull-out” L2
English Lessons
Segregationist Language Minority Apartheid Monolingualism
Minority Language
(forced, no
choice)
Transitional Language Moves from Assimilation Relative
Minority Minority to Monolingualism
Majority
Language
Mainstream Language Majority Limited Limited
with Foreign Majority Language with Enrichment Bilingualism
Language L2/FL Lessons
Teaching
Separatist Language Minority Detachment/ Limited
Minority Language Autonomy Bilingualism
(out of choice)
Strong Forms of Education for Bilingualism and Biliteracy
Type of Typical Type Language of Societal and Aim in Language
Program of Child the Classroom Educational Aim Outcome
Immersion Language Bilingual with Pluralism and Bilingualism &
Majority Initial Emphasis Enrichment Biliteracy
on L2
Maintenance/ Language Bilingual with Maintenance, Bilingualism &
Heritage Minority Emphasis on L1 Pluralism and Biliteracy
Language Enrichment
Two-Way/Dual Mixed Language Minority and Maintenance, Bilingualism &
Language Minority & Majority Pluralism and Biliteracy
Majority Enrichment
Mainstream Language Two Majority Maintenance, Bilingualism &
Bilingual Majority Languages Pluralism and Biliteracy
Enrichment
Note: (1) L1 = rst language; L2 = second language; FL = foreign language.
(2) Formulation of this table owes much to discussions with Professor Ofelia Garcia. This
typology is extended to 14 types of bilingual education in Garcia (1997, p. 410).
Table 1. A Summary of Different Types of Bilingual Education Programmes
(from Baker, 2001, p. 194)
Bilingual Education in Different Contexts: Principles and Practice 15
1. Maintenance programmes
2. Transitional programmes
3. Enrichment programmes
These three types of programme were rst differentiated by Fishman (1976). For Fishman,
Maintenance bilingual education programmes aim at maintaining linguistic minority/
immigrant students’ rst language while providing them with access to the dominant
language (L2) of the society through using the students’ rst language (L1) as a medium
of instruction in the early years of schooling or, in Two Way/Dual Language programmes,
through using both the students’ L1 and L2 as mediums of instruction for different subjects
or on alternate days. Transitional programmes, in contrast, aim at helping linguistic
minority/immigrant students to go through a more or less gradual transition from using
some of their L1 to using only the mainstream language as the medium of instruction. The
aim of transitional programmes is the assimilation of linguistic minorities/immigrants
into the monolingual mainstream society.
While both maintenance and transitional programmes have arisen from the needs of
the linguistic minority/immigrant students, enrichment programmes (also called “additive
bilingual” programmes), in contrast, have been designed for the majority language
students. Typically, the parents of these majority language students want their children to
master a high level of prociency in a socio-economically important L2 in addition to,
not in replacement of, the dominant societal daily life language (L1). This is done through
using the L2 as a medium of instruction in all or some of the subjects (e.g., total or partial
French immersion programmes for English-speaking students in Canada). Different
variants of the enrichment programme model are likely to be relevant to the situation
of Hong Kong, where bilingual education aims at enabling the majority Cantonese-
speaking group to acquire English as an additional language without weakening their
rst language.1 As the enrichment model is realized as different variants of immersion
education, the rest of this chapter is devoted to a critical review of the following three
successful programme models in immersion education: (1) Canadian French Immersion,
(2) European Schools, and (3) Two Way/Dual Language Immersion. The following
review draws on major references in bilingual and multilingual education which include:
Cummins, 1995, 2000; Baker and Jones, 1998 ; Johnson and Swain, 1997; Tung, 1992,
1. It has to be pointed out that the rst language or mother tongue of the majority of students
in Hong Kong is in fact Cantonese, not Modern Standard Chinese or Putonghua (the spoken
form of Modern Standard Chinese) (Tung, 1992). The daily life language of most Hong Kong
students is a colloquial variety of spoken Cantonese and Cantonese-style written Chinese
which is quite different from Modern Standard Chinese (e.g., Cantonese-style Chinese is used
in comics, popular youth magazines, colloquialized newspapers). Modern Standard Chinese
literacy is, therefore, not strongly supported in the daily life of the students in the larger
societal context. The educational implications of this complex sociolinguistic phenomenon in
Hong Kong are discussed in this chapter.
16 Bilingual Education: Southeast Asian Perspectives
1996; Cummins and Corson, 1997; Baker, 2001; Day and Shapson, 1996; Beardsmore,
1993, 1995; Bernhardt, 1992; Harley, Allen, Cummins and Swain, 1990; Genesee, 1987;
Swain and Lapkin, 1982; Freeman, 1998; Baker and Hornberger, 2001; Lindholm-Leary,
2001.
Canadian French Immersion
As prototypes of immersion education, the Canadian French immersion programmes
have often been cited as successful examples of using a second or foreign language as a
medium of instruction for achieving high levels of prociency in the L2 without sacricing
the L1 and academic learning. It is, therefore, important to consider the origins, design
features, programme outcomes and necessary conditions for the success of the Canadian
French Immersion Model, to derive possible implications for Hong Kong schools.
Origins
The rst French immersion programme was started in the mid-1960s as an innovative
educational experiment, when a group of vocal, middle-class English-speaking Canadian
parents in St. Lambert, Quebec, lobbied their school board for improvements to the teaching
of French as a second language. These parents had read accounts of different forms of
bilingual education that might serve as superior alternatives to the traditional French-as-
a-subject programme (“Core French”) which focused on grammar, memorization, and
drill and had not provided their children with sufcient skills to work in French, or to
socialize with French speakers. Collaborating with scholars in bilingualism at McGill
University, the St. Lambert parents proposed to their school board a radical departure from
any existing FSL (French as a second language) programme in Canada: a programme in
which their unilingual English-speaking children were taught entirely in French from
Kindergarten or Primary 1, with English language arts formally introduced in Primary 2
or 3 and about half the time devoted to each language from Primary 4 through 6. By the
late 1960s, the rest of Canada was becoming aware of the socio-economic and political
value of achieving a high level of prociency in French, and various French immersion
programmes modelled on or adapted from the original St. Lambert programme have
spread to other provinces. By the 1990s, French immersion programmes were offered
optionally by some school boards. In several school boards, enrollment may be as high as
50%, the rest of the students going to rst language medium schools. Across the country,
however, “only approximately 7% of the entire student population attends an immersion
programme” (Johnson & Swain, 1997, p. 2).
Design Features
The following sections aim at presenting an overview of the design and structure of
Canadian French Immersion Programmes.
Bilingual Education in Different Contexts: Principles and Practice 17
n Goals of Canadian French Immersion Programmes
The goals of Canadian French Immersion are typical of those of the enrichment model of
bilingual education. As summarized in Baker (2001, p. 204), the stated aims of Canadian
French Immersion are for students who are English-speaking Canadians:
1. to become competent to speak, read and write in French;
2. to reach normal achievement levels throughout the curriculum including the English
language; and
3. to appreciate the traditions and culture of French-speaking Canadians as well as
English-speaking Canadians.
It can be seen from these goals that additive bilingualism is the ultimate goal of Canadian
French Immersion programmes. It is expected that students in these programmes will
become bilingual and bicultural without any loss of academic achievement and rst
language competence.
n Principal Programme Types in Canadian French Immersion
Based on different combinations of values for the two design variables of (1) extent
of immersion and (2) beginning level of immersion, four types of Canadian French
Immersion Programmes which are commonly found can be outlined as follows:
1. Early total immersion
L2 is used in all lessons right from Kindergarten or Grade 1. L2 use gradually decreases to
approximately 80% in Grades 2 to 5 and approximately 50% in Grades 6 to 8. Ultimately
in Grades 9 to 12, approximately 40% of lessons are taught in L2.
2. Early partial immersion2
Early partial immersion is characterized by approximately 50% of L2 use from
Kindergarten through Grade 8. L2 use decreases to approximately 35% in Grades 9 and
10, and ultimately to approximately 30% in Grades 11 and 12.
3. Middle (delayed) partial immersion
L2 is learnt only as a subject from Kindergarten through Grade 3. However, starting in
Grade 4, approximately 80% of lessons are taught in L2. L2 use decreases to approximately
50% in Grades 7 and 8, and to approximately 40% in Grades 9 through 12.
4. Late partial immersion
L2 is learnt only as a subject from kindergarten through Grade 6. However, starting
in Grade 7, approximately 80% of lessons are taught in L2. L2 use decreases to
approximately 50% in Grades 9 through 12.
2. The “mixed mode” design often mentioned in public discourses in Hong Kong is an example
of partial immersion programmes. Also, the European Schools Model and the Two Way/Dual
Language Model reviewed in this chapter can be seen as examples of the mixed mode design.
18 Bilingual Education: Southeast Asian Perspectives
K 1 2 3 4 5 6 7 8 9 10 11 12
0
10
20
30
40
50
60
70
80
90
100
Percentage
Year/Grade in School
K = Kindergarten
% of English medium lessons % of French medium lessons
K 1 2 3 4 5 6 7 8 9 10 11 12
0
10
20
30
40
50
60
70
80
90
100
Percentage
Year/Grade in School
K = Kindergarten
% of English medium lessons % of French medium lessons
Second Language
French Lessons
Figure 1. Major Types of French Immersion Programme (from Baker, 2001, p. 206)
The early total immersion programme (closest to the original St. Lambert experimental
design) is the most popular programme, followed by late and then middle (also called
delayed) immersion. There is also a rare type of “late, late partial immersion” under
which L2 is rst introduced as a medium of instruction as late as university level, and in
one or two subjects only (e.g., at the University of Ottawa, there is an option of studying
Introductory Psychology in the students’ L2; see Burger, Wesche & Migneron, 1997).
Figure 1 compares and contrasts the relative proportions of lessons taught in L1
and L2 across different grade levels under the four major types of French immersion
Bilingual Education in Different Contexts: Principles and Practice 19
programmes in Canada described above. Table 2 outlines a description of some French
immersion programmes in Ontario to Grade 8. It can be seen from Figure 1 and Table
2 that there is considerable exibility in designing and implementing variants of French
immersion in Canada. Different school boards in different districts have made adaptations
to the generic French Immersion Model (originating from the St. Lambert experiment) to
suit the needs of their own student population. The implication for Hong Kong seems to
be that language education planners can explore different immersion models apart from
the late total immersion model traditionally practised in Hong Kong.
Percentage
K 1 2 3 4 5 6 7 8 9 10 11 12
0
10
20
30
40
50
60
70
80
90
100
Year/Grade in School
K = Kindergarten
% of English medium lessons % of French medium lessons
K 1 2 3 4 5 6 7 8 9 10 11 12
0
10
20
30
40
50
60
70
80
90
100
Year/Grade in School
K = Kindergarten
% of English medium lessons % of French medium lessons
Percentage
Also Second Language
Lessons in French
Figure 1. (Continued)
20 Bilingual Education: Southeast Asian Perspectives
n Levels of L2 Prociency Achieved and Amount of Classroom Exposure to L2
Swain and Lapkin (1982), based on their extensive research experience with immersion
education, have outlined the expected L2 learning outcomes in relation to different amounts
of classroom exposure to L2 (including exposure gained in both French immersion and
French-as-a-subject) as follows (p. 14):
Table 2. French Immersion Programmes in Ontario (from Swain & Lapkin, 1982, p. 49)
Program Board of Board Grade Description Accumulated
Education Terminology Program Hours of French
Begins at End of Grade 8
Peel County Late Partial 8 Grade 6: core French 625–780
Immersion (30 minutes daily)
Grade 7: core French
(20 minutes daily)
Grade 8: 55–70% of
curriculum in French
Late Toronto Late 7 Students have 700–870
Immersion Extended varying core French
backgrounds prior to
entering programmes
and have accumulated
from 90 to 315 hours
of core French
instruction to end of
Grade 6 Grade 7:
25–30% French
Grade 8: 40% French
Ottawa Late-Entry 6 K–Grade 5: core 2,145
Immersion French (20 minutes
daily)
Grade 6: 100% French
Grade 7: 50% French
Grade 8: 50% French
Carleton Late-Entry 7 K–Grade 6: core 1,845
Immersion French (20 minutes
daily)
Grade 7: 80% French
Grade 8: 80% French
Early Elgin County Early Partial 1 Grades 1 to 8: 50% 3,330
Partial Immersion French
Immersion
Early Total Ottawa, Early K K–Grade 1: 100% 4,450–4,985
Immersion Carleton Immersion French
Grades 2 to 4: 80%
French
Grade 5: 65%-80%
French
Grades 6 to 8: 50%
French
Bilingual Education in Different Contexts: Principles and Practice 21
1,200 hours This should enable students to reach the basic level of prociency, i.e., “a
fundamental knowledge of the language, the ability to participate in simple
conversation, the ability to read simple texts, and the ability to resume the
study of French in later life”.
2,100 hours This should enable students to reach the middle level of prociency, i.e.,
they should be able to “read newspapers and books of personal interest with
occasional help from a dictionary, to understand radio and television, to
participate adequately in conversation, and to function reasonably well in a
French-speaking community after a few months’ residence”.
5,000 hours This should enable students to reach the top level of prociency, i.e., the
student can “continue his or her education using French as the language of
instruction at the college or university level, to accept employment using
French as the working language, and to participate easily in conversation”.
Table 2 shows the different amounts of L2 exposure time in hours under different
immersion programmes. From this table and the above information, one might infer
that early immersion models yield the greatest amount of L2 exposure time and should,
therefore, yield the highest level of L2 learning outcomes, compared with late immersion
models. However, Swain and Lapkin (1982) have also made some important observations
on the relationship between L2 exposure time and L2 prociency outcomes. Their views
are summarized as follows:
1. Cumulative exposure alone (e.g., spread thinly over a long time) may be less
important than intensity of exposure.
2. Other things being equal, older students tend to be more efcient second language
learners than younger students, especially in the areas of grammar, abstract concepts,
classifying and drawing generalizations. They are more able to consciously reect
on and make use of language learning strategies. Their cognitive maturity and
foundation in L1 literacy enables them to perform better in L2 literacy-related tasks
than younger learners.
3. Programmes that use class time effectively (e.g., with a variety of comprehensible
input and uninhibited, motivated practice opportunities) can achieve more in less
time; i.e., the rich, interactive quality of L2 exposure is more important than the
mere amount of L2 exposure.
Programme Outcomes of Canadian French Immersion
Cummins (1999a) has provided a succinct summary of major programme outcomes of
Canadian French immersion. The following account is taken from Cummins (1999a, pp.
4–5; emphasis added):
Consistent ndings have been obtained from French immersion programme evaluations
across Canada. In early immersion programmes, students gain uency and literacy
in French at no long-term cost to their English academic skills. Within a year of the
22 Bilingual Education: Southeast Asian Perspectives
introduction of formal English language arts students catch up in most aspects of English
test performance. Usually students require additional time to catch up in English spelling
but by grade 5 there are normally no differences in English test performance between
immersion students and comparison groups whose instruction has been totally through
English. One potential limitation of these ndings is that standardized tests do not assess
all aspects of English academic skills; in particular, writing development is usually
excluded from such tests. However, the few studies that have examined English writing
development specically show no evidence of problems among immersion students in
this regard (Swain, 1997). There is also no evidence of any long-term lag in mastery of
subject matter taught through French in early, middle or late immersion programmes.
With respect to French skills, students’ receptive skills in French are better
developed (in relation to native speaker norms) than are their expressive skills. By the
end of elementary school (grade 6) students are close to the level of native speakers in
understanding and reading of French but there are signicant gaps between them and
native speakers in spoken and written French (Harley, Allen, Cummins & Swain, 1990).
Similar ndings are obtained for late immersion programmes. French skills
develop well in the rst two years of the programme and differences between students in
intensive forms of late immersion (100% French in grades 7 and 8) and those who have
come through an early immersion programme are relatively minor. The early immersion
programme students are generally more uent and comfortable in French but the late
immersion students show somewhat greater accuracy in their mastery of grammatical
constructions. As Swain (1997) notes in her review: “The evidence emerging from the
variety of immersion programmes with different starting ages suggests that older students
may possess cognitive characteristics which give them an advantage in learning certain
aspects of a second language” (p. 266).
As seen from Cummin’s account, the learning outcomes of different versions of Canadian
French Immersion seem to be largely positive. However, since the context of Canadian
French Immersion is likely to be very different from other societal contexts, it is important
to analyze the conditions for the success of Canadian French immersion before any
generalization of its effectiveness can be made.
Conditions for Success in Canadian French Immersion
The research literature strongly indicates that the success of Canadian immersion
programmes depends on a number of important conditions (Swain & Lapkin, 1982; Baker,
2001; Johnson & Swain, 1997; Cummins, 1999a). These conditions, further elaborated in
Part II of this volume, can be summarized as follows (Tung, 1996):
1. Parental involvement is important, for attention and material support, as well as
providing a home environment which is rich in support for L1 and L2 linguistic and
literacy development, e.g., rich print environment at home.
2. Both students and parents are members of the majority group in the society; i.e.,
the students’ L1 is not at risk but secure and prestigious in the society. The larger
Bilingual Education in Different Contexts: Principles and Practice 23
sociolinguistic context also supports the use and development of the students’ L1
speech and literacy.
3. The immersion programme is optional. Students can choose to leave the programme.
Students remaining in the immersion programme are therefore likely to be those
motivated to study in the L2.
4. Both students and parents hold positive attitudes towards French and French-
Canadians.
5. High quality of teachers is necessary, which means high standards of professional
training and high prociency levels in both languages.
6. An interactive style of teaching (rather than a teacher-fronted, didactic style of
teaching) provides a variety of high-quality input (e.g., in a range of language
functions) as well as rich opportunities for students’ productive language use.
Of the above-listed conditions/factors, teacher professional preparation and instructional
approach/teaching methodology seem to be areas which are, relatively speaking, most
amenable to the teacher training/curriculum planning efforts of government and language
education planners. So, here, we devote some more space to an examination of classroom
strategies used by effective French immersion teachers and of the instructional principles
upheld in Canadian immersion education.
For instance, Swain and Lapkin (1982) have delineated the instructional principles
of early total immersion, summarized as follows:
1. Students should receive the same type of education as they would in the regular
English programme, but the medium of instruction through which content is
presented and discussed is French. This provides a naturalistic setting for second
language acquisition.
2. The teacher accepts and starts from the existing language, interests and skills of
the children. The language acquisition process is seen as “production owing from
comprehension”.
3. The teacher’s focus is on conveying the content to the students and on responding to
the content of what the students are saying, whether it is said in broken French or the
home language, English.
4. The early emphasis is on teaching relevant vocabulary in the context of conveying
real messages through the use of pictures, gestures, and other body language cues.
5. Explicit instruction in grammar is given when the children get older (e.g., Grade 3)
in separate lessons.
Snow (1990, quoted in Baker, 2001, p. 337) has also provided a list of ten specic
techniques that tend to be used by experienced and effective immersion teachers:
1. Providing plenty of contextual support for the language being used (e.g., by body
language — plenty of gestures, facial expressions and acting).
2. Deliberately giving more classroom directions and organizational advice to
immersion students. For example, signalling the start and the end of different
routines, more explicit directions with homework and assignments.
24 Bilingual Education: Southeast Asian Perspectives
3. Understanding where a child is at, thereby connecting the unfamiliar with the
familiar, the known with the unknown. New material is linked directly and explicitly
with the child’s present knowledge and understanding.
4. Extensive use of visual material. Using concrete objects to illustrate lessons, using
pictures and audio-visual aids, giving the child plenty of hands-on manipulative
activities to ensure all senses are used in the educational experience.
5. Obtaining constant feedback regarding the level of a student’s understanding.
Diagnosing the level of a student’s language.
6. Using plenty of repetition, summaries, restatement to ensure that students understand
the directions of the teacher.
7. The teacher being a role model for language emulation by the student.
8. Indirect error correction rather than constantly faulting students. Teachers ensure that
the corrections are built into their language to make a quick and immediate impact.
9. Using plenty of variety in both general learning tasks and in language learning tasks.
10. Using frequent and varied methods to check the understanding level of the
children.
One would notice, however, that many of these instructional strategies and principles seem
to be more appropriate for early rather than late immersion learners, and for dealing with
early grade-level academic content, which tends to be less abstract and context-reduced
than higher grade-level content. Most research in Canada has also been conducted on
early immersion at the elementary level (Grades K–6) whereas, for the most part, in
Hong Kong English-medium instruction begins at the secondary level. We therefore need
to exercise caution when interpreting the Canadian research ndings in the contexts of
Hong Kong or other societies.
European Schools
Another famous successful example of immersion education is the European Schools
Model. The European Schools have, however, sometimes been charged with elitism for
their expensive and exclusive nature. Here is what Beardsmore (1993, pp. 3–4) said of
the European Schools:
The least productive model included in this collection is that of the European Schools,
given that it is not destined for expansion, is expensive to operate and could be taxed
with elitism. Nevertheless, the immense practical experience gained from this complex
form of multilingual education and the many insights it offers on how to handle mixed
populations on an equal footing should provide elements of inspiration. As a model,
it is unlikely to be adopted elsewhere. It differs signicantly from many so-called
international schools, however, in that, unlike the latter, it is genuinely multilingual
both in programme and in outcome, whereas most so-called international schools are
only international in population make-up, to some extent in curriculum, but rarely so in
languages on offer.
Bilingual Education in Different Contexts: Principles and Practice 25
The European Schools Model (Beardsmore, 1993, 1995; Cummins, 1995; Tung, 1996)
is therefore outlined here mainly for the insights that Hong Kong language education
planners might gain regarding the question of how to effectively design and run a small
number of top-quality, innovative schools which can provide the society and economy
with a top-notch workforce that is fully biliterate (in English and Standard Chinese) and
fully trilingual (in Cantonese, English and Putonghua), albeit small in size. Nevertheless,
some of the principles drawn from the practical experience of the European Schools can
also be adapted for application in less elitist blingual education models. The following
sections on the features of the European Schools are based on Beardsmore (1995),
Cummins (1995) and Tung (1996).
Origins and Goals of the European Schools
The European Schools were rst established in 1958 by the education authorities of
the twelve member states of the European Economic Community for children of civil
servants working for one of the supra-national European institutions. The schools use
multilingualism as a tool to promote a European identity, but at the same time ensure
the development of the students’ rst language and cultural identity. Students are taught
through at least two languages, and are required to learn a third language as a subject.
Organization of the European Schools
Each school consists of various linguistic sub-sections so that children are enrolled
initially in the sub-section using their rst language. In the primary school, therefore, the
children are mainly educated through their rst language. A second language (English,
French or German) is taught as a subject from the very beginning. Teachers are seconded
from schools of the various countries. New teachers are assigned a mentor to help them
adjust to the special circumstances of the school. There are also seminars and retraining
sessions for teachers. Unpaid co-ordinators are elected from teachers to ensure co-
ordination of activities across grades and languages. Remedial teachers are employed to
provide additional L2 instruction for students joining the school after Grade 1.
Curriculum of the European Schools
Primary school education lasts for ve years (see Table 3). All children follow the
same programme, irrespective of the language of instruction. When L2 is used as the
medium of instruction, it is used in cognitively undemanding and contextualized subjects
such as physical education and the European Hours. The latter involves children from
different linguistic sub-sections in co-operative activities such as sewing and cooking.
The intention is to let children from different countries interact before prejudices about
people of different origins can be formed.
26 Bilingual Education: Southeast Asian Perspectives
Secondary school education consists of three phases:
1. Grades 6–8: the “observation cycle” (see Table 4)
The role of L2 is extended in this phase, but it is still used in activities that are
relatively context-embedded and cognitively undemanding. Examples of
complementary activities are electronics, computer science, photography, painting,
typing and needlework.
Table 3. Primary School Curriculum in European Schools (from Beardsmore,
1995, p. 40)
1st and 2nd grades 3rd, 4th and 5th grades
L1 as a subject 16 × 30 mins L1 as a subject 9 × 45 mins
Mathematics 8 × 30 mins Mathematics 7 × 45 mins
L2 as a subject 5 × 30 mins L2 as a subject 5 × 45 mins
Music 3 × 30 mins Environmental studies 4 × 45 mins
Art 4 × 30 mins Art 1 × 45 mins
Physical education 4 × 30 mins Music 1 × 45 mins
Environmental studies 2 × 30 mins Physical education 1 × 45 mins
Religion or ethics 2 × 30 mins European Hours 3 × 45 mins
Recreation 7 × 30 mins Religion or ethics 2 × 45 mins
Total 33 × 30 mins Total 33 × 45 mins
per week per week
Note: All subjects are taught through the L1 of the section except those in italics.
Subject Grade 6 Grade 7 Grade 8
L1 as a subject 5 × 45 mins 5 × 45 mins 4 × 45 mins
Mathematics 4 × 45 mins 4 × 45 mins 4 × 45 mins
Latin (optional) - - 4 × 45 mins
Integrated science 4 × 45 mins 4 × 45 mins 4 × 45 mins
Religion or ethics 2 × 45 mins 2 × 45 mins 2 × 45 mins
Human sciences* 3 × 45 mins 3 × 45 mins 3 × 45 mins
L3 as a subject - - 4 × 45 mins
L2 as a subject 5 × 45 mins 4 × 45 mins 4 × 45 mins
Graphic & plastic arts 2 × 45 mins 2 × 45 mins 2 × 45 mins
Music 2 × 45 mins 2 × 45 mins 2 × 45 mins
Physical education 3 × 45 mins 3 × 45 mins 3 × 45 mins
Complementary activities 2 × 45 mins 2 × 45 mins 2 × 45 mins
Total 32 × 45 mins 32 × 45 mins 31/33 × 45 mins
* Taught through the medium of the L2 in Grade 8.
Note: a) All subjects are taught through the L1 of the section except those in italics.
b) Those who select Latin may drop either music or graphic and plastic arts.
c) In Grade 6, two optional complementary activities must be selected but may be
dropped in Grades 7 and 8.
Table 4. Secondary School Curriculum (Grades 6–8) in European Schools
(from Beardsmore, 1995, p. 43)
Bilingual Education in Different Contexts: Principles and Practice 27
Subject Grades 9 and 10 Elective courses
L1 as a subject 4 × 45 mins Latin 4 × 45mins
Religion or ethics 1 × 45 mins Greek 4 × 45mins
Biology 2 × 45 mins Economics & social sciences 4 × 45mins
Chemistry 2 × 45 mins Plastic arts 2 × 45mins
Physics 2 × 45 mins Music 2 × 45mins
Mathematics* 4 or 6 × 45 mins L4 as a subject 4 × 45mins
L2 as a subject 3 × 45 mins
Physical education 2 × 45 mins
History 2× 45 mins
Geography 2 × 45 mins
L3 as a subject 3 × 45 mins
Total 27 or 29 × 45mins
* The medium of mathematics depends on the pupil’s choice
Note: a) All subjects are taught through the L1 of the section except those in italics.
b) Enough elective subjects must be chosen to guarantee a curriculum of minimum 31 and
maximum 35 periods per week.
Table 5. Secondary School Curriculum (Grades 9–10) in European Schools
(from Beardsmore, 1995, p. 45) (align Beardsmore with Secondary)
2. Grades 9–10: the “semi-specialization cycle” (see Table 5)
Students have more options in this phase. Depending on their choice of elective
courses, over half of the curriculum can be conducted in the students’ L2.
3. Grades 11–12: the “specialization cycle” (see Table 6)
The specialization cycle leads to the European Baccalaureate, the internationally
recognized European school leaving certicate. This means that written examinations
are identical, irrespective of the language they are written in.
Conditions for Success and Learning Outcomes of the European Schools
The conditions leading to the success of the European Schools are summarized as
follows:
1. All teachers are bilingual, but they teach only in their native language. All other staff
members are also bi- or multilingual.
2. The learning of an L2 is given relevance in that there are children in the same
school speaking the L2 as a native language and that the L2 is used as a medium
of instruction from Grade 3. All children go through the same process of transition
from using L1 as a medium of instruction to using L2 as a medium of instruction.
28 Bilingual Education: Southeast Asian Perspectives
3. In the primary school, oral competence in the L2 is emphasized over written
competence, which is the focus of the secondary school. The focus in the primary
school is on basic vocabulary and sentence structures.
4. Examinations are not important until after Grade 8. Students in early grades can,
therefore, focus more on developing language and thinking skills rather than studying
mainly for examinations, as is often the case in Hong Kong.
5. The learning of a second language takes place in a non-threatening environment,
because all students have to use a weaker language at some time.
6. L1 is maintained both as a subject and as a medium of instruction for at least some
content subjects throughout primary and secondary schooling.
7. Teaching in the L2 moves gradually from cognitively undemanding and context-
embedded activities (e.g., physical education, music, European Hours, sewing and
cooking) to cognitively demanding and context-reduced activities (e.g., history,
geography, social sciences).
Table 6. Secondary School Curriculum (Grades 11–12) in European Schools
(from Beardsmore, 1995, p. 46)
Compulsory subjects Grades 11 & 12
L1 as a subject 4 × 45 mins
Philosophy 2 × 45 mins
Mathematics 3 or 5 × 45 mins
L2 as a subject 3 × 45 mins
History 2 × 45 mins
Geography 2 × 45 mins
Physical education 2 × 45 mins
Religion or ethics 1 × 45 mins
Elective courses
Latin 5 × 45 mins
Greek 5 × 45 mins
Physics 4 × 45 mins
Chemistry 4 or 5 × 45 mins
Biology 4 or 5 × 45 mins
L3 as a subject 3 × 45 mins
L4 as a subject 3 × 45 mins
Advanced course in L1 as a subject 3 × 45 mins
Advanced course in mathematics 3 × 45 mins
Advanced physics and chemistry 2 × 45 mins
Economics 5 × 45 mins
Advanced course in L2 as a subject 3 × 45 mins
Advanced course in geography 2 × 45 mins
Advanced course in history 2 × 45 mins
Plastic arts 2 × 45 mins
Music 2 × 45 mins
Sociology 2 × 45 mins
Other subjects 2 × 45 mins
Note: a) All subjects are taught through the L1 of the section except those in italics.
b) A pupil must have a timetable with a minimum of 31 and a maximum of 35 periods per
week.
Bilingual Education in Different Contexts: Principles and Practice 29
The learning outcomes of the European Schools are summarized as follows:
1. After 1,300 hours of instruction in French as an L2 at the European School of
Brussels (where French is the dominant social language), the students’ competence
in the language is comparable to Canadian French Immersion students after 4,500
hours of instruction. This shows the importance of the larger social and interaction
context of students in language learning.
2. When the students graduate, their best friends in school are often from a different
linguistic sub-section. Most students have developed positive attitudes towards the
L2 culture and L2-speaking people.
3. When students nish secondary school, their productive competence in L1 and L2
is of a standard that would allow them to take examinations in either language. The
most important factor was found to be the relevance and authentic purpose students
felt in learning and using the L2.
It can be seen from the above summary that European Schools seem to provide an even
more effective way of developing L2 prociency than Canadian French Immersion.
The key factor seems to be the mixed or multi-ethnic and multicultural nature of the
student population. The L2 takes on authentic communicative functions for the students
in the school environment, and this greatly enhances students’ L2 learning motivation,
L2 opportunities for use and thus L2 development (a similar favourable condition also
obtained for the Two Way/Dual Language Model discussed below).
The European Schools’ early emphasis on building a solid foundation in L1 literacy
also seems to be an important factor, as valuable literacy-related skills can be transferred
to the L2 later in the students’ school career (Cummins, 2000). The emphasis on early
exposure to L2 both explicitly as a subject and implicitly as a natural social language
among students from different ethnic groups also appears to be an important factor for
success.
Finally, the way in which L2 as a medium of instruction is introduced also seems
to be a key factor for success: L2 is introduced after the student has developed a solid
foundation in the L1 and has studied the L2 as a subject for some years.3 It is also
introduced gradually, rst in a few subjects, which are cognitively undemanding and
context-embedded (e.g., art work, music, European Hours), and only later gradually
to cognitively demanding and context-reduced subjects (e.g., history, social sciences).
However, throughout the school years, L1 remains an important school subject as well as
a medium of instruction in a number of key content subjects.
3. However, this is not to say that early L2 immersion cannot be effective; see discussion in
chapter 3 on early and late immersion.
30 Bilingual Education: Southeast Asian Perspectives
Two Way/Dual Language Programmes
The third successful example of immersion education is the Two Way/Dual Language
model. In Two Way/Dual Language schools there are approximately equal numbers of
language minority and language majority students in the same classroom. This is a more
appropriate model than total immersion in situations where students’ L1 is not widely
supported in the larger society (e.g., Spanish-speaking Cuban immigrant children in
Florida, US) and thus needs to be maintained as a medium of instruction for at least half
of the lessons in school.
There are a variety of terms used to describe such schools: Two Way schools, Two
Way Immersion, Two Way Bilingual Education, Developmental Bilingual Education,
Dual Language Education, Bilingual Immersion, Double Immersion and Interlocking
Education. The rst Two Way schools in the US appear to date from 1963 in Dade
County, Florida, and were developed by the US Cuban community in that area. During
the 1960s, fteen Two Way schools were established in Dade County. Since that time,
there has been a steady rise in the number of Two Way schools in the US. There are over
170 Two Way schools in the US now. The languages of instruction are predominantly
Spanish/English (over 90% of such schools) but with the following combinations also
represented: Cantonese/English, Portuguese/English, Haitian Creole/English, Korean/
English, Russian/English, Japanese/English, and French/English. Around 85% of all
these schools operate from Kindergarten to Grade 6 (Baker, 2001).
Lindholm-Leary (2001) has documented in detail the critical features of successful
Dual Language Education programmes. These critical features are found in programme
design, implementation and its larger institutional and sociocultural contexts. Following
is a summary of the features described in Lindholm-Leary (2001).
1. Effective Leadership in School
(a) Administrative and principal support: This support is demonstrated in the
structural and functional integration of the programme within the total school
system. Resources are allocated to the programme for staff training and for
curriculum materials development in each language.
(b) Instructional leadership: This leadership may come from a vice-principal,
programme coordinator or resource teacher. The individual has extensive
knowledge of the dual language education model being implemented, second
language development, bilingual and immersion education theory and research,
instructional methodologies, effective classroom practices and the belief that the
selected language education model can work once it is implemented correctly.
The individual needs to be the programme spokesperson, oversee programme
development, planning, coordination and staff training.
2. School Environment
(a) A positive school environment: (i) An orderly, safe and warm environment that
facilitates learning; (ii) an instructional focus and commitment to achievement
that is shared by all; (iii) high expectations for all students; (iv) students who
Bilingual Education in Different Contexts: Principles and Practice 31
feel pride in their school; and (v) students (including minority students) who
participate in extracurricular activities.
(b) Additive bilingual environment: All students are given the opportunity to
acquire a second language at no cost to their home language and culture (i.e.,
additive bilingualism rather than subtractive bilingualism).
(c) Co-operative and reciprocal instructional climate: A co-operative and reciprocal
interaction model of teaching is more benecial than the traditional transmission
model of teaching in which the teacher’s task is to impart knowledge or skills to
students. Co-operative learning with appropriate grouping will produce positive
learning outcomes. Successful grouping requires:
• that students work interdependently;
• clearly conceived individual and group accountability for all group
members;
• social equity in the group and in the classroom (Cohen & Lotan, 1995;
Johnson & Johnson, 1990; Johnson et al., 1985; Slavin, 1995).
(d) Cross-cultural components: Effective schools have staff that are committed to
equality, respect students from multi-ethnic and multicultural backgrounds,
and use curriculum materials that represent ethnic and cultural diversity. Every
student is expected to participate in a variety of roles in the classroom, including
leadership roles, and students are socialized to treat each other with respect and
equity.
3. Teachers and Staff
Effective language education programmes have the following characteristics:
(a) High-quality instructional personnel: Teachers have appropriate teaching
credentials and are native or native-like in either or both of the languages in
which they are instructing, to provide cognitively stimulating instruction. They
must be able to at least understand the child’s mother tongue in the initial stages
of language learning, in order to respond appropriately in the second language to
the child’s utterances in the mother tongue, to provide comprehensible input and
maintain linguistic equity in the classroom. Many classrooms have an instructional
assistant for part of the time, and these assistants also require training.
(b) Staff training: There must be pre-service and in-service training in:
• the dual language education model, including bilingual and immersion
research and theory;
• second language development;
• instructional strategies in second language development;
• multicultural and educational equity training; and
• co-operative learning.
4. Instructional Design and Features
(a) Duration of instructional treatment: At least four to six years are required for
students to attain academic prociency, albeit not native-like prociency, in the
L2.
32 Bilingual Education: Southeast Asian Perspectives
(b) Exposure to optimal dual language input: Optimal input is adjusted to the
comprehension level of the learner, is interesting, challenging and relevant,
and is in sufcient quantity. This requires careful planning in the integration of
language instruction and subject matter presentation and, in the initial stages,
the use of slower, more expanded, simplied and repetitive speech oriented to
the “here and now”; e.g., highly contextualized language and gestures.
(c) Language output — promotion and opportunities: Students should be
encouraged to produce extended discourse, in which they are forced to make
their language coherent, accurate and sociolinguistically appropriate. This
requires both structured tasks and unstructured opportunities involving oral
production. It also requires enforcing a strong language policy in the classroom
that encourages students to use the instructional language.
(d) Focus on academic curriculum: The curriculum should be equivalent to that
for students at the same grades not enrolled in the dual language immersion
programme.
(e) Integrated language arts instruction: Explicit language instruction of grammar
is needed, but it should not follow the traditional translation and memorization
curriculum. It is important to develop a language arts curriculum that species
which linguistic structures are needed (e.g., conditional verb forms) and how
these structures are incorporated into the academic content (e.g., including
imperfect forms of verbs in history subject matter and conditional, future, and
subjective tenses of verbs in mathematics and science content).
(f) Meeting distinct needs during language arts instruction: Teachers need to
understand that they need to cultivate cognitive/academic language prociency
necessary for literacy, particularly for minority students who do not have
literacy support in the home. This requires cross-grade co-ordination planning.
(g) Literacy instruction in two languages: Should children be taught literacy in
their native language rst, and then have the second language added later? Can
children be taught literacy simultaneously in two languages, or will they be
confused? Understanding the familial and community context of the student
population is important in determining which model to use. Research has shown
that the less socially prestigious and powerful language in a society is the one
most subject to language loss. To promote this less powerful language, it should
receive more focus in the early stages of immersion. Also, non-English speaking
students whose parents cannot provide English literacy-related experiences
in the home should receive literacy instruction in their native language rst.
However, if the student’s native language is the societal prestigious language
and well supported in the home and in the community (e.g., language majority
Anglophone students in Canada and the US), early immersion in a second
language (e.g., French) does not place the student at risk of losing her or his
rst language. Research has shown that a gifted student population with ample
familial support can benet from early simultaneous literacy instruction in
both languages (e.g., English and Spanish), but initial instruction in the rst
Bilingual Education in Different Contexts: Principles and Practice 33
language (e.g., Spanish) may be more benecial for students coming from less
well-supported and linguistic minority families.
5. Students
(a) Classroom composition: The ratio of speakers of the two languages should not
be more than 2:1.
(b) Although there is little research on the literacy and achievement of African
American children in immersion programmes, there is some research that
indicates that African American children are not negatively affected and may
have positive outcomes (Holobow et al., 1991; Lindholm-Leary, 1994).
(c) Equality in treating students from different social class and language
backgrounds: In many schools, the English speakers come from middle-class
and educated families while speakers of the other language come from working-
class and undereducated families. These differences must be acknowledged to
ensure equal educational opportunity in the classroom for all students.
(d) Students with special learning needs: Students who have learning disabilities
are also accepted in dual language immersion programmes, except those with a
serious speech delay.
(e) Home/school collaboration: Effective programmes tend to incorporate a
variety of home/school collaboration activities. Children’s academic progress
increases when parents support their children’s literacy development at home
(e.g., activities such as reading and listening to children read).
In concluding her summary of the features of effective dual language immersion
programmes, Lindholm-Leary (2001) pointed out that not all features will be appropriate
for all programmes, particularly those serving more homogeneous student populations. It
is therefore important to judge the relevance of these features by taking into consideration
the specic contexts of different student populations. To gain a more concrete picture of
examples of effective dual language immersion programmes, we describe one specic
dual language programme.
Cummins (1999a) has provided a very useful review of one of the longest-
standing Two Way Immersion programmes in the United States: the Oyster Bilingual
School in Washington, DC. Its relevance for the Hong Kong context resides in its clear
demonstration of what can potentially be achieved through bilingual education even for
students who are from relatively low socio-economic backgrounds (Cummins, 1999a).
The following description of the Oyster Bilingual School is taken from Cummins (1999a,
pp. 8–9; emphasis added).
Oyster Bilingual School (Washington, DC). The bilingual programme was started
in 1971 and involves instruction in both Spanish and English for about 50% of the
time in each language from kindergarten through grade 6. Each class is taught by two
teachers, one responsible for English-medium instruction and one for Spanish-medium
instruction. This instructional organization is achieved by means of creative management
of resources rather than by additional external funds (personal communication Elena
Izquierdo, former principal of Oyster School).
34 Bilingual Education: Southeast Asian Perspectives
Students read in both languages each day so there is simultaneous development
of literacy in the two languages. The student body is comprised of approximately 60%
Spanish L1 (primarily Salvadorean) and 40% English L1 (about half African American,
half Euro-American).
The academic results of this programme have been outstanding. For example, at the
grade 3 level reading, mathematics, language and science scores were 1.6–1.8 median
grade equivalents above norms (percentiles 74–81). The grade 6 grade equivalents were
4.4–6.2 above norms (percentiles 85–96) (1991 data reported in Freeman, 1998).
According to Freeman (1998), the school has evolved a social identities project that
positively evaluates linguistic and cultural diversity and communicates this strongly to
students. In the words of one of the teachers: “it’s much more than language”. Freeman
provides detailed discourse analyses that illustrate how the interactions between educators
and students in Oyster Bilingual School “refuse” the discourse of subordination that
characterizes the treatment of minorities in the wider society and in most conventional
school contexts. She points out that the discourse practices in the school “reect an
ideological assumption that linguistic and cultural diversity is a resource to be developed
by all students, and not a problem that minority students must overcome in order to
participate and achieve at school” (p. 233). Specically, educators have choices in the
way they organize discourse practices and these choices entail signicant consequences
for both language minority and majority students. The school requires all students to
become bilingual and biliterate in Spanish and English, and “to expect, tolerate, and
respect diverse ways of interacting” (p. 27).
In other words, the school “aims to promote social change on the local level by socializing
children differently from the way children are socialized in mainstream US educational
discourse” (p. 27):
Rather than pressuring language minority students to assimilate to the positively
evaluated majority social identity (white middle-class native English-speaking) in order
to participate and achieve at school, the Oyster educational discourse is organized to
positively evaluate linguistic and cultural diversity. ... this socializing discourse makes
possible the emergence of a wide range of positively evaluated social identities, and
offers more choices to both language minority and language majority students than are
traditionally available in mainstream US schools and society. The Oyster educators
argue that students’ socialization through this educational discourse is the reason that
[limited English procient], language minority, and language majority students are all
participating and achieving more or less equally. (p. 27)
The themes that Freeman emphasizes run through virtually all the programmes
for language minority students that have proven successful in elevating academic
achievement (Cummins, 1996). Respect for students’ language and culture is strongly
communicated to students, and they are encouraged to see themselves as potentially fully
bilingual and biliterate. Many two-way bilingual immersion programmes have reported
similarly positive outcomes and these programmes are even endorsed strongly by those
who claim to be opposed to bilingual education. Programmes that are less successful
Bilingual Education in Different Contexts: Principles and Practice 35
(e.g., most quick-exit transitional programmes) tend to see the students’ L1 as simply a
temporary bridge to English and do not aspire to bilingualism and biliteracy.
Although generalization from an elementary to a secondary context should be
cautious, a possible implication for Hong Kong and Southeast Asian contexts is the
desirability of communicating strong positive messages to students about the importance
of both their languages. A programme that emphasizes only the more socially prestigious
language to the virtual exclusion of students’ home language risks reinforcing the
educational legacy of colonialism. Research in a number of other contexts suggests that
this pattern results potentially in either ambivalence among students among their L1
and home culture (as a result of internalizing the negative message) or, alternatively, a
rejection of the school language as a potential threat to their identities (Cummins, 1996).
The North American and international data are abundantly clear that this “either-or” forced
choice process is unnecessary: the programmes that work best in developing English
academic skills are those that provide students with the support and encouragement to
become uently bilingual and biliterate.
From Cummins’s account, it can be seen that the Two Way/Dual Language model,
like the European Schools model, upholds strong beliefs in equally respecting and
developing both L1 and L2 prociencies as well as L1 and L2 cultural identities. The
implication for Hong Kong is that, traditionally, far more socio-economic prestige is
attached to English and English-medium education than to Chinese and Chinese-medium
education in the Hong Kong society, and this situation is counter-productive to additive
bilingual and biliterate development of Hong Kong students.
Conditions/Factors for Success in Immersion Education and
Implications for Hong Kong
From the above sections on the three successful examples of immersion education
(Canadian French Immersion, European Schools and Two Way/Dual Language), it can
be seen that immersion education (i.e., using an L2 as a medium of instruction for all or
some content subjects) depends on a number of conditions/factors for its success. These
conditions/factors and their implications for Hong Kong are delineated as follows.
Sociolinguistic Factors
It is important for the students’ rst language (and particularly rst language literacy) to
be well supported by the larger sociolinguistic environment, in both (1) prestige/status
and (2) its widespread and daily use for authentic communicative purposes.
Firstly, if the wider community views the L1 of the student as having a lower social
and/or economic status than that of the second language, then the student is likely to
develop low feelings of self-worth. These feelings will inuence the L1, L2 as well
as academic achievement. The student will have little motivation to learn the L1 (and
36 Bilingual Education: Southeast Asian Perspectives
L1 literacy) well and will not have a solid L1 literacy foundation, which research has
shown to be essential for L2 literacy development. The student will also be likely to
develop ambivalent or resistant feelings towards L2 and L2 literacy, which is likely to be
perceived as an imposed “Other” language (Cummins, 1996).
Secondly, if the L1 is not used widely both in speech and in print in the larger
society, the student has little exposure to L1 literacy. If she or he attends an L2 total
immersion programme, her or his L1 literacy will be at risk due to the further lack of
classroom exposure to L1 literacy. The student is unlikely to develop L1 literacy and L2
literacy well, resulting in subtractive bilingualism.
For the majority of Hong Kong students, there is a high risk of subtractive
bilingualism. Firstly, English still seems to enjoy a much higher socio-economic status
and prestige than Chinese, even after 1997. Parents and employers seem to place a greater
emphasis on English prociency and literacy than on Chinese literacy.
Secondly, many Hong Kong students seem to have failed to develop a strong
foundation in Modern Standard Chinese literacy (i.e., L1 literacy), not to mention English
literacy (i.e., L2 literacy). It has to be pointed out that, although the spoken language of
Cantonese is widely used in the student’s daily life, it is often not a “high variety” of
Cantonese that is used. Sociolinguists (e.g., Luke and Nancarrow, 1991) have pointed
out that “high” varieties of Cantonese (e.g., Cantonese used in news broadcasts, formal
public speeches and debates) are closer to Modern Standard Chinese in syntax and lexis.
In view of the current widespread sociolinguistic practices of the mass media (e.g., the
language of TV, newspapers and popular magazines is heavily inuenced by colloquial
Cantonese syntax and lexis, and is quite different from Modern Standard Chinese) and the
fact that the majority of Hong Kong students do not speak Putonghua (the spoken form
of Modern Standard Chinese) in their daily life (Tung, 1992), the typical Hong Kong
student’s exposure to Modern Standard Chinese and high/formal varieties of Cantonese
in the wider community is rather limited. Many students, as a result, have difculties
expressing themselves clearly in written Chinese, not to mention English. While the
standards of reading in Chinese among Hong Kong students are considered “relatively
high” internationally, “particularly in the domains of expository and documentary texts”
(Johnson, 1997, p. 182; Cummins, 1999a, p. 15), many students’ inadequate expressive
and writing skills in both Modern Standard Chinese and English are a sign of subtractive
bilingualism and subtractive biliteracy and a cause for concern.
To remedy the situation, language planners need to develop measures to raise the status
and prestige of the Chinese language in Hong Kong (e.g., raising the Chinese language
prociency requirements for civil servants and university admissions), to improve the
quality of the Chinese language used in the mass media, as well as to strengthen the
quality of both Chinese and English language teaching in schools, especially at primary
and junior secondary levels, where foundations in L1 and L2 literacy are laid. More
professional and prociency training opportunities and incentives need to be provided
for pre- and in-service teachers to upgrade their language prociency as well as teaching
methodology (e.g., to move away from a teacher-fronted, non-interactive, classroom
teaching style). In addition, having “a greater focus on language knowledge in the
Bilingual Education in Different Contexts: Principles and Practice 37
secondary school programme (e.g., contrastive study of Chinese and English, project
work focused on language and its use in society)” would also help to strengthen students’
“understanding of both Chinese and English literacy” (Cummins, 1999a, p. 15). It is,
however, recognized that language planners/the government alone cannot accomplish
all these tasks; wide societal/community/school/parental support has to be actively
solicited.
In view of the above-mentioned sociolinguistic factors, which are highly non-
conducive to the development of Chinese literacy in Hong Kong, English immersion
education should be introduced after a solid foundation in Chinese literacy skills has been
built in school, especially among student populations who have little literacy support at
home. Partial English immersion should also be introduced gradually, and a balance should
be maintained between the number of subjects taught in English and the number of subjects
taught in Chinese (realized in the classroom as a high/formal variety of spoken Cantonese
and Standard written Chinese ). This is to enable students to continue to develop both
Chinese literacy and English literacy skills in school, as the current wider society does not
strongly support the development of Modern Standard Chinese literacy in Hong Kong.
This stands in contrast with the situation of Canadian French Immersion, in which L1
English literacy development is widely supported in the larger society and the reduction of
English exposure in school due to total French immersion is compensated for by the rich
English exposure and use opportunities in the wider community and the home.
Parental/Home Support Factors
All successful immersion programmes are characterized by high levels of parental and
home support in attention and material support, as well as providing a home environment
which is rich in support for both L1 and L2 linguistic and literacy development, e.g., rich
print environment at home.
The majority of student home environments in Hong Kong, especially those of
disadvantaged students coming from low socio-economic backgrounds, are characterized
by a general lack of quality Chinese reading materials, not to mention the availability of
English print other than English textbooks (Lin, 1996b). Parents, in general, do not know
how, and are seldom involved by schools, to support students’ school language learning,
e.g., by helping to foster language learning strategies and extracurricular reading habits
in their children.
For English immersion education to succeed in Hong Kong, the current level of
parental and home support for the majority of students seems to be inadequate. To remedy
the situation, stronger links between schools and parents need to be developed and more
parental support for students’ learning needs to be solicited. It is recognized that this is
not an easy task, albeit not an impossible one. For instance, the government can capitalize
on parents’ strong desire to enrol their students in an English immersion programme, e.g.,
by making parental involvement a prerequisite for admission into English immersion.
Schools will also need to develop seminars, workshops or school-parent sharing sections
to communicate to parents the important role they play in supporting the Chinese and
38 Bilingual Education: Southeast Asian Perspectives
English development of their children and the things they can do to achieve that (e.g.,
tuning in to English TV channels some of the time, encouraging their children to do
extracurricular reading in Chinese and English, making available quality reading materials
at home, encouraging their children to do storytelling). For the importance of improving
communication and developing workshops for parents in Singapore preschool immersion
education, see Sharpe, 1991.
Educational/School Programme Factors
One major factor under this category is the professional preparation of English immersion
teachers as well as English-as-a-subject teachers. This appears to be an area which is,
relatively speaking, more amenable to government/language planning efforts than
the above-mentioned areas. All the successful examples of immersion education are
characterized by strong professionalism and high bilingual prociency levels of both
immersion and language teachers. An interactive, collaborative, project-, activity- and
meaning-based teaching methodology and curriculum provide the rich L2 input as well
as opportunities for student productive language use (output) necessary for L2 acquisition
to take place (Cummins, 1999a). Even schools situated in low socio-economic areas
(e.g., the Two Way/Dual Language school described by Cummins (1999a)) can make a
signicant difference in students’ language learning, given high levels of professional,
innovative commitment and enthusiasm on the part of teachers and school administrative
personnel. This is an area in which much can and still needs to be done in Hong Kong.
Another important programme factor concerns the way in which immersion is
introduced and at what level it is introduced. In the Hong Kong context, where neither
English nor Modern Standard Chinese literacy is widely supported by the larger
sociolinguistic environment and the home environment for the majority of students,
the following efforts seem especially important for additive bilingualism to develop for
the majority of students, many of whom come from low SES (socio-economic status),
disadvantaged families: (1) to improve the quality of the teaching of both Chinese and
English, especially during the primary and junior secondary levels, (2) to continue to
maintain Chinese as a medium of instruction in some content subjects even after English
immersion has been introduced (e.g., to allow for a partial immersion, what is termed
“mixed mode”, programme in Hong Kong). Hong Kong has traditionally forced itself
into the inexible dichotomy of either having total immersion or having no immersion
at all. Breaking away from this either-or mentality in language education planning and
exploring other partial immersion options as viable alternatives should be encouraged
(Cummins, 1999a).
Coda
The above three categories of factors appear to be the most important conditions necessary
for the success of immersion education. Other important factors include learner attitude,
Bilingual Education in Different Contexts: Principles and Practice 39
commitment, motivation and threshold prociency level. Factors regarding the linguistic
and cultural distance between the L1 and L2, however, do not seem to be essential if other
favourable conditions exist. It appears that, given the presence of the other categories of
conditions and factors for success, a linguistically and culturally distant L2 can also be
learnt well in immersion programmes.
Questions for Discussion
1. The authors classify bilingual education programmes into three broad categories:
maintenance, transitional and enrichment. Which category do you think the bilingual
programme in your area/country falls into with regards to its aims and contexts of
implementation?
2. The three immersion programmes elaborated in this chapter — Canadian French
Immersion, European Schools and Two Way/Dual Language Programmes — were
developed in different social, political and economic contexts. Write a summary to
compare these contexts. Which context is closest to your own societal context? Does
it mean that a similar programme can be developed, or does it require adaptation?
3. Three types of factors are summarized to explain the success in immersion education:
sociolinguistic, parental/home support, and educational/school programme. How do
these factors positively or negatively inuence the planning of language policy and
practice in your own contexts?
Blank page
Is Immersion Suitable for All?
The well-documented success of French immersion graduates (those who remain in the
programme tend to be successful) might have masked a problem that has characterized
early French immersion ever since its inception: the relatively high rate of student
dropout from the programme due to academic or behavioural problems. Keep (1993),
for example, reports that, in the province of Alberta between 1983 and 1984 and between
1990 and 1991, attrition rates from immersion ranged from 43% to 68% by Grade 6, 58%
3Key Issues in Immersion Education:
Implications for Hong Kong
In this chapter, key issues in immersion education are considered and their
implications for Hong Kong discussed. These key issues include the following
critical questions often asked about immersion education:
1. Is immersion suitable for all, or only for the elite? What should be the
criteria for selection of students into immersion programmes? Or, should
there be selection at all?
2. How are bridging and immersion-servicing programmes designed, and
how effective are they? Is there a bridging role for the use of some L1 in
immersion classrooms? If so, are there any clear guidelines on how and
when L1 should be used?
3. What factors need to be considered when selecting between early and late
immersion, and between total and partial immersion (mixed mode)?
4. Is immersion the only way to achieve a high level of L2 prociency? Can
L2 be effectively taught as a subject? Is teaching L2 as a separate subject
necessarily a less effective alternative to immersion?
42 Bilingual Education: Southeast Asian Perspectives
to 83% by Grade 9, and 88% to 97% by Grade 12. On this, Cummins made the following
comments:
The signicant drop-out from immersion has been debated for about 25 years in Canada
and considerable research has been conducted on this issue but without any denitive
answer to the question of whether French immersion programmes are suitable for all
students. Researchers have not been successful in specifying any set of attributes that
would identify students likely to have difculty in immersion but not in the regular
English programme. It has been suggested that the teacher-centred pedagogical
orientation in some French immersion programmes may have contributed to the
difculties experienced by some students since this form of teaching makes it difcult
for students to relate the instruction to their experience outside the school. (Cummins,
1999a, p. 5)
As Cummins (1999a, p. 5) alluded to, researchers to date have not found denite answers
to the question of what specic criteria can be used to select students into immersion
programmes to guarantee their future success, or whether the dropout rate is more due
to the didactic, non-interactive teaching methodology used in some immersion classes,
rather than merely student factors.
The problem of nding a denite set of criteria for the selection of students into
English immersion is also apparently a current dilemma of policymakers in Hong Kong.
Complaints from vocal parents and schools about the “unfairness” of the selection
procedures by which 114 out of over 450 secondary schools were selected as English-
medium instruction (EMI) schools have recurrently been heard since 1998.
Faced with this difcult dilemma and the lack of denite and precise criteria
established by the research literature for selection of students into immersion programmes,
the following might be what policymakers in Hong Kong could do at present:
1. to continue to conduct research on ongoing English-medium and Chinese-medium
programmes, to identify student attributes (e.g., levels of L1 prociency, L2
prociency and academic attainment; student attitudinal/motivational variables)
and teacher attributes (e.g., level of English prociency, classroom teaching style
and methodology, amount of English immersion teaching experience and type of
professional qualications) that are predictors or correlates of student difculties
or success in English immersion in Hong Kong, to develop programmes to help
students to adapt to English immersion, to compare student performance in English
and Chinese medium programmes; and
2. to examine the immersion theories and principles as well as prerequisite conditions
for success, to arrive at a set of principles informed by theory and research to guide
current policymaking.
In the following sections we examine some immersion theories and principles and revisit
the prerequisite conditions for success, with a view to informing policy. Specically,
we look at (1) the threshold and interdependence hypotheses, and (2) prerequisites for
success in immersion education.
Key Issues in Immersion Education: Implications for Hong Kong 43
The Threshold and Interdependence Hypotheses
Cummins (2000) claries his threshold and interdependence hypotheses as some basic
general principles regarding bilingual students’ academic literacy development and
cautions against misusing them as rationale for specic programme arrangements (e.g.,
delaying the English literacy instruction of linguistic minority students in New Zealand;
see below). To have a clearer idea of Cummins’s latest thinking on his earlier theories, let
us look in detail at his recent account (Cummins, 2000, pp. 174–7):
The threshold and interdependence hypotheses were proposed more than 20 years ago as
an attempt to account for research data showing that:
1. Many bilingual students experience academic failure and low levels of literacy in both
their languages when they are submersed in an L2-only instructional environment;
however, bilingual students who continue to develop both languages in the school
context appear to experience positive cognitive and academic outcomes.
2. Contrary to what the time-on-task notion would predict, instruction through a
minority language does not appear to exert any adverse consequences on students’
academic development in the majority language. This holds true for students from
both minority and majority language backgrounds in various kinds of bilingual
programs.
… I believe that the interdependence hypothesis is of crucial importance in
understanding the nature of bilingual students’ academic development and in planning
appropriate educational programs for students from both minority and majority language
backgrounds. The threshold hypothesis, by contrast, is less relevant to policy and
practice. I noted … that the hypothesis remains speculative and is not essential to the
policy-making process. What is relevant is the well-supported nding that the continued
development of bilingual children’s two languages during schooling is associated with
positive educational and linguistic consequences. This “additive bilingualism enrichment
principle” (Cummins, 1996) highlights the fact that bilingualism is not just a societal
resource, but it is also an individual resource that potentially can enhance aspects of
bilingual children’s academic, cognitive and linguistic functioning.
The threshold hypothesis went further than this, however. … Simply put, students
whose academic prociency in the language of instruction is relatively weak will tend to fall
further and further behind unless the instruction they receive enables them to comprehend
the input (both written and oral) and participate academically in class. A student whose
academic prociency in the language of instruction is more strongly developed is less
vulnerable to inappropriate instruction (e.g., English submersion programs). In other
words, educational treatment interacts with students’ academic language prociency to
produce positive or negative educational and cognitive outcomes.
The threshold hypothesis was necessarily vague in many respects; for example, with
respect to the lower threshold necessary to avoid adverse developmental consequences,
the extent to which students need to attain strong prociency in both L1 and L2 as opposed
to just the predominant language of their environment and school instruction was never
specied, for the simple reason that these conditions will vary so extensively. …
44 Bilingual Education: Southeast Asian Perspectives
My concern in this chapter is to clarify the empirical and theoretical status of both
the threshold and interdependence hypotheses and also their respective relevance to the
education of bilingual students. Why bother focusing on the threshold hypothesis when I
have said that it is largely irrelevant to policy and practice? Simply because both advocates
and opponents of bilingual education have frequently misinterpreted this hypothesis and
its implications. The major misinterpretation derives from conating the threshold and
interdependence hypotheses. For example, some educators in countries as diverse as the
United States and New Zealand have invoked the “threshold hypothesis” as justication
for delaying the introduction of English literacy instruction for a considerable period.
They assume that transfer of knowledge and academic skills across languages will
happen automatically. … I see these extrapolations of the threshold and interdependence
hypotheses as problematic and unsupported by research evidence.
What principles, then, can we draw from Cummins’s theories with regard to bilingual
education planning and programme design? The following seem to be the basic principles
that Cummins (2000, p. 182) afrms:
My point in reviewing these data is not to argue for the threshold hypothesis. … What
is highly relevant for policy is that the continued development of academic prociency
in bilinguals’ two languages is associated with enhanced metalinguistic, academic, and
cognitive functioning. However, the data summarized above and considerably more
research summarized elsewhere (e.g. Baker, 1996; Cummins, 1984) demonstrate that
the threshold hypothesis is not an unreasonable idea. My view has not changed from that
which I expressed in 1991:
The threshold hypothesis, on the other hand, while it has had valuable heuristic
impact, goes beyond what can be empirically validated at this point, given our limited
ability to specify and operationalize the nature of language prociency. The empirical
evidence continues to mount that bilingualism is associated with enhanced metalinguistic
functioning (and possibly with advantages in other aspects of cognitive performance),
but whether or not there are specic “thresholds” associated with these metalinguistic
and cognitive outcomes is unclear. The issue may be of only academic interest since
the practical implication of the data is the same: schools should attempt to encourage
minority students to develop their L1 abilities to as great an extent as possible both to
stimulate transfer to L2 and to reap the signicant personal and more subtle educational
benets of additive bilingualism. (1991, p. 86)
Based on Cummins’s latest account (2000) on his interdependence and threshold
theories, we can conclude that his theories can be summarized into the following basic
principles:
1. The continued development of academic prociency in bilinguals’ two languages is
associated with enhanced overall metalinguistic, academic and cognitive functioning.
2. Schools should attempt to encourage linguistic minority students to develop their L1
literacy knowledge and skills to as great an extent as possible to stimulate transfer to
L2 literacy.
Key Issues in Immersion Education: Implications for Hong Kong 45
3. Students whose academic prociency in the language of instruction is relatively weak
will tend to fall further and further behind unless the instruction they receive enables
them to comprehend the input (both written and oral) and participate academically
in class.
4. Specic bilingual education programme designs (e.g., should early or late L2
immersion be implemented; who should receive L2 immersion) must take into account
specic features of particular school populations and their situated contexts, and both
cognitive and sociocultural factors must be considered (e.g., whether the students’ L1
linguistic and cultural identities are supported in the larger society; whether students’
emotional and cultural development is adversely affected by L2 immersion).
In light of the above principles, we can say that early total L2 immersion is a feasible
bilingual education programme type for the development of additive bilingualism and
biliteracy in certain kinds of contexts (e.g., where students’ L1 literacy is well supported
in the family and the larger society). However, in view of the general lack of familial
and community support for Modern Standard Chinese literacy development in the larger
sociolinguistic context of Hong Kong for the majority of students (since the dominant
daily language of the society is Cantonese, which is linguistically quite different from
Modern Standard Chinese), helping the majority of Hong Kong students to build a solid
foundation in Modern Standard Chinese literacy in school, especially at primary and junior
secondary levels, seems to be very important for the development of additive biliteracy
instead of subtractive biliteracy. Although it takes further research to identify in more
precise terms the specic Chinese literacy level (e.g., specic language functions, targets,
goals) that constitutes a strong Chinese literacy foundation for additive Chinese-English
biliteracy to develop, some tentative principled language targets can be specied and tested
out. For instance, one can reasonably expect a student with a solid foundation in Modern
Standard Chinese literacy to be able to read formal academic texts and achieve high levels
of comprehension, as well as to write short formal essays summarizing key points in the
texts, or to write short formal answers to questions about the texts, in key content subject
areas at the age-appropriate grade level. Records of students’ Attainment Test results across
different form levels can be analyzed in the research process to ascertain the grade level at
which the majority of Hong Kong students have achieved such a Modern Standard Chinese
literacy foundation. Research studies can also be designed in which a Chinese academic
literacy task, consisting, for instance, of the specic task requirements mentioned above, is
tried out by random samples of students at different grade levels to identify the grade level
at which the majority of Hong Kong students do well on the task.1
1. It must be pointed out that the Chinese literacy task used in such a research study must have
content validity, i.e., it is a valid measure of students’ ability to function effectively when
reading and writing in Chinese for the academic subjects that they are taking at their age-
appropriate grade level. Indirect testing instruments modelled on the Chinese Aptitude Test,
for instance, are unlikely to be valid measures of students’ Chinese academic literacy skills.
46 Bilingual Education: Southeast Asian Perspectives
Cummins’s theories also speak to the importance of enriching the current English
(L2)-as-a-subject curriculum in Hong Kong schools to place some emphasis on the
development of academic English skills to help students to acquire the necessary L2
threshold academic prociency to benet from (some form of) L2 immersion at some
point in their school career. Chinese literacy alone will not guarantee transfer to English
literacy, unless the student also has attained a certain threshold level of English academic
prociency. The precise description of the threshold level has to be researched in specic,
situated contexts of particular student populations and programme types, and Cummins
cautions against the articial specication of a xed, universal threshold level, as
different contexts will require different thresholds. This brings us to a consideration in
the next section of the specic conditions that would seem to be required for successful
immersion to take place in Hong Kong.
Prerequisites for Success in Immersion
As discussed, the professional preparation of English immersion teachers is an essential
factor for success in immersion. All the successful examples of immersion education are
characterized by strong professionalism and high bilingual prociency levels of both
immersion and language teachers. An interactive, collaborative, project-, activity- and
meaning-based teaching methodology and curriculum will provide the rich L2 input
as well as opportunities for student productive language use (output) necessary for L2
acquisition to take place (Cummins, 1999a). It would appear that the majority of schools
in Hong Kong at present do not yet possess these prerequisite conditions for success in
immersion (Johnson, 1997), and they will need to be helped to improve their quality of
teaching if immersion education is not to be limited to a small number of elite schools or
gifted/elite students.
Should the Current Streaming Policy be Continued?
The question of selection and of who qualies for immersion has long been as much
a concern for North American researchers and policymakers as it is for Hong Kong
policymakers, due to both its educational and socio-political consequences. Genesee,
for instance, pointed out in as early as 1987 the social stratication/social inequality
consequences of making immersion exclusive:
The suitability of immersion for all students is a question of more than academic interest.
First of all, there is the danger that immersion programmes could become elitist. That
is to say, without substantive evidence to the contrary, it might be thought that such an
extraordinary form of education is suitable for only the most capable academic students,
thereby excluding less capable students. Indeed this charge has been levelled against
immersion. Such charges do not serve the educational community well because they
threaten to sour the relationship between different sectors of the community — those
involved in immersion and those not involved. Such charges may threaten the very
Key Issues in Immersion Education: Implications for Hong Kong 47
existence of immersion because they call its continuation into question — if immersion
is not good for all, it should not be offered to only a few; therefore, it becomes available
to none. Second, in many parts of Canada bilingualism is not a privilege but a necessity
for economic and social survival. To deny any or all children in these communities
effective means for achieving bilingual competence through public education is
ethically questionable. It is imperative that educational decisions concerning exclusion
of subgroups of students from immersion be founded on systematic and objective
investigation, and not on speculation or “common sense.” (Genesee, 1987, p. 78)
The Hong Kong educational dilemma is, however, further complicated by the following
historical, institutional and socio-economic factors:
1. English medium education has long been associated with a high SES (i.e., wealth and
power) in pre-1997 colonial Hong Kong, and the public hold persistent beliefs about
the strong association of English-medium education and future socio-economic
success.
2. The public in general seem to consider that Chinese literacy and Putonghua, albeit
important (or becoming more important after 1997), do not yet have as important an
SES and professional education/job market value as English.
3. The public’s beliefs are reinforced by the institutional fact that most current university
and higher professional institutes require English as a medium of instruction. They
also fear, quite understandably, that their children will not be able to cope with the
English medium in university/professional education if they have pursued all of their
secondary school studies in the Chinese medium. The lack of articulation between a
Chinese-medium secondary education and a mostly English-medium university and
higher professional education has been a great concern for the general public under
the current streaming policy.
4. The public in general have only partial understanding of the principles of immersion
education. Many of them are not aware of the principle of interdependence of L1
and L2 and the importance of having a solid foundation in Chinese literacy for the
successful development of Chinese-English bilingualism and biliteracy. They seem
almost willing to sacrice Chinese literacy development for the acquisition of an
English-medium education, if only in name, as English-medium education is still
deeply perceived by many parents as a ticket to future socio-economic success of
their children in the Hong Kong society.
Public beliefs are hard to change over a short time. Government, nevertheless, has the ethical
responsibility to inform the public of a scientic and research-informed basis for selecting
the best educational programmes for their children. It can be seen from the sections on the
threshold and interdependence hypotheses, on conditions for success in immersion, and in
this section that the factors we can be certain about at present are as follows:
1. Given the general lack of support for both Modern Standard Chinese development
and English development in the larger sociolinguistic context of Hong Kong, the
majority of students will not easily develop additive Chinese and English bilingualism
and biliteracy.
48 Bilingual Education: Southeast Asian Perspectives
2. The prerequisite conditions (e.g., bilingual prociency of content teachers, quality
of teaching) are not available in many schools in Hong Kong for successful English
immersion to happen.
Parents need to be informed of the negative educational consequences of enrolling their
children in English immersion schools where both teacher prociency and teaching
quality are not up to the standard for additive bilingualism to take place (see recent
ndings in Marsh, Hau & Kong, 2000; Lee, 2002). And, policymakers need to move
away from merely focusing on screening students to see who is t for immersion and
who is not. In the long run, policymakers need to focus on helping more schools and
teachers to improve their quality of teaching as well as helping more students to acquire
the solid Chinese literacy foundation and the necessary academic English skills to be able
to benet from English immersion, as it has been established in the research literature
that some form of L2 immersion is the most effective way of acquiring a high level of
prociency in an L2 (for details, see the section on immersion and achieving additive
bilingualism below). As Cummins pointed out:
...the question should not be: “Is French immersion suitable for the language impaired,
learning disabled, or low-IQ child?”, but should be changed to “How can French
immersion programmes be made more suitable for these children?” (Cummins, 1986,
quoted in Keep, 1993).
To help more students to benet from immersion education, policymakers can consider
the experience of different bridging and “immersion-servicing” programmes (Nuttall &
Langhan, 1997). It is to this topic that we turn in the next section.
Bridging and Immersion-Servicing Programmes
In this section, we review two successful examples of such programmes, one at junior
primary level in South Africa for students from low socio-economic backgrounds, and
one at university level in Canada. The review of the Molteno Project is mainly based
on and excerpted from Nuttall and Langhan (1997). The description of the University
of Ottawa’s late, late partial immersion programme is based on Burger, Wesche and
Migneron (1997).
The Molteno Project in South Africa: A Bridge Series to Help Limited-
English-Procient Children to Benet from English Immersion
In the following sections, the background, origins, goals and principles of the Molteno
Project are discussed with reference to implications for Hong Kong. An illustrative
example of the bridge programme is also described.
Key Issues in Immersion Education: Implications for Hong Kong 49
n Background
The general “decolonization ideologies” and antagonism towards English among many
postcolonial African governments (Blommaert, 1999) were, ironically, much less evident
in South Africa. Furthermore, native-language instruction was not popular in South Africa,
as it was closely related to the former political oppression of apartheid. Instead, a policy
of sudden English immersion was practised in most schools: according to government
gures in 1990, 98% of all higher primary schools opted for English-medium instruction
from Primary 5 onwards after four years of native-language instruction. Somewhat
similar to the case of Brunei (Jones, 1999), the sudden transition to English immersion
has created educational problems for the disadvantaged, e.g., Black children from low
socio-economic backgrounds. As one researcher put it:
The pronounced weakness that we discovered with the children’s English leads us to
believe that the current generation of junior primary children are not competent in terms
of the demands of the medium... (Macdonald, 1990, p. 41, quoted in Nuttall & Langhan,
1997).
This sounds all too familiar to the Hong Kong reader, though in Hong Kong, the sudden
transition from L1 to L2 medium took place at Secondary 1 for the majority of students
before the 1998 streaming policy was implemented. Some of the unfavourable institutional
factors/conditions leading to the educational difculties witnessed in the South African
classroom also seem to be similar to those found in Hong Kong. Those unfavourable
conditions under which English immersion takes places in South African primary schools
can be described as follows (summarized from Nuttall & Langhan, 1997, pp. 216–7;
emphasis added):
1. No preschool programmes
There is no formal preschool programme, meaning that most African children
entering school do not have the early learning experiences that prepare middle-class
children for school.
2. No sustaining environment for English outside school
(a) English is not related to any of the African languages (the languages are not
cognate).
(b) The majority of children usually have no contact with native speakers of
English, who form a small minority of the population.
(c) The children’s use of English stops at the classroom door. There is virtually no
English in their social environment, and the rest of their communication is in
one or more African languages.
3. Poor African language programmes
(a) There are very few books in African languages that can ease children into a
literate culture.
(b) African-language teaching programmes are very poor, focusing almost
exclusively on traditional formal grammar, and do not develop language skills
that can be transferred to the learning of English.
50 Bilingual Education: Southeast Asian Perspectives
(c) Pupils are usually not functionally literate in their native language by the time
they are required to begin English immersion.
4. Poor teacher training
(a) Teachers who teach English in African schools are ill-prepared by their training
for teaching learners English for immersion education; they have at best
been exposed to a very structural view of language and language use and to a
mechanistic view of the “four skills” as separate and discrete.
(b) Teachers are expected to conduct classes in English, but pupils are unable to
comprehend explanations in English.
(c) Teaching methods emphasize rote learning and, often, meaningless
memorization.
5. Inadequate preparation for the sudden transition to EMI
(a) There is a serious dislocation between the English taught as a subject in the
junior primary school and the demands of EMI in the higher primary school. As
a result, the quantity and quality of English teaching are not adequate to prepare
learners to cope with the sudden transition.
(b) African languages may not use concepts equivalent to those being taught in
English, and children often do not have appropriate background knowledge
that the teacher can activate through the use of quick digressions in the L1.
In addition, formal teaching/learning materials developed in isolation from
classroom realities generally fail to provide critical background information.
6. Serious textbook problems
(a) Pupils are unfamiliar with and unprepared for the expository texts that are used
in the content subjects.
(b) The majority of the prescribed content subject textbooks, written in English,
are wholly inappropriate in that they often contain badly constructed discourse,
language that is beyond the pupils’ competence and unfamiliar concepts remote
from their life experience.
It might be concluded that Hong Kong is much more fortunate than South Africa in
the availability of resources. However, a critical reection on our current educational
provision and school contexts might force us to admit that we could have done much
better, given our greater educational resources. Conditions (2), (3), (4) and (5), especially
those regarding the lack of exposure to English outside the classroom, the inadequate
quality of teacher-training and classroom teaching methodology as well as the lack of an
appropriate English-as-a-subject curriculum to prepare students for English immersion,
might not really be very remote from the classroom and school realities witnessed in
Hong Kong.
In view of the many problems of the sudden transition policy, non-government
organizations, educationists and groupings such as the African National Congress
called for a more exible language policy. They campaigned for the right of the local
communities to select their own language-medium policies and they proposed the
following six options:
Key Issues in Immersion Education: Implications for Hong Kong 51
1. L2 immersion from Primary 1,
2. a sudden transition from L1 medium to L2 medium at Primary 5 (i.e., the previous
policy),
3. a gradual change from L1 medium to L2 medium (gradual transition),
4. a modied form of gradual transition model,
5. alternative bilingual models (e.g., parallel-medium, dual-medium, multilingual), or
6. L1 medium throughout the schooling process
In late 1992, the South African government offered school communities a choice from the
rst three options listed above. Parents were invited to decide on the policy for individual
schools through a system of voting, and the results showed a distinct parental preference
for gradual transition from L1-medium to English-medium education (54%), while only
22% voted for early total English immersion. Interestingly, similar student preferences
have been found in Hong Kong: students support a gradual transition from Chinese to
English as the medium of instruction (Tung, Tsang, & Lam, 1997).
n Origins, Goals and Guiding Principles
The Molteno Project is an independent, non-prot-making, non-government organization,
funded primarily by donations and grants from the private sector and trusts in South
Africa. The project has been successful in helping poor African Black children to both
build a foundation in L1 literacy and acquire the initial English literacy skills for English
immersion. To date, some 26,000 teachers, teacher trainers, education advisers, school
inspectors and other educators have been trained in the use of the project’s methods and
materials. On any normal school day, more than half a million children in South Africa
are being taught through the project’s materials.
The Molteno Project’s initial research (Rodseth, 1978) focused on an analysis of
problems associated with the teaching and use of English as medium in African schools
and the recommendations of methods for preparing lower primary children for its use.
This early research has led to the development of both language-teaching and language-
learning materials and in-service teacher-training programmes.
Based on identied problems and associated needs in Black primary schools,
the Molteno Project adopted as its primary objective the preparation and support of
pupils and teachers for English-medium instruction through the development and
implementation of methods and materials that would, in the longer term, also result
in curriculum transformation. A primary focus was the development of linguistic and
cognitive competencies that equip disadvantaged learners to cope with the demands of
EMI.
To this end, the project’s research has resulted in the following guiding principles for
the development of its language teaching and language learning programmes:
1. The importance of L1 literacy as a foundation for the effective learning of an L2:
The project has always favoured a gradual transition from L1- to EMI when English
is not the learners’ home language.
52 Bilingual Education: Southeast Asian Perspectives
2. Bilingual instruction as a means of facilitating learning in a second language: The
project incorporates a two-pronged approach. The rst refers to the development
of the learner’s L1 and L2 competencies through separate but parallel African-
and English-language programmes that have been developed to t into the formal
curricular for each language. The second refers to the integration of the use of the
learner’s L1 into the English programme as a means of introducing and/or explaining
difcult or unfamiliar vocabulary and concepts (see also discussion in the section on
the role of L1 use in immersion classrooms, for the potential, bridging, role of L1 in
immersion classrooms).
3. English-across-the-curriculum: Integrating content-subject matter in the English
lesson as a means of introducing cognitive academic language learning. The project’s
language-across-the-curriculum (LAC) principle aims at providing opportunities
to relate English language learning materials to learners’ immediate and near-
future content-subject learning needs. This principle is based on both a researched
understanding of learners’ needs and a synthesis of relevant ESL theory.
4. An eclectic approach to language teaching and learning: This emphasizes the
functional-communicative approach in the earlier years, incorporating appropriate
aspects of a structural approach, and gradually progressing to a more cognitive
academic language-learning approach.
5. An eclectic approach to methodology: This is based on a learner-centred, small-
group approach, which also incorporates structured teacher-input and whole-class
activities. To ensure that the particular needs of teachers and learners in the South
African context are met, basic interactive communication skills and cognitive
academic language prociency are emphasized and achieved by means of both co-
operative and individualized learning through a variety of activities (see illustrative
example in the Bridge Series).
6. Meeting teachers’ linguistic, methodological and classroom management needs.
This is achieved through supportive materials and extensive in-service teacher
education and training. A distinctive feature of the project’s teachers’ manuals is
the incorporation of extensive teacher support, which is intended to assist in the
teacher’s own development through “guided effective teaching experiences”. This
support includes the language needed for conducting lessons in English, effective
methods of facilitating meaningful learning, suggestions and guidelines for effective
group-work classroom management, and suggested tasks and activities for learners.
All of the project’s courses are accompanied by intensive, ongoing in-service teacher
training in the use of its materials and the implementation of a learner-centred
methodology.
It is important to bear in mind that the guiding principles outlined above underlie
the project’s critique of those aspects of the existing formal curriculum that require
fundamental change. Essentially, it has been the project’s aim to establish an alternative
curriculum and teaching approach that enables learners to become, ultimately, able to
compete on an equal footing with L1 speakers of English in an English medium context.
Key Issues in Immersion Education: Implications for Hong Kong 53
The implication for Hong Kong seems to be that developing a programme on similar lines
can help to reform Hong Kong’s existing English language teaching practices, which
often tend to be didactic, structural-drill based and teacher-centred. More importantly, it
will help to infuse English-as-a-subject in Hong Kong schools with an important purpose:
that of helping students to learn to use English for authentic (content learning) purposes.
It will also help to develop in students the necessary English literacy skills to benet from
English immersion.
n Comprehensive Primary School Language Programme
The project’s ongoing research and curricular development have resulted in two major
types of programme:
1. An African-language literacy programme, to be followed by an African-language
development programme for the primary school. One of the aims of this programme
is to lay L1 cognitive and conceptual foundations that are prerequisites for subsequent
learning through the medium of English.
2. An English language programme for the primary phase. A key objective of this
programme is to anticipate and support the linguistic, cognitive and conceptual
needs created by English-medium instruction.
In these programmes, the project has developed the following course series:
1. Breakthrough to Literacy: an L1 literacy course for the rst year of primary school,
in eleven southern African languages.
2. The Bridge to English Series: a primary English language programme for the rst
ve years of school.
In addition to these series, the project is developing a series of parallel African- and
English-language programmes that complement each other in the preparation and support
of ESL learners for English immersion.
n An Illustrative Example from the Bridge Series
In this section, a progression of teaching and learning activities in the Bridge Series that
exemplies the guiding principles of the Molteno Project is described. It is hoped that, by
examining the curriculum design and implementation of the Bridge Series, curriculum
planners in Hong Kong can see if we can learn from the experience of designing bridging
programmes to prepare students for English immersion.
In the Bridge Series, simple, context-embedding communicative activities in earlier
courses lead towards more cognitively demanding, context-reduced communicative
activities in the later courses, in preparation for the meaningful learning of geography,
mathematics, science and history. In the following paragraphs, an example is taken from
the geography topic area to illustrate how content subject-specic language and tasks are
introduced, and how attention is given to the equally important development of “English
for everyday life”.
54 Bilingual Education: Southeast Asian Perspectives
1. Bridge Plus One: the third year of the programme
(a) Preparing to understand maps
Building on a foundation of perceptual skills with regard to size, shape and
spatial relations developed in the LAC components of the Project’s Bridge to
English course for Primary 2, a thirteen-lesson unit in Bridge Plus One lays the
foundations for understanding maps.
The rst two lessons in this unit introduce pupils to the notion of perspective, by using
concrete objects and simple pictures to sensitize them to “what things look like from
different sides”. The point of these lessons is to lead learners to an awareness of the
perspective from above, or “from the top”, which is critical for understanding and
interpreting maps. Support to the teacher is provided in the lesson plan, which includes
steps to follow, suggestions for classroom organization and examples of ways of saying
the kinds of things required to conduct the lesson in English.
In Lessons 2 and 3, the view “from the top” is reinforced through reading a story
about four friends who climb a mountain and look down on the village where they live.
The story introduces the expression “bird’s-eye view” and engages the reader in the
characters’ observations of how different their village, their school, the railway station
and so on look from high up on a mountain. The illustrations in the story are designed to
enable the learner to identify the buildings and places the characters point out, and in so
doing, begin to acquire a new understanding of perspective (see Figure 1).
Figure 1. Story Context for Introduction to Map Reading in the Molteno Project (From the
Molteno Project, 1986, p. 4)
Key Issues in Immersion Education: Implications for Hong Kong 55
The third lesson provides teacher support in the form of steps for the teacher to
follow in teaching reading. These are based on the schema-theoretic/interactive approach
to reading. The steps include suggestions on when to ask prediction questions and ways
to encourage learners to work out the meanings of unknown words from the context of
the story.
In Lessons 4 to 7, the learners read and discuss two more stories. The rst, The Four
Friends Hunt for Treasure, reinforces the view from above and introduces learners to the
way in which the four friends draw a map of a part of their village based on their view of
it from the top of the mountain in the previous story. The story then focuses on how one
of the friends hides a treasure, which the other friends must nd by using the map.
The second story, The Four Friends Play a New Game, introduces learners,
indirectly and very simply, to the use of symbols through a game of hide-and-seek that
the characters play. Clues, in the form of arrows drawn in the sand or made with sticks,
are the signs (symbols) that help them to nd each other. The aim of this story is to
anticipate the introduction of map symbols at a later stage in the programme.
Lessons 8 and 9 exploit the learners’ exposure to the four friends’ map of their
village, and engages them in a collaborative exercise of “reading a map” of their school
that is drawn by the teacher. This is followed by a “discovery exercise” in which pupils,
in groups, help each other to draw a map of their school from above.
The remaining lessons in this unit reinforce the view from above by means of
another story, in which the four friends win a competition for which the prize is a ride
in a helicopter. The readers are led, through the story, to seeing and understanding the
effect of height above the town on the size of the buildings. The illustrations in the story
reinforce the view from above that was introduced earlier.
(b) Integrating the development of writing skills
In addition to the wide range of language, cognitive and skill-developing
activities provided in the unit, attention is given to the integration of other
skills, such as writing, so as to reinforce what has been learned in the lessons.
For example, in a writing exercise based on a story they have read in a previous
lesson, learners are required to reorganize jumbled sentences into a logical,
coherent paragraph. The use of the familiar content of a story for such a task
ensures that the focus on developing the learners’ writing skills is not lost in the
effort to decode new information in an unfamiliar story. Such tasks also provide
the teacher with an indication of how well learners have actually understood the
meaning of the story.
n Implications of the Molteno Project for Hong Kong
The possible implications of the Molteno Project are outlined as follows:
1. Equal attention to L1 and L2 literacy development at primary level.
2. Integrating content learning with L2 learning: L2 learning is given purpose and
substance.
56 Bilingual Education: Southeast Asian Perspectives
3. A thematic, story-based, inquiry approach that provides engaging context for both
content and L2 learning.
4. A culturally sensitive and locally contextualized curriculum to arouse interest in
students (e.g., illustrations and story backgrounds are both attractive and culturally
indigenized to ensure that they are well suited to the imagination and interest of
young students in the local culture).
5. Curriculum design and materials development going hand in hand with intensive
in-service teacher training; detailed teachers’ manuals providing extensive support
to teachers, including suggestions for both practical classroom activities and useful
classroom language for the activities.
In summary, the Molteno Project is more than just a bridging course. It aims at reforming
the traditional teacher-fronted, drill-based, language-isolated-from-content approach to
L2 teaching. It seems that, when Hong Kong curriculum planners design the new English
syllabus for Hong Kong secondary school students, the experience of the Molteno Project
is instructive. If the majority of our English teachers follow the traditional, teacher-
fronted, textbook drill-based, language-isolated-from-content, language-teaching
approach, even the best-designed thematic, project- and task-based syllabus will stand
a small chance of motivating students. Materials development, therefore, needs to go
hand in hand with intensive in-service teacher training. As well, the topics and contexts
chosen for the English projects should draw on the culture and experience of young
Hong Kong students, in order to appeal to them. Too often in Hong Kong textbooks, we
see topics more appealing to middle-class adults than to young learners, many of whom
come from low socio-economic backgrounds. A culturally sensitive English language
course, for instance, might mean that curriculum planners and teachers will think twice
before asking these students to do an English project on travelling to Europe or nding
out about the development and preparation of different kinds of gourmet wine in France.
The Molteno Project’s sensitive care to develop culturally compatible, scientic, topics
that would appeal to imaginative, inquiring, adventurous young minds is instructive to
us. The Molteno Project’s work among poor Black children in South Africa demonstrates
to us that, given the right kind of language support and high-quality language teaching
in the school (i.e., favourable school/programme factors), even disadvantaged children
with little home and community support for their L2 skills can gain adequate L2 skills
for L2 immersion. However, that comes only with a well-planned English course with
well-trained and committed teachers who do not see language learning isolated from
content learning and who see the value of an interactive, thematic, contextualized inquiry
approach to L2 teaching.
In the next section, we look at a successful “immersion-servicing” or bridging
programme at the other end of the educational ladder — L2 Sheltered and Adjunct L2
Courses for University partial immersion students.
Key Issues in Immersion Education: Implications for Hong Kong 57
Late, Late Partial Immersion and Bridging Programmes at the University
of Ottawa, Canada
At the University of Ottawa, a few subjects (e.g., Introduction to Psychology, The History
of Canada since its Discovery) are offered in parallel mediums, French and English. To
help university students with high intermediate levels of L2 prociency to benet from L2
immersion in those subjects, bridging L2 courses have been developed. The experience of
the University of Ottawa has special relevance for universities in Hong Kong, which are
likely to face the challenge of helping an increasing number of Chinese-medium secondary
school graduates to cope with English instruction in the university if the current Mother-
Tongue-Education policy is carried through to Secondary 7. The following account of the
Ottawa bridging programmes is excerpted from Burger, Wesche & Migneron, 1997.
n Evolution of the Bridging Programme: Sheltered and Adjunct Courses
Several formats for discipline-based language instruction have been tried out over the past
decade, with the aim of providing university students with a transition between second
language courses and academic course work for native speakers. Sheltered courses, sometimes
called “late, late immersion”, were introduced at the University of Ottawa in 1982. Classes in
the original sheltered format were composed only of second language speakers.
The rst sheltered courses were set up by the Second Language Institute and the
School of Psychology in the winter of 1982. In a special second semester of the course,
Introduction to Psychology/Introduction à la psychologie, was offered in French for
Anglophones and in English for Francophones. Students who had successfully completed
the rst semester of the course in their rst language and who had high intermediate
levels of L2 prociency were recruited for the programme. They received their second
semester psychology instruction entirely through the medium of their second language in
special FSL (French as a second language) or ESL sections taught by a regular psychology
professor. These classes followed the prescribed curriculum, used the recommended
textbook (available in French translation) and wrote the same nal (bilingual) multiple-
choice examination as students in rst language sections. During each class period, a
language teacher spent about twenty minutes reviewing and previewing current material
with an emphasis on language use, and was available to help students with their language
problems outside of class.
Introduction to Psychology/Introduction à la psychologie was selected for the pilot
project for several reasons. Members of the School of Psychology were interested in the
project, including several professors who regularly taught these courses. The professors
were willing to take students’ limited prociency into consideration in their lecture
delivery, explanations and evaluation procedures. The existence of parallel sections in
English and in French allowed students the possibility of transferring back into a rst
language section during the term, without worrying about having missed something. The
format was lecture style and therefore did not require oral production on the part of the
students (although participation was encouraged). The course was highly structured with
58 Bilingual Education: Southeast Asian Perspectives
a well-organized, uniform outline and audio-visual study materials, regular quizzes and a
common multiple-choice nal examination.
This rst experiment had many positive results, and research revealed that at the end
of the semester students had learnt the subject matter well, made progress in their second
language equal to that made by similar students in regular second language classes and
expressed greater condence in their ability to use the target language as well as the
intention to use it in the future.
The experiment was repeated the next year and in 1984 was expanded to two
semesters. The history course, L’histoire du Canada depuis les découvertes, was added to
the sheltered course offerings for FSL students in 1983–84, and its English counterpart,
The History of Canada since Its Discovery, was added in 1984–85. A subsequent three-
year evaluation found consistent positive results in subject-matter mastery, language
gains and changes to language use patterns and related attitudes.
In 1985, budgetary restrictions meant that it was no longer nancially possible to
sustain the sheltered lecture sections, given their lower enrolments than in sections for
native speakers. An adjunct format was adopted in which L2 students join an existing
class for native speakers and take a linked ninety-minute-per-week second language class
for separate academic credit. The adjunct language class is nanced by offering it as a
separate credit course corresponding in level to advanced second language reading and
writing courses.
The adjunct course not only provides students with the language support they need
to succeed in the discipline class but offers opportunities to practise the language in a
“safe” environment. Exposure to challenging tasks in the L2 in both courses leads to
development of high-level language abilities for the academic context, and introduces
students to the discourse of the chosen discipline.
n Aspects of the Programme
In the following sub-sections, different aspects of the programme such as teachers,
students, instructional principles and programme outcomes are described.
Teachers
Most professors are native or native-like speakers. None receive special training, but the
language instructors confer with them during the initial arrangements for adjunct courses,
to make sure that they know that having second language students in their class may
require adjustments. The use of visual support (the chalkboard or transparencies) helps
second language learners understand lectures and provides them with key concepts for
note-taking. The professor should provide a good language model for students. Although
humorous anecdotes and a amboyant style may be motivating for native speakers,
second language learners are better able to cope with a straightforward explanation of
the content at hand.
Since the language instructors attend all lectures and work closely with the second
language students, they may suggest to the professors that they proceed more slowly,
Key Issues in Immersion Education: Implications for Hong Kong 59
offer more explanations, use the board or overheads frequently, follow the course outline
more closely, make some allowances for inadequate language skills when setting tests
and marking student essays, and in general do what is necessary to ensure that second
language speakers can follow the course and receive fair treatment.
The language instructors who team-taught the sheltered courses or who currently
teach the adjunct second language courses volunteered for the assignment. Without
exception, they have had broad experience in second language teaching at the university
level; hold a master’s degree in applied linguistics, second language methodology or
education; are native speakers of the target language and know the students’ rst
language; and have either background knowledge or a strong interest in the subject matter
of the discipline course. Since they attend the lectures with the students, they are in an
excellent position to understand the difculties posed by the discipline content as well as
the language-learning challenges facing the students in the given context. In most cases,
language teachers have made a long-term commitment to their particular sheltered or
adjunct course. This ensures continuity and the preparation of extensive materials for
their courses for their own use or later to be passed on to successors.
The language instructor’s role has evolved since the beginning of the programme. In
the original sheltered course it was largely determined by the primary goals of ensuring
that students develop adequate language skills for success in the discipline courses. The
language instructors regularly veried student understanding of the subject matter, gave
supplementary explanations when students had difculties, and taught strategies for
carrying out successful academic work through the L2. Although the language instructors
continue to full the same functions in the language adjunct format, there is also time
to devote to oral activities such as debates, discussions and presentations based on the
subject matter, and to various writing activities.
Students
In most cases, students voluntarily enrol in these courses and are motivated to achieve
high levels of academic ability in their L2. In the English adjunct courses, most students
are French-speaking Canadians, but some are immigrants. In the French-language
courses, the majority of students are English speakers born in Canada. Students from
bilingual high school and immersion programmes often take one of these courses before
making the leap to courses for Francophones, without language support.
An important issue is the L2 prociency level required if students are to benet
from sheltered and adjunct courses. In the sheltered courses, a minimum intermediate
L2 prociency level was set according to scores on the university’s English and French
prociency tests, corresponding to a level attainable by a beginner in four to ve semesters
of language courses at the university level. This level was found to be sufcient, at least
partly because discipline professors spontaneously made linguistic adjustments for the
sheltered students to ensure their comprehension of the subject matter.
In the adjunct format, because the L2 students attend large lecture classes with
native speakers, discourse adjustments by the professor cannot be expected. It has thus
60 Bilingual Education: Southeast Asian Perspectives
been necessary to raise the prociency requirement for this format. With experience it has
also become clear that different minimum prociency levels are desirable for different
courses, reecting the language skills that the respective subjects require of students.
Thus, while motivated students with intermediate-level second language prociency can
succeed in the textbook-based, highly structured introductory courses in linguistics and
psychology, which emphasize receptive L2 skills, a higher level of L2 prociency is
needed for courses such as Canadian History, in which students write summary essays
on varied readings throughout the course, or Sociology of the Family with its varied
readings and emphasis on oral presentations and discussions.
A ceiling level on the university’s English and French prociency tests has also been
in effect since the beginning of the programme, in recognition that students with very
advanced skills in the second language do not need the adjunct language instruction when
they take disciplinary courses offered through their L2.
Instructional Principles
1. Simplied input
In sheltered courses, discipline professors modify their lecture presentations,
assignments and evaluation practices to lighten the language demands, adapting their
language to ensure comprehension by second language students, and emphasizing
assignments that require receptive skills. However, the basic course content is
maintained, and L2 students take the same nal examinations as those taken by L1
students.
In the adjunct courses, L2 students take the course together with L1 students. A
higher L2 prociency is required of the immersion students. However, the adjunct
language class gives them the language support, teaches them the strategies and
develops the condence they need to succeed in the content course.
2. Well-organized course structure
The features of the discipline course selected for L2 immersion are as follows. The
professor is a good language model and is sympathetic to the learning difculties of
L2 students; the lectures are well structured and clear, with multimodal presentation
of material; the emphasis is on receptive language skills at the initial stages; there
is coherent organization and content redundancy between lectures, readings and
assignments. The following features are avoided: cognitively demanding subject matter,
a difcult lecturing style, the absence of a textbook, complex or inappropriate assigned
readings or study materials, evaluation procedures requiring production skills.
3. Building prociency step by step
(a) The rst few weeks of the language class are devoted to building condence
in students and giving them whatever tools they need to be successful in the
course.
(b) Once students feel fairly comfortable with their performance in the discipline
course, work on L2 development begins, initially with attention to receptive
skills in which students increasingly focus on formal aspects of the language.
Key Issues in Immersion Education: Implications for Hong Kong 61
(c) Gradually, students are given more opportunities to practise their L2 in
production activities, with discussion and short oral presentations at rst, then
longer presentations requiring supplementary reading. In this way, they extend
their active vocabulary and gain exposure to written material other than their
textbook. They will then be able to take part in spontaneous discussions on a
topic they have prepared. Writing assignments are usually controlled at rst,
providing practice of a particular discourse type, such as the research summary.
Exercises are then expanded to require synthesis of material from the discipline
course, and sometimes a formal term paper for the discipline course.
Programme Outcomes
The overall success rate for students is very high. Very few participants have dropped out
of the programme over the years. Over the rst ve years of the programme, of the 175
students enrolled, only three withdrew and two failed to pass courses in psychology and
other disciplines.
Evaluations of learning outcomes found consistent gains in L2 prociency in both
the sheltered and adjunct formats that were similar to or greater than gains by students
at the respective prociency levels in well-taught ESL and FSL classes. In both formats,
students were almost without exception successful in subject-matter learning and compared
favourably with L1 students. Students generally also reported greater self-condence and
lower anxiety in using the L2 and indicated greater readiness and determination to use
the L2 out of class for a variety of purposes. Substantial gains are, however, found more
in L2 listening and reading skills than in speaking and writing. One must also note that
language class activities in which course content was reviewed were more valuable for
disciplinary learning (e.g., discussion of lectures, quizzes on course content and the use
of review charts) than for language learning. In the latter case, informal grammatical
correction, oral activities (discussions, presentations) and writing exercises were more
valuable; that is, those activities that involved the reworking of information in alternative
formats and channels, emphasizing production skills and accuracy.
Implications for Hong Kong
EAP (English for academic purposes) and English enhancement courses (providing
remedial English instruction in basic English grammar and the four skills) have been
growing rapidly in universities in Hong Kong to help university students to cope with
English-medium studies especially since the expansion of the university education sector
in the early 1990s when more and more university entrants were lacking in English skills to
successfully pursue English-medium studies. A more serious scenario has been witnessed
in countries such as Malaysia, where there has been a serious lack of articulation between
a predominantly L1-medium (Malay) secondary school sector and a pre-dominantly
L2-medium (English) university education since the Malaysian government abolished
traditional English-medium secondary schools in the postcolonial era. This motivated
62 Bilingual Education: Southeast Asian Perspectives
the recent Malaysian ofcial policy to readopt English as the MOI for science and math
subjects in secondary schools (more discussion on this in Part III). The same problem can
be expected to happen in Hong Kong if the current Mother-Tongue-Education policy is
carried through to Secondary 7 for the majority of schools. In this regard, it is important to
improve the quality of the EAP and English enhancement courses offered at universities
to ensure that they serve the bridging purpose.
The sheltered and adjunct courses implemented at the University of Ottawa can
offer us some important insights. One such insight is that successful L2 immersion at the
university level is possible, but certain conditions have to be obtained:
1. The English language courses offered at university should be closely related to the
discipline courses that the students are taking. There needs to be close collaboration
between EAP language instructors and discipline course professors (e.g., EAP
language instructors actually attend some of the discipline course lectures). EAP
courses developed in isolation or not closely supporting the immediate language
needs of students taking discipline courses stand a small chance of success in helping
students to benet from L2 immersion.
2. Language instructors should aim at helping students to gradually develop the
receptive skills and then the productive skills required to participate actively in
content classes. A variety of activities that involve the reworking of information
in alternative formats and channels, emphasizing production skills and accuracy
should be developed. Also, both discipline course lecturers and language instructors
should provide good language models for students.
3.