English in Singapore: Modernity and Management (Asian Englishes Today) [Illustrated] 9888028421, 9789888028429

Citation preview

Asian Englishes Today Series Editor: Kingsley Bolton Chair Professor of English, City University of Hong Kong

The volumes in this series set out to provide a contemporary record of the spread and development of the English language in South, Southeast, and East Asia from both linguistic and literary perspectives. They reflect themes related to English that cut across national and linguistic boundaries, including the study of language policies; globalization and linguistic imperialism; English in the media; English in law, government and education; ‘hybrid’ Englishes; and the bilingual creativity manifested by the vibrant creative writing found in many societies in Asia. The editorial advisory board comprises a number of leading scholars in the field of world Englishes, including Maria Lourdes S. Bautista (De La Salle University-Manila, the Philippines), Susan Butler (Macquarie Dictionary, Australia), Braj B. Kachru (University of Illinois, Urbana-Champaign), Yamuna Kachru (University of Illinois, Urbana-Champaign), Shirley Geoklin Lim (University of California, Santa Barbara), Tom McArthur (founding editor of English Today), Anne Pakir (National University of Singapore), Larry E. Smith (co-founding editor of World Englishes), and Yasukata Yano (Waseda University, Japan).

Also in the series: Hong Kong English: Autonomy and Creativity Edited by Kingsley Bolton Japanese English: Language and Culture Contact James Stanlaw China’s English: A History of English in Chinese Education Bob Adamson Asian Englishes: Beyond the Canon Braj B. Kachru World Englishes in Asian Contexts Yamuna Kachru and Cecil L. Nelson Philippine English: Linguistic and Literary Perspectives Edited by Ma. Lourdes S. Bautista and Kingsley Bolton

Hong Kong University Press 14/F Hing Wai Centre 7 Tin Wan Praya Road Aberdeen Hong Kong © Hong Kong University Press 2010 Hardback Paperback

978-988-8028-42-9 978-988-8028-43-6

All rights reserved. No portion 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 permission in writing from the publisher. Secure On-line Ordering http://www.hkupress.org British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. This digitally printed version 2010.

Contents

Series editor’s preface Acknowledgements List of contributors Map of Singapore and the region

vii ix xi xiv

Part I: The Ecology of English in Singapore

1

1

English in Singapore: Policies and prospects Lisa Lim, Anne Pakir and Lionel Wee

3

2

Migrants and ‘mother tongues’: Extralinguistic forces in the ecology of English in Singapore Lisa Lim

19

Part II: Reconceptualizing ‘English’

55

3

Singapore Standard English revisited Anthea Fraser Gupta

57

4

The Speak Good English Movement: A web-user’s perspective Paul Bruthiaux

91

5

Hybridity in ways of speaking: The glocalization of English in Singapore Lubna Alsagoff

109

vi

Contents

Part III: Ethnicity and Ownership

131

6

Whose English? Language ownership in Singapore’s English language debates Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff

133

7

Language and social capital in Singapore Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang

159

8

Language policy and planning in Singaporean late modernity Christopher Stroud and Lionel Wee

181

Part IV: English in Education

205

9

Problematizing the implementation of innovation in English language education in Singapore Rani Rubdy

207

10 Sounding local and going global: Current research and implications for pronunciation teaching Ee-Ling Low

235

11 English as a lingua franca: Negotiating Singapore’s English language education Anne Pakir

261

Part V: Research Bibliography

281

12 Researching English in Singapore: Bibliographic sources

283

Index

305

Series editor’s preface

Of all Asian societies, the role of English in Singapore — at least from an outsider’s perspective — has typically been regarded as most successful and least contentious, compared to other English-using societies in the region. In contrast to Hong Kong, for example, the societal space for English as an interethnic lingua franca has meant that the rationale for English has often foregrounded its utility as a ‘neutral’ language of education and social administration. Singapore has also gained a strong reputation regionally for the relatively high proficiency of its English users, a reputation that has risen in pace with the city-state’s recent branding of itself as a knowledgebased community and arts and education hub. Ironically, however, at the same time that Singapore has worked hard to such ends, its domestic linguistic complaint tradition has striven less to emphasize its strengths, and somewhat more to bemoan the community’s collective mastery of English. While linguists have been fascinated by the emergence of local varieties of English, both educated and informal, the government has expressed much concern about the existence of ‘bad English’, which has often been equated with ‘Singlish’, however vaguely defined and described. Meanwhile, English continues to spread as a language of the home in a society where ‘mother tongue’ — for a number of official purposes — may only refer to such heritage languages as Mandarin, Malay, and Tamil. This present volume, edited by Lisa Lim, Anne Pakir, and Lionel Wee, is ground-breaking in the way in which it is able to account for and explain at least some of these contradictions and tensions. By engaging an ensemble of truly expert commentators on English in Singapore, this book succeeds in providing an insightful account of the interplay of linguistic ecology, language policies, and sociolinguistic realities of the Singapore community, which cumulatively offers a rich and fine-grained account of the sociolinguistics of English in this context. Part I (with a chapter from Lim, Pakir and Wee, and another from Lim) deals with the ecology of English in Singapore, where an integrated ‘ecological model’ requires an

viii Series editor’s preface understanding of the dynamics of both migration and official language policies. Part II, with contributions from Gupta, Bruthiaux, and Alsagoff, then highlights the need to (re)conceptualize ‘English’ in the Singapore context, with particular reference to both the Speak Good English Movement, and the thorny issue of Singlish. Part III — with three chapters by Bokhorst-Heng, Rubdy, McKay and Alsagoff; Vaish, Tan, Bokhorst-Heng, Hogan and Kang; and Stroud and Wee — tackle the issue of ‘ethnicity and ownership’ focusing on such topics as language debates, language and social capital, and the shifting ground for traditional language policies, and the theorization of language planning, in a late-modern consumer society such as Singapore. Part IV then deals with the issue of English language education, with chapters from Rubdy on the history of syllabuses, Low on pronunciation, and Pakir on the implications of the world Englishes (WE) and English as a Lingua Franca (ELF) approaches for English language education. Finally, Part V provides an immensely useful guide to published books and articles in this field. The level of discussion throughout is impressive, and the layering of linguistic description and sociolinguistic commentary that is at the heart of this volume not only serves to illuminate the Singapore situation, but will also impact the conceptualization and discussion of the sociolinguistics of English in Asia in the international frontline of research. This work provides a major contribution to the Asian Englishes Today series and to the description and analysis of English in Singapore, and, as such, it is likely to serve a standard reference work (for both students and researchers) for many years to come.

Kingsley Bolton April 2010

Acknowledgements

Our appreciation extends first and foremost to all the contributors for making this volume what it is: thank you for being part of this project. We also express our heartfelt thanks to Kingsley Bolton, editor of the Asian Englishes Today series, for his enthusiasm and support for a volume on English in Singapore in the first place, his sage advice on its contents, and his aiding and abetting throughout its incarnation. We are much obliged to Nicole Wong, our research assistant, for the initial formatting of chapters and checking of references, as well as for her work in the preliminary collation of the research bibliography, and to the National University of Singapore’s Faculty of Arts and Social Sciences Staff Research Support Scheme for the grant (C103-000-222-091) that funded the research assistantship. We are grateful too to Chris Leung for his work on the cover design and map. We thank the two anonymous reviewers appointed by Hong Kong University Press for their constructive comments on the chapters, and the staff at the Press for their advice, support and help throughout the preparation of this volume.

Contributors

Lubna Alsagoff is Associate Professor and Head of the English Language and Literature Department at the National Institute of Education, Nanyang Technological University, Singapore. Her research interests include the socio-cultural grammar of English as an international language, and the teaching and development of pedagogical grammars of English. Wendy D. Bokhorst-Heng has held appointments as Assistant Professor in the Centre for Research in Pedagogy and Practice at the National Institute of Education, Singapore, and in the Sociology Department of American University, Washington, DC. Her research interests include literacy, language policy and ideology, multilingualism and multiculturalism, and comparative education. Paul Bruthiaux has a PhD in linguistics from the University of Southern California. He has worked in California, Texas, Hong Kong, Singapore, and Thailand. His research interests are English as a global language, the place of language education in economic development, and language variation across social settings. Anthea Fraser Gupta is Senior Lecturer in the School of English at the University of Leeds, England. Her main area of research is in English as a world language, and on the language situation in Singapore. David Hogan is Professor and Dean of the Office of Education Research at the National Institute of Education, Nanyang Technological University, Singapore. Trivina Kang is Assistant Professor in the Policy and Management Studies Academic Group at the National Institute of Education, Nanyang Technological University, Singapore, and is presently seconded to the Centre for Research in Pedagogy and Practice (CRPP) there.

xii Contributors Lisa Lim is in the School of English at the University of Hong Kong, where she directs the Language and Communication Programme, and teaches in that and the English Studies Programme. Her current research areas include New Englishes, especially Asian, postcolonial varieties, with particular interest in contact dynamics in the ecology metaphor, as well as identity, endangerment, shift and revitalization in multilingual minority communities. Ee-Ling Low is Associate Professor in the English Language and Literature Academic Group, and Associate Dean for Programme and Student Development at the National Institute of Education, Singapore. Her current research interests include the linguistic features of world Englishes, language education, and research into initial teacher preparation and teacher professional development. Sandra Lee McKay is currently Visiting Professor at the University of Hawai‘i, Manoa. Her most current works include a text on the social context of English as an international language, as well as a text on L2 classroom research. Presently she is editing a reader on sociolinguistics and language teaching. Anne Pakir is Associate Professor in the Department of English Language and Literature at the National University of Singapore. Her research interests lie in language policy and planning, world Englishes and English in Southeast Asia. Rani Rubdy is Associate Professor in the English Language and Literature Academic Group at the National Institute of Education, Nanyang Technological University, Singapore. Her research interests include English as a world language, language planning and policy, teacher education and the management of educational innovation. Christopher Stroud is Professor and Head of the Department of Linguistics at the University of the Western Cape, South Africa, and Affiliated Professor of Bilingual Research in the Centre for Research on Bilingualism at Stockholm University. His research includes multilingualism, language and education, socio-ethnographies of literacies, and language ideological discourses. Teck Kiang Tan is Research Associate in the Centre for Research in Pedagogy and Practice at the National Institute of Education, Nanyang Technological University, Singapore. His research interests include school effects, multi-level modelling, and longitudinal data analysis.

Contributors xiii Viniti Vaish has a PhD in educational linguistics from the University of Pennsylvania. She is Assistant Professor at Singapore’s National Institute of Education, Nanyang Technological University. She teaches in the English Language and Literature Academic Group and does research for the Centre for Research in Pedagogy and Practice. Lionel Wee is Associate Professor and Head of the Department of English Language and Literature, National University of Singapore. His research interests include the relationship between metaphor and discourse, new varieties of English, language policy, and general issues in pragmatics and sociolinguistics.

Map of Singapore and the region

Part I

The Ecology of English in Singapore

1

English in Singapore: Policies and prospects Lisa Lim, Anne Pakir and Lionel Wee

The Republic of Singapore, independent since 1965, makes an interesting case study for various issues in sociolinguistics, not least because it is an ethnically and linguistically diverse society with a strong history of attempts at social engineering by the state. Language policies instituted by the state, follow-up language campaigns aimed at ensuring conformity to these policies, the ongoing tensions between what the state envisions for the general population and their actual language practices are just some of the phenomena that provide the grounding for a host of analyses. A common thread throughout these phenomena concerns the role of English. As a former British colony, Singapore’s leaders inherited not only a system of government that relied heavily on the use of English, but also a population already given to viewing the language as an important resource for socio-economic mobility. So, rather than dispense with English, Singapore’s leaders decided to retain it as an official language. This is characteristic of the state’s ‘pragmatic’ approach to government, where social and economic policies have been formulated with the intention of keeping Singaporean society open to global and regional forces, whilst retaining a sense of stability and connection to a historic past, however imagined. In this context, the status of the English language is an important barometer of how successful Singapore is in its attempt to stay relevant to, and engaged with, the world ‘outside’, even as it insists on portraying itself in terms of a resolutely Asian identity. It is therefore no exaggeration to suggest that English is implicated everywhere in Singaporean society, as much by its varied manifestations as by its absence. What we intend to do in this brief introduction, then, is to spell out a number of the issues that are implicated by the place of English in Singapore. To do this, we first provide an overview of Singapore’s language policy. We then discuss some relevant lines of investigation before closing with an outline of the various chapters that comprise the present volume.

4

Lisa Lim, Anne Pakir and Lionel Wee

Language policy in Singapore In order to make more manageable Singapore’s ethnolinguistic diversity, the state divides the population into four ethnically based categories: Chinese, Malay, Indian and Others. The state takes it as self-evident that the major ethnic categories (Chinese, Malay, Indian) and their relevant members can be unproblematically identified using a mix of historical language affiliation, cultural practices and racial characteristics. In the case of individuals with mixed heritage, the state’s response is to classify them according to the ethnic membership of the father. These moves are motivated partly on historical grounds (see also Lim’s chapter in this volume), partly on the basis of ideological assumptions, and partly as a matter of administrative expedience. The result is that Singapore’s population of approximately 3.2 million is categorized as 76.8% Chinese, 13.9% Malays, 7.9% Indians, and 1.4% Others (Census of Population 2000). Singapore recognizes the Chinese, Malays, Indians and Eurasians as the ‘founding races’ that have contributed significantly to its independence (Hill and Lian 1995). The decision to maintain this ethnic heterogeneity rather than pursue a strong policy of assimilation was (and still is) motivated by the belief that ethnic distinctiveness is a primordial fact. Consequently, any attempt to quash this distinctiveness can only lead to unnecessary social tensions. We see this explained in the following statement by Singapore’s first prime minister and current minister mentor, Lee Kuan Yew (cited in Han, Fernandez and Tan 1998: 163–5): The Indians have their own method. So do the Malays. The Malays: Islam and also the kinship ties … I don’t think you can erase all that. That’s for hundreds of years, or thousands of years. You can’t erase it. Because I recognised it, I decided you cannot change it. Or if you tried to change it, you’d change it for the worse … In every culture, there is a desire to preserve your distinctiveness. And I think if you go against that, you will create unnecessary problems, whether it is with the Indians and their caste or with the Chinese and their clans.

Despite this desire to recognize the sanctity of ethnic distinctiveness, however, only the first three founding races are considered numerically significant enough to be consistently accorded recognition as ethnic communities in their own right. The Eurasians are sometimes absorbed under the Others category because the state views their number as too small to warrant a distinct category.1 This is especially clear with regard to Singapore’s language policy, which insists that Singaporeans be bilingual in English and their official mother tongue. The official mother tongue is the language assigned by the state to an ethnic community as representative

English in Singapore

5

of that community’s identity and ethnocultural heritage. It is presented in official discourses as a crucial cultural anchor that connects an ethnic community to its ancestral repository of traditional values. With the exception of the Eurasians, the other three ethnic groups each have their own official mother tongue: Mandarin for the Chinese, Malay — the national language — for the Malays, and Tamil for the Indians. The Eurasians have no official mother tongue because the most likely candidate would have to be English. The relatively small size of the Eurasian community notwithstanding, English is not acceptable by the state as an official mother tongue on the grounds that it needs to remain ethnically neutral for a number of reasons. First, English is the major language of socio-economic mobility. The state actively encourages proficiency in English by instituting it as the medium of instruction in the education system. However, it also insists on positioning the language as ethnically neutral so that the distribution of economic advantages is not seen as being unduly associated with a specific ethnic group, which would otherwise raise the danger of interethnic tension. But even as English is successfully seen as being ethnically neutral by most Singaporeans, it is becoming clear that it is not at all socially neutral. Speakers of ‘good’, ‘standard’ or ‘proper’ English are more likely to have come from more affluent homes, have better educational qualifications, and have access to better jobs. As a result, one might reasonably explore the implications for Singapore of a class divide between the English ‘haves’ and ‘have-nots’. Second, English also serves as an interethnic lingua franca. This is needed in order to ensure that cross-ethnic communication is facilitated, so that the different ethnic groups do not become segregated from each other. Ethnic segregation would have highly negative consequences for attempts at cultivating a sense of national unity. However, actual crossethnic communication sometimes makes use of languages that carry little or no state approval. For example, military service (National Service) is obligatory for Singaporean males. Life in the army brings together young men of varied ethnic and socio-economic backgrounds, and usefully creates opportunities for them to (learn to) interact with each other. These men quickly become sensitized to class differences, and, in order to neutralize rather than exacerbate such differences, ‘standard’ or ‘proper’ English is eschewed in favour of Singapore English or Singlish, the colloquial variety of English that is denigrated by the state, or one of the Chinese dialects such as Hokkien, which, unlike Mandarin, has no official standing (Stroud and Wee 2007). What this means is quite simply this: the state correctly realizes that language can serve as a key social lubricant; but there is perhaps an unbridgeable gap between the language that the state would prefer and the language that can most effectively do the job.

6

Lisa Lim, Anne Pakir and Lionel Wee

Third, English is treated as a language that essentially marks a nonAsian ‘other’, and therefore cannot be bestowed the status of official mother tongue. This goes to the crux of why Singapore’s language policy is an aggressively bilingual one. Singapore cannot do without English; attempting this would mean disengaging itself from the global economy, with predictably disastrous consequences. At the same time, it cannot do with only English; attempting this would mean compromising Singapore’s ‘Asianness’ by allowing a Western language to play a constitutive role in local identity politics, a role that is reserved for the mother tongues. As a result, the language policy treats both English and the mother tongues as equally important, though for different reasons. This situation nicely illustrates the highly politicized status of English. A specific ethnic group is denied the possibility of claiming English as its mother tongue, because the language must serve the entire country. As an official language, English is valuable because it provides access to technological and scientific knowledge, and helps maintain economic competitiveness in an increasingly globalized world. English can of course serve as a lingua franca, but Singaporeans are not generally expected to develop a sense of identification with the language. At both the national and communal levels, then, tensions exist in the functions that English fulfils, and in how these are managed. The foregoing description of Singapore’s language policy suggests rich possibilities for investigations along various lines, and it is to these that we now turn.

From past to present Though the practice does little harm if we are careful to bear in mind the inevitable oversimplifications, it is clear that it is not enough to merely speak of ‘English’ as being present at different times in Singapore’s history, up to and including the present. The recurrent use of this label in different contexts can in fact obscure significant changes and variations in actual language use, since what counts as ‘English’ — that is, what kinds of usage activities are considered to merit being categorized as ‘English’ — is itself ultimately a matter of social and political negotiation. The names that we give to languages reflect invented understandings of language (Makoni and Pennycook 2007: 2), as the following brief example illustrates. As Makoni and Pennycook (2007: 9) point out, Sir George Abraham Grierson’s linguistic Survey of India, which was completed in 1928, had to face the problem of deciding on the boundaries between languages and dialects. To do this, Grierson openly admitted the need to invent language names while ignoring the complexity of actual language use (1907: 350, quoted in Makoni and Pennycook 2007: 10):

English in Singapore

7

… nearly all the language-names have had to be invented by Europeans. Some of them, such as Bengali, Assamese, and the like, are founded on words which have received English citizenship, and are not real Indian words at all, while others, like ‘Hindostani’, ‘Bihari’, and so forth, are based on already existing Indian names of countries and nationalities.

The significance of this, as Makoni and Pennycook (2007: 10) observe, is that ‘these were not just new names for existing objects … but rather the invention and naming of new objects. The naming performatively called the languages into being’. Returning to the case of English in Singapore, we therefore have to be careful that we do not get too carried away with labels such as ‘English’, ‘Singlish’ or even ‘Singapore English’.2 The linguistic diversity of Singapore means that English inevitably comes into contact with other languages present in the sociolinguistic environment such as the different varieties of Malay, Chinese and (to a lesser extent) the Indian languages. This clearly points to the possibility of structural changes in the language, which tend to occur in concert with developments in the socio-political arena (see Schneider 2007). Given the significantly growing presence of English in the lives of many Singaporeans, two important questions then need to be asked. Firstly, how are the various structural features of English related to its presence in the changing landscape of Singapore’s sociolinguistic history? Secondly, does English have the potential to displace other languages (such as the official mother tongues), and would such displacement (where it is perceived to occur) be uniformly distributed across the society or skewed along specific ethnic identities or cultural activities?

Revisiting key assumptions and concepts Discussions about English, implicitly or otherwise, typically draw on various assumptions that are all too often taken for granted in public discourses. Such assumptions include the questions of ownership and standardization, and the relationship between language and culture. The question of language ownership (Widdowson 1994; Jenkins 2000) is especially relevant for a global language such as English, whose spread across the globe has led to the rise of different varieties — ‘New Englishes’ — as well as an increase in the number of non-traditional native speakers. In the light of such developments, it has become necessary to consider the extent to which non-traditional native speakers of English should feel that they too have a legitimate say in what counts as grammatical or appropriate. Related to this is the question of standardization. Here, it is not so much standardization as the process of trying to eliminate variation (Milroy

8

Lisa Lim, Anne Pakir and Lionel Wee

and Milroy 1999) that is at stake, but rather, the more commonplace notion of what sorts of standards non-traditional native speakers ought to have in mind. For many ordinary Singaporeans as well as the state, there is often some ambivalence in this matter, whether the choice is exonormatively towards an American or British standard, or endonormatively towards the variety associated with the local elite. Finally, while many Singaporeans seem to be quite convinced that English is essentially a Western language that serves as a vehicle for Western values, and that they ought instead to be fluent speakers of their own mother tongues, this remains a desideratum rather than a sociolinguistic reality. In fact, the state’s promotion of English can itself be credited with contributing to massive language shift over a period of thirty years (Li, Saravanan and Ng 1997: 368; Stroud and Wee 2007). While the rise of English is most pronounced in Chinese and Indian homes, Malay homes too show a similar shift, albeit less pronounced, possibly due to the close affiliation between the Malay language and the religion of Islam (Pakir 1993: 75; Kwan-Terry and Luke 1997: 296). This leads to the question of whether or not it makes sociolinguistic sense to continue positioning English as a language of the non-Singaporean ‘Other’ rather than as a language that belongs to Singaporeans themselves. This question, of course, intersects with heated debates over what it means to be a native speaker of English, whether ownership of the language can only ever reside with ‘traditional native speakers’, and whether accepting English as a Singaporean language compromises Singapore’s claim towards an Asian identity. All of these developments make it imperative that Singaporeans open up a dialogue on the ideologically loaded question of whether English is intrinsically a Western language.

The gap between policy and practice Consider the observation in the preceding section that English is increasingly becoming the home language for many Singaporeans. Such a development is perhaps not surprising since the language is the medium of instruction in schools, which makes knowledge of English particularly critical in considerations of socio-economic mobility. The status of English as a home language is perhaps most clearly the case among Chinese Singaporeans, where English has overtaken Mandarin as ‘the primary language used in homes of Primary 1 Chinese pupils’ (Ministry of Education [MOE] 2004a). This has forced the state to acknowledge that a significant number of Chinese Singaporeans actually have great difficulty coping with Mandarin, despite the fact that it is supposed to be their official mother tongue. Consequently, in early 2004, the Ministry of Education (2004a)

English in Singapore

9

announced a number of changes to the mother tongue policy. Especially interesting was the introduction of a ‘B’ syllabus for the Chinese language, to cater to students with learning difficulties. These changes indicate that, contrary to its earlier expectations, the state no longer believes that the majority of Chinese Singaporeans are likely to be highly proficient in Mandarin. It is only a minority, an elite estimated at about 10% of the student population (The Straits Times, 26 November 2004), who are expected to be fully bilingual in English and the mother tongue. As a result, in a subsequent press release in September (MOE 2004b), the Ministry of Education announced the start of a Chinese Bicultural Studies Programme, aimed at the minority of students who are able to cope with both English and Mandarin at a high level of competence. And at the end of that same year, the government announced that it would undertake a review of the Malay and Tamil language curricula to see if measures similar to those taken for Mandarin are indeed called for (MOE 2004c). At the same time that the increased use of English has impacted the state’s general expectations regarding the bilingual capacity of the population, it has also led to growing fears that a local variety, known as Singlish (and in scholarship usually referred to as colloquial Singapore English), is gaining popularity and legitimacy, and that this might jeopardize Singaporeans’ ability to improve their command of standard English. Such fears were the main motivations behind the initiation of the Speak Good English Movement, making English the only other language (in addition to the promotion of Mandarin in the Speak Mandarin Campaign) to have been the object of a national language campaign. In rationalizing this campaign, the state has often made the argument that it is not possible to ‘go global’ with Singlish. A choice, it argues, has to be made between the ‘ghettoization’ (Freeland and Patrick 2004: 17) that supposedly accompanies Singlish and the economic prosperity that comes from speaking ‘good’ English. However, supporters of Singlish, such as writer Hwee Hwee Tan, have responded by arguing — on the public stage that is Time — that it is an authentic reflection of a national identity (Tan 2002). To this, the Speak Good English Movement has suggested that a more appropriate linguistic expression would be ‘a standard Singaporean English’, even though it is unclear what such an endogenous standard would or should look like. In many ways, Singlish is an inevitable development. As a population becomes more proficient and more comfortable with using English, it will necessarily make the language its own. And this of course means that a nativized, restructured variety will emerge, reflecting the population’s greater ease with and wider use of the language in the naturalistic environments of home and informal peer interactions. But it is an unfortunate fact that new varieties such as Singlish are all too often viewed

10 Lisa Lim, Anne Pakir and Lionel Wee as a corrupt version of standardized ones. There is also some irony in all this. Recall that the state has long promoted English as an interethnic lingua franca. It might therefore seem that with the emergence of Singlish, the role of English in cutting across Singapore’s ethnic mix has come to fruition. But unfortunately, Singlish has been often disparaged by Singapore’s leaders — to quote the then prime minister in his National Day Rally Speech — as a ‘type of pidgin English’ (Goh 1999). Public discourses surrounding Singlish are therefore highly contested (Fong, Lim and Wee 2002; Lim 2009). Supporters of Singlish claim it to be a colloquial variety used mainly to reflect and build up social solidarity. Opponents of Singlish, including the state, prefer to see it as clear evidence of a drop in standards of usage. And in fact, the state’s concerns about Singaporeans’ ability to speak and write standard English have prompted the Ministry of Education to announce a slew of educational initiatives, including greater emphasis on oral communication skills, continued attention to grammar throughout the educational stream, and the introduction of linguistics as a pre-university subject, among others. In making these announcements, the ministry’s decision to employ foreign language experts has raised concerns that this represents another indicator of the state’s inability to free itself of the hegemony of the native speaker ideology. A more useful approach would be to find ways of helping students learn the standard variety whilst not forcing them to disavow whatever knowledge of English they already possess — even if this knowledge consists of a variety that is stigmatized. Any exploration of this particular avenue, however, would first require putting aside deeply ingrained language prejudices, and it is unclear at present whether the political will exists for such an initiative.

Contributions in this volume The issues raised in this sketch of the English language situation in Singapore are explored in detail by the contributors to this edited volume.

The ecology of English in Singapore This volume starts with a section which sets the stage by outlining the ecology in which English exists, providing valuable socio-political and historical overviews, and considering some of the more crucial factors whereby English has evolved to reach the state it is in Singapore today. This first and current chapter has already provided a sketch of Singapore’s language policies and related issues and implications.

English in Singapore 11 Complementing this is Lim’s consideration of ‘Migrants and “mother tongues”’, which examines the linguistic ecology of Singapore, focusing on two external factors of migration and language policies over different eras, and demonstrates the significance of these for a better understanding of the development of a contact language such as Singlish (Singapore English, SE), and the implications these hold for policy and education. In the different eras distinguished by differing migration patterns and language policies, different sets of languages can be seen to be dominant. In the era dating from pre-colonial times through to early post-independence years, characterized by natural immigration and vernacular maintenance, two main original immigrant languages, Bazaar Malay and Hokkien, are dominant. During early independence where there was controlled immigration and new language and educational policies instituted, the official languages, in particular Mandarin, gain prominence. In the era of late modernity, with foreign manpower and a relaxation with regard to nonofficial languages, global-media languages such as Cantonese see a rise in prominence. An examination of structural features of SE does indeed show the influence of these various languages at different stages of SE’s evolution. Lim also examines the current era, which is replete with new practices in immigration and language policies, and identifies other languages which she predicts may soon play a significant role in Singapore’s ecology. Combining a sensitivity to historical eras with an examination of linguistic features in SE and those of the various contact substrate languages, Lim suggests, allows the sources of various linguistic features of SE to be discerned with greater precision. This then contributes to the establishment of the more likely substrate sources, and in turn a better appreciation, not only for the structure of SE, but also for the social forces that have shaped it.

Reconceptualizing English Given that English in Singapore has evolved over the decades, the next section of chapters considers what English has become — in a process of reconceptualization which involves reviewing fundamental notions which are usually taken as given, such as ‘standard English’ and ‘good English’, as well re-evaluating what ‘Singapore English’ itself entails — and the implications this holds for policy and pedagogy. Gupta’s contribution, ‘Singapore Standard English revisited’, takes as a starting point her early discussion in the 1980s with Mary Tay of the possibility of a Singapore Standard English. At the time, arguments for an endonormative standard for Singapore English were seen as revolutionary. In the thirty years since, however, it has become widely accepted that local words and local accents are necessarily part of standard English.

12 Lisa Lim, Anne Pakir and Lionel Wee The concept of a local standard grammar, though, remains problematic. In her latest thinking on the matter, Gupta suggests that it is no longer appropriate to predetermine what is and is not standard by the methods adopted in the 1980s. The chapter therefore considers the meaning of standard English for Singapore in the twenty-first century, and develops the concept of International Englishes, arguing that the main variety of International English is ordinary standard English, which is one dialect of English with minor regional differences (mostly in lexis). While all Englishusing communities participate in the maintenance of International English, they also usually have local non-standard dialects of English, which are used in specific contexts, especially informal speech, literature, and humour, and Singlish fills this slot in Singapore. In an analysis of the way in which Singaporeans use English in formal and informal written material, and the language advice given online and in The Straits Times in connection with the Speak Good English Movement (SGEM), Gupta suggests that Singaporeans participate in the same uncertainty about standard English as users of English from other English-using nations. The SGEM, which has figured so prominently in recent public discussions of English, is also the focus of Bruthiaux’s chapter ‘The Speak Good English Movement: A web-user’s perspective’, which examines the campaign’s claim to be based on sociolinguistic scholarship, its policy goals, and its principal activities. Bruthiaux makes a number of key observations. First, the campaign suggests a conflict within Singapore government policies in that this top-down approach to language management cannot be easily reconciled with the simultaneous encouragement of critical skills and informed choice, especially in education. Second, far from equipping the nation with the linguistic tools it needs to flourish in the twenty-first century, the campaign is an outdated attempt to perpetuate increasingly irrelevant postcolonial preoccupations with exogenous standards, suggesting a lingering lack of self-confidence among the Singapore leadership. Third, the publicly unchallenged claim that both international and internal communication can only be effectively transacted through standard English (however defined) is manifestly false given current self-reports of actual usage. Fourth, despite the apparent academic backing the campaign enjoys, it betrays a profound misunderstanding of the nature of language variation and of the dynamics of language in use across a range of social settings. Bruthiaux then argues that an appropriate policy response should abandon the misconceived and ineffectual effort to campaign Singlish out of existence. Instead, public resources should be devoted to helping all Singaporeans become comfortable along the entire continuum of English (from the colloquial to the standard) through a sustained educational effort while letting the Singlish end to look after itself.

English in Singapore 13 In ‘Hybridity in ways of speaking’, Alsagoff’s target of reconceptualization is Singlish itself. From the context of recent discussions of English in Singapore being pulled in two opposing global-local directions, she offers a model of variation of English in indigenized contexts. Originally conceived as the Cultural Orientation Model (COM), which explains variation in relation to the global-local contrast of the cultural orientations of speakers, Alsagoff develops this further in relation to the concept of ‘glocalization’, which emphasizes the simultaneity of the global and the local in the process of globalization. In this light, she suggests that, given the co-presence of features of both local and global in the speech of Singaporeans, a change in the approach to describing language variation in Singapore is required. Glocalization, which presents language and identity as intertwined and fluid, offers a more dynamic orientation for understanding the ways in which people appropriate English for their own purposes, but who are at the same time constrained by institutional discourses and policies favouring standardization and conformity. Singlish is thus seen more as a range of lingua-cultural resources that speakers use in order to identify or mark a change in cultural orientation or style.

Ethnicity and ownership On the basis of the previous section, involving a recognition and acceptance that English needs to be reconceptualized, in terms of what it is and how it is used in multilingual societies such as Singapore, this section takes up the question of where the ownership of such new varieties of English lies. The chapters here also explore the ideologies and claims that underlie current policies, as well as the disjuncture between this and Singaporean sociolinguistic reality, reflecting on the potential repercussions of policy for the status of (the varieties of) English as forms of linguistic capital. The first chapter in this section, by Bokhorst-Heng, Rubdy, McKay and Alsagoff, takes us straight into the debate, asking ‘Whose English?’ in considering language ownership in Singapore’s English language debates. It addresses the idealization of the so-called native speaker found within the native-speaker (NS)/non-native-speaker (NNS) dichotomy, as well as the unwillingness to recognize the different varieties of world Englishes as legitimate languages. Within Singapore’s English language debates, this NS/NNS dichotomy and ownership discourse is evident particularly in the government’s steadfast denial of allowing Singaporeans native speaker membership, even though more and more Singaporeans do in fact regard English as their first language and the primary or only language of the home. Instead, the officially preferred model is British RP, and the Inner Circle speakers of English continue to be regarded as the true

14 Lisa Lim, Anne Pakir and Lionel Wee owners of English. However, by comparing the findings of two recent studies conducted by the authors, the chapter aims at a model of language ownership that moves away from the limitations of NS/NNS discourses, and focuses instead on speakers’ orientations towards English norms to foreground speakers’ degree of ownership of the English they speak. The chapter ‘Language and social capital in Singapore’ by Vaish, Tan, Bokhorst-Heng, Hogan and Kang focuses on language maintenance as an outcome of social capital within particular ethnic groups. Starting with the All Party Report on Chinese Education of 1956, language planning in Singapore has managed to maintain ‘mother tongues’ in the nation’s linguistic ecology against the onslaught of global English. However, these gains are not evenly spread across social groups. The authors therefore ponder the following questions: What are the differences in language use on the basis of ethnicity? How are these differences materialized in the various socio-spatial domains? The authors hypothesize that some ethnic groups are able to retain their mother tongue languages through social ties, which can be conceptualized as social capital created within the group by strengthening common cultural values and beliefs to achieve resource sharing. The analysis is based on the ongoing Sociolinguistic Survey of Singapore (SSS 2006), which surveys one thousand children stratified by race and class, linked to twenty-four qualitative follow-up case studies, and documents patterns of language use in the domains of school, family and friends, religion, public space and media, and asks questions about attitudes and ideology based on such language use. In the final chapter of the section, Stroud and Wee evaluate ‘Language policy and planning in Singaporean late modernity’. They note that Singapore’s language policy attempts to manage the tension between modernity (construed as a global orientation achieved through the medium of English) and tradition, where each mother tongue is supposedly the cultural repository of values for its associated ethnic group. They then go on to argue that a sociolinguistic ordering around notions of ethnicity and nation does not fit easily with the multilingual dynamics of late modern societies. Societal development in late modernity is generating linguistic hierarchies of value that are reconfiguring issues of language and ethnicity into questions of language and class. Despite this, Singapore’s language policy continues to be firmly shaped by (conventional) ethnolinguistic frameworks. The chapter thus argues that Singapore’s language policy needs to appreciate that patterns of multilingualism are increasingly constructed around the dynamics of language choice and change in terms of a logic of lifestyle consumption. Specifically, the sociolinguistics of multilingualism needs to recognize the consumer as a specific type of (linguistic) identity. Their proposal, which gives greater prominence to autonomy, choice, and reflexivity — notions that seldom figure in conventional language policies

English in Singapore 15 — is thus for a reconceptualization of the notion of language in terms of sociolinguistic consumption, an understanding of identity as involving not only processes of recognition but also of (re)distribution, and the deconstruction of the category of mother tongue in discourses of language planning.

English in education Continuing the thrust of reconceptualizing English in Singapore, the final section focuses on the domain of education, where English often faces the most controversy. The contributions here provide an examination of the potential for innovative methods in English language education, and also consider the model of lingua franca, and the tension between exonormative and endonormative practices in teaching. The point of departure for Rubdy’s ‘Problematizing the implementation of innovation in English language education in Singapore’ is the intensely proactive management of educational policies and the decisiveness and expedience with which these policies are generally implemented. Most studies of Singapore’s English language policies have, however, focused on the ‘what’ (i.e., the goals and content-based changes) rather than the ‘why’ (i.e., the pedagogical assumptions and beliefs or ideological rationales underlying them) or the ‘how’ (i.e., the means employed and the general approach adopted in their implementation) of these reform initiatives. Given this current state of affairs, Rubdy’s chapter provides a much-needed critical review of the structures and practices involved in the English language syllabus over the years, identifying the distinct stages — the curriculum having been revised approximately every decade — in its evolution. In so doing, she deconstructs how the assumptions and ideological beliefs underlying them have helped create, on the one hand, the prevailing educational culture and, on the other, a docile workforce that serves the country’s economic targets but lacks the creativity and critical mindset integral to the New World Order. Low then focuses on a specific area of pronunciation in her chapter ‘Sounding local and going global’. She provides an overview of research in the phonetics, phonology and prosody of Singapore English, and highlights the principle of intelligibility, following Jenkins (2000) in identifying a core for the teaching of pronunciation. Stressing the importance of preserving both global and local orientations in pronunciation, she then proposes a number of principles and practices that may be adopted in developing a pronunciation syllabus for English in Singapore. Rounding up this section and the volume, Pakir’s ‘English as a Lingua Franca: Negotiating Singapore’s English language education’ reflects on

16 Lisa Lim, Anne Pakir and Lionel Wee English language education in the twenty-first century, drawing on current discussions within the paradigms of world Englishes, originating first in the US and gaining currency in Kachru’s Outer and Expanding Circles of English, and English as a Lingua Franca (ELF), originating in Europe, which have developed distinct theoretical models of ‘lingua franca’. Pakir then looks at how ‘lingua franca’ as a theoretical concept may help us understand the range of interpretations of English in localized contexts, including the possibility of experimentation that can and must go on as English transplants itself into different soils. The mobility and portability of English and its changing functions, values and meanings in localized contexts create hybrids and mixed varieties, some desired and some less so. These pose challenges to applied linguistics, English language education and pedagogy, and the discussion of identity for English-knowing bilinguals.

Research bibliography Obviously no single volume can encompass all the interests and angles of research on a particular intellectual or geographical area. The current volume takes as its angle an exploration of the implications which have arisen as a result primarily of the language policies that have been instituted in Singapore over the decades, and thus provides a very focused and coherent collection of the most current thinking and research on issues in this regard. 3 For those whose interests are piqued, there is a wealth of work that has been conducted on other areas of research on English in Singapore, and the closing chapter of this volume provides a valuable resource for readers in this respect, comprising a selective bibliography of such research.

Prospects The next few decades in this era of modernity will see Singapore facing various challenges. Among them are the following: pursuing foreign talent and encouraging such talent to take up Singaporean citizenship, retaining ties with those Singaporeans who have migrated overseas by cultivating a sense of a Singaporean ‘diaspora’, and narrowing a potentially devastating class divide between relatively well-off Singaporeans and their less affluent counterparts. In trying to manage these challenges and others, it is clear that discussions over the role of English in Singapore and for Singaporeans will continue to be relevant. In this regard, we are optimistic that the chapters in this volume have a significant contribution to make to these discussions.

English in Singapore 17

Notes 1.

2.

3.

There are occasions when the Eurasians are acknowledged as an ethnic group in its own right (Rappa and Wee 2006). For example, the Eurasian Association is treated as one of four ethnically based self-help groups, alongside the Chinese Development Assistance Council, Mendaki (for the Malays), and the Singapore Indian Development Association. These groups are all Institutions of Public Character (IPC) and each receives dollar-for-dollar matching from the government for funds that are raised. This is not the same as saying that such labels should be ignored. They reflect metalinguistic assumptions about how language practices cluster together, as well as how such practices index particular in-group and out-group identities. See also Fong, Lim and Wee (2002). We would in any case like to place on record our regret that we were unable in particular to include a section on English in Singapore literature, including Singapore films. Scheduling conflicts and prior commitments made it difficult, if not downright impossible, for the potential contributors who had been invited to complete their manuscripts on time.

References Fong, Vivienne, Lim, Lisa and Wee, Lionel (2002) ‘Singlish’: Used and abused. Asian Englishes, 5(1), 18–39. Freeland, Jane and Patrick, Donna (2004) Language rights and language survival: Sociolinguistic and sociocultural perspectives. In Language Rights and Language Survival: Sociolinguistic and Sociocultural Perspectives. Edited by Jane Freeland and Donna Patrick. Manchester, UK: St. Jerome Publishing, pp. 1–33. Goh, Chok Tong (1999) First-world economy, world-class home. Prime Minister’s National Day Rally Speech 1999. Ministry of Education, Media Centre, Speeches. Singapore: Ministry of Education. http://www.moe.gov.sg/media/ speeches/1999/sp270899.htm. Accessed January 2008. Han, Fook Kwang, Fernandez, Warren and Tan, Sumiko (1998) Lee Kuan Yew: The Man and His Ideas. Singapore: Times. Hill, Michael and Lian, Kwen Fee (1995) The Politics of Nation-Building and Citizenship in Singapore. London: Routledge. Jenkins, Jennifer (2000) The Phonology of English as an International Language. Oxford: Oxford University Press. Kwan-Terry, Anna and Luke, Kwan-kwong (1997) Tradition, trial and error: Standard and vernacular literacy in China, Hong Kong, Singapore and Malaysia. In Vernacular Literacy: A Re-evaluation. Edited by Andrée Tabouret-Keller, Robert B. Le Page, Penelope Gardner-Chloros and Gabriella Varro. Oxford: Clarendon Press, pp. 271–315. Li, Wei, Saravanan, Vanithamani and Ng, Lee Hoon Julia (1997) Language shift in the Teochew community in Singapore: A family domain analysis. Journal of Multilingual and Multicultural Development, 18(5), 364–84.

18 Lisa Lim, Anne Pakir and Lionel Wee Lim, Lisa (2009) Beyond fear and loathing in SG: The real mother tongues and language policies in multilingual Singapore. In Multilingual, Globalizing Asia: Implications for Policy and Education. AILA Review, 22, 52–71. Edited by Lisa Lim and Ee-Ling Low. Makoni, Sinfree and Pennycook, Alastair (2007) Disinventing and reconstituting languages. In Disinventing and Reconstituting Languages. Edited by Sinfree Makoni and Alastair Pennycook. Clevedon: Multilingual Matters, pp. 1–41. Milroy, James and Milroy, Lesley (1999) Authority in Language. 3rd ed. London: Routledge. Ministry of Education (2004a) Refinements to Mother Tongue language policy. Ministry of Education, Media Centre, Press Releases. Singapore: Ministry of Education. 9 January 2004. http://www.moe.gov.sg/media/press/2004/ pr20040109.htm. Accessed January 2008. Ministry of Education (2004b) Nurturing a core of students with advanced knowledge of Chinese language and culture. Ministry of Education, Media Centre, Press Releases. Singapore: Ministry of Education. 3 September 2004. http://www. moe.gov.sg/media/press/2004/pr20040903.htm. Accessed January 2008. Ministry of Education (2004c) Malay Language and Tamil Language Curriculum and Pedagogy Review Committees set up. Ministry of Education, Media Centre, Press Releases. Singapore: Ministry of Education. December 2004. http://www. moe.gov.sg/media/press/2004/pr20041214a.htm. Accessed 28 September 2008. Pakir, Anne (1993) Two tongue tied: Bilingualism in Singapore. Journal of Multilingual and Multicultural Development, 14(1/2), 73–90. Rappa, Antonio and Wee, Lionel (2006) Language Policy and Modernity in Southeast Asia: Malaysia, The Philippines, Singapore, Thailand. New York: Springer. Schneider, Edgar W. (2007) Postcolonial English: Varieties Around the World. Cambridge: Cambridge University Press. Stroud, Christopher and Wee, Lionel (2007) Consuming identities: Language policy and planning in Singaporean late modernity. Language Policy, 6(2), 253–79. Tan, Hwee Hwee (2002) A war of words over ‘Singlish’. Time, 22 July. http://www. time.com/time/magazine/article/0,9171,322685,00.html. Accessed January 2008. Widdowson, Henry G. (1994) The ownership of English. TESOL Quarterly, 28(2), 377–89.

2

Migrants and ‘mother tongues’: Extralinguistic forces in the ecology of English in Singapore1 Lisa Lim

It has long been recognized that the history and fortunes of Singapore have been closely intertwined with migrants and migration (e.g., Yeoh 2007). In this chapter, I suggest that the fortunes of the various languages in the ecology of Singapore — their various rises and falls — can also be seen to be not only intertwined with migrants and migration but also very much affected by politicians and policies. Rather than simply consider these as distinct factors in the scenario, however, I represent them as components of an integrated ecological model for understanding the dynamics of the evolution of English in Singapore.2 Such an approach in linguistic study — widely associated with work by Mufwene (e.g., 2001, 2008) — uses ecology for language as a metaphor from biology and population genetics. Ecology encompasses both internal and external aspects (or intra- and extralinguistic features): this includes not only internal factors, such as the typology of languages in the feature pool, from which an emerging linguistic variety draws its features through a process of competition and selection, and the frequency of particular linguistic features; it also includes external factors, such as power relations between speakers of the languages, relative prestige of languages, and so on, all of which contribute to the relative dominance of a language. The significance of ecology in the investigation of linguistic features of Singapore English (SE) has been demonstrated in previous work, addressing discourse particles (Lim 2007a, 2007b, 2009a) and the presence of tone (Lim 2008b, 2009b, forthcoming), focusing in particular on internal factors of the ecology, namely the typological dominance of Sinitic varieties in the feature pool — Cantonese, in particular, where particles are concerned, for their richness in number, tone and meaning. This present chapter focuses primarily on the external factors in the ecology of Singapore, examining in some detail what I identify as the two major forces that play a role in raising the dominance of certain languages over others.

20 Lisa Lim The first force, as identified at the outset of this chapter, comprises the changing trends in immigration patterns which are an important factor in Singapore’s dynamic ecology, and which are particularly significant in the early British colonial period, in more recent decades, and in these current and future years. The other significant force involves the implementation of various — again, swiftly changing — language policies which impact on the importance that the different languages have at different periods of time. To some extent then, this approach can be seen to be echoing what is expressed in Bloom (1986: 359), where ‘up to the early part of [the twentieth] century, the linguistic situation of Singapore, in particular the division between English speakers and non-English speakers, was determined largely by settlement patterns and colonial policy, or the lack thereof’. However, this chapter aims to go further than this, and not just in terms of the time line. The first half of the chapter will provide an analytical account of Singapore’s immigration patterns and policy implementation from the (pre-)colonial era to the present day (for the period up until the 1980s, see also Bloom 1986 for a more comprehensive and critical survey than what is possible here). What is important to bear in mind what these patterns and policies translate to in ecological terms. By this I mean how we can understand the policy decisions and immigration patterns in terms of how they lead to different communities and/or their languages becoming more or less dominant in the ecology during a particular period — either because certain languages have been given institutional support or are seen as having certain capital (after Bourdieu 1984), or because the communities that speak the language are more dominant because of their greater numbers or economic strength or prestige. The identification of dominance contributes, in turn, to explaining the dynamics of language evolution: the more dominant a community or language in the external ecology, the greater the likelihood that features of that language are dominant in the competition process and are selected from the feature pool into the emergent linguistic variety.3 Further, patterns of both immigration and policy change over time, and indeed do so relatively swiftly and distinctively in Singapore’s case. This chapter shows how, by proposing a periodization for Singapore’s ecology using these two factors, we can recognize several distinct eras, each with relatively stable characteristics. Based on this, we can then identify which languages are dominant in the ecology in the different eras. Finally, in the second, shorter part of this chapter, these eras of immigration and policies are related to the linguistic ages first outlined in Lim (2007a), and suggestions are made for the influence on SE that the dominant languages have had in each age.4 Recognizing that migration and policies have been crucial factors in Singapore’s history and development is, of course, not novel; what this

Migrants and ‘mother tongues’ 21 chapter does is to show how such factors can be viewed in an integrative model of ecology, as well as provide an illustration of the value of an ecological approach in an analysis of the structure and evolution of a linguistic variety.

Population and policies In this section, I propose a periodization in which we divide Singapore’s history into a number of identifiable socio-historical eras, each distinguishable in terms of two factors: (1) the trends in immigration patterns, and (2) the implementation of various language policies,5 both of which impact on the importance that different communities and/or languages have at different periods of time. The eras identified are: (a) the colonial era (c. 1800s to 1960s); (b) independence (1960s to late 1980s); (c) late modernity (late 1980s to 2000); and (d) a new world order (2000 to the present), which will be elaborated on in the rest of this section. It should be noted that these eras should not be seen as coinciding precisely with historical landmarks or incidents, and the dividing dates are meant to indicate an approximate point in time. Further, such a division into eras does not suggest that the eras change abruptly and are discrete, but rather that the situations in one period transform gradually into the next. In what follows, for each of the eras, I highlight the significant languages that would have been spoken and been in contact, and their position in the linguistic ecology of Singapore.

The colonial era: Natural immigration and vernacular maintenance Long before Singapore was claimed for the British East India Company by Sir Stamford Raffles in 1819, the ancient Indian Ocean trade routes had already seen extensive contact in the region between sailors and traders from south, east and southeast Asia. From as early as the fourteenth century, Singapore — known then as Temasek — was a node in the trading networks along with Riau and other islands in the archipelago. It is reported to have been inhabited at the beginning of 1819 by a few families of Orang Laut (‘sea people’ or Proto-Malays), pirates, a settlement of more than 30 Chinese cultivators of pepper and gambier, and about 100 Muslim Malay fisherfolk led by the Temenggong of Johore, who had moved there in 1811 (Bloom 1986: 349). The establishment of Singapore as a British trading post by Raffles in 1819 meant a rapidly expanding economy, which was coupled with a liberal, open-door immigration policy (Yeoh 2007); this resulted in an even more

22 Lisa Lim rapid influx of immigrants, the majority of them from southern China, Malaysia, the Indonesian archipelago, and South Asia. The population quickly grew from a few hundred to half a million by the 1931 census, and in fact, prior to World War II, population increase was primarily due to immigration. Already with a capitan (‘captain’) system in place — invented pre-colonially in Malacca — which divided the community into three basic groups of ‘Malays’, ‘Chinese’, ‘Indians’ and ‘Others’, whereby each ethnic community had in effect its own legal system (Bloom 1986: 352), the growing population continued to be settled according to Raffles’ Town Plan of 1822, which, as part of the divide-and-rule policy of the British, involved the allotment of separate areas within the central urban area to the different ethnic communities. The Chinese were located in the entire area south of the Singapore River, itself subdivided according to different dialect groupings (Kwok 2000: 203). The Indians were in a small area on the south bank of the river (which later became more scattered with several distinct enclaves according to language, religion, caste and trade, Liu 1999: 133), and the Malays were in two areas, each around one of the leaders who had signed the documents ceding Singapore to the British (Liu 1999: 19– 20, 61–6). The different groups built their own temples, formed their own religious and clan-based welfare associations, and set up their own trade and occupational guilds (Liu 1999: 143). Table 2.1 Percentage distribution of total population by ethnic group (from Kwok 2000: 200) Year

Total Pop.

Chinese

Malay

Indian

Others

Total

1824

10,683

31.0

60.2

7.1

1.7

100

1830

16,634

39.4

45.9

11.5

3.2

100

1836

29,984

45.9

41.7

9.9

2.6

100

1840

35,389

50.0

37.3

9.5

3.1

100

1849

52,891

52.9

32.2

11.9

3.0

100

1860

81,734

61.2

19.8

15.9

3.1

100

1871

97,111

56.2

26.9

11.8

5.0

100

1891

181,602

67.1

19.7

8.8

4.3

100

1901

226,842

72.1

15.8

7.8

4.3

100

1911

303,321

72.4

13.8

9.2

4.7

100

1921

418,358

75.3

12.8

7.7

4.2

100

1931

557,745

75.1

11.7

9.1

4.2

100

1947

938,144

77.8

12.1

7.4

2.8

100

1957

1,445,929

75.4

13.6

8.6

2.4

100

1970

2,074,507

76.2

15.0

7.0

1.8

100

1980

2,413,945

76.9

14.6

6.4

2.1

100

Sources: Cheng 1985; Saw 1970; Singapore Census of Population 1990; Demographic Characteristics 1992.

Migrants and ‘mother tongues’ 23 The population in this era exhibited a richness and diversity in terms of origin and multilingual repertoire. The structured outline that follows describes the pattern of immigration and settlement of the various communities; each numbered paragraph addresses a particular community, distinguished by their region of origin, namely, from the Malay archipelago, China, and (then British) India, or by their being a particular mixed group, namely the Straits Chinese and the Eurasians. In some cases, a paragraph about a community encompasses a number of different linguistic varieties. a.

While there was a fairly homogeneous Malay-speaking Malay community, the ‘Malay’ group also included immigrants from various parts of the Malay archipelago who spoke other Austronesian languages such as Javanese, Buginese, Boyanese. These early immigrants were from the Riau Islands, Malacca and Sumatra, and later also included peoples especially from Java and Bawean Island, as well as Sulawesi and other islands. The ‘Malays’ also included the English-speaking Jawi-Peranakans, and a small but economically important group of Arabs (Bloom 1986: 353), with late nineteenth-century community leaders including the wealthy Alsagoff and Alkaff families (Liu 1999: 154). The ‘Malays’ certainly constituted the majority of the population in the first part of the nineteenth century (60% in 1824), but their numerical dominance was swiftly displaced by the growing Chinese population (see point b) and decreased steadily to less than 15% by the 1900s (see Table 2.1).

b. While the Malays initially formed the bulk of the population, the Chinese population started growing swiftly, as can be seen in Table 2.1, and came to form the largest ethnic proportion of Singapore’s total population within two decades of British colonization (45.9% by 1836). The Chinese population continued to grow steadily and rapidly over the century, most dramatically after the 1880s, not least as a result of the establishment of the Chinese Protectorates in the Straits Settlements which controlled labour abuses and freed up the labour markets (Kwok 2000: 201), to reach its current proportion of three-quarters of the population by 1921. The vast majority of the ethnic Chinese immigrants hailed from cities and provinces on the southern coast of China, mainly Xiamen in southern Fujian (the Hokkiens), Chaozhou in the east of Guangdong (the Teochews), and Guangdong itself (the Cantonese), though there was sizeable representation of Hakkas and Hainanese, as well as small numbers from the coastal counties of northern Fujian and ‘Three Rivers People’ from northern and central China. Although the Teochews constituted the largest proportion of Chinese in the early nineteenth century, being twice as numerous as the Hokkiens in 1848 (Kwok 2000: 204), they were however economically weak, being involved primarily in agriculture, and were also perceived to be

24 Lisa Lim more conservative and risk-averse (Li, Saravanan and Ng 1997). The Hokkiens by contrast were a strong economic power in Singapore, especially from the late 1800s, establishing themselves first as traders and go-betweens, and then as importers, exporters, manufacturers and bankers, and virtually monopolizing commercial activities by the end of the nineteenth century. As a result, the Hokkiens became the most powerful bang ‘clan’, and played a leading role within the Singapore Chinese Chamber of Commerce, set up in 1906, as well as within the Chinese community at large (Li et al. 1997). As a consequence, Hokkien was the most frequently understood and spoken Chinese language (note that it is mutually intelligible with Teochew, both being subvarieties of Southern Min), followed by Cantonese and Mandarin, up until the 1970s (Lock 1982: 302), and, more crucially, was the de facto lingua franca for intraethnic communication within the Chinese community (Platt and Weber 1980), which by 1840 comprised half the population (Table 2.1). Cantonese is also suggested to be important in terms of input in the ecology of Singapore and the development of SE (Gupta 1994: 41), as the Cantonese are reported to have taken up English education with more enthusiasm than the Hokkiens (Chia 1977: 160). Mandarin came into the picture only from the 1920s in the Chinese-medium schools, once the Chinese republic was founded (Bloom 1986: 359ff; Kwok 2000), but would still have filled the role of High (H) variety, fulfilling more formal functions, in the diglossic situation (Ferguson 1959; or polyglossic, Platt and Weber 1980) said to be found in Singapore then. c.

A distinct and important group of Chinese were the Straits-born Chinese or Babas (for a detailed socio-historical account, see Ansaldo, Lim and Mufwene 2007; Lim 2010). Descendants of eighteenth/ nineteenth-century south Chinese immigrants and local (Malay or Indonesian) women, largely arriving in Singapore from Malacca and Penang, their vernacular was a contact Malay variety, Baba Malay, restructured to show in particular Hokkien elements in syntax and lexis (see e.g., Lim 1988; Ansaldo and Matthews 1999); they also usually spoke one or more Chinese languages, in particular Hokkien, along with much code-mixing (Pakir 1986, 1989; Lim 1988). Moreover, the Babas were (and are still) also noted, particularly for that time, for having English (increasingly) in their repertoire: they held a high regard for English-medium education and sent their children to English-medium schools. By the mid-nineteenth century, their ability to converse in the colonial language had strengthened their prominent socio-economic position compared to other local communities, to the point where they were in fact sometimes referred to as ‘King’s Chinese’ (Tan 1988: 53). Their English is a particular variety, showing influences from (Baba) Malay, known as Peranakan English (Lim 2010; also,

Migrants and ‘mother tongues’ 25 ‘English, with a Straits accent’, Bloom 1986: 449). Their multilingual repertoire — comprising Baba Malay, Hokkien, and possibly one or two other Chinese languages, plus English — afforded them an important role as intermediaries between Europeans and locals, thus allowing closer contact with British administrators and merchants (Nathan 1922: 77); this together with their business acumen gave them predominance in the commercial sectors (also see Ho and Platt 1993: 8–9). Although they comprised a mere 9.5% of the Chinese population in 1881, their social and economic influence was strong in comparison, and they formed an important sector of the local elite (Kwok 2000: 202, 204), and were seen to be the best educated, wealthiest and most intelligent section of the Chinese community (Nathan 1922: 77). d. Indian contact, trade and migration to Southeast Asia had been occurring since ancient times, including significant influence of the Sri Vijaya empire in the eleventh and twelfth centuries in the region, and the transporting of Dravidian cultural, religious and literary influence to the Malay world (Wignesan 1995). This, however, bears less directly on the influence of South Asian languages in Singapore’s contact dynamics from the nineteenth century. At the founding of Singapore, the first Indians who arrived with Raffles were 120 sepoys, lascars and several assistants (Liu 1999: 82). The migration from the subcontinent that followed involved South Asians who were geographically and linguistically diverse, and represented a number of different castes, each filling a niche in the early days. For example, Ceylonese Tamils tended to work as clerks, junior civil servants and in the professions; Christian Malayalis from Kerala were English-educated and worked mainly in the civil service; Punjabi Sikhs were the backbone of the armed forces and the police force, and worked as private security guards; Tamil Muslims, Sindhis and Gujaratis were often small traders; and the Tamil Chettiar caste from Tamil Nadu were moneylenders and currency changers. The largest group of immigrants were South Indian, mainly from the Tamil, Telugu, and Malayalam language communities, with the Tamils in the majority. A minority were well-educated entrepreneurs and professionals — such as those from Jaffna, many of whom were educated in English and recruited by the British for government and educational services (Bloom 1986: 353) — who would have had some status. The majority were, however, recruited from South India as indentured workers in low-status jobs such as rubber-tapping (Bloom 1986: 353). Overall, given the pattern of migration and settlement, as sketched above, the Indian community was fragmented and dispersed unevenly along various cultural and professional lines. Apart from prominent individuals in the form of Parsi, Tamil and North Indian businessmen, the Indian population had no community leadership and

26 Lisa Lim were divided in background, language and religion (Liu 1999: 82). Furthermore, they represented in total only a small proportion of the population, peaking in the mid-1800s at 16%, but otherwise ranging between 7% to 9% from the 1900s onwards. Their various languages — Tamil, Telugu, Malayalam, Punjabi, Bengali, and so on — were used in the home domains of the respective communities, but not in interethnic communication; nor did any of them really serve as lingua franca between the different Indian groups. e.

By the turn of the nineteenth century, Singapore was a cosmopolitan city, with small groups of Armenians, Jews and Japanese, and Europeans, and growing numbers of Anglo Indians and Anglo Chinese (Liu 1999: 164). The population of Eurasians — a community from Malacca with a complex heritage of Portuguese, British or Dutch, mixed with Indian, Macanese, Malaccan, Burmese, Siamese, and/or Ceylonese origins — was also expanding. By 1931, there were almost three thousand Eurasians, three-quarters of whom were born in Singapore (Liu 1999: 242). These groups spoke a variety of vernaculars, such as Arabic or Malay in the Jewish community in the family domain (Geoffrey Benjamin p.c. cited in Gupta 1994: 41), and Portuguese creole for the Portuguese Eurasians. Soon, however, most of these groups had English as a mother tongue (Liu 1999: 164). Where the Eurasians are concerned (Braga-Blake 1992: 12–13), theirs was a particular Eurasian English (Gupta 1994: 37, 39, 44; Wee 2010).

Thus far we have seen the main Asian communities and their vernacular languages which would have been present — and in some cases, dominant — in the linguistic ecology of Singapore’s colonial era. Two further issues are important to address. First, and very significantly, beyond each community’s own vernacular, there was Bazaar Malay, one of the local forms of restructured Malay, the lingua franca in the region for centuries (Holm 1988; Adelaar and Prentice 1996; Ansaldo 2009a), which certainly served as the interethnic lingua franca in Singapore (Platt and Weber 1980; Bloom 1986: 360). The Chinese, for example, that came to Singapore in the 1820s and 1830s would either have had intimate contacts with the Malay and Thai worlds across several generations of residence in the region, or be already acquainted with the British and Dutch administrations through sojourns in the other trading centres in the area, or would at least have relatives with decades of trading experience (Wang 1991), and would have had Bazaar Malay in their repertoire (also see Ansaldo 2009a). Bazaar Malay in the Chinese and Indian communities has certainly been documented (Khin Khin Aye 2005; Sasi Rekha 2007). Baba Malay, the vernacular of the Straits Chinese mentioned above, has also been equated in some literature to Bazaar Malay

Migrants and ‘mother tongues’ 27 (see Ansaldo, Lim and Mufwene 2007 for details; but cf. Lim 1988 and Ansaldo 2009a who explain why and how they are distinct varieties). The English-speaking Eurasians also spoke Malay (Gupta 1994: 37). Even after the growth of the Chinese population, Bazaar Malay was the second most understood language, after Hokkien, in early 1970s Singapore, and was still the most important language for interethnic communication, with all Indians and 45% of the Chinese claiming to understand it. The second issue is where and how English came into the picture. Most scholars argue that it was ‘exclusively through the schools that English spread’ (Bloom 1986: 348, crediting Platt and Weber 1980 to be the first to note this in print),6 and it is to the domain of the school that we now turn. The school system of the Malay community was most certainly established before Raffles’ arrival, and the Malays were encouraged by the British to attend their vernacular schools (Gupta 1994: 34). Where the Chinese group was concerned, traditional Chinese schools were funded by the community through clans, voluntary associations or philanthropic individuals. Although initially somewhat fragmentary, ‘staffed by teachers from a particular clan, who used their own dialect as the medium of instruction’ (Koh 2006), they developed after the 1911 Revolution in China into high-quality establishments, involving Mandarin as a medium of education. These Malayand Chinese-medium schools did not teach English at all in the early years, though by the 1920s and 1930s, many were teaching English as a subject. Meanwhile, the first English-medium school was established in Singapore in 1834, whose enrolment, while starting small, began to rise gently in accordance with population, towards the last decade of the nineteenth century, and then rose steeply at the turn of the century through to the 1930s. During the nineteenth century, the pupils formed a relatively stable proportion as follows: equal numbers of Europeans/Eurasians and Chinese (each 40%), and smaller numbers of Malays and Indians/Others (each 10%). The enrolment of Chinese pupils increased significantly at the start of the twentieth century to constitute two-thirds of students within twenty years. Where teachers were concerned, until the early 1920s, the largest single racial group of teachers was the Eurasians. Interestingly, their variety of English was noted and frequently deplored in education reports (Gupta 1994: 43). This was followed by roughly equal numbers of Europeans and Indians. In 1935, out of 161 European teachers, there were 12 American, 15 French, and 14 German, Portuguese or Italian; the remaining majority were British, with not few Irish or Scottish (Gupta 1994: 43). Teachers from then-Ceylon (Sri Lanka) and India, recruited during colonial times, were also well represented in the teaching profession (Ho and Platt 1993: 6), though not mentioned in the reports (Gupta 1994: 44): they would have contributed another variety of English, Indian English.7 From the 1920s, the proportion of Chinese teachers rose rapidly, they having been the students the decade before (Gupta 1994: 39–40).

28 Lisa Lim The significant point to note here is this: such a predominance in the early nineteenth century of both Eurasians and Babas as teachers and/ or students in the local English-medium schools would mean that there would also have been extensive use of Malay there (Gupta 1994: 41), since, as mentioned earlier, in addition to English, both groups had a variety of Malay as a dominant language in their repertoire. This is testified in the comments in early reports on the extensive use of Malay in Englishmedium schools, not just outside but also within the classroom (Gupta 1994: 41–2). It should be noted that this would have been Bazaar or Baba Malay — references are to ‘spurious Malay’ (Report 1874, cited in Gupta 1994: 42) and ‘the wretched town Malay generally spoken by the Chinese’ (Kynnersley Report 1902, cited in Gupta 1994: 42) — and not the Malay variety spoken by the Malay community. The presence and dominance of Malay in the schools is evidenced by the fact that even British children in English-medium local schools, who were not few in the nineteenth century, are reported to have been able to speak Bazaar Malay (Gupta 1994: 38). In other words, Bazaar/Baba Malay would have been in significant contact with English, even within the schools. With the presence of the Straitsborn Chinese, there would also have been Hokkien, and a number of other Chinese languages such as Teochew and Cantonese, in particular with the increase in Chinese teachers and students in the twentieth century. After World War II, the British still promoted English-medium schooling among the elite, with about 32% of students enrolled in Englishmedium schools in 1947 (Tickoo 1996: 434), and, by the 1950s, education became effectively universal and English-medium education increasingly the norm. By 1952, 43% of school enrolment was English-medium, with the numbers registering for English-medium education overtaking those for Chinese-medium education by the end of that decade (Doraisamy 1969). The period of free immigration, which lasted over a century, came to an end when the colonial government passed the 1928 Immigration Restriction Ordinance (Yeoh 2007). After a dwindling during the Great Depression of the 1930s, followed by a temporary halt of migration during the Japanese occupation of Singapore (1942–45) during World War II, new immigrants came during the post-war boom years, accompanied by a new immigration ordinance which came into force in 1953 that admitted only those who could contribute to the social and economic development of Singapore (Yeoh 2007). In an attempt to ensure the availability of jobs and a certain standard of living among local residents, the inflow of manual workers was stemmed, while priority was given to those who could contribute specialized services in scarce supply, such as professional and managerial expertise. In this period (1945–65), immigrants came primarily from Peninsular Malaysia, since Singapore and Malaysia were part of the same political entity then (i.e., British Malaya from 1946 to 1948, the Federation of Malaya from

Migrants and ‘mother tongues’ 29 1948 to 1963, and from September 1963, the Federation of Malaysia) which allowed free movement (Yeoh 2007), and, until Singapore separated from Malaysia and achieved independence in 1965, people from the Federation faced no entry restrictions. However, the immigration of the earlier decades which created the pluralist mix that formed the foundation of Singapore’s population to this day was a phenomenon that was not to recur.

The independence era: Population stabilization and language institutionalization When the People’s Action Party (PAP) government came into power — first in 1959 with Singapore self-governed, followed by a brief period of unification with Malaysia, and finally independence in 1965 — and, additionally, with the withdrawal of British military in 1967, economic and socio-political insecurity called for the creating of national unity and forging of national identity and consciousness that transcended ethnic boundaries (Chiew 1983: 45ff). We see this manifested in both the management of the population and the overhaul of the education system (Tickoo 1996: 436). In this period, as a result of the new immigration laws mentioned above, no significant immigration took place from the early 1960s (Gupta 1994: 1), and natural increase was the more dominant contribution to population growth, as opposed to immigration in the previous era. The immigrant communities were becoming more stable, and consequently interacting increasingly outside their dialect groups (Bloom 1986: 359). This was additionally aided by measures in public housing, primarily to improve living conditions, which commenced in the late 1950s: as a consequence of the establishing of required ethnic proportions in each set of housing board flats, this also ensured the end of ethnic enclaves and promoted greater interaction between groups. Where the community organization of the various Chinese dialect groups in earlier decades was along clan lines, this was now achieved via national bodies such as community centres, which fostered instead a more pan-Chinese sense of community. In the realm of education, a number of measures were implemented, one of which was the institutionalization of English as a compulsory language in schools. This was initially either as a first language or, in the vernacular (Chinese-, Malay- or Tamil-medium) schools, as a second language, in the bilingual education system advocated in the 1956 Singapore White Paper on Education. Eventually, English was implemented as the medium of instruction in all schools in 1987. As is by now well recognized, the economic consideration of this move was that a usable competence in English, the language of science and technology, and of international trade and commerce, was seen as a basic need. The political consideration was that English, being a neutral non-native language, not

30 Lisa Lim associated with any of the Asian cultures, and not the mother tongue of any of the ethnic groups, gave none of the ethnic groups an advantage (Kuo 1980: 59ff). At the same time, in order to maintain Asian values, children learnt their ‘Mother Tongue’ (MT) — the official language assigned to the ethnic group that one is categorized as belonging to, the term notably identified with initial capitals — as, eventually, their ‘Second Language’. Thus an English-knowing bilingualism system (Pakir 1991) was established. With regard to the vernacular languages, however, the bilingual education programme was assessed by the Goh Report (Goh et al. 1979) as having failed, as the Chinese dialects were still being used at home. This prompted the annual Speak Mandarin Campaign, launched in 1979, designed to convince Chinese Singaporeans to shift from Chinese ‘dialects’ to Mandarin in all domains (see e.g., Bokhorst-Heng 1998). The four official languages were also implemented in the mass media, again with the suppression of all other non-official languages (see e.g., Bokhorst-Heng 1998). Such aggressive institutionalization of certain languages over others and active implementation of language policies and practices had an immediate and long-reaching impact on Singapore’s linguistic ecology during this period.8 a.

English started displacing Hokkien and Bazaar Malay as lingua franca from the late 1970s to early 1980s, especially among the younger and more educated, with some 70% of Primary 1 children in 1990 having English as a dominant language (Lim and Foley 2004: 5–6). b. Mandarin became the language most frequently spoken at home for the Chinese as a whole, increasing substantially from 10% (1980) to 30.1% (1990) to 45.1% (2000), displacing other Chinese languages (decreasing from 81.4% to 50.3% to 30.7% in the same years). Some 87% of the Chinese population claimed to be able to understand Mandarin by 1988, and it became the language of choice for many younger Chinese Singaporeans’ intraethnic communication in all domains (Lim and Foley 2004: 6).9 c. The three main dialect groups — out of more than twenty Chinese dialect groups in Singapore — are still the Hokkiens, Teochews and Cantonese, who comprised 41.1%, 21% and 15.4% respectively in 2000, making up three quarters of the Singapore Chinese population. In spite of the shift to Mandarin outlined in (b) above, Hokkien and Cantonese were in 2000 still dominant home languages in the Chinese community — ranking as the third and fourth languages most frequently spoken at home in the Chinese community as a whole, after Mandarin and English. d. The use of Bazaar Malay as a lingua franca declined, except in the older generation (see e.g., Khin Khin Aye 2005) and perhaps in the lower social strata. Malay, even while Singapore’s national language since

Migrants and ‘mother tongues’ 31

e.

independence, is really only used within the Malay speech community itself, though in 2000 it was still by far the most frequently spoken language at home (91.5%) for the ethnic Malays, and in some 12% of Indian as well as Other households. Tamil is spoken by a minority of the population. Some 4.2% of the total population are Indians of Tamil-speaking origin, even if they form the greatest proportion, of slightly more than half, within the Indian racial grouping, with the rest of the Indian population being Hindi, Gujarati, Malayalam and Punjabi speakers. Even with its official status, Tamil has continued to be perceived as having little economic value, and seen very much to be a classroom language, with very little functional use elsewhere (Saranavan 1993; Schiffman 2003). It has been replaced largely by English (as well as Malay) in the community in interethnic as well as intraethnic communication, even in the home domain, with a steady shift over the years (52.6% to 43.7% from 1980 to 1990), particularly in the higher classes and the better educated. The situation is similar with the other Indian languages. (See also Vaish et al. in this volume.)

As is evident from the figures above, this is the era where the official languages which represent the official races in Singapore (Chinese, Malay, Indian, and ‘Other’) — that is, those associated with ‘higher-order ethnicity’ (Gupta 2001: 5) — attained prominence, and started displacing the other languages which were in fact languages of actual separate dialect groups (‘lower-order ethnicity’, Gupta 2001: 5), and which, poignantly, were essentially the real mother tongues of the population (see Lim 2009c). Overall, there was a large-scale shift from societal multilingualism in a multiplicity of languages to a small number of official languages, or, as suggested by Murray (cited in Bloom 1986: 359), the distribution of languages was no longer to be described in terms of pluralism but rather of heterogeneity.

The late modernity era: Foreign manpower and ‘dialect’ ascendance In the last decades of the twentieth century, to overcome the limits of local resources, Singapore began to nurture a policy of attracting and relying on ‘foreign manpower’ at both the high and low ends of the spectrum, a practice which has directly contributed to an increasing proportion of the nonresident/noncitizen population in recent decades (Yeoh 2007). As is evident in Table 2.2, while non-residents comprised a small 3% of the total population in 1970, from 1980 this started to rise steadily through to 20% in 2001. The average annual growth of non-residents peaked at 9.3% between 1990 and 2000, which contrasts sharply with the growth of

32 Lisa Lim Singapore residents, which never goes beyond 1.8%. In line with population and growth figures, we see foreigners constituting approximately 29% of Singapore’s total labour force in 2000, comprising the highest proportion of foreign workers in Asia. The most rapid increase occurred over the last decade, an increase of 170%, from 248,000 in 1990 to 670,000 in 2006 (Yeoh 2007). Such a significant proportion of foreigners in Singapore clearly does ‘potentially change a nation’s demographics’ and has implications for policy (Wee and Bokhorst-Heng 2005: 159); more crucial to this chapter, it must be taken seriously in the consideration of the balance of languages in Singapore’s linguistic ecology and the impact on the evolution of varieties such as Singapore English. We can differentiate between two objects of interest in this respect, namely the status and impact upon Singapore’s linguistic ecology of two types of foreign manpower: foreign workers (lower skilled) and foreign talent (highly skilled). Table 2.2 Population and annual growth (from Tan 2002: 2; Lee and Yeo 2003: 10) Year

Number (’000)

Total pop.

S’pore residents

1970

2,074.5

2,013.6

1980

2,413.9

1990

3,047.1

2000

Average Annual growth (%) (growth over previous decade; for 2001 and 2002, growth over previous year) Nonresidents

Total pop.

S’pore residents

Nonresidents

60.9

2.8

n.a.

n.a.

2,282.1

131.8

1.5

1.3

8.0

2,735.9

311.3

2.3

1.7

9.0

4,017.7

3,263.2

754.5

2.8

1.8

9.3

2001

4,131.2

3,319.1

812.1

2.8

1.7

7.6

2002

4,171.3

3,378.3

793.0

1.0

1.8

-2.4

(a) The lower skilled (foreign workers). In the affluence of recent decades, because Singaporeans have been reluctant to fill low-skilled low-wage jobs, the country has had to turn to foreign workers to fill such positions; these workers came largely from Malaysia, Thailand, the Philippines, Sri Lanka and India. However, believing that too much permanent, low-skilled migration is disruptive to society, the government has imposed an immigration policy since the 1970s which maintains low- and unskilled migrants as a transient workforce (Yeoh 2007). This is achieved via a series of measures, such as the regulation of the proportion of foreign to local workers, and the work-permit system, by which foreign workers are only allowed to work for the employer and in the occupation as reflected in the work permit, with

Migrants and ‘mother tongues’ 33 employment contracts of a maximum of two years (subject to a oneoff renewal), and therefore cannot gain access to the local labour market. Foreign workers are also subject to repatriation during periods of economic downturn, or in the case of females, upon becoming pregnant. Furthermore, they are not allowed to bring their spouses and children with them, nor can they marry Singaporeans or permanent residents (PRs). Consequently, while foreign workers have formed a substantial proportion of the population — approximately 86.5% of the non-resident workforce in 2006 (580,000 out of 670,000) — their languages are in fact not likely to be dominant in Singapore’s ecology for a number of reasons: (i) foreign workers are mainly concentrated in the construction industry, in domestic maid services, and in service, manufacturing and marine industries, and by definition hold a lower status in the country; (ii) as mentioned above, they are not permitted to marry Singaporeans or PRs, nor do they have their own families with them to engage in a ‘normal’ lifestyle. They tend to live in districts or enclaves for foreign workers, and are thus relatively segregated, with little or no intense or intimate contact with SE speakers. One exception might be the community of Filipina, Indonesian and Sri Lankan domestic workers who do have intensive and intimate contact with Singaporeans — one in seven households employs a foreign maid (Piper 2005); more crucially, they are in close contact with children, being, in many cases, the primary caregiver of children from their birth or early years. While it has been suggested (e.g., Yip and Matthews 2007 for the situation in Hong Kong) that their languages or variety of English would not have a significant influence in the ecology, since the community would be deemed of low status and little prestige (and we can think of the Teochews, for example, in the first era), substantial research has yet to be conducted on this question though it has indeed been recognized (e.g., Lim 2009d). It has been noted nonetheless that the Filipina domestic workers, in particular, who tend to have a reasonable proficiency in English and speak Philippine English, often serve as English-language teaching auxiliaries to the children of middle-class families (McArthur 2002). (b) The highly skilled (foreign talent). Given its aspirations to become a major player in a globalized world, Singapore’s main economic strategy has been to invest heavily in information technology and human capital, focusing on developing Singapore into the ‘talent capital’ of the global economy. To this end, immigration policies have been liberalized for some: skilled immigrants are allowed to bring their family members and receive permanent residency or citizenship more easily, and various programmes aimed at attracting talent have been launched, such as

34 Lisa Lim company grant schemes to ease costs of employing foreign skilled labour, and recruitment missions by government agencies (Yeoh 2007). This sector of foreign labour — professional and managerial workers, often working for multinational corporations — usually referred to as ‘foreign talent’ in both government and public discourse, accounted for 13.4% (about 90,000) of Singapore’s total non-resident population in 2006. While traditionally from the United States, Britain, France, Australia, Japan and South Korea, recent years have seen a majority of them coming from China, India and Malaysia, due to policies instituted in the 1990s targeting the highly skilled in non-traditional source countries (Yeoh 2007). A particular group of immigrants that warrants special mention in this respect are the Chinese from China and Hong Kong, whose immigration was encouraged in the 1980s and 1990s in order to widen the talent pool. In 1989 in particular, Singapore mounted a campaign to attract skilled professionals from Hong Kong, offering a Chinese cultural environment with lower living costs, and accepted 25,000 individuals from Hong Kong as permanent residents (translating in reality to potentially 100,000, if one includes in the calculation the families of these individuals, Anne Pakir, p.c. 2007), as well as an undisclosed number from China (Kwok 2000: 201). In an opposite trend, significant emigration occurred in the late 1980s. Between July 1987 and June 1988, some 4,200 Singaporeans emigrated to Australia, New Zealand, Canada and the USA. This emigration was of culturally marginalized English-speaking minority communities, mostly middle-class, including Indians, Eurasians, Peranakans, a large proportion of them universityeducated professionals, and is seen to be the result of the bilingual policy and the particular emphasis on Mandarin, favouring the Chinese community. The 1990s also saw the immigration of a fairly large expatriate Indian community of well-educated and wealthy professional and business people; interestingly — and perhaps significant to local ecology — the interaction between the local and expatriate Indian communities is noted to be ambivalent rather than natural. Clearly, these patterns affect the balance of proportions of the population and subsequently the relative dominance of relevant languages. Where language policies are concerned, while the official languages are still upheld in official discourse and education, a phenomenon involving increased prominence of previously non-sanctioned linguistic varieties may be observed. Five Non-Tamil-Mother-Tongues (NTMTs), namely Bengali, Gujarati, Hindi, Punjabi and Urdu, were included in the school system in the 1990s, though it may be noted that these are largely community- and not state-funded. In spite of the continuing official discourse on Mandarin

Migrants and ‘mother tongues’ 35 versus the other Chinese languages, a relaxation in the domains of use of Chinese ‘dialects’ has occurred: in the lead-up to the general elections in 2001 and 2006, for example, ministers of the ruling political party, including the then prime minister, gave election rally speeches not only in the official languages (English, Mandarin, etc.) but also in Hokkien, Teochew, and Cantonese, to better connect with their voters. In parallel, Singlish — while still played against ‘good English’ in official discourse, such as the Speak Good English Movement — has similarly slipped through the cracks to figure in supposed H domains. During the SARS epidemic of 2003, in a bid to promote healthy practices to prevent the spread of SARS, the popular local sitcom character Phua Chu Kang — who in 2000 had been reprimanded for speaking Singlish and in subsequent episodes had to attend grammar classes to improve his English — performed in Singlish in a music video, ‘The SAR-vivor Rap’, released by the Ministry of Health and endorsed by the government. (See e.g., Lim 2009c for an account of such uses of the Chinese ‘dialects’ and Singlish.) For this era then, the dominance of various languages in the ecology is evaluated as follows: 1. Mandarin is still the language most frequently spoken at home for the Chinese as a whole, compared to other Chinese languages (45.1% vs. 30.7% in 2000). 2. In spite of the shift to Mandarin outlined in (1) above, the other Chinese varieties Hokkien and Cantonese are still fairly dominant in the Chinese community, ranking as the third and fourth most frequently spoken language at home, after Mandarin and English, in the Chinese community as a whole. However, what is more significant and revealing is that, when each dialect group is examined separately (see Lee 2001) — a more appropriate practice in such an investigation — Cantonese is in fact still spoken by the Cantonese at home more frequently than Mandarin (36% vs. 32% in 2000), even if there is an overall decrease in usage (from 51.5% Cantonese and 20.2% Mandarin in 1990). This contrasts with both the Hokkiens and Teochews who speak more Mandarin (46.3% and 43.4%) than their own language (29% and 25.7%) at home in 2000 than in 1990.

A new world order: Betting on work and play, and regional language recognition This fourth era, I argue, has begun — we may set the start date around the end of the old millennium. We may view it as the continuation and expansion of the patterns of the third era, but I suggest that there is a qualitatively distinct thrust in immigration policy, which is in the process of gathering momentum.

36 Lisa Lim Singapore of the twenty-first century is generally recognized (e.g., Arnold 2007; Kingsbury 2007) as facing challenging economic prospects — as a consequence of competition from low-cost countries such as China in high-tech manufacturing jobs, once crucial to economic growth — as well as shrinking population, due to numerous younger Singaporeans seeking employment overseas and others having fewer children. These two trends mean that the population is set to shrink in 2020, which subsequently means stagnating economic growth and a declining standard of living. The government’s solution is to boost the population by 25% to 6.5 million over the next few decades, through a radical increase in the foreign population. The prime minister has been urging Singaporeans to ‘change our mindset towards foreign talent’ (Goh 2000) in his National Day Rally speeches since 2000, with the increase in population to be achieved in two main thrusts. The first is by increasing what may be seen as the ‘fun factor’. Since drawing in such a large number of high-income foreign talent requires more than being one of the best places to live, which Singapore regularly accomplishes in regional surveys (e.g., international human resource company Mercer 2006 in their Quality of Living Survey ranked Singapore as the most attractive Asian city for expatriates to live, work and play in), in terms of efficient government, first-world infrastructure, solid educational system, and clean crime-free streets. This additional ‘buzz’ is being realized, among other plans, by the following: a. nightclubs being allowed to be kept open 24 hours from 2002; b. gambling being legalized in 2005, and two integrated resorts (IRs) (casino/resort developments) completed by the end of this century; c. the running of the Formula One Grand Prix circuit — and its first night race — from 2008; d. the promotion of private banking from 2002 — which has resulted in nearly 40 private banks having regional operations in Singapore — through bolstering banking secrecy laws (already in 2001), offering generous tax incentives, and modifying its trust laws that guarantee the right of trust holders to determine who inherits the estate — the last especially attractive to clients from Europe and the Middle East (Arnold 2007); e. the establishing of economic partnerships, involving attractive tax treaties, with Qatar, Bahrain, Egypt, Israel, Kuwait and Oman, numbering 56 at present (Bakhda 2007); f. the passing of a new property law in 2006 allowing land on Sentosa Cove (see next point) to be owned by foreign individuals without special government clearance; g. the development of Sentosa Cove, Singapore’s first waterfront property development of 2,500 luxury homes, positioned as an ‘international resort community’, targeted at not just wealthy locals but also expatriate residents and overseas investors; completed in 2010, some 60% of the

Migrants and ‘mother tongues’ 37 buyers so far are foreigners, including those from Europe, Malaysia, Hong Kong, Indonesia and Japan (Channel News Asia 2006; Arnold 2007); h. the development and sale of other luxury residential projects in the city, whose buyers, in at least one case, comprise three-quarters from Europe, the USA and the Middle East; for prime districts landed homes, the main buyers were UK nationals, Australians, Americans, Malaysians and Indians (Business Times 2007). In this sense, immigration policy has come full circle, with the thrust in this current era being, as declared by Prime Minister Lee Hsien Loong, ‘to promote immigration into Singapore’ (Lee 2006). The difference from the first era lies in the composition of the communities, and thus their languages, which consequently make for a different composition in the feature pool. The second thrust is the goal of establishing Singapore as an education hub, focusing on its strengths, including its English-speaking environment and high educational standards as well as its reputation for public order and safety. World-class educational institutions such as INSEAD and Johns Hopkins University have already established Singapore campuses for some years now. International students comprise this third group of foreigners (the other two being foreign workers and foreign talent) which has indeed been growing in importance. While Singapore has long attracted foreign students from Malaysia and Indonesia, since 1997 the country has made specific efforts to develop Singapore into an international education hub for primary- to university-level students (Yeoh 2007), promoting Singapore as ‘The Global Schoolhouse’ of Asian school systems with Western-styled education practices. In 2005, 66,000 foreign students (amounting to about 10% of all students in the country) came to Singapore, with the target, as recommended by a government economic review panel, of 150,000 foreign students by 2012, being more than double the 2005 figure. This is estimated to not only create 22,000 jobs but also raise the education sector’s contribution to the gross domestic product from the current 1.9% (S$3 billion/US$1.9 billion) to 5% (Yeoh 2007). Since Asian students are expected to dominate the increasing global demand for international higher education, the main targeted markets include China and India, as well as neighbouring Southeast Asian countries. It is estimated that there are now approximately 36,000 students from mainland China studying in Singapore’s local schools, an increase from 32,000 in 2006 (Toh 2008), making up half of the foreign students in Singapore. What is also noteworthy is that often accompanying the students are their mothers — known locally as study mamas or peidu (Mandarin ‘accompany study’) mamas — who, estimated at 5,000, have in many cases significant contact with the local population, working as salespersons, tutors, workers in local

38 Lisa Lim coffeeshops, cleaners, and, in the sex trade (Toh 2008). The possibility of (good Indian) education is also said to be a draw, even the factor for choosing to relocate to Singapore, for the professional expatriate Indian population (Global Indian International School 2008), which has also been growing, as evidenced by the rapid increase in enrolment in the Global Indian International School (from 48 students in 2002 to more than 4,000 in three campuses four years on) and DPS International (increasing from 169 students in 2004 to more than 1,700) (Ng and Sengupta 2008). To this end, success may already be noted. Singapore’s population saw a significant increase in June 2008, the biggest annual spike since the collection of such data in 1837 (as in Table 2.1). Notably, the increase is not due to citizens’ birth rate (rising by 1%) but primarily due to foreigners (increasing by 19% to 1.2 million), with the increase in new citizens and permanent residents increasing by about 25% (Li 2008). More specifically, the number of ‘new’ migrants from China is estimated to be close to 100,000 (Chan 2006: 9), and Indian nationals in Singapore number some 200,000 (Ng and Sengupta 2008); the proportion of Indians in the population has in fact increased to 8.9% (Li 2008), a level the group has not had for the past half century (see Table 2.1 for comparison). Where language policies are concerned, we see continued support for the official languages, but also new emphases. In addition to the traditional choices of French, German or Japanese as a third language (i.e., a foreign language) to be studied in school, another trio has recently been introduced as third language choices — this time of regional languages Mandarin, Malay/Indonesian and Arabic — an incentive to do so being the awarding of two additional bonus points for university entry (announced during the prime minister’s National Day Rally 2007). The continuing emphasis on Mandarin is not surprising, given the continuing growth of China’s economy. Mandarin has for some years now been seen as most instrumental of all the Mother Tongues offered in education, with other ethnic groups wanting to study it as a second language rather than their own Mother Tongue (see e.g., Wee 2003; Wee and Bokhorst-Heng 2005). While the argument for the instrumentalism of the other Mother Tongues is suggested to be doubtful (Wee 2003; Wee and Bokhorst-Heng 2005), the option of Malay/Indonesian may be traced back to the Indian Ocean tsunami and other earthquakes in the Indonesian archipelago from 2004 through 2006, when Singaporean aid teams found it difficult to communicate with the populations in the affected neighbouring countries, in this case, Indonesia. Singapore’s government then stated that measures in the education policies would be implemented to bridge the ‘Bahasa gap’, with Minister Mentor Lee Kuan Yew expressing a desire for 10–15% of non-Malays to learn Malay/Indonesian as a third language, with the Malay language curriculum to be made more convenient and accessible.

Migrants and ‘mother tongues’ 39 On the basis of what we see unfolding in the fourth era in terms of neoimmigration and language policy, we may predict the following scenario. There are languages which will continue the dominance that they have already seen in the third age: a. Mandarin, for its continuing status as the most instrumental of the Mother Tongues and its position as a language that can be studied by non-Chinese students, and as a result of the growing proportion of Mandarin-speaking new immigrants who have been targeted and who have arrived in Singapore in recent years and will continue to grow, due to the luxury lifestyle and education that Singapore offers its foreign talent; what is notable here is the fact that the new immigrants come into everyday contact and interaction with the local population; b. Cantonese, as a result of increased presence of Hong Kong population for business and residential purposes. Then there are the languages which (potentially) come into prominence in this era: c. Malay/Indonesian, as a consequence of the increased presence of Indonesians, for business, education and residence, as well as its position as one of the official third languages to be studied by nonMalay students; d. Arabic, as a result of the various government-level agreements that have come into being during this period, as well as residential possibilities, both of which have attracted and continue to attract a substantial population of Middle East individuals and families; this is also supported by Arabic being made one of the new official third languages available to students; e. while the increased presence of Indians is surely significant in terms of numbers and prestige, the significant point to note is that these communities do not appear to have as much contact with the local population (cf. the Chinese immigrants); for instance, they send their children not to local schools but global Indian ones; furthermore, the community tends to be (Indian) English-speaking; their Indian languages are thus assumed not to have such an influence in the ecology as a result of this. (See also Bruthiaux 2009 on the potential influence of Indian English in the region.) To sum up the preceding discussion, we have thus far surveyed the patterns of immigration over time as well as the language policies implemented, using them to identify four socio-historical eras which are distinguishable in terms of their linguistic ecology, that is, the languages which may be seen to be relatively more dominant in the various eras. This is summarized in Table 2.3, which has been adapted and developed further from Lim (2007a).

I

Era

early 1900s

before early 1900s

• Free immigration: immigrants from southern China, India, Malay archipelago • Rapid increase in Chinese population

• Indian Ocean trading patterns

pre-1800s Malay Sultanate

1819–1965 British colony

Immigration patterns

Time period and historical circumstances

• sharp rise in Chinese students in Englishmedium schools • dramatic increase in Chinese teachers in English-medium schools (c.1927) • Mandarin replaces other languages in Chinesemedium schools (c.1912)

• first English-medium school (1834) • English-medium schools: majority of Eurasian teachers, smaller equal numbers of European and Indian teachers; majority of Eurasian and Straits-born Chinese students; extensive use of contact Malay varieties • Chinese-medium schools use other Chinese languages, not Mandarin

Language policies

• contact varieties of Malay are main lingua franca • Hokkien also serves as lingua franca • numerous southern Chinese, Malay/ Indonesian and South Asian languages spoken by respective communities • English spoken by communities such as Eurasians, Armenians, Ceylonese, Babas, British, Americans, Europeans

Language situation

Table 2.3 Some landmarks in the various eras of immigration patterns and policy implementation in Singapore

40 Lisa Lim

1965 leading to independence and after

c.1980–2000

c. 2000 and after

II

III

IV

• New immigration: banking, integrated resorts, education

• New immigration: foreign talent, e.g., Cantonese

• Controlled immigration of foreign manpower: foreign workers and foreign talent

4 official languages (1956) English as compulsory school language (1965) Speak Mandarin Campaign (1979) all schools English-medium, ‘mother tongue’ taught as L2 (1987)

• Mandarin, Malay/Indonesian and Arabic as third languages

• • • •

• Cantonese used more than Mandarin in Cantonese homes

• English sees increase in literacy and use as dominant home language, starts becoming main interethnic lingua franca especially in younger generation • Mandarin becomes most frequently used home language for Chinese, and preferred intraethnic lingua franca for Chinese, especially in younger generation

Migrants and ‘mother tongues’ 41

42 Lisa Lim

Linguistic ages in the dynamic ecology of Singapore The preceding section has provided an account of immigration patterns and language policies over the decades, which allows us to distinguish distinct eras of characteristic practices. In an ecological perspective, these practices are recognized as significant factors in a community’s external ecology, which allow us to define correspondingly a number of linguistic ages, i.e., periods in which a (number of) language(s) can be seen to be dominant in the ecology.10 These linguistic ages in turn provide a structured view of dominant languages in the ecology, and thus help to account for the influences of different languages at different times on the evolution of a contact variety such as Singapore English. As the main aim of this chapter is the account of the external factors of the ecology, this section will present only a very brief illustration of the influence of different languages in different periods. 11 At the outset, it should be borne in mind that because Bazaar Malay and southern Sinitic languages show a substantial degree of typological similarity in terms of morphosyntax (e.g., in features such as zero copular, predicative adjectives, topic prominence, aspect rather than tense systems, and so on; see e.g., Platt and Weber 1980; Ansaldo 2004, 2009a, 2009b), it is at best challenging (Ansaldo and Matthews 1999; Ansaldo 2004, 2009a), if not completely unrealistic from a historical linguistics perspective (Umberto Ansaldo p.c. April 2009), to identify which of these substrates is the sole or main source of these SE features. In what follows then, for simplicity’s sake, lexical items and discourse particles are used as illustration.

The age of the immigrant languages As already sketched above, in the first socio-historical era, the natural patterns of immigration and settlement, including, crucially, the relative proportions and status of the various ethnic and linguistic groups, result in Bazaar Malay and Hokkien being prominent languages. The former is the interethnic lingua franca not just in Singapore but in the region, for centuries, and is possibly reinforced by the prominence of the Baba Malay-speaking Straits Chinese. The latter is the main language of the Chinese population due to the Hokkiens being economically and socially most powerful. This is coupled with educational and language policies that afforded in schools some leeway in the language of education, and in society the possibility of vernacular maintenance. This is thus the age of the original immigrant languages. At the same time, English, introduced via English-medium schools as well as spoken as mother tongues of some ethnic groups, starts spreading through the population, and starts

Migrants and ‘mother tongues’ 43 undergoing nativization (Schneider 2003, 2007) through contact with the other languages. We can find quite clear indication of which languages must have been dominant in the feature pool in this age in three areas. The first is found in SE lexical items that are derived from the background languages: for example, of the fourteen words or expressions identified as typical (Platt and Weber 1980: 83–5), eleven are from Malay, two from Portuguese, and one from Hokkien, quite a convincing indication of the dominance of (Bazaar) Malay in the era up to the 1970s. The second area in which an indication of language influence is quite clear comprises the SE discourse particles. The only particles documented in scholarship during this period are lah, ah, and what, with the former two demonstrated to be from Hokkien and/or Bazaar Malay (Platt and Weber 1980: 76–7; Lim 2007a). A third area is the influence of Bazaar/Baba Malay prosody on the prosody of SE. Unlike Sinitic-dominant Hong Kong English, in which high tones are located on stressed syllables and low tones on unstressed syllables, SE instead shows patterns of pitch prominence at word- and phrase-final positions, quite likely reflecting such prosodic patterns found in Malay varieties (Lim 2009b, forthcoming).

The age of the official languages In the second socio-historical era starting from independence, the emphasis is on the forging of a stable and coherent national identity of the resident population through the institutionalization of official languages to represent each of the three main ethnic groups, manifested in the bilingual education policy and reinforced by the Speak Mandarin Campaign. The result is the rise in prominence of Mandarin over all the other Chinese languages, as well as over the other two ‘Mother Tongues’, i.e., Mandarin becomes dominant in the ecology. English too increases in use as the language for interethnic communication, replacing Bazaar Malay and Hokkien, especially in the younger and more educated generation. This is thus the age of the official languages. As explained in Lim (2007a: 467), the role of this age is really dual. On the one hand, it sees ‘the damping of the original, once-prominent substrates: where Bazaar Malay and Hokkien are replaced as languages of inter- and intraethnic interaction’ (2007a: 467). Further, with (Singapore) English having become a language with increasing numbers of native speakers and speakers for whom it is a dominant language, this age also means that the features which had started to develop in SE now see frequency of usage which leads to ‘crystallization and consolidation’ (2007a: 467).

44 Lisa Lim

The age of the global-media languages The third era of late modernity, starting from around the late 1980s, has, as its more important force, the immigration of foreign talent, as detailed above. One of the significant immigrant groups in this period is that of Hong Kong Cantonese; this brings about a paradigm shift (Lim 2007a) in the Chinese population not only in terms of the demography of immigration patterns, but subsequently of language use in the home (and elsewhere), such that Cantonese is more dominant in the external ecology in terms of relative number of speakers and amount of usage. Bloom (1986: 397) also suggests that the identification with Hong Kong with its especially modern, vital and fun brand of ‘Chineseness’ could contribute to a confident sense of ‘Chineseness’ that is not threatened by English education, resulting in a preponderance of Cantonese children in the English stream (already in the 1970s) and their overall success in English. Coincident with this change in population make-up are the golden years of Cantopop (Wong 2003) and Cantonese cinema (Teo 1997; Bordwell 2000). Bloom (1986: 397) also notes Hong Kong’s ‘lively and democratic cultural life’, in contrast to the ‘drab ideological products of China and Taiwan [that] are in Mandarin’. As argued in Lim (2007a: 457ff), it is important to recognize the significance of the media and pop culture in the linguistic ecology, which in this case contributes to raising the profile of Cantonese in the diaspora. This can be seen as dominance of Cantonese in terms of cultural prestige. Finally, Cantonese dominance can also be seen in terms of the internal ecology, where typologically it is rich in terms of tone and particles. This is thus conceived as the age of the global-media languages. Turning to SE’s discourse particles once more, we see evidence for the selection of features from Cantonese during this era. While the particles lah and ah have been present in SE since the earliest descriptions, a larger group of particles only start being documented from the late 1980s, which include hor, leh, lor, ma, meh. These particles are demonstrated to show exact matches, not only in their segmental form and meaning but most crucially in their tone, with the same particle as in Cantonese itself (Lim 2007a, 2009b). In contrast, while similarities can be seen with Hokkien and/or Mandarin particles at least partially in segmental form and meaning, the neutral or weak tone that the particles have in these languages cannot account for the full tone in the SE particles. This latter fact comprises quite convincing evidence that these are Cantonese particles, acquired in a large set (see Lim 2007a for a full account), and subsequently reinforces the account of the dominance of Cantonese in the feature pool in this period.

Migrants and ‘mother tongues’ 45

The age of the regional languages Finally, the fourth socio-historical era can perhaps be seen as the third era writ large, where there is an even greater impetus on attracting foreign talent, but of a different, more regional origin, in particular from China and India. In addition, language policies serve to support regional languages Mandarin, Malay/Indonesian and Arabic. What appears in this section is naturally speculative. However, based on what we have seen, rather convincingly, in the previous eras, how a language that becomes dominant in a particular period in the linguistic ecology exerts significant influence on the evolution of a developing variety in terms of some structural features, the speculation offered here, I believe, should be quite credible. There is a potential shift towards a renewed, increased dominance in the Chinese languages in Singapore’s ecology, in particular Mandarin and Cantonese, and this is perhaps a more effective shift, where, for example, Mandarin is not just a top-down imposed ‘Mother Tongue’ but one that is perceived in reality as desirable, and where there is now a real presence of Mandarin speakers. This implies a potential in their influencing the structural features of Singapore English in the future. For instance, there is the likelihood of the development of a more tone-language prosody. This is a phenomenon that has only recently been observed and documented (e.g., Lim 2008a, 2008b, 2009b, forthcoming), and may be viewed as a possible result of the current and future dominance of the Chinese languages. Such dominance would be both in the external ecology, in terms of the increased proportion of speakers, and their increased prestige, as well as in the internal ecology, since the proportion of Chinese languages in the ecology also means typological dominance in the feature pool which predicts greater likelihood of selection of features. Another possibility is a renewed influence of Malay on SE. The presence of (Bazaar/Baba) Malay in Singapore’s ecology from the beginning, in particular its dominance in the first age, has already been suggested to have had a significant and persistent influence on SE features, as predicted by the founder principle (see Mufwene 2001) and seen in certain prosodic patterns (Lim 2008a, 2009b, 2009d, forthcoming), also briefly mentioned above. The resurgence of the Malay/Indonesian language, for not only the traditional Malay speakers in the population, which may mean an increase in prestige and dominance in the feature pool, may very well serve to reinforce such influences. Based on the analysis for the fourth era, we now reinforce what was tentatively suggested in Lim (2007a: 469), in the form of a fourth age, which takes the framework from the present into the future:

46 Lisa Lim I.

The age of the original immigrant substrates, starting from pre-colonial centuries (pre-1800) through post-independence years (mid-1970s) II. The age of the official languages, from the mid-1970s up until the present III. The age of the global-media languages, beginning around the late 1980s through to the present IV. The age of the regional languages, beginning around the new millennium through to the present Figure 2.1 (adapted and developed from Lim 2007a: 458) summarizes what has thus far been outlined, with regard to the significance in the different ages that the various languages have in the Singapore speech community as a whole and in the emergence of Singapore English. What is conceptualized as the third age, that is, the age of the global-media languages, has begun, even if the second age, that of the official languages still continues. Similarly, the fourth age may be seen to be overlaid on the second and third ages. This is not a contradiction in the model, but may be seen as coherent if we understand the practices in a community as being negotiable at different levels, or as different linguistic markets, e.g., a national one (in the second age) and, at least in some respects, a more global one (in the third age). This does not only mean that the primary language players and practices are concurrent and layered, but more significantly, it also implies constant competition within the ecology. Figure 2.1 Representation of the relative significance of languages in the different ages of linguistic history in Singapore Languages English

------- - - - - - - - - - - - - - - - - - - - - - -===============================================

(Bazaar) Malay ====================--------------------------- -- -- -- -- -- =============== Hokkien

====================- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Mandarin

- - - - - - - - - ====================================================

Cantonese

------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ==========================

Arabic

-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --============= | The first age

Key: ---- -----====== -- -- --

| The second age

present but not dominant in the ecology dominant in the ecology marginal or waning in the ecology

time | The third age | The fourth age

Migrants and ‘mother tongues’ 47

Concluding thoughts This chapter has focused on two significant forces in the external ecology of Singapore, namely immigration patterns and language policy, which have been shown to lead to the dominance of certain communities and/or their languages, and consequently result in an impact on the development of English in Singapore. This chapter has also demonstrated that, in order to appreciate the evolving nature of Singapore English, both in terms of structure as well as the status and functions it serves in society, it is important to distinguish between different socio-historical eras in Singapore’s ecology, in order to tease out the prominence of different languages in different periods. This is not merely an exercise in intellectual speculation. When combined with an examination of linguistic features in SE and those of the various contact substrate languages, structural similarity echoes the socio-historical hypotheses, and in turn is supported by the very same socio-historical facts. Such a two-pronged investigation has been demonstrated in the analysis, for example, of SE’s discourse particles, to show influence from Baba/Bazaar Malay and/or Hokkien in the first age, and Cantonese in the third age (Lim 2007a). Just as SE has evolved over the decades — from a variety which had dominant input from Baba/Bazaar Malay and/or Hokkien in the first age and which used to be seen as similar to Malaysian English until around the 1980s (e.g., Platt and Weber 1980), to one which saw dominance of Cantonese in the feature pool from the late 1980s and subsequently evolved to acquire elements of that language — I suggest that it will continue in its evolution, and continue to draw on whichever languages there are in its ecology. As sketched at the end of the previous section, the major players present in Singapore’s current ecology would appear to be Mandarin, Cantonese, Malay/Indonesian and Arabic. The linguistic features of these languages are thus present and potentially dominant in the feature pool because the communities that speak them have some numerical, economic or cultural dominance, and/or because the languages are dominant from a typological perspective; e.g., Cantonese is still dominant for its tones and/ or particles; with Mandarin helping to reinforce them; Malay/Indonesian is being reinforced by Bazaar/Baba Malay which has been present and persistent in the feature pool from the earliest age (see Lim 2008a, 2009b, 2009d, forthcoming). What specifically happens in the future with English in Singapore is of course what nobody can predict. After all, as evidenced in all the work on other contact varieties around the world (e.g., see Ansaldo 2009a for examples in Monsoon Asia), ecologies and the dynamics — both internal and external — that go on are complex matters. Moreover, most recent events such as the ‘credit crunch’ and subsequent falls in economic output

48 Lisa Lim and trade across the region must certainly affect the balance of the ecology and the make-up and dominance of the communities and languages therein (Bruthiaux 2009). Nonetheless, what has been presented in this chapter, I believe, provides a systematic framework which helps define more clearly the shifting dominance of languages in such a dynamic multilingual ecology as one that exists in Singapore. This then contributes to the establishment of the more likely substrate sources, and in turn a better appreciation, not only for the evolution of English in Singapore, but also for the social forces that have shaped it.

Notes 1.

2.

3.

4.

5.

I am grateful to colleagues for their comments on my paper given at IAWE in Regensburg (Lim 2007b) in which I explore some of these ideas on ecology, in particular: Salikoko Mufwene for his positive response to my establishing of eras in ecology, as well as his comments on the role of the Peranakans in the evolution of Singapore English; Rajend Mesthrie for his reminder of the significance of the early Indian and Ceylonese English teachers; and Anne Pakir for her more astute knowledge of the Cantonese immigration from Hong Kong in 1989. I also thank Umberto Ansaldo for his novel takes and our constant exchanges on issues of ecology and contact languages of Asia. In previous scholarship on the evolution of Singapore English (SE) and its structural features, the influence of the languages with which SE has come into contact has naturally been considered. For instance, Mandarin has been turned to as a substrate in SE’s relexification process (e.g., Bao 2005); in the examination of reduplication patterns, arguments have been made for Malay and/or Chinese (e.g., Wee 2004) as well as Hokkien (e.g., Ansaldo 2004) as providing the substratal source; discourse particles have been ascribed largely to Southern Chinese varieties (e.g., Gupta 1992). I suggest that a more integrated ecological approach provides a more comprehensive picture of the evolution of a contact variety of English. The notion of ecology has also used previously for the Singapore context (e.g., Gupta 2001), but only as a way of capturing the external environment, i.e., the communities and languages in Singapore. More specific examination of the relative dominance of one or more communities and thus their language(s) was not made in detail, nor was there consideration of the internal ecology and the implications of these for the evolution of SE. This is of course a vast simplification of a competition-and-selection process that is far more complex and takes into account many more factors. Dominance in the external ecology is but one aspect influencing the outcome in the evolution of a new variety. That such an endeavour is crucial in teasing out the precise sources of structural features of a contact language has been demonstrated for, e.g., Sranan (Arends 1989), Hawai‘i Creole English (Roberts 2004), Singapore English (Lim 2007a). For comprehensive and critical surveys of the field, particularly rich in research based on primary sources including census and education material, see in

Migrants and ‘mother tongues’ 49 particular Platt and Weber (1980), Bloom (1986) and Gupta (1994: 32–47); some of the material presented here is also drawn from Lim and Foley (2004) and Lim (2007a). 6. Gupta (1994: 33) also argues for an additional locus of SE development in certain racially mixed, English-focused areas, such as the districts where Eurasians, Jews, Armenians and, in later years, Straits Chinese lived. One of the central Eurasian enclaves of Waterloo Street and Queen Street was also the area where the principal English-medium schools were located. Similarly mixed English-dominant areas also developed in suburban areas later, especially in Katong on the east coast. 7. Their input is said not to have had a significant influence on SE development except for a few aspects of syntax and lexis (Ho and Platt 1993: 8), but given their presence in the education system, as outlined here, this issue perhaps requires further investigation. 8. The source of data pertaining to the specific Chinese dialect groups is the newsletter of the Singapore Department of Statistics (Lee 2001), with all other statistics in this section derived from the Population Census over the years (Ng 1995; Leow 2001) unless otherwise specified. We should note that there are problems with comparability of census results with regard to sampling, self-report, variation in question type, terminology, categorization of ethnic affiliation, acknowledgement of multilingualism, and so on; these are critically evaluated by Bloom (1986: 389ff) and Gupta (1994: 24–32). 9. An exception may be the use of Hokkien and Singlish by young Singaporean males doing their National Service (see Lim, Pakir and Wee in this volume), but there the language choice is motivated by interethnic communication. 10. The linguistic ages for Singapore were first established and outlined in detail in Lim (2007a), with particular reference to its relevance to establishing the ages and origins of SE particles, though the socio-historical eras were not clearly defined there. 11. An account of the first three periods is presented in greater detail in Lim (2007a), and thus a more summarized version of these is presented here.

References Adelaar, Karl Alexander and Prentice, D. J. (1996) Malay: Its history, role and spread. In Atlas of Languages of Intercultural Communication in the Pacific, Asia and the Americas. Edited by Stephen A. Wurm, Peter Mühlhäusler and Darrell T. Tryon. Berlin: Mouton de Gruyter, pp. 673–93. Ansaldo, Umberto (2004) The evolution of Singapore English: Finding the matrix. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam/ Philadelphia: John Benjamins, pp. 127–49. Ansaldo, Umberto (2009a) Contact Languages: Ecology and Evolution in Asia. Cambridge: Cambridge University Press. Ansaldo, Umberto (2009b) The Asian typology of English: Theoretical and methodological considerations. In Special Issue on The Typology of Asian Englishes. Edited by Lisa Lim and Nikolas Gisborne. English World-Wide 30(2), 133–48.

50 Lisa Lim Ansaldo, Umberto, Lim, Lisa and Mufwene, Salikoko S. (2007) The sociolinguistic history of the Peranakans: What it tells us about ‘creolization’. In Deconstructing Creole. Edited by Umberto Ansaldo, Stephen Matthews and Lisa Lim. Amsterdam/Philadelphia: John Benjamins, pp. 203–26. Ansaldo, Umberto and Matthews, Stephen J. (1999) The Minnan substrate and creolization in Baba Malay. Journal of Chinese Linguistics, 27(1), 38–68. Arends, Jacques (1989) Syntactic developments in Sranan. Unpublished PhD dissertation, University of Nijmegen. Arnold, Wayne (2007) In Singapore, a local Switzerland for Asia’s wealthy. International Herald Tribune, 24 April. http://www.iht.com/articles/2007/04/24/ business/singapore.php?page=1. Accessed 3 November 2007. Bakhda, Saish (2007) Singapore — Qatar extending the economic horizon, says Singapore Company Registration firm. Free Press Release Distribution, 29 October. http://www.prlog.org/10035824-singapore-qatar-extending-the-economichorizon-says-singapore-company-registration-firm.html. Accessed 10 November 2007. Bao, Zhiming (2005) The aspectual system of Singapore English and the systemic substratist explanation. Journal of Linguistics, 41(2), 237–67. Bloom, David (1986) The English language and Singapore: A critical survey. In Singapore Studies: Critical Surveys of the Humanities and Social Sciences. Edited by Basant K. Kapur. Singapore: Singapore University Press, pp. 337–452. Bokhorst-Heng, Wendy (1998) Language planning and management in Singapore. In English in New Cultural Contexts: Reflections from Singapore. By Joseph A. Foley, Thiru Kandiah, Bao Zhiming, Anthea F. Gupta, Lubna Alsagoff, Ho Chee Lick, Lionel Wee, Ismail Talib, and Wendy Bokhorst-Heng. Singapore: Oxford University Press, pp. 287–309. Bordwell, David (2000) Planet Hong Kong: Popular Cinema and the Art of Entertainment. Cambridge, MA: Harvard University Press. Bourdieu, Pierre (1984) Distinction: A Social Critique of the Judgement of Taste. London: Routledge & Kegan Paul. Braga-Blake, Myrna (1992) Eurasians in Singapore: An overview. In Singapore Eurasians. Edited by Myrna Braga-Blake. Singapore: Eurasian Association/Times Editions, pp. 11–23. Bruthiaux, Paul (2009) Multilingual Asia: Looking back, looking across, looking forward. In Multilingual, Globalizing Asia: Implications for Policy and Education. Edited by Lisa Lim and Ee-Ling Low. AILA Review 22, 120–30. Business Times (2007) Foreign power has landed in Singapore. The Business Times, 30 March 2007. Available at Singapore Property News Upfront. http:// singaporerealestatenews.blogspot.com/2007/03/singapore-property-newsupfront_29.html. Accessed 10 November 2007. Chan, Brenda (2006) Virtual communities and Chinese national identity. Journal of Chinese Overseas, 2(1), 1–32. Channel News Asia (2006) Sentosa Cove development attracts strong interest from buyers. Channel News Asia, 16 July 2006. Available at http://www. channelnewsasia.com/stories/singaporelocalnews/view/219537/1/.html. Accessed 10 November 2007. Cheng, L. K. (1985) Social Change and the Chinese in Singapore. Singapore: Singapore University Press.

Migrants and ‘mother tongues’ 51 Chia, Siew Hock (1977) An investigation into language use among Secondary Four pupils in Singapore. In The English Language in Singapore. Edited by William Crewe. Singapore: Eastern Universities Press, pp. 157–70. Chiew, Seen Kong (1983) Ethnicity and national integration: The evolution of a multi-ethnic society. In Singapore: Development and Trends. Edited by Peter Chen. Singapore: Oxford University Press, pp. 29–64. Doraisamy, T.R. (ed.) (1969) 150 Years of Education in Singapore. Singapore: Publications Board, Teacher’s Training College. Ferguson, Charles A. (1959) Diglossia. Word, 15, 325–40. Global Indian International School (2008) Education is key consideration for expats when relocating. GIIS Newsletter, vol. 225, 22 August. http://www.giissingapore. org/news/newslink.asp?newsid=402&type=L. Accessed 28 September 2008. Goh, Chok Tong (2000) Transforming Singapore. Prime Minister Goh Chok Tong’s National Day Rally 2000 Speech in English. Singapore: Singapore Government Media Release. http://www.gov.sg/nd/ND00.htm. Accessed 27 September 2008. Goh, Keng Swee and the Education Study Team (1979) Report on the Ministry of Education 1978. Singapore: Ministry of Education. Gupta, Anthea Fraser (1992) The pragmatic particles of Singapore Colloquial English. Journal of Pragmatics, 18, 31–57. Gupta, Anthea Fraser (1994) The Step-Tongue: Children’s English in Singapore. Clevedon/Philadelphia/Adelaide: Multilingual Matters Ltd. Gupta, Anthea Fraser (2001) English in the linguistic ecology of Singapore. Paper presented at the GNEL/MAVEN conference on The Cultural Politics of English as a World Language, Freiburg, Germany, June 2001. Ho, Mian Lian and Platt, John (1993) Dynamics of a Contact Continuum: Singaporean English. Oxford: Claredon Press. Holm, John (1988) Pidgins and Creoles. Vol. II. Cambridge: Cambridge University Press. Khin Khin Aye (2005) Bazaar Malay: History, grammar and contact. Unpublished PhD dissertation, National University of Singapore. Kingsbury, Kathleen (2007) Singapore’s new look. Time, 24 May. http://www.time. com/time/magazine/article/0,9171,1624897,00.html. Accessed 3 November 2007. Koh, Tommy (ed.) (2006) Singapore: The Encyclopaedia. Singapore: Editions Didier Millet. Kuo, Eddie C.Y. (1980) The sociolinguistic situation in Singapore: Unity in diversity. In Language and Society in Singapore. Edited by Evangelos A. Afendras and Eddie C.Y. Kuo. Singapore: Singapore University Press, pp. 39–62. Kwok, Kian Woon (2000) Singapore. In The Encyclopaedia of the Chinese Overseas. Edited by Lynn Pan. Singapore: Chinese Heritage Centre; Cambridge, MA: Harvard University Press, pp. 200–17. Kynnersley, C.W.S. (1902) Report by the Commission of Enquiry into the System of English Education in the Colony. Singapore: Government Printing Office. Lee, Eu Fah Edmond (2001) Profile of the Singapore Chinese dialect groups. Statistics Singapore Newsletter. Singapore: Singapore Department of Statistics, pp. 2–6. Lee, Eu Fah Edmond and Yeo, Yen Fang (2003) Singapore’ demographic trends in 2002. Statistics Singapore Newsletter. September 2003. Singapore: Singapore Department of Statistics, pp. 10–13.

52 Lisa Lim Lee, Hsien Loong (2006) Prime Minister Lee Hsien Loong’s National Day Rally 2006 on Sunday, 20 August 2006, at the University Cultural Centre, National University of Singapore. National Day Rally Videos and Speeches, Prime Minister’s Office. Singapore: Government of Singapore. http://www.pmo.gov. sg/NDRVideosandSpeeches. Accessed 27 September 2008. Leow, Bee Geok (2001) Census Population 2000 Statistical Release 2: Education, Language and Religion. Singapore: Department of Statistics, Ministry of Trade and Industry. Li, Wei, Saravanan, Vanithamani and Ng, Lee Hoon Julie (1997) Language shift in the Teochew community in Singapore: A family domain analysis. Journal of Multilingual and Multicultural Development, 18(5), 364–84. Li, Xueying (2008) Population up a record 5.5%. The Straits Times, 27 September. http://www.straitstimes.com/Prime+News/Story/STIStory_283114. html?sunwMethod=GET. Accessed 27 September 2008. Lim, Lisa (2007a) Mergers and acquisitions: On the ages and origins of Singapore English particles. World Englishes, 26(4), 446–73. Lim, Lisa (2007b) Not just an ‘Outer Circle’, ‘Asian’ English: Singapore English and the significance of ecology. Invited Focus Paper presented at the 13th International Association for World Englishes conference, Regensburg, Germany, 4–6 Oct 2007. Lim, Lisa (2008a) Dynamic linguistic ecologies of Asian Englishes. Asian Englishes, 11(1), 52–5. Lim, Lisa (2008b) English can be tone language meh55? Singapore English what21! Sinitic particles and the hybrid prosody of a contact variety of English. Ms, University of Amsterdam. Lim, Lisa (2009a) Not just an ‘Outer Circle’, ‘Asian’ English: Singapore English and the significance of ecology. In World Englishes: Problems, Properties, Prospects. Edited by Thomas Hoffman and Lucia Siebers. Amsterdam/Philadelphia: John Benjamins, pp. 179–206. Lim, Lisa (2009b) Revisiting English prosody: (Some) New Englishes as tone languages? In Special Issue on The Typology of Asian Englishes. Edited by Lisa Lim and Nikolas Gisborne. English World-Wide 30(2), 219–40. Lim, Lisa (2009c) Beyond fear and loathing in SG: The real mother tongues and language policies in multilingual Singapore. In Multilingual, Globalizing Asia: Implications for Policy and Education. Edited by Lisa Lim and Ee-Ling Low. AILA Review 22, 52–71. Lim, Lisa (2009d) The diversity less considered: On the role of minority groups in Asian Englishes. Paper presented at the 15th International Association for World Englishes conference, Cebu City, The Philippines, 22–24 Oct 2009. Lim, Lisa (2010) Peranakan English in Singapore. In The Lesser-Known Varieties of English. Edited by Daniel Schreier, Peter Trudgill, Edgar W. Schneider, and Jeffrey P. Williams. Cambridge: Cambridge University Press, pp. 327–47. Lim, Lisa (forthcoming) Tone in Singlish: Substrate influences from Sinitic and Malay. In Creoles, Their Substrates, and Language Typology. Edited by Claire Lefebvre. Amsterdam/Philadelphia: John Benjamins. Lim, Lisa and Foley, Joseph A. (2004) English in Singapore and Singapore English: Background and methodology. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam/Philadelphia: John Benjamins, pp. 1–18.

Migrants and ‘mother tongues’ 53 Lim, Lisa, Pakir, Anne and Wee, Lionel (2010) English in Singapore: Policies and prospects. In English in Singapore: Modernity and Management. Edited by Lisa Lim, Anne Pakir and Lionel Wee. Hong Kong: Hong Kong University Press, pp. 3–18. Lim, Sonny (1988) Baba Malay: The language of the ‘Straits-born’ Chinese. In Papers in Western Austronesian Linguistics, no. 3. Edited by Heinrich Steinhauer. Pacific Linguistics A-78. Canberra: The Australian National University, pp. 1–61. Liu, Gretchen (1999) Singapore: A Pictorial History 1819–2000. Singapore: National Heritage Board & Editions Didier Millet. Lock, Graham (1982) The Chinese language in Singapore — Status and features. Language Learning and Communication, 1(3), 233–344. McArthur, Tom (2002) English as an Asian language. ABD, 33(2), 3–4. Mercer (2006) Mercer Singapore Watch — Q3 2006. Mercer. http://www.mercer. com/referencecontent.htm?idContent=1246080. Accessed 27 September 2008. Mufwene, Salikoko S. (2001) The Ecology of Language Evolution. Cambridge: Cambridge University Press. Mufwene, Salikoko S. (2008) Language Evolution: Contact, Competition and Change. New York: Continuum. Nathan, J.E. (1922) The Census of British Malaya 1921. London: Waterlow and Sons. Ng, Jane and Sengupta, Nilanjana (2008) 3rd campus for Indian School. The Straits Times, 28 August. Available at GIIS Newsletter, vol. 228. http://www.giissingapore. org/news/newslink.asp?newsid=405&type=L. Accessed 28 September 2008. Ng, Poey Siong (ed.) (1995) Singapore Facts and Pictures 1995. Singapore: Ministry of Information and the Arts. Pakir, Anne (1986) A linguistic investigation of Baba Malay. Unpublished PhD dissertation, University of Hawai‘i. Pakir, Anne (1989) Linguistic alternants and code selection in Baba Malay. World Englishes, 8(3), 379–88. Pakir, Anne (1991) The range and depth of English-knowing bilinguals in Singapore. World Englishes, 10(2), 167–79. Piper, Nicola (2005) Migrant labor in Southeast Asia. Country study: Singapore. Ms, prepared for the Friedrich Ebert Stiftung project on Migrant Labor in Southeast Asia. Singapore: Asia Research Institute. Platt, John and Weber, Heidi (1980) English in Singapore and Malaysia. Kuala Lumpur: Oxford University Press. Roberts, Sarah J. (2004) The emergence of Hawai‘i Creole English in the early 20th century: The sociohistorical context of creole genesis. Unpublished PhD dissertation, Stanford University. Saravanan, Vanithamani (1993) Language and social identity amongst Tamil-English bilingual speakers in Singapore. In Languages in Contact in a Multilingual Society: Implications for Language Learning and Teaching. Edited by Rosemary Khoo, Ursula Kreher and Ruth Wong. Singapore: SEAMEO Regional Language Centre, pp. 135–60. Sasi Rekha d/o Muthiah (2007) A description of Singapore Indian Malay: A pidgin’s pidgin. Unpublished MA thesis, National University of Singapore. Saw, Swee Hock (1970) Singapore Population in Transition. Philadelphia: University of Pennsylvania Press.

54 Lisa Lim Schiffman, Harold F. (2003) Tongue-tied in Singapore: A language policy for Tamil? Journal of Language, Identity and Education, 2(2), 105–25. Schneider, Edgar W. (2003) The dynamics of New Englishes: From identity construction to dialect birth. Language, 79(2), 233–81. Schneider, Edgar W. (2007) Postcolonial English: Varieties Around the World. Cambridge: Cambridge University Press. Tan, Chee Beng (1988) The Baba of Malacca. Culture and Identity of a Peranakan Community in Malaysia. Petaling Jaya/Selangor: Pelanduk Publications. Tan, Yeow Lip (2002) Current population trends. Statistics Singapore Newsletter. September 2002. Singapore: Singapore Department of Statistics, pp. 2–6. Teo, Stephen (1997) Hong Kong Cinema: The Extra Dimensions. London: British Film Institute. Tickoo, M.L. (1996) Fifty years of English in Singapore: All gains, (a) few losses? In Post-imperial English: Status Change in Former British and American Colonies, 1940– 1990. Edited by Joshua A. Fishman, Andrew W. Conrad and Alma Rubal-Lopez. Berlin/New York: Mouton de Gruyter, pp. 431–56. Toh, Mavis (2008) Tough life for ‘study mamas’. The Straits Times, 28 September. http://www.straitstimes.com/News/Home/Story/STIStory_283550.html. Accessed 28 September 2008. Vaish, Viniti, Tan, Teck Kiang, Bokhorst-Heng, Wendy D., Hogan, David and Kang, Trivina (2010) Language and social capital in Singapore. In English in Singapore: Modernity and Management. Edited by Lisa Lim, Anne Pakir and Lionel Wee. Hong Kong: Hong Kong University Press, pp. 159–80. Wang, Gungwu (1991) China and the Chinese Overseas. Singapore: Times Academic Press. Wee, Lionel (2003) Linguistic instrumentalism in Singapore. Journal of Multilingual and Multcultural Development 24(3), 211–24. Wee, Lionel (2004) Reduplication and discourse particles. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam/Philadelphia: John Benjamins, pp. 105–26. Wee, Lionel (2010) Eurasian Singapore English. In The Lesser Known Varieties of English. Edited by Daniel Schreier, Peter Trudgill, Edgar W. Schneider and Jeffrey P. Williams. Cambridge: Cambridge University Press, pp. 313–26. Wee, Lionel and Bokhorst-Heng, Wendy D. (2005) Language policy and nationalist ideology: Statal narratives in Singapore. Multilingua 24, 159–83. Wignesan, T. (1995) The extent of the influence of Tamil on the Malay language: A comparative study. Paper presented at the VIIIth World Tamil Studies Congress, Tanjavur, India, January 1995. Wong, James (2003) The rise and decline of Cantopop: A study of Hong Kong pop music. Unpublished PhD dissertation, University of Hong Kong. Yeoh, Brenda S.A. (2007) Singapore: Hungry for foreign workers at all skill levels. Migration Information Source. Country Profiles. Washington, DC: Migration Policy Institute, January. http://www.migrationinformation.org/Profiles/display. cfm?ID=570. Accessed 22 September 2007. Yip, Virginia and Matthews, Stephen (2007) The Bilingual Child: Early Development and Language Contact. Cambridge: Cambridge University Press.

Part II

Reconceptualizing ‘English’

3

Singapore Standard English revisited Anthea Fraser Gupta

In the 1980s, Mary Tay and I outlined what we thought were the features of a Singapore Standard English (Tay 1982; Tay and Gupta 1983; Gupta 1986). At the time, our preference for a local (or ‘endonormative’) standard for Singapore English was seen as revolutionary, because the policy then was that the English taught in Singapore should be British Standard English with an RP accent. This was a policy in theory rather than one that could actually be delivered. Delivery of an RP accent was impossible because almost no teachers were (or ever had been) speakers of RP. In any case, it is evident that accents of Standard English are diverse and that to impose a particular foreign accent on a population is unnecessary, unpopular and impossible. It was also tacitly recognized that the English of Singapore needed words to meet the needs of expressing Singapore culture. By the end of the century, as a result of sociolinguistic research, it had become widely accepted that local words and local accents are necessarily part of local Standard English, and that it is neither possible nor desirable to look to a foreign country for all vocabulary, or for an accent. Where grammar is concerned, even in 1986 I realized that ‘St[andar]d S[ingapore] E[nglish] would differ little from general St[andar]d E[nglish], and that by and large its features would exploit possibilities within standard which ... are ... to be found in the spoken English — at times in the written English — of users ... who are thought of as standard users’. No-one has suggested that there are major grammatical differences between regional forms of Standard English. The way in which the minor differences are negotiated continues to be a source of discussion all over the world, and is an area of great concern in the teaching of English as a foreign language, especially in terms of the choice between ‘British’ and ‘American’ English. In these settings, differences are often exaggerated, for example in the form of a false belief that some forms (such as ‘I ate already’) would be wrong in British English, or in the false belief that the present tense is always replaced by the present continuous in Indian English.

58 Anthea Fraser Gupta By the 1990s, school textbooks knowingly incorporated local cultural terms, and there was a generally confident, tolerant, and empowering approach to Singapore English. Books and websites celebrating the local non-standard dialect, Singlish, appeared. In writing, Singlish was confidently used in informal communication and in creative writing, especially in dialogue and humour, in the same way as other non-standard English dialects are used in other parts of the English-using world. Then, in August 1999, speeches were made by Lee Kuan Yew (as senior minister) and the prime minister, Goh Chok Tong, which initiated a drive to promote Standard English in all contexts of use. This led to the establishment of the Speak Good English Movement (SGEM) in Singapore. Its purpose was ostensibly to promote Standard English and discourage the use of Singlish: one of the things it had to do was to interpret the meaning of ‘Standard English’. It soon became clear that the SGEM was promoting a narrow concept of Standard English, which did not allow for anything local, or informal, rejecting even words that had been unproblematically accepted since the 1980s. At official levels, at least, there has been a move back to the attitude of the 1970s, when the local was seen as bad. Once again Singaporeans are being told to look overseas for correction of their English, and are being given advice that is often based on the strictest possible concept of correctness. Even the notion that a Singaporean accent is wrong has been resurrected. Since the 1980s, my own theorization of Standard English has changed too, but towards greater tolerance of variation, not less. I have observed that the differences between what I used to call ‘Singapore Standard English’ and the Standard English of other regions is minimal: the very few differences that have been suggested as distinctively Singaporean (or British, or Australian, or Nigerian ...) are not enough to warrant its being a distinct dialect. Most variation within Standard English is variation due to text type. Standard English includes an element of regional variation, especially in lexis. But when we look at non-standard dialects (Singlish, broad Yorkshire, Jamaican Patwa, etc.) we see that there are grammatical features that distinguish them from each other, and from Standard English. These include substantial differences in inflectional morphology and the construction of the verb (e.g., Merlion kena hit by lightning; He were that punctual; mi done go a di show already). No such grammatical features relating to major linguistic systems have ever been suggested as distinguishing one region’s Standard English from another’s. Important areas of grammar, including inflectional morphology and the structure of the verb, are unified in Standard English across all regions. The division of Standard English into multiple local Standard Englishes is not justified on linguistic grounds.

Singapore Standard English revisited 59 I no longer feel it is appropriate or possible to predetermine what is and what is not standard, and do not accept the methods we adopted in the 1980s. No-one can say what should or should not be Standard English, either in Singapore or anywhere else. Standard English is an organic and changing variety, not something fixed by fiat, and Singapore, like everywhere else that uses English, participates in its maintenance and development. Whatever is part of the Standard English of one region is part of Standard English as a whole. Standard English should be conceptualized as a single dialect that includes a small amount of regional (and a large amount of functional) variation. It is important to understand that this is an inclusive concept of Standard English that places all English-using nations on the same footing and that recognizes that there is (a little) diversity of practice from one place to another. Because Standard English is a living dialect, controlled by unclear consensual processes, there is dispute and negotiation within Standard English. Areas of dispute are generally global, and only a few relate to regional differences. Although it would be foolish to deny that some regions have more clout within the politics of English — the USA has the most, as in other areas — the way power relations work within English are by no means straightforward. Few issues within Standard English are regionally motivated. The concerns expressed about Standard English in Singapore relate to areas of dispute within Standard English everywhere it is used, and not to circumstances peculiar to Singapore. This chapter considers the meaning of Standard English for Singapore in the twenty-first century, and develops the concept of Standard English as a global dialect. I examine the current public discussion of English in Singapore and argue that it shows that Singapore’s concerns about correctness in English are the same as those of the rest of the English-using world.

What is Standard English? Varieties of English can be divided into those which differ little from one place to another and those which are associated with specific regions of the English-using world. The main variety of English globally is Standard English, which is a single dialect of English with a small number of minor regional differences, most of which are lexical. To say that a variety of English is global or international means simply that it is used all over the world. It does not mean that it is used only in communication across national boundaries or between people from different countries. My stance here is different from the more usual stance in the study of English as a world language, and different from the stance I used to take.

60 Anthea Fraser Gupta It is usually argued that English is pluricentric, with a number of local standards. Bex and Watts (1999: 5) say that they are ‘quite clear that the notions of “Standard English” vary from country to country, and not merely in the ways in which such a variety is described but also in the prestige in which it is held and the functions it has developed to perform’. I disagree with these claims. Notions of Standard English vary little from one country to another, and it is held in high prestige and performs similar functions in all the countries in which it has some official status and where it is a widespread medium of instruction in schools. Standard English is a concept that I regard as central to the understanding of English in a social context. Sociolinguists have a choice whether to focus on the shared features and uses of Standard English across the English-using world, or to focus on the differences. There are both political and linguistic issues here. We have spent too much time focusing on the rare features that distinguish the Standard English in one region from that in another, and too little time on the much more widespread features that unite Standard English all over the world. Standard English is a single dialect of English, used all over the world with minor regional differences. The same Standard English dialect is used in most written text types all over the world. All over the world, some text types require Standard English. This applies whether communication is within a country or across national boundaries: Standard English is not used only for speaking and writing to people from another country; it dominates every English-using community internally as well, especially in writing. It is in that sense local to nowhere and local to everywhere. It belongs to the entire English-using world and in it, on the whole, the regional origins of the writers are neutralized.

What are the linguistic features of Standard English? What is and what is not standard is clear in some areas, especially in orthography and inflectional morphology (e.g., when to use am vs. is, did vs. done, teacher vs. teachers vs. teacher’s vs. teachers’; how to build a complex verb group). The concept of Standard English is most applicable to written English. We still do not fully understand the grammar of speech (Carter 2004), but we do know that there is more opportunity to express the self, more room for playfulness, and more regional and personal variation in speech than there is in writing. All of these are expressed in writing too, though mostly through variation within a dialect rather than across dialects. There is no standard accent of English: Standard English can be spoken in any accent. For example, the word dance can be pronounced with a vowel anywhere from a mid front vowel /d‫ܭ‬ns/, through /dæns/, to /dans/, and all points in between to a low back vowel, /d‫ܤ‬ns/, and the

Singapore Standard English revisited 61 vowel can be either short or long. However, some pronunciations, not associated with accent, are generally regarded as incorrect (for example, starting choir with the same sound as chip, or ending picturesque with /kju:/). There are choices in the pronunciation of some words (such as schedule beginning with /‫ݕ‬/ or /sk/). Good pronouncing dictionaries will try to reflect areas of consensus and disagreement in such issues. We may use these — and we may use other people — in order to find out how to pronounce words that we may have seen only in print. Speakers of English will accept correction of word pronunciation in this way, but will not accept correction of their accent. There are substantial differences among accents of English around the world, and it is essential for all speakers of English to understand and be tolerant of accent variation and to develop techniques for understanding unfamiliar accents. Sensitive users of English also need to be tolerant of the ‘mistakes’ we all make, especially in the pronunciation of words we have mainly encountered in writing. English is welcoming to new words, which enter English all the time from other language, from coinages, from informal English, and from other dialects of English. Within Standard English, there are minor differences in vocabulary from one place to another. Most of these differences are local words for local things, so that, for example, the seat of government might be called parliament in one place and senate in another. Such words are used by anyone referring to that institution, regardless of where they come from themselves. Others are what Görlach (1990) calls heteronyms: different words for the same thing (for example, boot/trunk/dicky — the same part of a car). All English users can expect to learn a few new words when they move around the English-using world, or when they refer to other parts of the English-using world than their own. Food names often retain regional associations. For example, ravioli, pirogi, and won ton (all in the Oxford English Dictionary) are similar foodstuffs, but the cuisines of Italy, Poland and China are sufficiently different to mean that English-speaking eaters retain the culturally specific words: and are they all dumplings? HDB estates, council estates, and projects might, similarly, be structurally similar, subsidized and governmentally organized housing schemes, but the way in which social housing is managed and its social and political associations are not the same in Singapore, the UK and the US, and nor are the words. Governmental institutions are one of the areas of language use where local terms have to be used by both insiders and outsiders. But words as a whole are eminently portable and a word regional one year can easily spread across the world next year. I do not see Standard English as being only what every region shares. Standard English includes a small element of regional variation. Boot, dicky and trunk are all part of the single global dialect of Standard English and are best seen as choices within it.

62 Anthea Fraser Gupta Standard English is tolerant of internal variation in pronunciation and lexis, but there is almost no tolerance of variation in inflectional morphology. The structure of the verb group is especially strict in Standard English, where the ordering of auxiliaries (modal, perfective, progressive, passive) and the forms of each word are absolute (e.g., He might have been being pursued by the police). This is one of the most strongly definable aspects of the grammar of Standard English. There are a handful of forms where there is some choice (for example, the plural of appendix can be appendixes or appendices; the past tense of dive can be dove or dived). Such words are rare: most texts would include no words allowing choice. In contrast, dialects of English are distinguished from one another by the way in which they manage inflections, and by the way in which the verb is constructed (for example, the difference between I am happy, I be happy, I happy, mi happy). The areas of choice within Standard English are too small to justify any dialectal divisions within Standard English. Orthography is also very strict in Standard English, with only a small number of words in which there is a choice (e.g., colo(u)r, jail/gaol, adviser/ advisor, café/cafe), some, but not all, of which are the well-known differences between the ‘US’ and ‘British’ traditions, mostly arising from spelling variants of the eighteenth century. Such spellings account for less than 0.5% of words in any given text. This small degree of difference in orthographic practice is not sufficient to justify reference to different spelling systems: there is one spelling system which includes some areas of choice. In speech it is mainly the grammar that distinguishes one dialect from another, but in written English, dialectal differences can also be signalled by using conventional non-standard spellings that relate indirectly to features of the accent that is associated with the dialect: this creates an especially sharp distinction between writing in Standard English (unified orthography and grammar) and writing in another dialect (different spellings and grammar). In addition to ordinary Standard English, there is one set of contextually restricted text types that I would like to exclude from discussion here. There are a set of text types which, all over the world, use an abbreviated form of Standard English in which structural elements (such as BE , first person subjects, articles) are routinely omitted, and in some of which particular abbreviations and alternative spellings may be used (such as l8, GSOH, brb, appt., gonna). Such abbreviated forms of English have always been used where space or time are precious (such as in notes, diaries, postcards, telegrams). In his great grammar of English, Jespersen (1948: 124ff) said that in some of these text types, such as headlines, a sentence is understood by ‘transposing it into ordinary language with its usual grammar’ while others, such as book titles, cannot be said to be sentences. A skilled user of Standard English has to learn when to use this grammar, in order to construct SMS messages, advertisements, and

Singapore Standard English revisited 63 so on. Despite many complaints that sending SMS messages is destroying children’s ability to write correctly, users seem to have little or no problem understanding when abbreviated English is appropriate or acceptable. Communities in which English is used as part of everyday life often have local non-standard dialects of English, which are used in specific contexts (especially informal speech, literature, and humour). Singlish fills this slot in Singapore. The differences between Standard English and the non-standard dialect are generally well known, and users of English are conscious of them. Features of the non-standard dialect seldom appear in written texts where Standard English is the target (Qualifications and Curriculum Authority 1999; Gupta 2006a; Rubdy 2007). This is true of everywhere that has a well-known local dialect, such as the Singlish of Singapore, the Cockney of London, the Geordie of Newcastle upon Tyne, the Patwa of Kingston, and the Nigerian Pidgin English of Lagos. There is no evidence that the presence of a non-standard dialect in a community prevents the learning of skills in Standard English. The majority of the lexis is shared with Standard English. The features of the local dialects are generally salient and at a high level of consciousness and, because young children are sophisticated shifters of style, they do not find it hard to learn which features separate the local dialect from world-wide varieties. Local dialects can be seen as a sign that English is a living language in a place, with a range of cultural and stylistic expression. Even the SGEM pages occasionally use Singlish used in this way: in August 2007 there was a link from the SGEM pages to ‘Sayang Singapore’, the untranslatable Singlish name of Singapore’s storytelling festival (it means something like ‘Love to Singapore’ — sayang is a loan from Malay). There are aspects of Standard English that are more contentious, some of which I will discuss in this chapter. There is some variation within Standard English in certain aspects of English. These include: the way in which the present perfective is used (e.g., This is the first time I have seen/am seeing this play); the use of the definite article (e.g., to have (the) flu; in (the) university); the way in which prepositions are used in some fixed expressions (e.g., comprise (of), different to/from/than); and the way in which the number of certain nouns is treated (e.g., The audience was/were, accommodations). On these features and others like them, there are differences between regions in terms of relative frequency of one form over another, rather than presence or absence (Gupta 2006b). Some of these differences are the subject of public discussion and individuals may have strong feelings about their correctness. The discourse of correctness in English is part of a normative tradition that goes back at least 300 years (discussed in different ways by Leonard 1962; Mittins et al. 1970; Milroy and Milroy 1985; Cameron 1995; Wardaugh 1999). I will use the term ‘disputed usage’ to refer to usages that are found

64 Anthea Fraser Gupta in Standard English texts but that some users regard as incorrect, and whose degree of correctness is a common topic of overt comment. In discussion of English usage, items which are disputed usage are more visible than, and at a higher level of consciousness than, items which are undisputed. For example, users of English in Inner and Outer Circle countries are more likely to be able to articulate that ‘prepositions at the end of sentences are wrong’ (which is disputed usage) than to explain why there is an ‘s’ on the verb in She likes swimming but not in I like swimming (a basic feature of Standard English concord about which there is no dispute). Linguists have engaged with the normative tradition for many decades. One of the basic tenets of linguistics (stemming from desperate reiterations by Jespersen in introductions to various volumes of his grammar) is that ‘linguistics is descriptive not prescriptive’, from which it follows that linguists should not be concerned with telling speakers what is right and what is wrong, but rather should examine what speakers do and then infer the rules of the language from the practice of its speakers. Quirk (1958) long ago recognized that linguists had to be prescriptive in some contexts, part of the responsibility of teachers of English being to tell their students how to use the language, but that prescription should be firmly based on description. Cameron (1995) emphasized that it is important for linguists to engage with those promulgating notions of correctness: this is a debate that must be entered. The major problem with the definition of Standard English is that we have to engage both with usage and with form. My starting point is to look at edited written texts in which Standard English is required. How do such texts compare formally from one region to another? Do they share criterial linguistic features that are central to Standard English? But all users of Standard English make mistakes. When I write accomodation or concensus, it does not become Standard English because I have written it. There are processes of monitoring, self-correction, correction by others and editing that writers of Standard English do in order to keep what they write in line with the prevalent practice in Standard English. There is a circularity here that cannot be avoided: it is the writing practices of writers of English all over the world that determine what is and what is not Standard English; but those writers respond to notions of correctness based on tradition and wider practice. But we are not in an ‘anything goes’ situation, and there are specific linguistic features that can be used to identify the dialect, and, other than in lexis, choices are made and change occurs within linguistically specific confines. There is very substantial variation within Standard English related to text type. Bhatia (2004: xiv) describes ‘the real world of discourse’ as ‘complex, dynamic, versatile and unpredictable’. Variation linked to text type is far greater than variation linked to regionality. Some text types are

Singapore Standard English revisited 65 prone to be playful, especially literary writing, which is the main domain in which non-standard regional dialects can be found. Although there is large variation relating to text type in Standard English, with the exception of the abbreviated texts, the variation does not relate to what I regard as the core areas of Standard English: dialect is consistent across text types. Differences generally relate to differences in degree or frequency (such as sentence length, proportion of passives, amount of nominalization, particular organizational features) rather than to qualitative differences in orthography, grammar or lexis. In this chapter the term ‘Standard English’ should be understood as referring to Standard English other than in its grammatically reduced abbreviated variety. Standard English is of great importance. It is accepted as the norm for written English, and the overwhelming majority of all written texts are in Standard English, and have been in a single shared standard dialect since around 1500 (the form has changed a little since then). Writers go to great lengths to correct errors. Schoolchildren are examined on their skill in it. It is essential, and universally accepted, that Standard English is what is taught in schools and that editors should do their best to make sure that edited texts are written in Standard English (except in the case of those rare texts that deliberately use a non-standard dialect). Standard English is not some distant and unreal kind of English, but something present and immediate for all of us who learn or use English, at any level. I have made calls for realism in the teaching and assessing of the use of Standard English (Gupta 2001). The concept of Standard English is strongest in writing, and English users vary in their skills. There is no English-using community in which everyone has a high level of skill in Standard English. While all Singaporeans (like residents of all English-using countries, and many others) need to have some skill in speaking, reading and writing Standard English, the production of high-quality written text requires input from specialist editors. High-prestige printed and electronic text is edited: this means it has been looked at by several skilled writers who work together to eliminate as many errors in Standard English as possible. Not every user of English can be expected to have these high-level skills. Everything I have said about Standard English applies to all Englishusing countries, including the United Kingdom, the USA, Australia, Nigeria, India, Jamaica and Singapore. Increasingly, it is coming to apply to speakers of English in countries where English has had little internal use in the past, such as Japan and France (some of the issues of English in these areas are discussed by Jenkins 2007). As I will show in this chapter, the way in which Singaporeans use English in formal and informal written material, and the language advice given in connection with the Speak Good English Movement, suggests that Singaporeans participate in the same uncertainty about Standard English as do users of English from other English-using nations.

66 Anthea Fraser Gupta There is no tradition in English of any central control of the standard. As is well known, there has never been an academy for English, and it is hard to imagine how there now could ever be one, given the range of countries that have made English one of their own languages. Even those languages ostensibly policed by academies have to face the onslaught of usage (Schiffman 2002). And in English, usage is all. What is and what is not Standard English is maintained by a mysterious process of worldwide consensus. There are important agencies, especially newspapers and publishing houses, but even these agencies are participants in a wider world and have to respond to innovation from many quarters, both social and regional. Change in English can take place rapidly. New words enter English constantly and become part of Standard English. Sometimes, when new technology comes along, the same word is used all over the world (e.g., DVD); on other occasions regional differences develop (e.g., cell, mobile, handphone). No-one makes a decision. When a word is used enough in texts written in Standard English, it becomes part of Standard English. Dictionaries and grammars react to English: they do not control it. However, dictionaries, grammars, and usage all participate in the feedback loop by which means users of Standard English monitor their usage and thereby maintain a unified Standard. When Samuel Johnson planned his dictionary, the first on scientific principles, he thought English could be fixed and perfected. But the process of writing his dictionary taught him that language change was inevitable: hoping ‘to enchain syllables’ was like wanting ‘to lash the wind’ (Johnson 1755–56: 10). This has not prevented many of us, including myself, from trying to say what should and should not be regarded as correct. Jenkins’s (2000, 2007) promotion of a more tolerant approach to the pronunciation of ‘English as a Lingua Franca’ — that is, English as used between non-native speakers — has been criticized by those who still hope to enchain syllables (the reaction to the earlier book is discussed in the later one). I feel that Jenkins does not go far enough because she makes recommendations that learners of English are taught some specific features that appear to make English more comprehensible. Among other things, she recommends that ELF speakers maintain a distinction between long and short vowels, and pronounce /r/ everywhere it appears in the spelling. Neither of these features is shared by all dialects of English. It seems to me that varieties of English in places like Germany, Japan, China and Thailand are also part of the organically unpredictable nature of English and we cannot expect to require specific features from speakers there any more than we can require it of speakers of English in London or Singapore. The unclear way in which English is maintained applies to English as a whole, and also to English locally. When I came to live in the UK in 1996 after over 20 years’ absence, one of the many words I had to learn was

Singapore Standard English revisited 67 manky, which was then an informal word meaning ‘bad, inferior, defective; dirty, disgusting, unpleasant’ (Oxford English Dictionary). OED’s first example of this word is from 1958. Although OED labels it as ‘British colloquial’, this word has been moving into Standard English over the years. Whereas the earlier examples in OED are all from representations of speech, the most recent two are in contexts that predict Standard English (a fishing handbook and a newspaper). The word has also spread beyond the UK: by 2007, Solomon Lim, writing in Singapore’s Electric New Paper was sitting in soft, slightly-manky swivel chairs. In 1999 bling spread around the world, moving in twelve months from Black slang (in the form of bling bling) to global Standard English — one company in Singapore is even selling fridge bling. Why these words and not others? Decisions like this are made by the unclear process of the interaction of the gatekeepers (such as newspapers) with the body of users. Standardness emerges; it is not predetermined. The whole world participates in the maintenance of Standard English. It is as much the property of Singapore and Nigeria, and perhaps even of Germany and Thailand, as it is of Australia and the United Kingdom. We need to understand that Singapore is not a passive recipient of Standard English but participates in the process of shaping Standard English, both locally and globally.

International assessment of standard English Whenever the attainment of schoolchildren in Singapore is compared to the attainment of children of the same age in other regions, Singapore comes out top or near top (Dixon 2005: 626). In 2003, Singapore’s children were, as usual, the best in the world in maths and science (TIMSS 2003). In 2006 (Mullis et al. 2007: 42) ‘[t]he Russian Federation, Hong Kong SAR and Singapore were the three top-performing countries’ of those countries that took part in tests of literacy in schoolchildren aged 9 to 10 years. The tests were administered in the local language or languages of education, which means that in five countries (England, Scotland, Singapore, Trinidad and Tobago and the United States) they were administered only in English (Mullis et al. 2007: 287–8). The results mean that Singapore’s primary schoolchildren could reasonably be said to be better at Standard English than children in the other four countries. Whereas over 70% of children in all the other four countries spoke only English at home, in Singapore only 21% spoke only English at home (only 5% spoke no English at home): bilingualism is clearly no handicap to excellence. Nor is bidialectalism: it can be assumed that the Singaporean children were competent in Singlish as well as in Standard English.

68 Anthea Fraser Gupta In all English-using countries, there are people who do not develop sufficient skill to write in Standard English. In both Singapore and the UK national examinations are taken at around the age of 16 years: in England, Wales, and Northern Ireland these are GCSEs and in Singapore they are O-Levels. In 2007, 62% of those taking the GCSE examination in English (taken by all children in England, Wales and Northern Ireland) attained a grade A–C (BBC News 2007). In Singapore, the success rate in the equivalent examination (O-Level) is much higher, at 87% in 2006 (Ellis 2007), but the examination is taken by an estimated 80% of the age group (Dixon 2005: 627). The proportion of the age group who pass is still slightly higher (at around 70%) in Singapore than it is in the UK, as one would expect from the 9-year-olds’ better performance. We need to remember those 37% of children in the UK who did not pass at the required level. The vast majority of those children are British born and bred, native speakers of nothing but English, who have not developed much skill in producing Standard English. There is no reason why Singapore should expect that the proportion of children who can reach this level in Singapore should be any more (or less) than the proportion who succeed in the UK. The figures, the baseline for proficiency in Standard English, are similar in these two countries, and probably reflect the proportion that could be reasonably expected to succeed in learning to read and write Standard English to a societally agreed threshold for adults in any English-using country where education is effectively delivered. Producing Standard English is a skill, like playing the piano or designing furniture: it has to be learnt and it is tested. We should see and promote it as a desirable skill, but one that some will succeed at better than others. Singapore is doing rather well at teaching it to its children.

The idea of a local standard When Mary Tay and I looked at Singapore Standard English in the 1980s, we were using the old technology of paper and the old theory of deviance from an imagined Standard English. The methodology that I used in 1983 was to identify in texts published in Singapore those features that I believed would not be used in the English of the USA or UK, and then to indicate which of those features could be said to be standard. There were two problems with this methodology: (a) It is possible to say that something is used in a given variety. Attested use is concrete and important. It is much harder to say that something is not used in a variety. I identified this as a problem fairly early (Shields 1977) and discussed it further in 1986, but I persisted in using the expression ‘nonstandardism’ as though I knew what it meant. I applied

Singapore Standard English revisited 69 this term, in some cases, to features attested in Singapore English that I later found to be also used in other varieties, such as British and/or American English (including accommodations, This is the first time I am doing …, and less workers). I thus identified as ‘Singapore English’ many features that are geographically widespread, some of which are disputed usage. At least I made the right predictions — that some of the features I identified as Standard Singapore English would spread to other varieties. (b) It is not possible to pick and choose what should and should not be standard. This can only be established retrospectively by the test of time based on the pattern of usage. Something becomes Standard English if it is often used in Standard English texts. Since the 1980s, the internet has made it possible to scrutinize the English of the world as a whole. This has made it even clearer that (a) Standard English is one dialect and remarkably uniform; (b) most written English, of all text types, is in Standard English, while writing in local nonstandard dialects is rare and is severely restricted in terms of text type; and (c) the patterns of usage are complex and no single person can rely on personal judgement to determine what is and what is not Standard. It is not appropriate to take texts from Singapore and compare them to some idealized, fictitious, Standard British (or American, or Australian …) English. Meaningful comparison of the English of different countries has to be based on the comparison of real texts of the same sort. So Singapore newspapers could be compared to newspapers from New Zealand; university essays by Singaporeans could be compared to university essays from the USA; charity websites from Singapore could be compared with charity websites in India; and so on. When this sort of comparison is done (Tottie 2002; Gupta 2006b; Paulasto 2008), we see that there is very little difference from one place to another in the features of English used in text types in which Standard English can be expected, and that most differences are differences of relative frequency of forms rather than qualitative differences. There are few categorical differences (things that occur in one region but not in another), but there are regionally patterned differences in preference, where there are alternatives within Standard English (such as don’t have/haven’t/haven’t got). All writers make errors in writing Standard English: where the criteria for what is and what is not standard are clear, we should not be afraid of identifying errors. Texts that are edited (such as newspaper articles) have very few errors, though they do have some errors, everywhere in the world. When it comes to unedited texts, there are more errors, but not noticeably more in one place than in another. We do start to see individual difference, however, and evidence that individuals vary in skill in writing Standard English. Some make more mistakes than others.

70 Anthea Fraser Gupta All users of English interact with English both locally and transnationally. Even if all face-to-face contact were with people from one’s own country, reading and viewing is likely to include material from other countries. Enclosure within an exclusively Singaporean world is impossible. Pedagogy and policy need to be based on the real Standard English and not on some imagined perfection. The reality of Standard English is that it crosses borders. It does not make sense to talk about discrete local standards: there is one Standard English which incorporates some regional differences. Singapore needs to accept that its local words are part of Standard English as a whole, as are the local words of the USA, UK, Australia, Nigeria, and so on.

Kiasu English is porous and global. I would like to give an example how a word can move into the mainstream. In Singapore, the word kiasu moved from informal use into Standard English when it started to be used in parliament and newspapers and is now in the Oxford English Dictionary (OED). Kiasu has already gone through the early stages of global spread that manky and bling went through a decade earlier, but its starting location was Singapore. The Dictionary of Singlish and Singapore English documents this movement into standardness. In the early days, its spelling was uncertain. Its first attestation, in the erroneous form kian su, is in 1978, in the glossary of a sociological study of national service which presents it as Army Slang. It was little known then: it did not appear in the early studies of Singapore words (including my own, Shields 1977), but soon moved to general colloquial use. Then it moved towards the standard: it was first used in the official report of parliament in 1990 (‘I wish that the Government Ministers do not become infected with the same kiasu syndrome that they themselves have advised other people against’), quickly followed by multiple appearances in the Singapore press. Early examples are often glossed, but later ones are less likely to be. The next step for kiasu is use outside Singapore, with reference to Singapore. The closely related English-using neighbours of Singapore, Malaysia and Brunei, start to use it, and when they do it is often, but not only, with reference to Singaporeans. In 2001, only a decade after a Singaporean newspaper used it for the first time, it begins to appear in the British press. The British newspaper, The Guardian, first used kiasu in 2001 (7 September), quoting, without any gloss, a Singaporean speaker (‘This pursuit of material wealth combined with the constant need to be No. 1 has created the Singaporean we hear so much about — the kiasu Singaporean’). In 2002 (27 August) The Guardian got it slightly wrong, making it an abstract noun:

Singapore Standard English revisited 71 The answer is simply that there are not enough hours in the day — if a child wants to have a remotely successful adulthood in what is such an incredibly pressurised society that the fear of losing or failing, known as ‘kiasu’, is all-pervasive. Indeed, it so [sic] ubiquitous that there is a popular cartoon character named Mr Kiasu.

In 2004 (7 August) it got it right: Risk aversion, a bureaucratic imperative, is valued in Singapore. It makes the place reassuringly predictable for businesses. Singaporeans have a word for it: kiasu (afraid to lose out).

And in 2007 (2 June) when Kiasu, a restaurant selling Singaporean food, opened in London, and by which time kiasu had made it to the Oxford English Dictionary, there was even more publicity for the word: As part of the ongoing, industry-wide drive to appeal to the widest possible demographic base, today’s review is targeted at those of you who hope one day to be a guest on Call My Bluff [a television show in which contestants have to guess which is the correct definition of an unusual word]. I don’t think it’s actually on telly at the moment, but when it is next revived, ‘kiasu’ will probably be one of the thrice-defined words.

We can trace this process, but the nature of the agency of standardization is obscure. Who made the decision to use kiasu in parliament? Who decided it could appear in The Straits Times? These decisions initiated the move towards Standard English. What was the process by which its spelling became finalized? When will it move the next step, of being used far from Singapore without reference to Singapore? This final step has been taken by many words that once were regionally restricted. For example, amok came into English (from Malay via Portuguese) meaning ‘a frenzied Malay’ in the sixteenth century (OED), but by the eighteenth century a person of any ethnicity, anywhere in the world, could ‘run amuck’.

The ‘Speak Good English’ Movement Singapore launched the Speak Good English Movement (SGEM) in 2000, following speeches attacking Singlish made in August 1999 by two senior political leaders, Lee Kuan Yew and Goh Chok Tong. This movement was supposed to focus on spoken English, and to promote Standard English at the expense of Singlish. However, it was not supposed to attack the Singapore accent. Rubdy (2001), writing in the early stages of the campaign, predicted success. Seven years later, Singlish continues to flourish, and the SGEM remains as an arena for information on English, and a place

72 Anthea Fraser Gupta where the linguistically insecure can seek answers. It has been associated with attacks on the Singapore accent, and has dealt with written as well as spoken English. Singapore has a long history of active and successful language planning, summarized with balance and clarity by Xu and Li (2002). Its ‘strong interventionist approach, or linguistic dirigisme’ (Xu and Li 2002: 275) has been applied (from time to time) to promoting Malay, English, and Mandarin. Promoting a specific dialect of English is the latest challenge. But managing a change in speech is difficult. One reason for this is that speech is immediate and contingent: we speak to real people in a real situation. Most of the SGEM materials ask Singaporeans to imagine a foreign interlocutor, who cannot understand Singlish, and who may perceive its rapid fire delivery as signalling rudeness, but in real life most Singaporeans principally use English to speak to other Singaporeans, who can understand Singlish and who perceive it as informal and warm. Promoting change in speech is difficult because when we speak we are doing our best to project a sense of personal identity (Le Page and Tabouret-Keller 1985). Some of the scenarios from the early years of the Speak Good English Movement promoted a stilted, colourless, over-formal kind of English, in situations where the Singaporean expected something more informal. Indeed, the ‘Standard English’ alternatives promoted are typically very formal for anywhere. There are a set of recommendations for phrases to use in service situations, such as: Avoid: Say: Avoid: Say:

No more blue colour. I’m sorry, we have run out of blue. You want to try? Would you like to try that?

The alternatives, like the downloadable lessons on the SGEM site, promote a textbook English of full sentences and extreme formality. To anyone from a community where English is a living daily language, as it is in Singapore, they seem stiff, like the English that learners of English, rather than speakers, might use in the classroom (but not, of course, in real life, as Jenkins has shown). It is as if the SGEM wants Singaporeans to speak as if they have learnt English only at school, from a rather old-fashioned teacher. And as if they have learnt only one, very formal style of speaking it. In real life Singaporeans are more sophisticated users of English than this. Like speakers of English in all societies where there is a daily use of English in a range of domains, they need to have different styles of English in their repertoire. The idea that a Singaporean accent is acceptable has also not been maintained in the SGEM materials. The SGEM website includes exercises

Singapore Standard English revisited 73 in pronunciation which address some of the core features of the Singapore English accent, such as the conflation of the vowel of pat and the vowel of pet. Website readers are invited to practise the distinction, as in: My cousin’s name is Ellen. My cousin’s name is Allan.

Sometimes recommendations are given for how to pronounce specific words. Generally, where there are alternatives, the one recommended is always the most old-fashioned kind of British English that can be found. For example, when one questioner asks how many syllables vegetable should have, the answer is three, while the correct pronunciation of Wednesday is said to be with two syllables (and no /d/). Cross-checking OED and pronouncing dictionaries will show that there is variation in both these words, even within British English. When there are alternative pronunciations, the SGEM always seems to promote the one that has maximal reduction. The high-prestige British accent, RP, is at one extreme of English accents, with a great deal of reduction, while Singapore English is at the other extreme, with very little. In RP, many unstressed syllables use a reduced vowel (/ԥ/ or /ܼ/) where other varieties, including Singapore English, have a full vowel. This applies both in words (as in the first syllables of consider and exam) and in sentences (in an RP rendition of ‘I gave it to him’ it is likely that only gave will have a full pronunciation). There is little doubt that as far as world-wide intelligibility is concerned (the supposed aim of the SGEM), the fuller pronunciations are clearer (Jenkins 2000, 2007; Gupta 2005): the only motivation for recommending these extremely reduced forms is to promote a rather old-fashioned type of RP. The ‘Phone-in Lessons’ (developed by the British Council in association with the SGEM), which have been part of the SGEM from the start, are built around a scenario of a Singaporean company employing foreign staff. In the first lesson, a worker from Britain is met at the airport by a coworker from India. They speak together in the usual SGEM textbook style of English, with a slow and careful delivery: Jane: Hello, I’m Jane. Are you Jaya by any chance? Jaya: Yes, I am. Hi, Jane. I’m Jaya from HotDotCom. Pleased to meet you. Jane: Ah yes. You’re the webmaster, aren’t you? Pleased to meet you too.

When they get into a taxi, however, Jaya and the driver have a discussion about the route in more natural-sounding Singlish: Jaya: This way can. Driver: No lah, this way cannot.

74 Anthea Fraser Gupta Jane thinks Jaya and the driver are speaking Chinese and wonders how long it will take her to learn. It is true that on hearing a new nonstandard dialect of English for the first time, a hearer might think it is another language. I was once asked by a visiting American in Singapore what language a woman behind us on an escalator was speaking to her child ‘(it was Singlish)’, but a Botswanan visiting Newcastle also once asked me what language the revellers in the streets on a Friday evening were speaking (the Newcastle dialect, Geordie). Tuning in to other dialects is part of what we have to do as English speakers, and foreigners who come to live in Singapore learn to understand Singlish easily enough. And most Singaporeans who encounter foreigners can adapt their English too. It is not clear why this dialogue is built around a non-Singaporean speaking Singlish: Are the writers suggesting that Indians are more able than British people to pick up the local dialect? Or that they are more likely to be corrupted by it? It is all rather neo-colonial. It is foolish to expect that stilted, unreal English, of a kind associated with foreign learners, could replace a living dialect. Even if Singlish does diminish in use over the years, it will be replaced by a form of speech that allows its users to express a Singaporean identity, and to express degrees of formality. It is generally accepted that all English-using countries need their own vocabulary to deal with elements of local culture. But the commentary on the first SGEM phone-in lesson attacks even this, giving ERP as an example of something to avoid. Instead of using ERP, should Singaporeans say ‘Electronic Road Pricing’ every time? Or should they perhaps adopt the London expression, ‘Congestion Charge’, even though Singapore had this idea many years before London? Local cultural terms are essential for all communities. Much of this is a return to the old days of the 1980s and before, when Singapore looked to Britain (an imagined, unreal Britain) for a model of excellence in English. The world has moved on, and the colonial cringe is over. Singapore is one of the success stories of the world, and can be confident about its English as well as about other areas of its success. Singapore’s SGEM movement seems to have given the world the impression that the standard of English in Singapore is poor, and that, as the current (March 2009) Wikipedia entry on the SGEM says: The Speak Good English Movement is a program launched by the government of Singapore. Singaporeans usually speak a variant or dialect of English known as Singlish, which has heavy influences from Chinese and Malay in pronunciation, grammar, and vocabulary, and can be incomprehensible to non-Singaporeans. According to the government[citation needed] , this causes problems when Singaporeans need to deal with people from other countries in business and trade, so the campaign aims to discourage the use of Singlish and encourage the increased use of a more

Singapore Standard English revisited 75 standardized form of English (i.e. generally modelled on the British standard). The launch of the program has been reportedly[citation needed] unable to effectively change the diglossia (two parallel languages for formal and informal situations) among Singaporeans. Singlish still remains as a ‘common language’ among Singaporeans, and is mostly used in informal communication, while a more standardized form is used for formal communication.

Note that this entry assumes that even in ‘formal communication’, Standard English is not actually achieved in Singapore. Singapore should be attacking, not fostering, the erroneous view, still sadly widespread, that while people from the UK and USA (etc.) are perfect at Standard English, people in places like Singapore and India (etc.) never quite make it. Such a view is not supported by linguistic analysis. Of course, criticizing the Standard of English (or maths, or behaviour) in a country is perennial. No country in the world is immune from this selfflagellation. Here, for example, is a claim from Clive James (Australian by birth, long-time resident in the UK, writing in an Australian magazine): In which English-speaking country is the English language falling apart fastest? Britain. Are things as bad in Australia? I hope not. In Britain, in 2006, the Labour government is still trying to fix Britain’s education system, but surely one of the reasons it’s so hard to fix is that most of the people who should know how are themselves the system’s victims, and often don’t even seem to realize it. They need less confidence. Even when they are ready to admit there might be a problem, few of them realise that they lack the language to describe it. (The Monthly, May 2006, repr. on Clivejames.com)

The rest of the world is not very conscious of Singapore and does not know the extent to which it is an English-using country. When I came to work at the University of Leeds from the National University of Singapore, one of my colleagues commented that it must be nice for me to be teaching students whose English was better: the truth is that the English of the students I taught in Singapore and the students I teach in Leeds is very much the same. Foreigners reading official sources will not get a good (or an accurate) impression of the real standard of English in Singapore. They might not realize that the notion that standards are falling is a myth as old (at least) as Plato — Bolton (2003) discusses the same myth in the context of Hong Kong. I fail to understand why the very real successes of generations of Singaporean politicians and teachers have not been recognized. It is time for Singaporeans to understand that Singapore is no longer a British colony, and that the standard of English in Singapore has risen steadily since independence, is still rising, and has, for some years, been excellent by any reasonable measures. Singapore’s school students regularly top

76 Anthea Fraser Gupta international tests of educational excellence, and of English. It is hard to understand what is to be gained by wrongly presenting Singapore’s education system as a failing one, and by wrongly presenting Singapore as a place where people speak ‘bad English’. People unfamiliar with Singapore are likely to believe it.

Areas of uncertainty The traditional areas of disputed usage continue to give rise to uncertainty and insecurity in most of those who use English, and continue to be the focus of diatribes from the language police. These areas loom large in the minds of users of English, and are the subject of a disproportionate amount of attention in schools and in the media. Most overt discussion of Standard English takes place not around those areas where the Standard is strictest (spelling, inflectional morphology, structure of the verb), but around these peripheral areas, where there is dispute about alternatives within Standard English. The better style guides address the issues and explain the disagreement, but there are also pundits who set themselves up as experts in judging the disputes, often making capricious decisions not based on the linguistic or social facts. Such people are sometimes called ‘language mavens’. Language mavens do not refer to reputable dictionaries in their pronouncements, or to usage, or to history. They foster insecurity in ordinary users of English without any serious effort at improving their English, something that has been attacked by linguists for half a century (for example, Leonard 1962). Unfortunately, the SGEM appears to have fostered a maven culture for the English of Singapore, rather than empowering people to extend and develop their English. All over Singapore, there are advertisements for classes to improve English, many with a focus on pronunciation or ‘phonics’. Writing in The Straits Times on 21 October 2007, Janadas Devan reported that, at that time, ‘bad English’ was the second most popular topic of discussion on the Straits Times Interactive website. The SGEM invites the public to ask questions of their language experts. Questions are answered on the SGEM pages, in a weekly ‘English as it is broken’ column in Singapore’s leading newspaper, The Straits Times, and in the Straits Times Online Mobile Print (STOMP). The credentials of those who answer the questions are not on display, unlike the credentials of those who publish dictionaries, grammars, and style guides. Some of the SGEM answers appear to be provided by 938LIVE, a radio station, and others are reprinted from The Straits Times. The STOMP answers are provided by ‘the Ministry of Education’s English language specialists’, who do seem to be much better informed than the respondents from 938LIVE. I looked in detail at the

Singapore Standard English revisited 77 four months of questions and answers in STOMP (March–June 2007), when members of the public asked 198 questions (some people covered more than one topic in a single posting, which is why the figures below add up to more than this). Answers can be grouped into 17 categories, the first 2 of which are not so much questions as jokes and complaints. 1.

Look at this (44). Members of the public send in photographs of notices that they believe include errors in English (e.g., FOR EMERGENCY USED ONLY, figurine for a life-size statue) or that they find funny (e.g., GM FOOD CENTRE). The supposed peculiarity is not always identified, and some of them seem to me to be just ordinary notices (e.g., STOP BEWARE OF CHILDREN). In one case a notice is photographed that is likely to have been a knowing use of Singlish (INSIDE GOT MORE DRINKS AND BEER). One of these pictures begins every page of questions.

2.

This is appalling (6). Members of the public complain about English they have seen in textbooks that they regard as wrong, about features of Singapore English they do not like (using ‘t’ for ‘th’ and pronouncing ‘w’ wrongly), and, sometimes, about previous answers to questions. For example: Many adults above 45 years pronounce the letter ‘w’ as ‘dub-dew’ which should be pronounced as ‘double-u’ (quickly). I hope the younger generation is taught correctly. I am not talking about American pronounciation [sic] of ‘z’ as ‘zee’ and British slang as ‘zed’ or ‘h’ as pronounced as ‘hedge’ or ‘edge’. Go and ask those over 40+ yr teachers how they pronounce ‘w’. THE LIST GOES ON!

The answers to these complaints, especially those on pronunciation, were generally tolerant of the feature that caused the complaint. 3.

Analysis and convention (10). Three questioners ask for technical terms to assist in linguistic analysis, one asks whether off is a part of the word offer (relevant to the rules of a game), two questioners ask what ‘full sentence’ means in the context of comprehension questions, and four ask whether other school tests were correctly marked.

4.

American and British English (3). British English (or English English) is referred to occasionally as a reference point in both questions and answers, but only twice does the question put this at the centre of concern. One questioner asks for information on where to learn American English, while another shows a clear sense of British English as the model, asking whether spellings

78 Anthea Fraser Gupta such as ‘subsidised’ and ‘criticised’, believed by the questioner to be American, are ‘allowed as we need to follow the trend’. The answer comes down firmly in favour of British English: ‘Both are acceptable as both are British, not American.’ A third questioner asks about the past tense of dive, and the answer addresses the US/UK difference. In other categories, most of the other issues raised are irrelevant in terms of the British/American differences, though some of the questions on pronunciation suggest that pronunciation differences between British and American English are behind them. The extent to which the answers address this is variable. 5.

Concord (10). Six of the questions about subject-verb concord relate to unclear areas of concord in which there is some mismatch of notional and strict concord, as in collective nouns (troupe, pair, team), each, and X as well as Y structures. One questioner was puzzled by the subjunctive (‘If he were ...’) while one asks a much more basic question, apparently thinking that if always occurs between a singular noun and a verb concord will be plural (Mary likes vs. Mary always like). There was an additional question on the choice of pronoun after someone (the answer promoted the oldfashioned and controversial singular he).

6.

Determiners (5). Three of these are about the choice between a and an (an unusually regular feature of English). Two of the questions relate to ‘an earring’ and reveal a belief that earring begins with a consonant (like year). The answer corrects this (unexpected) assumption. The third, more predictably, relates to ‘a useful answer’. The answer says that the ‘usual rule does not apply to u’ rather than saying that ‘useful’ really does begin with a consonant like ‘year’. The other two questions about determiners relate to the idioms on (the) air and (a) sense of humour.

7.

Prepositions and particles (11). These questions all relate to the choice of one or another preposition or particle, usually where choice is possible within Standard English (in/ on the bus (twice); run over/down/into/out; (to) home; through/across the underpass; prefer tea to/or coffee; between 9 to/and 5; go to/for lunch; good at/ in; upon in fractions).

8.

Singular and plural nouns (10). Of these, five relate to uncertainty about mass nouns (lingerie(s), flurry of activity/-ies, discrepancies, nuclear plant(s), water(s,) translation(s), feeling(s)), and one to the plural of fish. The other five are all about the choice between singular and plural in noun modifiers (Academy Award(s) Nominee, goody/ies bags, 2-room(s) flats, Aesthetic(s) department).

Singapore Standard English revisited 79 9.

Tense (12). Some of these are very broad. The majority of the specific ones relate to the use of either will/would, can/could or the use of tense in reported speech, especially in minutes of meetings.

10. Other inflections (17). As with the questions on concord, with one exception (about obsoleted) these are questions about areas where there is choice or variation within Standard English, or where the structure gives cause of confusion (let it begin(s)). Three questions relate to the comparative, with two questioners checking that double comparatives (more cheaper) are incorrect, and a third querying the comparative adverb faster. One question addresses you are welcome(d). Three are about the possessive (goat(’s) milk; the possessive of James; choice between inflectional and periphrastic possessive) and six are concerned with pronouns. The areas of uncertainty in pronouns are whom (the answer promoted it); the comparative (taller than he/he is/him) and the choice of I/me in coordinated subjects and objects (John and I/me). 11. Word order (2). Both relate to the placement of adverbials (the answers drew attention to choice). 12. Other grammatical (9). The questions relate to other assorted aspects of grammar, including two on the reflexive, four relating to idioms that affect the choice of verb form in structures such as saw C shout(ing) and our mission to bring/ of bringing, one asks about the use of But although … One is about the use of both with not (the answer approves it), and one just puts down a sentence (with errors of concord and punctuation) and asks if it is correct. 13. Orthography (5). There are few worries about spelling, it seems. Of the five questions about orthography, two are about apostrophes, one about the choice of words or figures for numbers, one about the placement of full stops when a sentence ended with a quotation that was a full sentence, and one about stationary/ery. 14. Words (6). There are six questions about the use of specific words. Concerns include whether impact can be a verb (yes) and whether reservist can be used to mean reservist duty (the answer is that this is ‘a grammatically incorrect shortening which has become commonly accepted’). The expert is less modern in the case of email:

80 Anthea Fraser Gupta Language grow and develops so e-mails may emerge as a plural of e-mail just as electronic mail become e-mail (now an accepted word) which will probably in time lose the apostrophe and become email (Did you know fortnight was originally fourteen nights?)

15. Confusing words (31). A stunning 31 questions deal with words that either have similar meanings (e.g., speak/talk, fetch/take/bring, lend/borrow, prohibition/ban) or are similar both in meaning and form (e.g., among(st), orient(ate), ironic(al)). Two questions are about What/how about. Answers to these questions often draw attention to the presence of choice in English, but sometimes come out in favour of one alternative or another, not necessarily for good reasons. Which alternative is unpredictable. Sometimes the answer comes out in favour of the more Singaporean alternative: Question: Can I ‘eat my dinner’? Or should it be ‘take my dinner’? Answer: I would rather use ‘take’ for meals.

In other cases, answers make a spurious distinction between words with substantial overlap, as in this answer, which defines chicken in a way not supplied by OED: Chick would refer to the cute, fluffy, yellow baby chicken. In fact, the young of any bird can be called chicks. Chickens are the fully matured females which taste so good when they are fried.

16. Pronunciation (9). All of these relate to specific words. Three of them are linked questions about the vowel in the first syllable of colleague. Other issues are: the two pronunciations of the; the number of syllables in itinerary (the answer said there should be four, going as usual for a maximally reduced alternative), the initial sound of Thailand, the pronunciation of film, tuition and laboratory. It is surprising that the answer to laboratory does not address the large US/UK difference in this word, especially as a much smaller US/UK pronunciation difference is addressed in one of the answers on colleague. 17. Miscellaneous (11). Eight questions relate to specific issues about particular idioms, mostly to do with collocation or idiom: Should a ‘worst tyrant’ really be a ‘best tyrant’? (Back) home? Is it tautologous to say someone is a ‘dishonest liar’? Can meat be ‘freshly frozen’? Two questioners are concerned with ways of talking about colours (red colour(ed), colour red). One questioner raises

Singapore Standard English revisited 81 the Singaporean invariant tag is it asking whether it is true that ‘native speakers’ are now using it. The answer is ‘Personally, I haven’t heard native speakers follow suit.’ In fact invariant tags are common in a range of Englishes, and there is an invariant (negative) tag used by some speakers in colloquial British English, usually spelt as innit. The topics of these questions suggest that those who send questions in association with the SGEM are not confused about the difference between Singlish and Standard English. There are no questions about well-known features of Singlish such as pragmatic particles (such as lah), or about omission of the verb BE (He so stupid) or about the use of the base form of the verb where standard English required some inflection (Fatimah finish already). No translation is invited of any sentence in Singlish. These questions are questions from within Standard English, and the vast majority of them deal with areas of difficulty and dispute that beset all users of English. They address areas that are intrinsically difficult or irregular in English, or where English users do not agree on what is and what is not ‘correct’. These topics are the subject of advice directed at the insecure all over the English-using world. Clive James gives examples from the British (and American) press to illustrate his claim about the poor standard of English in Britain and elsewhere among ‘people who have learned English as a first language’. These are the usages he regards as appalling, attached to the numbers in my list of the STOMP issues: • dangling participles (not mentioned by any of the STOMP questioners); • mixed and dead metaphors (mentioned by some questions in 17); • confused words: hone it/home in; solecism/solipsism (15); • concord in complex subjects (5); • determiners (6); • lower case in name of a magazine (not mentioned by any of the STOMP questioners). The areas of disputed usage that worry the STOMP participants (and James) are covered by usage notes in dictionaries and style guides published and online. They are among the areas discussed by the linguists who have attacked prescriptivism for decades. What we see here are Singaporeans attempting to negotiate those areas of Standard English about which there is uncertainty. The presence of dispute and uncertainty within Standard English is part of the essence of Standard English as an organic dialect: these are the areas where the linguistic features of Standard English are not firmly defined. The features associated with the normative debate do not divide Standard English into multiple dialects: they are a debate about notions of correctness within a dialect in which concepts of correctness are central because it is a Standard and is required in certain domains. Engaging

82 Anthea Fraser Gupta in areas of dispute is part of what advanced writers of English have to do, making informed decisions through the process of feedback and monitoring that is the agent of maintenance and change in Standard English.

Deciding what’s right It is quite a challenge to answer some of the questions in a non-technical way, briefly, and accurately. While it is easy enough to tell a questioner (for example) the difference between its and it’s, it is much harder to explain the subtle difference between bring and take. We have seen that some of the STOMP answers are sensible, and some are capricious. None of them refer to any source of authority, either in works of reference from reputable sources, or in usage. The answers on the SGEM’s website are even worse. Of the 16 answers to questions in August 2007, only 5 could be verified as valid by checking against appropriate sources. Among errors in answers were the following very obvious ones: • learnt/learned. The SGEM answer claimed that learnt is past tense, while learned is the adjective. It is true that the form of the rare adjective can only be learned (pronounced as two syllables, as in ‘a learned doctor of the church’). However, any good dictionary (such as OED) will show that the past tense (and past participle) can be either learnt or learned. • inquiry/enquiry. The SGEM answer says ‘In British English, used here in Singapore, if you make an enquiry, you are asking for information. An inquiry is an investigation conducted by an organization, the police for example. However, in American English, it is common that the word “inquiry” is used for both instances.’ Most dictionaries (like OED) state that the two spellings are interchangeable, though some writers may try to make the distinction that SGEM wrongly claims is generally made by British English. • in/on a bus. The SGEM answer says it should be ‘on a bus’. OED has examples of both. Most of the erroneous recommendations in the SGEM site give a clear answer to an area of language where there is actually choice within Standard English. In many cases, they are not even disputed usage. These answers misrepresent English. They would constrain anyone who took them seriously, but would not result in someone saying or writing non-standard English. This is true of other answers to do with (e)specially, bring/take, fell for you/fell in love with you, and pressured/pressurized. The only damage done to a reader would be to make them feel insecure because they cannot trust the evidence of their instinct or what they read or hear.

Singapore Standard English revisited 83 But there is one answer where the recommended choice would take Singaporeans away from what is now considered best practice in writing formal English. This is an area where there has been change since the 1980s. Over the whole of the English-using world, the use of he to refer to a referent of either gender (‘If a student submits his essay late, he will be penalized’) became progressively less acceptable over the course of the last two decades of the twentieth century. Most reputable publishers (including Hong Kong University Press) do not allow authors to use this ‘generic he’. If you see this usage in a printed book from a major publisher now, you can guess it is pre-1985. This change was accompanied by a great deal of discussion. Style guides suggest alternatives to the use of he where referents might be of either gender or both, such as making the whole thing plural (e.g., ‘If students submit their essays late, they will be penalized’). The use of the singular they has become more acceptable than it used to be, especially with pronouns such as someone, but is still disputed because of an incongruity in number with verb concord (‘If a student submits [not submit] their essay late, they will be penalized’). The STOMP and SGEM answers promote the singular he with someone and no-one, as in this SGEM answer: Question: ‘Everyone wants to change the world but no one wants to change themselves.’ Should it be ‘themselves’ or ‘himself’? Answer: It should be ‘himself’ because ‘no one’ is singular.

This recommendation flouts all mainstream style guides and usage advice in modern dictionaries. Using he to refer to someone and no-one looks at best old-fashioned and at worst offensive. The Online Style Guide of The Times (London) takes the most traditional stand now possible without giving offence to anyone. It bans singular they, but does not even consider generic he as a possibility, recommending instead a compromise alternative — complete pluralization — that no-one objects to: they should always agree with the subject. Avoid sentences such as ‘If someone loves animals, they should protect them’. Say instead ‘If people love animals, they should protect them’. (Times Online)

In the SGEM response we find an example of Singapore not following the current trend in usage found in most countries, but holding out for the traditional maven-favoured alternative. On the same lines, the SGEM website identifies Professor Koh (a woman) as the ‘Chairman’, something that would be proscribed in edited texts from many publishers around the world. The SGEM has, as might be expected, been criticized from a number of quarters. The hardline approach to correctness lays the SGEM open to criticism from other mavens, who comb through its site and the speeches

84 Anthea Fraser Gupta of its leadership for real or imagined errors. Among those who have done this are the anonymous blogger ‘The Grammar Terrrorist’ in his (Unofficial, Unauthorized) English as it is Broken column, where he responds to The Straits Times column, and ‘Mr Brown’ (real name, Lee Kin Mun), in another blog, who on 26 July 2006 ‘corrected’ a speech by Koh Tai Ann, the chair of the Speak Good English Movement. Many of the corrections seem designed to render Koh’s elegant and idiomatic speech into the bland kind of English the SGEM seems to promote. The whole pernickety approach of the SGEM lends itself to the same response in others — this is how the SGEM is hoist by its own petard.

Solutions This focus on ‘correctness’, and the promotion of one alternative as right when there is in fact choice, has long been known to be damaging to insecure users of English (Mittins et al. 1970; Milroy and Milroy 1985; Wardaugh 1999). Instead, those promoting excellence in English would do well to look at how English actually operates in the world. Those who pronounce on usage should, before pronouncing, first check established and up-to-date dictionaries, grammars and style guides from major publishers, and then, if the information is not available from reputable published sources, they should investigate, by using internet searches intelligently, what the prevalence and the social and geographical pattern of usage of a given item is. There cannot be recourse to the ‘native speaker’. The idea that the native speaker is the sole and reliable source of judgement on Standard English is a damaging myth. There are many ways of defining ‘native language’, and the definition of the ‘native speaker of English’ has been especially politicized. Any definition of ‘native speaker’ that has recourse to race, ethnicity or citizenship is invidious and unjustifiable. Any definition that requires different criteria to be applied to native speakers of English than to native speakers of other languages also seems to me to be linguistically unjustifiable. Any definition that denies native speakerness to speakers of dialects other than the standard one, or to people who have more than one native language, is politically and linguistically unjustifiable. The most usual definition among linguists is that you are a native speaker of a language (any language, any dialect) if you acquired it naturally in childhood (Li 2000: 497). We could make this stricter by saying it must be a language spoken by the age of three years. Let us take that definition and see what facts it leads us to: 1.

Many Singaporeans (and most of those under 40) are native speakers of English, including Lee Kuan Yew, Singapore’s first prime minister.

Singapore Standard English revisited 85 2. 3.

4.

5.

Not all native speakers (in any region) are native speakers of Standard English. Not all native speakers of English (including native speakers of Standard English) can write Standard English skilfully. This applies whether the native speakers are Singaporean, British, or whatever (remember the GCSE failures in English in the UK). Success in writing Standard English is societally defined and is measured by educational attainment in English. Some non-native speakers of English are better at writing Standard English than are nearly all native speakers (e.g., Koh Tai Ann, Mary Tay, professors of English in Poland). The high-prestige edited texts that set the standard for English are produced by people skilled in the writing of Standard English, regardless of whether they are native speakers or not.

What matters is skill in performing Standard English, and we have traditional means of assessing this in the form of examinations, qualifications and ad hoc tests. These are the tests that should be applied, rather than any test based on accidents of birth. But any user of English with a high level of skills recognizes the need to use reference works, to keep in touch with trends in usage and to recognize personal limitations. My appeal for a realistic approach to Standard English is not to be seen as ‘anything goes’. Because this is not how Standard English works. Skill in writing Standard English is a central part of validation in English-using societies. Getting it right matters societally: Standard English has cultural capital (de Swaan 2001; Blommaert et al. 2005). But it is not the property of any one region: it is a truly global brand. English is not a classical language with clear rules that exist outside practice. It is a living language with all the areas of choice and flexibility that that implies. The use of English takes place in real social situations where negotiation of communication is what matters. Learning to use Standard English well can be thought of as progressing in stages: 1. 2.

3.

Speak some English (any dialect/dialects). Learn to read and write English. This is a process ideally, and usually, undertaken throughout childhood, assisted by teachers who are sensitive to the stage of the learner. Literacy is always developed in Standard English. Develop skill in producing and understanding Standard English, especially by intensive and attentive reading of and listening to a wide variety of texts from many times and places, and by collaborative writing. This is a lifelong task.

86 Anthea Fraser Gupta 4.

5.

Develop skill in producing a range of varieties of English, as appropriate to where you live, who you are, and what you do with English. Over your life you may live in different places and do different things with English: you will need to extend your range. Another lifelong task. Develop skill in understanding the spoken English of other regions. Learn about non-standard dialects as well as about variation in Standard English. This is also a lifelong task.

All users of English progress through these stages but the levels of skill attained in Standard English are variable. There is no evidence that a vigorous local non-standard dialect is detrimental to the development of strong skills in Standard English. There is every evidence that using English in a range of domains, making societally appropriate adaptations in style depending on context, interlocutor and text type, and taking pleasure in the use of English, promotes the development of strong skills in Standard English. A normative maven culture that promotes a judgemental approach to language is likely to create insecurity, possibly leading to less use of English, and a concomitant reduction in skill. The promotion of a single style is unrealistic and, if obeyed, would lead to a deterioration in the use of English. Especially in speech, all users of English need to be tolerant of a range of proficiency and to negotiate the construction of communication in reallife situations. We need to be prepared to adapt vocabulary, grammar and pronunciation to the needs of those with whom we speak face to face. There is immediate feedback from interlocutors to help us do this. We need to be tolerant of the differences between the way we speak and the way others speak, and hope this is mutual. When speaking to an unknown or mixed audience we need to assess the audience and do our best to meet their needs. There are different levels of skill at doing this. There are also different levels of skill in listening (Gupta 2005). We understand best the accents we hear most often. This means that for a Singaporean the easiest accent to understand is almost certainly a Singaporean accent. In the world as a whole, the most heard accent is almost certainly the accent known as General American, thanks to the popularity of American mass media. But that does not mean we should all strive to speak it. Variety is something to celebrate rather than deplore. In writing, things are simpler, because in many respects Standard English really is essentially monolithic. In any given text of Standard English (such as a newspaper article) more than 99.5% of words will be words spelled, inflected and used in the same way by Standard English everywhere. Standard English is so much a given that it is almost invisible. And it is not something remote and unattainable. It is something that Singaporeans are as good at as anyone else — indeed, possibly better than anyone else.

Singapore Standard English revisited 87 Singaporeans’ anxieties about English are shared by English users in the rest of the world.

References British Broadcasting Corporation (2007) Education. BBC News, 23 August 2007. http://news.bbc.co.uk/1/hi/education/6958992.stm. Accessed October 2007. Bex, Tony and Watts, Richard J. (1999) Introduction. In Standard English: The Widening Debate. Edited by Tony Bex and Richard J. Watts. London: Routledge, pp. 1–15. Bhatia, Vijay K. (2004) Worlds of Written Discourse. London: Continuum. Blommaert, Jan, Collins, James, and Slembrouck, Stef (2005) Spaces of multilingualism. Language and Communication, 25(3), 197–216. Bolton, Kingsley (2003) Chinese Englishes: A Sociolinguistic History. Cambridge: Cambridge University Press. Cameron, Deborah (1995) Verbal Hygiene. London/New York: Routledge. Carter, Ronald (2004) Language and Creativity: The Art of Common Talk. London: Routledge. de Swaan, Abram (2001) Words of the World. Cambridge: Polity. Dixon, L. Quentin (2005) The bilingual education policy in Singapore: Implications for second language acquisition. In ISB4: Proceedings of the 4th International Symposium on Bilingualism. Edited by James Cohen, Kara T. McAlister, Kellie Rolstad and Jeff MacSwann. Somerville, MA: Cascadilla Press, pp. 625–35. A Dictionary of Singlish and Singapore English (2004–). Edited by Jack Tsen-Ta Lee. http://www.singlishdictionary.com. Accessed August 2007. Ellis, N. (2007) Adding to the cultural diversity: A short history of expatriate teachers at Nanyang Girls’ High School. Nanyang Girls’ High School. http:// www.nygh.moe.edu.sg/RnD/research_2007_02.htm. Accessed July 2008. Görlach, Manfred (1990) Heteronymy in international English. English World-Wide, 11(2), 239–74. Gupta, Anthea Fraser (1986) A standard for written Singapore English? English World-Wide, 7(1), 75–99. Gupta, Anthea Fraser (2001) Realism and imagination in the teaching of English. World Englishes, 20(3), 365–81. Gupta, Anthea Fraser (2005) Inter-accent and inter-cultural intelligibility: A study of listeners in Singapore and Britain. In English in Singapore: Phonetic Research on a Corpus. Edited by David Deterding, Adam Brown, and Low Ee Ling. Singapore: McGraw-Hill Education (Asia), pp. 138–52. Gupta, Anthea Fraser (2006a) Singlish on the web. In Varieties of English in Southeast Asia and Beyond. Edited by Azirah Hashim and Norizah Hassan. Kuala Lumpur: University of Malaya Press, pp. 19–37. Gupta, Anthea Fraser (2006b) Standard English in the world. In English in the World: Global Rules, Global Roles. Edited by Rani Rubdy and Mario Saraceni. London: Continuum, pp. 95–109. James, Clive. Clivejames.com. http://www.clivejames.com/. Accessed October 2007. Jenkins, Jennifer (2000) The Phonology of English as an International Language. Oxford: Oxford University Press.

88 Anthea Fraser Gupta Jenkins, Jennifer (2007) English as a Lingua Franca. Oxford: Oxford University Press. Jespersen, Otto (1948[1909]) A Modern English Grammar on Historical Principles. (Seven volumes.) London/Copenhagen: George Allen & Unwin/Munksgaard. Johnson, Samuel (1755–56) A Dictionary of the English Language. 2nd ed. London. From Eighteenth Century Collections Online. Gale Group. http://galenet.galegroup. com/servlet/ECCO. Le Page, R.B. and Tabouret-Keller, Andrée (1985) Acts of Identity: Creole-Based Approaches to Language and Ethnicity. Cambridge: Cambridge University Press. Leonard, Sterling A. (1962) The Doctrine of Correctness in English Usage 1700–1800. New York: Russell and Russell. Li Wei (2000) Glossary. In The Bilingualism Reader. Edited by Li Wei. London/New York: Routledge, pp. 494–9. Milroy, James and Milroy, Lesley (1985) Authority in Language. London: Routledge & Kegan Paul. Mittins, W.H., Salu, Mary, Edminson, Mary and Coyne, Sheila (1970) Attitudes to English Usage. Oxford: Oxford University Press. ‘Mr Brown’ (Lee Kin Mun). http://www.mrbrown.com. Accessed July 2006. Mullis, Ina V.S., Martin, Michael O., Kennedy, Ann M. and Foy, Pierre. 2007. PIRLS 2006: International Report. TIMSS & PIRLS: International Study Centre. Oxford English Dictionary. 2nd ed. (1989) Edited by J.A. Simpson and E.S.C. Weiner. Additions 1993–7, edited by John Simpson and Edmund Weiner; Michael Proffitt, and 3rd ed. (in progress) Mar. 2000–, edited by John Simpson. OED Online. Oxford University Press. http://oed.com. Accessed August 2007. Paulasto, Heli (2008) Focus fronting as a feature of vernacular English. Paper presented at The Thirteenth International Conference on Methods in Dialectology, Leeds, England, 4–8 August 2008. Qualifications and Curriculum Authority (1999) Improving Writing at Key Stages 3 and 4. London: QCA. Quirk, Randolph (1958) From descriptive to prescriptive: An example. English Language Teaching 12. [Repr. with small alterations in Essays on the English Language: Medieval and Modern. Randolph Quirk (1968). London: Longman, pp. 108–13.] Rubdy, Rani (2001) Creative destruction: Singapore’s Speak Good English movement. World Englishes, 20(3), 341–55. Rubdy, Rani (2007) Singlish in the school: An impediment or a resource? Journal of Multilingual and Multicultural Development, 28(4), 308–24. Schiffman, Harold F. (2002) French language policy: Centrism, Orwellian dirigisme, or economic determinism. In Opportunities and Challenges of Bilingualism. Edited by Li Wei, Jean-Marc Dewaele and Alex Housen. Berlin: Mouton de Gruyter, pp. 89–104. Shields, Anthea Fraser (1977) On the identification of Singapore English vocabulary. In The English Language in Singapore. Edited by W.J. Crewe. Singapore: Eastern Universities Press, pp. 120–40. Speak Good English Movement (SGEM) (2000–2010) Speak Good English Movement. http://www.goodenglish.org.sg/. Accessed July 2007. The Straits Times Online Mobile Print (STOMP). http://www.stomp.com.sg/. Accessed July 2007.

Singapore Standard English revisited 89 Tay, Mary W.J. (1982) The phonology of educated Singapore English. English WorldWide, 3(2), 135–45. Tay, Mary W.J. and Gupta, Anthea Fraser (1983) Towards a description of Standard Singapore English. In Varieties of English in Southeast Asia. Edited by R.B. Noss. Anthology Series 11. Singapore: SEAMEO Regional Language Centre, pp. 173–89. Times Online. http://www.timesonline.co.uk. Accessed September 2007. TIMSS: Trends in International Mathematics and Science Study 2003. National Center for Education Statistics. http://nces.ed.gov/pubs2005/timss03/. Accessed July 2008. Tottie, Gunnel (2002) Non-categorical differences between American and British English: Some corpus evidence. In Studies in Mid-Atlantic English. Edited by Marko Modiano. Gävle: Högskolan, pp. 59–73. (The Unofficial, Unauthorized) English as it is Broken Column. http://englishasitisbroken. blogspot.com/. Accessed September 2007. Wardaugh, Ronald (1999) Proper English. London: Blackwell. Wikipedia. http://en.wikipedia.org/. Accessed August 2007. Xu, Daming and Li Wei (2002) Managing multilingualism in Singapore. In Opportunities and Challenges of Bilingualism. Edited by Li Wei, Jean-Marc Dewaele, and Alex Housen. Berlin: Mouton de Gruyter, pp. 275–95.

4

The Speak Good English Movement: A web-user’s perspective Paul Bruthiaux

There was a knock at the door. ‘Come in.’ A Chinese clerk, very neat in his white ducks, opened it. ‘Mr Crosbie is here, sir.’ He spoke beautiful English, accenting each word with precision ... Ong Chi Seng was a Cantonese, and he had studied law at Gray’s Inn. He was spending a year or two with Messrs Ripley, Joyce, and Naylor in order to prepare himself for practice on his own account. He was industrious, obliging, and of exemplary character. The Letter: W. Somerset Maugham (1927/2001: 117)

Set in Singapore, ‘the meeting place of a hundred peoples’ (Maugham 1927/2001: 116), this scene from one of Maugham’s Malayan stories has a curiously modern ring despite the colonial trappings that would be hard to spot in today’s city-state. Introduced to set the scene for the skulduggery that is to come, the suggestion that speaking ‘beautiful English’ correlates with good character would probably have seemed entirely plausible to many of Maugham’s readers. Attitudes have of course changed, in postcolonial settings as elsewhere, and not least in Singapore, a country now promoting itself as a model not only of rapid socio-economic growth but also of the successful integration of ‘a hundred peoples’ into a social grid in which the benefits of that growth can be shared by all. From pre-colonial times when those ‘hundred peoples’ began to interact, language has played a key part in that social integration — or in social division, as the case may be. Since it was ejected from newly independent Malaysia in 1965, Singapore has implemented a number of major language-related policies, including designating one language as the mother tongue of each of three official ethnic groups (Chinese, Malay, Indian), promoting Mandarin to replace Chinese dialects among Chinese Singaporeans, and making English the dominant medium of instruction. In

92 Paul Bruthiaux 2000, policymakers turned their attentions to English as it was being used (or not used, or, as they saw it, misused), not only within the relatively easily controlled institutional framework of education but also in the daily lives of Singaporeans. This resulted in one of Singapore’s trademark campaigns, the Speak Good English Movement (henceforth SGEM), a systematic attempt to influence the English language as used locally by steering it away from indigenized adaptations and closer to something internationally recognizable as standard English. SGEM now broadcasts its exhortations to Singaporeans to modify their linguistic behavior in the desired direction through a combination of booklets, billboards, public events, speeches by politicians, media coverage, and in particular a substantial website (Speak Good English Movement 2007). This website offers (among other things) word games, a story-telling competition, and a narrative competition in which successful efforts are promised an animated film version. It even boasts ‘Be Understood’ e-cards, with rulings such as ‘Wrong: Why you so like that? — Right: Why are you behaving in such a manner?’1 replacing the inspirational poems and risqué jokes normally favoured by users of these products, an initiative that will surely redefine birthday fun in Singapore. In this chapter, I propose to evaluate SGEM not as an academic might analyze it but as a web user might experience it through the content — and the subtext — of its website. Surveying the website, five major themes emerge. These are: intelligibility, the validity of linguistic descriptions, the prescriptivist tradition, the role of evaluation, and the role of government in language matters. I will review each of these themes in turn. Although the website stresses that SGEM is a movement and not a campaign, it does not elaborate on what this distinction might mean and why it might matter. To avoid taking sides unwittingly, I refer to SGEM as a project throughout this chapter.

SGEM and intelligibility The thinking behind SGEM is articulated on its website (as of 2007) in two speeches. One was given in 2006 by Minister of State for Education Lui Tuck Yew, the other in 2007 by SGEM Chair Professor Koh Tai Ann. Given the prominence given to these two documents on the website, it is fair to assume that they jointly represent SGEM philosophy, and I will refer to them collectively as SGEM without further distinguishing between the two. The two documents restate what has been a central SGEM tenet since its inception, namely that Singaporeans will gain collectively from being able to plug into an increasingly globalized world (for a discussion of some possible motivations — some overt, others less so — underlying this belief,

The Speak Good English Movement 93 see Rubdy 2001). Singapore will make itself more attractive to international investment, trade, and tourism if its population can communicate effectively with outsiders. This implies the ability to speak a variety of English broadly intelligible across English-speaking communities including the largest of all, that of non-native speakers. This, in a nutshell, is the message behind the SGEM slogan: ‘Be understood. Not only in Singapore, Malaysia, or Batam.’ (the last being the Indonesian island closest to Singapore, now undergoing rapid development as a result of investment from Singapore). One problem with slogans is that they often do all too well what they are meant to do, which is to galvanize through simplification. It is obvious from the moment a conversation between a Singaporean and a nonSingaporean starts that Singapore English sounds different. Today, a solid corpus of research details where phonological differences lie, especially at the segmental level (see, for example, Lim 2004a; Deterding 2005). Research also shows that differences affect suprasegmental aspects of Singapore English, including the prosody of given versus new information (Low 2006). However, difference does not entail unintelligibility, though it will figure among prima facie causes of unintelligibility when this occurs. As Low (2006) argues, intelligibility can be tested and its causes investigated. Whether SGEM pronouncements on how Singaporeans should speak are grounded in research of this type or in the untested hunches of its promoters is not clear from the website. Second, it is not immediately clear why an internationally intelligible variety of English should become the norm internally. Not only should Singaporeans ‘speak grammatically correct English that is universally understood’. They should also ‘speak English with confidence at work, at home, and at play’. This implies that communication across a range of settings in a highly multilingual society can be ordered in such a way that it will only take place in one of (for most Singaporeans) two approved languages, the designated ethnic mother tongue and standard English. At no point shall the twain meet, which of course it is in the nature of languages in contact to do, an aspect of bilingualism which SGEM concedes but sternly advises against. In this, SGEM appears to have softened its earlier position, at the cost, it seems to me, of greater confusion. On the one hand, SGEM states that it now accepts a Singapore accent as legitimate. Confusingly, however, it also urges Singaporeans to ‘pay attention to accurate pronunciation’, by which, judging by a number of items on the website, would appear to mean adhering to standard English in distinguishing between tense and lax vowels such as /i/ and /ܼ/, avoiding consonant cluster reduction (don’t want, not don wan), avoiding the devoicing of final stops (log in should not be homophonous with lock in), and adhering to standard stress placement (‘purchase, not pur’chase). Sensibly, SGEM is happy to sanction lexical imports denoting specifically

94 Paul Bruthiaux Singaporean entities and practices, as in HDB [public] housing or handphones. In matters of morphosyntax, however, the original prohibition stands. Singaporeans, we learn, should resist the temptation to ‘slip into inattentive, sloppy usage’, ‘lapse’ into hybridized English, or blame on ‘pressure of time’ a tendency to ‘cut corners’ with tense or agreement markers. They should, as I read it, emulate Maugham’s Ong Chi Seng and through their use of language prove themselves to be ‘industrious, obliging, and of exemplary character’. In practice, Singapore’s ‘hundred peoples’ have always had a range of interconnected speech forms that allowed them to transact business and bind themselves together as a regional community and more recently as a nation. As successive generations of immigrants settled in Singapore, they blended their own linguistic heritage with Malay, the dominant indigenous language, and Hokkien, the dominant tongue of the fast-growing Chinese immigrant community, into a patchwork of unregulated hybrids, with Bazaar/Trade Malay the main vehicle for interethnic communication. Gradually, English came to displace Bazaar/Trade Malay as the dominant lingua franca, especially after it was granted privileged status in the newly independent nation in 1965 and a privileged role as the sole medium of instruction in 1987 (for overviews of Singapore’s multi-ethnic and multilingual history, see Gupta 2007; Lim 2007, this volume; for a detailed study of the early ethnolinguistics of Singapore, see Miksic 2004). This, however, could not prevent (or perhaps even encouraged) continued crosshybridization between English and other local languages, with the resulting creole, popularly known as Singlish, settling increasingly comfortably into the role of preferred linguistic vehicle for informal communication as well as identity construction. In effect, this diglossic state of affairs created the raison d’être for SGEM, despite statements that SGEM ‘is not about Singlish’. In the very act of promoting standard English, SGEM aims to demote the competition, that is, Singapore English. This insistence that Singaporeans distance themselves from their home-grown variety of English in all areas of their lives implies that the linguistic kaleidoscope of Singapore must be reduced to a monochrome of English plus mother tongue, with no room for blending. The case for this view of multilingualism in action is not helped by justifications recycled from the stock in trade of colonial racism, as in ‘certain mispronunciations can lead to unintelligibility, for example, wishing someone a pleasant fright instead of a pleasant flight’. But apart from avoiding unfortunate statements such as this, SGEM needs to publicly address the nature of intelligibility beyond repeating that standard English is more intelligible than Singapore English. Evidence that departures from standard English do not necessarily hinder but may in fact enhance multilingual communication is offered by

The Speak Good English Movement 95 Deterding and Kirkpatrick (2006). In a study of interactions across ASEAN nationals speaking English, they found no evidence that departures from standard English such as the avoidance of reduced vowels in unstressed syllables led to unintelligibility provided the feature was shared among speakers. As Deterding and Kirkpatrick suggest, even native speakers unfamiliar with emerging contact varieties of this type may find that willingness on their part to depart from standard features may well assist intelligibility, not only in phonology but also (I would add) in morphosyntax. The grammar of standard English requires relatively frequent inflections, many of which add little or nothing to semantic content. For example, it seems entirely reasonable — as well as empirically testable — to suggest that yesterday she go may be more readily intelligible than yesterday she went since the stem-changing inflection adds nothing to the temporal reference already carried by yesterday while the morphologically unrelated lexical item went adds to the comprehension (and acquisition) load. Similarly, it is not self-evident that the item OK is by itself any more intelligible than the item can, a favorite Singaporean strategy for indicating ability and willingness to do something (Today cannot, tomorrow can, Sunday also can). What makes OK the preferred response is not inherent intelligibility, especially given its obscure etymology (discussed at length in numerous websites), but simply the fact that it has now passed into standard practice, first in US English, then more widely among other varieties of English, and now in many languages worldwide. Whatever SGEM is about, I suggest, it is not primarily intelligibility. Further, the notion that only one variety of English is acceptable in Singapore is at odds with the everyday reality of how multilinguals order their affairs, which is by shifting across codes as circumstances require (for a review of the vast literature on shift and accommodation, see Coupland 2007). What SGEM at no point addresses is the possibility that multilingual Singaporeans might be behaving entirely rationally by speaking standard English (with local characteristics) in some contexts, Singlish in others, and something in between in the rest. In other words, raising the competence of Singaporeans at the standard end of the continuum need not imply obliterating it at the other. The case for Singaporeans speaking standard English in multinational professional settings is obvious. But a rationale for insisting on standard English in private domains is missing, if only because informal spoken language cannot be controlled to this degree (Bruthiaux 2006). But this insistence is also absurd because if it were to succeed it would generate possibly the first recorded case of single-code, variation-free speakers of a language. I am not, of course, recommending that Singapore schools teach Singlish for academic purposes. But I am stressing that language policy

96 Paul Bruthiaux in Singapore needs to follow practice and recognize that there is room in Singapore for standard English, Singlish, and shades of Singapore English in between. Meanwhile, the suspicion remains that the main concern of SGEM is not intelligibility but respectability. Singaporeans, SGEM seems to be suggesting, need to see themselves and be seen as ‘industrious, obliging, and of exemplary character’, and this is incompatible with speaking Singapore English, let alone Singlish. In brief, SGEM needs to come clean in this respect and provide a coherent rationale for its activities beyond largely unsubstantiated claims that Singapore English is a major impediment to the involvement of Singaporeans in internal as well as international communication. What is at stake is that all Singaporeans — and not just an elite — should enjoy a range of codes that spans the entire continuum from standard English to Singlish (in addition to the mother tongue). Once this is in place, a quasi-instinctive human ability to choose from the full range to suit each communicative setting should be trusted to operate without institutional intervention.

SGEM and descriptive linguistics Despite being implicitly supported by a range of academic figures of impressive pedigree, some of the materials on the SGEM website betray an alarming lack of awareness of elementary descriptive linguistics. Recycling a ruling from an online source2 that the words although and but can be used interchangeably (on the vague grounds that ‘they both aim to connect two ideas inside the same sentence’), the website warns that ‘but can only be used in the middle of a sentence whereas although can be used either at the beginning or in the middle of a sentence’. While this is syntactically true, the statement entirely misses the fact that the two conjunctions connect clauses with radically different types of dependence. In the examples provided by the website, substituting although for but in She likes him but [*although] doesn’t want to get married leads to nonsense (as well as ungrammaticality since the subordinate although- clause requires a subject whereas the coordinate but- clause does not), as does substituting but for although in They did not eat although [*but] they were hungry. Even if these elementary issues had been spotted, understood, and explained in terms intelligible to most Singaporeans, this would still ignore the important issue of register variation. That is, but is a high-frequency item that covers the entire range from formal to informal discourse whereas although is less frequent and generally confined to formal, typically written contexts (Francis and Kuþera 1982). In the unlikely event of Singaporeans taking any notice of the ruling in question, its application would result in some very odd discourse indeed: ungrammatical, semantically obscure,

The Speak Good English Movement 97 socially inappropriate, or some combination of the above: in short, precisely the opposite of what SGEM aims to achieve, which is to promote Englishlanguage communication in Singapore. Linguistic descriptions should of course be helpful in practical ways to those who consult them, and, as in all pedagogy, a degree of simplification is to be expected. In this sense, it is reasonable for SGEM to state, for example, that ‘learnt is the past tense of learn and is used as a verb (she only learnt about the news yesterday)’. By contrast, ‘learned is used as an adjective to refer to someone who has much knowledge (we consulted the learned professors)’. However, an opportunity is missed to add two key specifications. One concerns dialectal variation. Given the growing presence of USsourced material in TV, music, media, and IT in Singapore, it seems probable that the user who posed the question (assuming it is authentic), ‘Is there a difference between these words — learnt and learned? When do we use each of them?’, will have been exposed to variation between the past verbal inflections learned (predominantly US) and learnt (predominantly UK). Warning users that variation of this type is rife in the Anglo-American twilight of Singapore English — the SGEM chair herself goes for the learned variant — would have done genuine service to the advancement of linguistic awareness among SGEM users. The other missed opportunity concerns phonology as no warning is given regarding the bisyllabic pronunciation of the adjective learned, a relic — incidentally — from the eighteenth-century shift from two syllables to one in verbal derivations of this type. This change caused Jonathan Swift a great deal of spleen and led him to denounce the resulting consonantal clusters (which he blamed on poets looking to drop inconvenient syllables, and which later spread from poetry to everyday speech) as ‘harsh unharmonious Sounds, that none but a Northern Ear could endure ... [and] so jarring a Sound, and so difficult to utter, that I have often wondred [sic] how it could ever obtain’ (1712/2007).3 If objections from no less a literary figure than Swift could not prevent or redirect phonological shift in eighteenth-century England, what hope is there that SGEM might do any better?

SGEM and prescriptivism Over the past two and a half centuries, attempts to prescribe language use in detail have tended to focus on randomly selected features on which to issue absolute rulings (for a review, see Finegan 1980; for an analysis of the role in this process of Bishop Lowth, every anti-prescriptivist’s bête noire, see Tieken-Boon van Ostade 2008). Evidently, the SGEM website is firmly anchored in this tradition. Usage is prescribed on an odd set of items SGEM prescriptivists appear to harbour special animus towards, often regardless of

98 Paul Bruthiaux actual usage. Indeed, an additional reason for doubting that intelligibility is a primary concern for SGEM is that much of its website is devoted to ruling on minute details of usage that will likely be missed by most even in writing (except perhaps by professional proofreaders). This is especially true since variation across written varieties of English is limited, often trivial (learned versus learnt, for example) and generally inconsequential in communication (Bruthiaux 2006). An example of this concern with the minutiae of written usage affects a verb that ends in the highly productive -ize derivational morpheme. For some reason, SGEM selects incentivize for special denunciation, claiming that ‘there is no such word’. However, a Google search, a ready and reasonably reliable measure of current usage by large numbers of users of English, yields over half a million hits for this verb in the -ize spelling alone, a further 360,000 in the -ise variant, and over half a million again for the combined -iz/sing, -iz/sed, and -iz/ses inflected forms. Disturbing, for a word that does not exist. The SGEM website recycles from other websites similar pronouncements on arcane points of usage widely ignored by modern speakers of English worldwide. ‘Farther’, we are informed, ‘is an adjective used to refer to a greater distance or a more distant place. It is only used when we are speaking of real places and distances … The adjective further is used to mean more, extra or additional.’ Leaving aside the fact that further/farther would be more accurately described primarily as adverbs and only secondarily as adjectives, a Google search quickly reveals the inaccuracy of this description, with further scoring 669 million hits compared to 28 million for farther. To modern web users, further is the term of choice, covering distance (‘How much further apart at the top are the support pillars of the Akashi Kaikyo Bridge in Japan than they are at the base due to the curvature of the earth?’), addition (‘Bush announces further Myanmar sanctions’), and the metaphorical gray area in between (‘Tennessee coach Phillip Fulmer says his 15th-ranked Volunteers are further along than he expected going into the season opener’). The efforts of prescriptivists notwithstanding, farther looks destined to follow farthermore into oblivion, a form already described as obsolete in the 1913 revision of Webster’s 1828 dictionary (Webster 1828/1913/2007) and today scoring a measly 9,600 Google hits, mostly — tangentially but intriguingly — thanks to a lingering fondness for this term among computer programmers (‘Converted figures are inserted into HTML without additional scaling, because they are properly scaled at conversion. Farthermore, avoiding additional scaling noticeably improves image quality’). Pronouncements to the contrary, which any Singaporean with online access can challenge on the basis of usage, suggest an alarming lack of awareness of current developments in English usage among authorities on language within SGEM.

The Speak Good English Movement 99 In fact, rather than being a case of ‘good English’ versus bad, the unequal co-existence of further and farther in modern English appears to be a leftover from earlier pronunciations of the stressed vowel as /ԥ/ versus /‫ܤ‬/, as in words such as university, with the /‫ܤ‬/ variant for the -ver- syllable now obsolete except in the linguistic fossil varsity. This earlier competition between phonological variants also remains audible in the US (/ԥ/) versus UK (/‫ܤ‬/) pronunciations of terms such as Berkeley, Derby, or clerk. Judging by current preferences, the fate that befell varsity will before long afflict farther, SGEM notwithstanding. Competition between variants is simply a manifestation of the inherent messiness of natural language, which, to the prescriptivists’ dismay, no amount of rationalization can affect substantially. In the unlikely event of Singaporeans taking notice of such rulings, promoting obsolete forms repackaged as ‘Good English’ would lead to Singaporeans being not more but less widely understood. A more likely outcome is that the SGEM watchdogs will meet with the fate of earlier prescriptivists: ignored in their time and forgotten soon after.

SGEM and evaluation Singapore is perhaps one of the most planned and monitored societies in the world, with little escaping the attention of the city-state’s vast governmental machinery. In addition to uncontroversial areas such as the economy and investment, housing, transport, or education, areas of Singapore life considered legitimate targets for government intervention have included courtesy in the service sector, flushing etiquette in public restrooms, or the matrimonial (dis)inclinations of Singaporeans (Romancing Singapore 2007). Best known among policy initiatives targeting language is the Speak Mandarin Campaign, initiated in 1979 to encourage Chinese Singaporeans to speak Mandarin at the expense of Chinese dialects. As in all policy implementation, it is to be expected that a policy tool such as SGEM will be the object of meticulous research, leading to continuous adjustments as results are processed and progress evaluated. What does the SGEM website, therefore, offer in this respect? In a section on ‘Background’, the website reports having been ‘especially successful in reaching out to Singaporeans’ with the help of ‘partner organizations’, and to have enabled Singaporeans ‘to learn from a wide range of activities and courses’. Reaching out to targeted participants and beneficiaries is of course a key step in any social program, but it does not guarantee that the program will be taken up, much less that it will be effective in bringing about the desired change. Yet, the effectiveness of SGEM goes unreported on the website.

100 Paul Bruthiaux Under the heading ‘Over the Years’ the website does review the first six years of the project’s existence in some detail. It reports that, a year into the project, a survey found that ‘9 in 10 Singaporeans agreed that it was important to speak good English’, with perhaps a hint of a causal relationship between the project itself and this happy outcome. In 2002, a survey by The Straits Times, a habitual government mouthpiece, found that ‘66% of Singaporeans who were aware of the Movement were motivated to speak good English’, a finding for which the line of causality — if indeed there was one — could have gone either way. Nothing further is reported regarding the effectiveness of the project for the rest of the period covered (2000–05). Given that, as mentioned above, the project can hardly have unfolded over this period without careful evaluation, this lack of publicly available information is puzzling, and it is left to an anonymous Wikipedia contributor to suggest, in the absence of hard evidence, that the project ‘may have failed to substantially alter the diglossic character of language use in Singapore’ (Speak Good English 2007). If this view is in error, the SGEM promoters are silent on the issue. Admittedly, it would be unfair to expect a project of this type to report its own ineffectiveness on what is in effect a promotional website. But if demonstrable evidence that the project is succeeding is available, it is reasonable to suppose that it would be given pride of place on the website. After seven years of SGEM fanfare, the jury cannot still be out. In the absence of a public statement of outcomes, there will remain a suspicion that the news is not all good.

SGEM and the role of government In an attempt to bring religious practices among Tibetans under tighter control, the Chinese government recently created a reincarnation management programme, in effect a requirement that announcements regarding the divine succession line must be pre-approved by Beijing (Haines 2007). Though not quite matching this development in ambition, the SGEM project betrays a fear of uncontrolled social behaviour, especially if that behaviour is undergoing change. In Singapore, it often appears, nothing is too unimportant to be left unplanned and unmonitored, and language form and language use are no exceptions. Evidence abounds on the website of a view of English in Singapore as a problem, to which, naturally and at government initiative, a solution will be found and implemented top-down among the population. Indeed, the fact that a localized, non-standard variety persists and is even associated with identity construction seems to frustrate those in charge given that the problem appears to them have a ready solution. Making standard English the norm in Singapore ‘should not be difficult as all Singaporeans … who

The Speak Good English Movement 101 have attended English medium schools or learned [sic] English as a second language will have been taught standard English’. All it should take is a willingness among Singaporeans to apply what they were taught. ‘Speaking good English among younger and all-English-speaking Singaporeans ... [is] mainly a matter of attitude, conscious effort, and attentiveness to the language that we have all learned at school.’ In this naively mechanistic view of teaching and learning, the blame for any failure rests squarely with the intended beneficiaries of the policy because they will not apply what they learned. To attack the problem at the source, government intervention reaches not only into schools but also into family life: ‘Take the time to help your child be sensitive to common mistakes made in spoken English. You can have him or her tell you about the things that happen in school. This will help sharpen his/her senses to spot common mistakes made in everyday conversations.’ One difficulty with this recommendation is that it rests on the questionable assumption that the older generation will have a better command of English than the younger. Given that English was made the sole medium of instruction in Singapore only in 1987 or less than a generation ago (Pakir 2004), it is likely that English language input within families will flow not from parents, many of whom may have missed out on English medium education, but from children, who will bring home greater exposure to the language and a confidence to communicate in it across the range of daily activities. To its credit, SGEM recognizes that some parents may not be up to the task of tutoring their children in standard English. However, the proposed solution is a non sequitur, and the result is conceptual incoherence: ‘If ... [parents] are not comfortable with English, they can speak their mother tongues with their children so that their children will develop the ability to communicate well in their mother tongue.’ Quite how this is supposed to help with the acquisition and use of standard English is not specified, an omission that does little to raise the overall credibility of the SGEM project. Another difficulty is that even if most parents feel sufficiently knowledgeable and confident regarding standard English (thanks, perhaps, to their regular perusal of SGEM output) to act as home tutors to their children and turn SGEM recommendations into family routines, this may further raise anxiety levels on both sides as the relentless testing culture that characterizes Singapore education penetrates further into the home. No matter: the British Council is at hand, the SGEM website informs us, to deal with any emotional side-effects through workshops designed ‘to get Singaporeans to speak up and talk about their feelings and emotions creatively’, as though creativity were a matter of training. In practice, postings on countless blogs show that Singaporeans need no encouragement to speak creatively. All they need is space. The ‘problem’ is that they tend

102 Paul Bruthiaux to do it in a variety of English that has become strongly localized. The difficulty for SGEM is that the proposed ‘solution’ cannot simultaneously give free rein to creativity and tighten the monitoring required to steer Singapore English closer to standard English. The suggestion that Singaporeans should monitor their language output closely targets the language behaviour not only of families but also of individuals: ‘Put yourself on the spot and audit the things you say daily. Once you’ve identified the non-standard English phrases you usually use, think of how you can rephrase them in Standard English.’ Here too, the underlying assumption seems to be that Singaporeans are conversant with standard English but are not making the effort to speak it. ‘There are no short cuts in speaking English well. Even if you are in a hurry, you need to remember that slipping into non-standard English out of convenience is unacceptable.’ In fact, far from being ‘unacceptable’, the ability to shift across varieties to suit different social settings is a sine qua non of inclusion in social networks. It is an ability that very young children, second language learners, and severely asocial individuals typically lack. But for well-adjusted adults, to ‘slip into’ another variety, as SGEM censoriously puts it, is an entirely appropriate response to shifts in the social context, especially in a linguistically complex society such as Singapore. Indeed, its absence would be evidence not only of lack of competence (you cannot switch into a variety you are not competent in) but more importantly, among fluent speakers, of social maladaptation bordering on pathology. In brief, Singaporeans know what they are doing when they shift across varieties. This communicative strategy is so fundamental that it should be regarded as a defining characteristic of every socially well-adjusted human, which no amount of intervention by language planners will suppress or even modify substantially. Moreover, the very existence of SGEM and many of the items on its website betray a mindset of condescension on the part of the Singapore leadership. Singaporeans, it seems, cannot be trusted to decide for themselves what is appropriate in language use, as though without benefit of governmental guidance a typical Singaporean might launch into a job interview with the words Eh, gimme job, leh? By SGEM’s own (unwitting) admission, a population that, judging by two items selected for remediation, is familiar with words such as alleviate and ricochet but expects governmentsponsored instructions on how to pronounce them has a problem not with English but its own passivity in the face of hyperactive government. Yet, not content with tweaking the minutiae of English phonology, SGEM also sets itself the task of equipping Singaporeans with a language that is ‘not just a tool for passing examinations and conducting business’. Many Singaporeans, we learn, ‘lack the ability to connect with another human being at a deeper level … We need emotional literacy to create the social glue that holds us together as a people and as a nation. The language

The Speak Good English Movement 103 of the heart, of feelings and emotions is the key to connecting with our parents, children, colleagues, our neighbours and especially, our loved ones.’ Many Singaporeans would retort that Singapore already has always had such tools in the shape of local languages: mother tongues, dialects, and of course Singlish, a localized hybrid that many Singaporeans — including teachers — cheerfully admit adequately meets every one of the social and emotional needs listed above. Singaporeans need no instruction in expressing ‘sadness, anger, fear and love’, and neither Singlish nor any of the mother tongues are known for lacking expressive potential. What SGEM presumably means is that English in Singapore cannot yet compete with Singlish or with local languages across the functional range but that the situation can be altered through policy. However, provided Singaporeans have access to standard English for communicative settings where it is appropriate, there is no reason why other languages or varieties of English should not coexist and fulfil more interpersonal needs, as individuals and groups may choose. If ‘our children tell us they are OK’ without specifying further, it is not, I suspect, because the term OK is the only tool in their linguistic kit — though it may at times seem that way. If a teenager were to stray far beyond OK and whatever in response to parental inquisitions over where the afternoon was spent and instead launched into a full report in standard English adorned with ‘expressive’ terms, that teenager would soon run out of friends to, like, hang and do cool stuff with. The ubiquitous reliance by teenagers on responses such as OK is not a linguistic issue but a socio-developmental one. It is after all in the nature of teenagers to communicate with parents reluctantly and minimally, especially when pressured to do this along norms they associate with authority and, in particular, with school. This may be regrettable, but it is probably beyond the capabilities of even the Singapore government to re-engineer the teenage mind. To be sure, a sense of omnipotence is not unusual in rulers. But for rule to be effective, rulers must temper this side effect of power — and especially of near-absolute power — with the ability to remain in touch with the ruled. To the extent SGEM policymakers seriously hope to achieve anything of substance in the matter of language in Singapore, they appear curiously out of touch. If they really believe that a valid model for standard English is ‘the lyrics of all kinds of music’, it is time they took an interest in the songs their children download to their iPods. If they really believe that communication with ‘the friend from Prague or Peru encountered online’ is conducted in standard English, it is time they looked at what goes on in chat rooms, instant messaging, blogs, forums, Facebook, Twitter, and the like. Clearly, it is an advantage for authority figures to be convinced of their own infallibility since it eliminates the need to regularly test the validity of their beliefs against public perceptions. Whether it makes their policies

104 Paul Bruthiaux effective is another matter, which is perhaps just as well since the best that can be hoped of the SGEM project is that it do no harm while implemented and quickly fade from public view once its irrelevance is recognized.

Conclusion This chapter has reviewed the Speak Good English Movement website (as of 2007), which is devoted to promoting standard English in Singapore. It has noted that five broad themes emerged from the claims and language-related activities on display. It argued that (1) a concern for respectability, not intelligibility, is a major driving force behind SGEM; (2) the linguistic descriptions used in support of SGEM aims are often incomplete, unconvincing, and occasionally plain wrong; (3) much of the detailed advice on usage harks back to a tradition of prescriptivism that takes little account of actual practice and in particular of linguistic hybridity in multilingual societies; (4) little evaluation appears to have been carried out (or is reported) on the effectiveness of the project; and (5) the project fits into a pattern of intervention by government in all spheres of Singapore life including language matters. In brief, the SGEM project is a slick but confused, confusing, outdated, and ultimately irrelevant case of governmental overreach. If not SGEM, then, what would be an appropriate policy response to English as it happens to be spoken in Singapore? First, the widely held view that both international and internal communication can only be effectively transacted through standard English should be publicly challenged. SGEM promoters may not like the Singapore variant, but if it did not work, it would have disappeared long ago. If English in Singapore needs to be reengineered, that aim should be backed up by a rationale based on evidence of language in use, not by unsupported assertions that too often rely for effect on awed references to former prime minister and SGEM initiator Goh Chok Tong that echo Swift’s reverential nods to ‘the Most Honourable Robert, Earl of Oxford’ (Swift 1712/2007). Second, facile dichotomies between correct versus incorrect English need to be finessed. Language, and especially language in Singapore, cannot be reduced to a vertical, ‘good’ versus ‘bad’ representation even for pedagogical purposes. English in Singapore is probably best understood (and explained) along an acrolectal-mesolectal-basilectal, English-Singlish continuum. In the mesolectal range, a growing body of research (for a theoretical framework, see Gupta 1998) shows that far for being an unseemly deviation from standard English indulged in by ‘sloppy’ speakers, Singapore English is highly rule-governed (for a demonstration of this key point, see the collection of papers in Lim 2004b; also Alsagoff and Ho

The Speak Good English Movement 105 1998a for syntax, and Wee 2003, Wong 2004, and Lim 2007 for discourse). Towards the basilectal end of the continuum, there is even a case for believing that the speech form in question is not primarily English but a contact variety built upon a Chinese substratum (Alsagoff and Ho 1998b; Bao and Wee 1999; Ansaldo 2004; Bao 2005; Bao and Hong 2006). At the practical level, the fallacy whereby standard English is the only appropriate linguistic vehicle for effective communication within a multilingual society should be exposed (Bruthiaux 2008). It is reasonable to assume that all Singaporeans will wish to enjoy the potential benefits provided by standard English in terms of increased employment opportunities, for example, provided those can be demonstrated and not simply believed in. Indeed, given that acrolectal English now functions as the prestige variety in Singapore, it would be highly reprehensible for a government to deliberately deny ownership of it to sections of its population. However, Singaporeans also bond with each other in endlessly shifting alliances, a process that takes place linguistically somewhere within the basi- to mesolectal range on the continuum even as a distinctly Singaporean identity emerges linked in part to non-acrolectal language use. (For a summary of seminal and recent research on language variation and identity, see Coupland 2007; for a detailed analysis of the mechanisms underlying these identity shifts in the specific context of Singapore, see Gupta 2000.) Since humans are generally adept at deciding for themselves (mostly subconsciously) where on the lectal continuum to pitch each message for optimal communicative effect, the issue becomes one of access to the lectal range. Those who are fluent across the entire continuum will glide along it effortlessly as each communicative setting requires. By contrast, given the linguistic makeup of modern Singapore, those whose lectal range is confined to the basilectal end of the continuum will most likely suffer. If, as Bao and Hong (2006) argue, more than mere register variation is at play and the basilectal end of the continuum owes more to a Chinese substratum than to an English one, those finding themselves confined to that end of the range face the added burden of having, in effect, to learn an additional language (that is, English) if they are to escape sociolinguistic isolation. The appropriate policy response is to abandon a misconceived and ineffectual attempt to subtractively campaign basi- to mesolectal English out of existence in Singapore and to devote public resources instead to additively helping less privileged or less lucky Singaporeans become comfortable at the acrolectal end of the continuum. As SGEM itself notes, many Singaporeans ‘are extremely proficient speakers and users of the English language’ despite the fact that they ‘come from non-English speaking homes’. Speaking standard English is clearly not the preserve of an elite but is achievable by large numbers of Singaporeans whose socio-

106 Paul Bruthiaux economic background would not have predicted success (Dixon 2005). Against the odds, thanks to lucky interactions with inspirational teachers or access to intensive reading, perhaps, these Singaporeans extended their linguistic range to the point where they can operate equally comfortably in settings where acrolectal resources are called for and in those where shifting to the meso- to basilectal range is not just preferable but socially required. For Singapore language policymakers, therefore, the challenge is to find out through research how these Singaporeans were able to extend their original range and then help all disadvantaged Singaporeans equip themselves with the acrolectal end of the range, not to deprive them of the basilectal end on the ground that it does not conform to some officially sanctioned variety or other, real or imagined. As the dénouement of Maugham’s story (featuring Ong Chi Seng’s complicity in blackmail and corruption) makes clear, speaking ‘beautiful English’ is no guarantee of respectability. Singaporeans fluent across the entire lectal continuum routinely demonstrate their ability to distribute their linguistic resources as each setting requires. In response, policymakers should accept that a wide lectal range is part of modern Singapore and work to enable disadvantaged Singaporeans to extend their own lectal range to include the acrolectal end. Policymakers should then trust Singaporeans to know where on the continuum to place each utterance, and leave Singlish to look after itself. Regrettably, in minutely planned Singapore, the notion that trust in the people’s ability to make sensible choices with little or no governmental guidance appears to be alien to the policy-making mindset. This predicts that, regardless of its apparent ineffectiveness, the SGEM project has a long and well-resourced future ahead of it.

Notes 1.

2.

3.

In this chapter, single quotes are used in the conventional manner to indicate text taken verbatim from academic or literary sources as well as to indicate text taken verbatim from the SGEM website itself. Italics indicate technical terms, individual lexical examples, or emphasis. The SGEM website relies primarily for information of this type on www. stomp.com.sg, an online offshoot of The Straits Times, and the English@work Newsletter, an online creation of Mediacorp, a local media provider offering (among other things) thirteen radio stations with output in four languages. Whatever his musical tastes, Swift would have been heartened, a full thirty years later, by Georg Friedrich Händel’s conservative setting of the libretto of Messiah, in which the aria He was despised, rejected, ... has both words sung trisyllabically.

The Speak Good English Movement 107

References Alsagoff, Lubna and Ho, Chee Lick (1998a) The grammar of Singapore English. In New Englishes: The Case of Singapore. Edited by Joseph Foley. Singapore: Singapore University Press, pp. 127–51. Alsagoff, Lubna and Ho, Chee Lick (1998b) The relative clause in colloquial Singapore English. World Englishes, 17, 127–38. Ansaldo, Umberto (2004) The evolution of Singapore English: Finding the matrix. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam: John Benjamins, pp. 127–49. Bao, Zhiming (2005) The aspectual system of Singapore English and the systemic substratist explanation. Journal of Linguistics, 41, 237–67. Bao, Zhiming and Hong, Huaqing (2006) Diglossia and register variation in Singapore English. World Englishes, 25, 105–14. Bao, Zhiming and Wee, Lionel (1999) The passive in Singapore English. World Englishes, 18, 1–12. Bruthiaux, Paul (2006) Restandardizing localized Englishes: Aspirations and limitations. Special Issue on Lingua Franca Communication: Standardization versus Self-Regularization, International Journal of the Sociology of Language, 177, 31–50. Bruthiaux, Paul (2008) Language education, economic development, and participation in the Greater Mekong Subregion. International Journal of Bilingual Education and Bilingualism, 11, 134–48. Coupland, Nikolas (2007) Style: Language Variation and Identity. Cambridge: Cambridge University Press. Deterding, David (2005) Emergent patterns in the vowels of Singapore English. English World-Wide, 26, 79–97. Deterding, David and Kirkpatrick, Andy (2006) Emerging southeast Asian Englishes and intelligibility. World Englishes, 25, 391–409. Dixon, L. Quentin (2005) Bilingual education policy in Singapore: An analysis of its sociohistorical roots and current academic outcomes. International Journal of Bilingual Education and Bilingualism, 8, 25–47. Finegan, Edward (1980) Attitudes Toward English Usage: The History of a War of Words. New York: Teachers College Press, Columbia University. Francis, W. Nelson and Kuþera, Henry (1982) Frequency Analysis of English Usage. New York. Houghton Mifflin. Gupta, Anthea F. (1998) A framework for the analysis of Singapore English. In Language, Society, and Education in Singapore: Issues and Trends. 2nd ed. Edited by S. Gopinathan, Anne Pakir, and Vanithamani Saravanan. Singapore: Marshall Cavendish Academic, pp. 119–32. Gupta, Anthea F. (2000) Bilingualism in the cosmopolis. International Journal of the Sociology of Language, 143, 107–19. Gupta, Anthea F. (2007) The language ecology of Singapore. In Encyclopedia of Language and Education: Ecology of Language. Edited by Angela Creese, Peter Martin and Nancy H. Hornberger. Dordrecht: Kluwer, pp. 99–111. Haines, Lester (2007) China slaps ban on reincarnation. http://www.theregister. co.uk/2007/08/29/reincarnation_ban. Accessed 15 November 2007. Lim, Lisa (2004a) Sounding Singaporean. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam: John Benjamins, pp. 19–56.

108 Paul Bruthiaux Lim, Lisa (ed.) (2004b) Singapore English: A Grammatical Description. Amsterdam: John Benjamins. Lim, Lisa (2007) Mergers and acquisitions: On the ages and origins of Singapore English particles. World Englishes, 26, 446–73. Lim, Lisa (2010) Migrants and ‘mother tongues’: Extralinguistic forces in the ecology of English in Singapore. In English in Singapore: Modernity and Management. Edited by Lisa Lim, Anne Pakir and Lionel Wee. Hong Kong: Hong Kong University Press, pp. 19–54. Low, Ee Ling (2006) A crossvarietal comparison of deaccenting and given information: Implications for international intelligibility and pronunciation teaching. TESOL Quarterly, 40, 739–61. Maugham, W. Somerset (1927/2001) The Great Exotic Novels and Short Stories of Somerset Maugham. New York: Carroll & Graf. Miksic, John N. (2004) 14th-century Singapore: A port of trade. In Early Singapore: 1300s–1819. Edited by John N. Miksic and Cheryl-Ann Low Mei Gek. Singapore: Singapore History Museum, pp. 41–54. Pakir, Anne (2004) Medium-of-instruction policy in Singapore. In Medium of Instruction Policies: Which Agenda? Whose Agenda? Edited by James W. Tollefson and Amy B.M. Tsui. Mahwah, NJ: Lawrence Erlbaum, pp. 117–34. Romancing Singapore (2007) http://www.romancingsingapore.com/index.php. Accessed 15 November 2007. Rubdy, Rani (2001) Creative destruction: Singapore’s Speak Good English. World Englishes, 20, 341–55. Speak Good English (2007) http://en.wikipedia.org/wiki/Speak_Good_English_ Movement. Accessed 15 November 2007. Speak Good English Movement (2007) http://www.goodenglish.org.sg/site. Accessed 15 November 2007. Swift, Jonathan (1712/2007) A proposal for correcting, improving, and ascertaining the English tongue. http://andromeda.rutgers.edu/~jlynch/Texts/proposal. html. Accessed 15 November 2007. Tieken-Boon van Ostade, Ingrid (2008) Eighteenth-century prescriptivism and the norm of correctness. In The Handbook of the History of English. Edited by Ans van Kemenade and Bettelou Los. Oxford: Blackwell, pp. 538–57. Webster, Noah (1828/1913/2007) Webster’s Revised Unabridged Dictionary. http:// machaut.uchicago.edu/websters. Accessed 15 November 2007. Wee, Lionel (2003) The birth of a particle: ‘Know’ in colloquial Singapore. World Englishes, 22, 5–13. Wong, Jock (2004) The particles of Singapore English: A semantic and cultural interpretation. Journal of Pragmatics, 36, 739–93.

5

Hybridity in ways of speaking: The glocalization of English in Singapore Lubna Alsagoff

English in Singapore has been described as serving both as a global language of trade, commerce, science and technology, as well as a local language for interethnic communication. Given the ideological incongruence and often conflicting histories and goals associated with these roles, researchers such as Pakir (1994, 1999, 2001), Rubdy (2001), Wee (2003), Bokhorst-Heng (2005), and Chew (2006) have talked about English in Singapore as being pulled in two opposing directions. Pakir, for example, discusses the competing norms that these divergent ideologies assume: the global orientation necessitates the adoption of an English that is internationally standardized, while the local advocates the development of an indigenized English. In this chapter, I extend the discussion of the global-local tension to offer a model of variation of English in indigenized contexts such as Singapore.1 The model, originally introduced as the Cultural Orientation Model (COM) in Alsagoff (2007), explains variation in relation to the global-local contrast of the cultural orientations of speakers. The discussion here develops the model further by looking at the global-local interactions in relation to the concept of ‘glocalization’ as originally formulated by Robertson (1995, 1997). Key to Robertson’s conceptualization of glocalization is the emphasis of the simultaneity of the global and the local in the process of globalization. Expanding on this concept, I discuss how the co-presence of features of both local and global in the speech of Singaporeans necessitates a change in the approach to describing language variation in Singapore.

Variation as glocalization The use of the notion of ‘glocal’ or ‘glocalism’ in characterizations of Singapore English is not new: both Pakir (e.g., 2000, 2001) and Bokhorst-

110 Lubna Alsagoff Heng (2005) have used the term in their discussions of Singapore English. Pakir’s (2000: 18) use of the term specifically relates to the ways in which English as a global language of international communication has taken root in ‘different soils’, and is therefore ‘subject to the forces of localization’ — as a result of which English is said to be glocal. The idea of English as glocal can, in fact, be found in Widdowson’s characterization of the ‘spread’ of English as an International Language (EIL) as English being ‘actualized in diverse ways, subject to local constraints and control’ (Widdowson 1997: 139–40). Bokhorst-Heng, on the other hand, uses the term ‘glocalism’ to refer to the juxtaposition of two perspectives. In speaking of the global, Bokhorst-Heng (2005: 185) refers to internationalism as perspectives of ‘the nation looking outside from itself at the world and global economy, and defining itself and its needs in relation to its position within the world’. This is in contrast with the local, or what Bokhorst-Heng (2005: 185) describes as issues relating to national identity, viz. ‘the nation looking within itself at its societal structure and composition, defining its needs in relation to national identity’. In an effort to present an even more fluid and dynamic understanding of language variation, the model explored here, to which I will refer as the Glocalization model, develops COM as a more articulated conceptualization of variation in terms of glocalization. The concept of glocalization, as a global-versus-local tension, is a well-established one (e.g., Block and Cameron 2002; Canagarajah 2005, 2006; Rubdy and Saraceni 2006; Pennycook 2007), although the actual neologism may be less frequently used. Instead, related terms and phrases such as ‘local globalization’, ‘global localization’ or hybridization (Pieterse 2006: 280) that offer a view of the two-way flows between the local and the global are more common. In adopting the term ‘glocalization’, I wish to emphasize an interpretation originating from Robertson’s (1995, 1997) formulation of the term, which characterizes the process as one in which both local and global exist in simultaneity, creating a fluid push-pull effect. Robertson’s use of this neologism reflects his understanding of globalization as a process which is grounded in the local even while taking into account global perspectives and actions. Robertson describes glocalization as ‘the simultaneity — the co-presence — of both universalizing and particularizing tendencies’ (Robertson 1997). In other words, the term ‘glocalization’ foregrounds the understanding that in the dichotomies of the global-local, universalistparticularist, homogeneity-heterogeneity, ‘indigenization is [simply] the other side of the coin of the homogenizing aspects of globalization’ (Robertson 1997, italics as in original). To offer insight into what such simultaneity of the global-local means, we look to the original Japanese term dochakuka that Robertson draws on in formulating an understanding of globalization as the meeting of global and local forces:

Hybridity in ways of speaking 111 Now, very briefly, to the term ‘glocalization’, as I derive it from the Japanese term dochakuka. As it is used in Japanese business practice, this term actually refers to the selling, or making of products for particular markets. And as I think most of us here know, Japanese business people have been particularly successful in selling their products in a variety of different markets … But the basic idea of glocalization is the simultaneous promotion of what is, in one sense, a standardized product, for particular markets, in particular flavors, and so on. (Robertson 1997)

The success of many Japanese multinational corporations has depended on their ability to take a standardized product globally to new markets — by adapting that product to the local needs of the users of the new markets, yet keeping to standard processes of manufacture, quality and customer service — hence, dochakuka re-frames our understanding of the process of globalization as glocalization to emphasize the local existing simultaneously with the global. In applying glocalization to a study of variation, I embrace two key features: the co-presence of the global-local, and the multidimensionality of the process of glocalization. The process of globalization and its effects relate particularly to the spread of English globally to a multiplicity of speech communities. Pieterse (2006: 278) suggests that there are essentially three paradigms for understanding the cultural effects or outcomes of globalization. I apply them here to the discussion of the different ideological perspectives on the international spread of English. The first is where the spread is seen as a homogenizing force that results in cultural sameness. This perspective, known as linguistic imperialism (Phillipson 1992), argues for a view of the globalization of English as hegemonic and destructive of local cultures and practices. Thus, proponents of this paradigm would see the spread of English as eroding local languages and cultures. While valid in many respects, I would argue that this position is not entirely tenable for a description of the use of English in Singapore, and other such contexts where English is spoken in a multilingual environment, and maintained as a working language because it is seen by its speakers (and not just by the government) as economically and instrumentally useful. Erosion, as I will argue shortly in my discussion of the third paradigm of globalization, is often accompanied by an accommodation of English to its new speakers and circumstances. The second paradigm described by Pieterse (2006) is where the spread is seen as surface deep, meaning that its adoption as an additional language by speakers does not deeply affect the lingua-cultural practices of the speech community. Consequently, such a spread is seen as non-assimilationist, since English is seen as existing separately and apart from the indigenous cultures, and it is in this sense that the spread is seen as ‘surface deep’ — not interfering with or impinging on the cultural differentiation and

112 Lubna Alsagoff variety in the local contexts. It is this perspective that is at the heart of the Singapore government’s attempt to differentiate the roles of the ‘mother tongue’ languages such as Chinese (Mandarin), Malay and Tamil from that of English, where the former are to serve as ‘local languages’ and the latter as a global language. As Wee (2003: 211) points out: In Singapore, this move towards a more pragmatic view of language can be seen in how the relationship between the endogenous Asian languages and English is constructed. The former, officially known as the ‘mother tongues’, are tasked with preserving ethnic cultural traditions while the latter is presented as the language of economic development.

In line with Singapore’s pragmatism in economic policy, and the critical recognition of the importance of English to making Singapore an international banking and finance centre, language policy in Singapore necessarily depicts English in purely pragmatic and economic terms. It is thus seen as ‘cultureless’ in the sense of it having no significance or role in the cultural traditions of Singapore (Ho and Alsagoff 1998; Alsagoff 2007). Such a position is clearly untenable, given what we know of the close relationship between language and culture, and our understanding of the ecological co-existence of languages in a speech community. Consequently, and not surprisingly, the Singapore government’s attempt to quell the growing popularity and use of the Singapore English vernacular, Singlish, through the Speak Good English Movement (Rubdy 2001) has given way to a quieter, qualified acceptance of it, with even the minister mentor Lee Kuan Yew conceding it as a de facto lingua franca: Up to the 1970s in our markets and hawker centres, Bazaar Malay was the lingua franca. Everybody could understand and speak some Malay. Because of our bilingual policy, today the lingua franca is English, or Singlish. (Lee 2005)

The third paradigm, which I argue is the most appropriate perspective in which to understand English in Singapore, is where the spread of English is seen as a force that leads to cultural mixing and hybridity. Closely aligned to this view is Kachru’s Three Circles Model of English (Kachru 1992, 1997). Typically, Englishes such as Singapore English, Nigerian English, and Indian English, i.e., those Englishes which have taken on an institutionalized role in the country, are seen as indigenized Englishes. However, while Kachru’s model emphasizes the pluralization of English through appropriation by local cultures, it is hybridization only in a limited sense (Bhatt 2005). Instead of a cultural mixing, the Three Circles model emphasizes the separateness and autonomy of these different Englishes, and presents the interaction between the local and the global as primarily a

Hybridity in ways of speaking 113 one-way flow from the local to the global, whereas cultural mixing entails a fluid bidirectional flow, in which heterogeneity encompasses the multiplicity of identities, its fluidity and interactions with discourse, and of a fluid bidirectional flow between the local and the global. Such a bidirectional flow is what glocalization seeks to emphasize in the translocal spread of language and culture. Just as local cultures change English, so does English change local cultures. The two, while in contrast, are co-existent forces. Whether one is a stronger force than the other is a point that can form the focus of debate. What is not debatable, however, is the two-way flow and interaction that is at work. It is not simply a case of there being a hegemonic domination of the global over the local, or an appropriation of the global by the local, but rather the co-existence and intermingling of both of these perspectives and cultures. Glocalization is useful in conceptualizing the interplay between the global and the local because it articulates a two-way flow of the local as a transformative force on the global, just as the global is seen to transform and shape the local. In representing the heterogeneity of the local and the homogeneity of the global as interwoven and co-present, glocalization is able to provide a richer basis for understanding the subtlety and fluidity of language variation.

Previous models of Singapore English In advocating glocalization as a perspective, my intention is to develop a more current model of language variation in Singapore. Previous models such as the lectal continuum model, and the diglossia model which have generally depicted variation as dialect shift or code-switching, while historically accurate, do not adequately describe contemporary patterns of use of English in Singapore. Patterns of acquisition and use of English in Singapore have continued to change at a rapid pace, and it is necessary to present a model of variation that more accurately reflects current patterns of use in today’s Singapore.2 The lectal continuum model developed by Platt and associates (Platt and Weber 1980; Ho and Platt 1993) was the first well-known model proposed for Singapore English. It described the variation of English in Singapore as a continuum of English that ranged from the standard, educated acrolectal variety to the mesolectal and the non-standard basilectal, and was an accurate depiction of English patterns of use during the 1950s to 1970s. During the time of British colonial rule, and for some years after that, it was clearly the case that the kind of English spoken depended on the socio-economic status (SES) and education of the speakers. The higher the SES of the speaker, the better educated they were, and thus the more likely their English approximated to the standard

114 Lubna Alsagoff dialect, and vice versa. But as English became more widespread and access to it more democratized, the language situation in Singapore changed. The use of non-standard English was no longer confined to the non-English educated, and was now also being used by the educated in a wider range of domains (Lim and Foley 2004: 5). These speakers commanded a standard variety of English — Standard Singapore English (SSE) — but also spoke a non-standard variety (Colloquial Singapore English — CSE) when the context moved from formal, institutional settings to more informal ones oriented around family or friends, suggesting that variation was changing to one patterned along function rather than education or SES. The diglossia model (Gupta 1994a, 1994b) ‘succeeded’ the lectal continuum model as the more widely referenced model for Singapore English — it focused on the use of English among educated Singaporeans, and presents the variation of English use in Singapore as register variation, in which speakers switch between two varieties, and in particular, use the non-standard dialect, not because they are constrained by their social positions (education and SES), but because of function or domain. 3 Diglossia gave a different ideological status to the non-standard local variety, because Singlish is presented as a language of choice, rather than constraint. Pakir’s (1991) expanding triangles model presented what was to be a more comprehensive model by combining the predictive scope of the lectal continuum model with the diglossia model, in which English is seen as varying along two clines — education and register. Pakir’s model is thus able to account for the type of English used by the significant percentage of the population of speakers in Singapore which still does not use English as a primary home language, something which the diglossia model omits in its theoretical span. However, even though the expanding triangles model talks about clines of variation, it still describes the use of English in Singapore in relation to two more or less well-defined varieties: SSE and CSE, the use of which is contextually determined by register or function. The primary difference between the expanding triangles model and the diglossia model lies in the former model’s development of the notion of the range of register that speakers with different educational backgrounds can command.

The Glocalization model In presenting the Cultural Orientation Model (COM) as an alternative to previous models, I have (Alsagoff 2007, forthcoming) argued that in modern Singapore, language variation cannot be simply determined by a matching of the social background of the user, and/or the function or register of the context. The profile of speakers in Singapore has changed in the last twenty years, with most Singaporeans now receiving at least ten

Hybridity in ways of speaking 115 years of education in English and an increasing number speaking English as their native language.4 In addition, as the range and domains of English spread to more personal and less institutional settings, English has taken on an increasingly local flavour, with speakers appropriating the language for their own communicative ends. This increased sense of ownership over the language has inevitably moved the language further away, even in formal contexts, from an acrolectal exonormative standard modelled on Standard British English. Consequently, the choice over what code to use is not simply one determined by function or domain. How and why people use language is, I believe, more adequately explored as style variation in relation to identity construction and cultural orientations. In COM, I therefore explored ideas of the global-local contrast as a tension between the forces of the global and the local in shaping the use of English in Singapore. Similarly, in the glocalization model presented here, variation in the form and use of English in Singapore is expressed as resulting from the different kinds of cultural demands made on English — one oriented towards global perspectives, the other towards the localist perspective; these are congruent and co-variant with the constructed identities of Singapore speakers who also see themselves in relation to local concerns versus global aspirations. On the one hand, the local(ist) perspective is separatist in nature. As such, its concern is separateness. As Singaporeans exert their national and their personal identities, there is an accompanying desire for a uniqueness of the English that ‘belongs’ to Singapore that must set its citizens apart from other English speakers. The local orientation is a result of Bakhtin’s centrifugal force, which pushes speakers away from the centre towards the periphery, emphasizing divergence (Bakhtin 1981). The globalist perspective, on the other hand, is assimilationist — consequently as Singaporeans forge identities that are global because of economic needs and aspirations, there is a tendency to foreground less differentiation and more congruence with perceived international norms. Thus, such an orientation is represented by the use of an English which is exonormatively defined, whose form and substance emphasize ‘same-ness’ with the rest of the world (Phillipson 1992), pushing speakers towards the centre. This idea is akin to the Bakhtinian concept of the centripetal force exerted on language by political and institutional forces that try to impose a standard variety over other varieties, so as to enforce the adoption of a unitary linguistic identity (Bakhtin 1981). For Singaporeans, the global and the local are counterpoised, and it is in the negotiations of identity between the global and the local that result in speakers varying their use of English to reflect such concerns. This variation, as I have argued in COM and as I argue here, is more aptly described in terms of stylistic shifts, where movement is not simply a binary shift between two well-defined varieties, but much more fluid and hybrid in

116 Lubna Alsagoff nature.5 Thus, in expressing the interplay and tension between global and local, speakers can vary their ways of speaking along a multidimensional continuum, which is defined by a range of variables. Table 5.1, modified from Alsagoff (2007), lists some of these variables. Table 5.1 The globalist and localist orientations Globalism

Localism

Economic capital

Socio-cultural capital

Authority

Camaraderie

Formality

Informality

Distance

Closeness

Educational attainment

Community membership

Relating the global and local identity orientations to stylistic variations along a specific set of social variables, as set out in Table 5.1, offers more concrete ways of understanding how language variation can be understood in terms of identity and culture, and is intended to obviate possible criticisms that identity as a sociolinguistic variable is too subjective and vague, something that has been levelled at Le Page and Tabouret-Keller’s (1985) model which explained code-switching in terms of linguistic ‘acts of identity’. Although COM attempts to present variation as style or more broadly ways of speaking, it does not go far enough in one respect — like previous models, COM still describes variation in relation to two distinct varieties, an international variety of English (SSE) that represents a globalist orientation, and a local variety of English (Singlish) that represents a localist orientation. Even though the intention was to present these two varieties as referential, i.e., idealized extremes of a multidimensional continuum of variation, rather than de facto codes between which speakers switched, their presence nonetheless weakens the explanatory force of the model. Clearly, if we are to offer a perspective of variation as style-shifting rather than dialect or code-switching, we need to move away from speaking in terms of varieties, and speak instead in terms of linguistic features.6 Hence, while both COM and the glocalization model presented here posit that speakers of Singapore English vary their style of speaking by exploiting the multidimensional space defined by the contrast between these two counterpoised cultural perspectives, a critical distinction between COM and glocalization lies in the way in which the variational space is linguistically defined. In COM, linguistic variation is described in relation to the contrastive use of referential varieties associated with contrasting macro-cultural perspectives, the glocalization model, on the other hand, references variation through the use of linguistic features which serve to

Hybridity in ways of speaking 117 index aspects of a speaker’s cultural orientations and identity orientations in the global-local spectrum. The use of linguistic and pragmatic features associated with establishing group or community membership and rapport, emphasizing solidarity as well as rootedness in indigenous culture, is aligned with a symbolic movement towards a local(ist) orientation that stresses socio-cultural capital (Bourdieu 1986, 1991). Contrastively, the use of features associated with formality, educational attainment, institutionalism and authority indicates a symbolic shift towards a global(ist) orientation, stressing economic capital (Bourdieu 1986, 1991). Speakers are seen as continually and constantly negotiating between the two contrastive cultural perspectives of the global and local, resulting in speech which displays degrees of variation, rather than poles of separation between two clearly defined varieties. The degree to which speakers vary their use of such features to reflect the different cultural orientations depends on the degree to which speakers wish to signal the associated ideologies, values or practices. Table 5.2 presents an example of the possible features that can be used to articulate the global-local variational continuum. Table 5.2 Globalist and localist features of Singapore English Globalism LexicoGrammar

Localism

Use of interrogative mood for expressing questions

Use of declarative mood for expressing questions

Fronted wh-question words in interrogative clauses

In-situ wh-question words in interrogative clauses

Use of verbal inflections (tense) to indicate time of event

Reliance on adverbs or contextual information to indicate time of event

Conditional clauses linked by subordinating conjunctions

Conditional clauses expressed without subordinating conjunctions

Use of disjuncts and parenthetic comments to express interpersonal meanings

Use of pragmatic particles to express interpersonal meanings

Absence of endogenous lexical items Inclusion of endogenous lexical or meanings, etc. items or meanings, etc. Phonology Presence of interdental fricatives in phonemic inventory

Absence of interdental fricatives in phonemic inventory

Consonant clusters present in careful speech

Final consonant clusters reduced even in careful speech

Stress-timed rhythm, etc.

Syllable-timed rhythm, etc.

Pragmatics Use of Western forms of address, etc.

Use of ethnic Asian forms of address, etc.

118 Lubna Alsagoff In re-presenting the stylistic variation of language use in terms of features, we are able to model variation as multiple-layered and incremental, i.e., as ‘thick variation’ (Geertz 1973), where speakers indicate or index degrees or dimensions of change in their ideological cultural orientations by selecting features associated with a localist or globalist orientation. Linguistic and pragmatic features associated with a particular orientation may cluster, or be used singly, depending on the degree to which the orientation is indexed, and on the feature used. For example, the pragmatic particle lah is an emblematic marker of community membership, and is often used at the end of utterances even in formal contexts to index ‘Singaporeanness’. Linguistic and pragmatic features may also not necessarily align in orientation. That is to say, speakers may choose features associated with a globalist orientation to indicate authority and yet include other features with a localist orientation that signal them as members of the community. For example, a speaker may choose to use the interrogative mood to ask a question, perhaps indexing educational attainment, but also append a pragmatic particle at the end of it to indicate closeness of the speaker to the addressee. The simultaneity or co-presence of linguistic and pragmatic features associated with both globalist and localist orientations is key to understanding variation in Singapore English. Kwek (2005), for example, shows that classroom teacher talk is the product of multiple forces at play, where the interconnectedness of possibly conflicting forces such as the need to maintain authority yet achieve rapport or the need to weave formality and informality as the moment necessitates, results in an English not easily cast as either standard English or Singlish because it contains both standard and non-standard forms within the same stretches of discourse (see Alsagoff 2007 for a fuller discussion). Such subtle and fluid shifts in ways of speaking can be explained only if we view variation in relation to the interweaving and co-presence of features associated with the two contrastive orientations, which give rise to differential degrees of variation within a multidimensional space.

What is Singlish? Moving away from speaking about variation in terms of a contrastive switch between a standard and non-standard variety offers insight into what Singlish is, and is not. In the lectal continuum model, Singlish is associated with the basilect, i.e., it is seen as the uneducated variety of Singapore English. In contrast, Singlish became associated with the colloquial or L variety in the diglossia model. Clearly, while the two entities — basilectal Singapore English and CSE — might exhibit similarities, I agree with Chew’s (1995: 165) observation that ‘it is unlikely that a Singaporean

Hybridity in ways of speaking 119 would mistake an educated English speaker speaking informally from an uneducated speaker’, and that ‘while they [educated English speakers] might use some lexical items associated with people with lower levels of education, they will never use others. They also use expressions which are only found in an educated repertoire.’ Kandiah (1998: 101) makes a similar point that CSE and basilectal Singapore English are neither identical entities nor ideologically homogeneous. Basilectal Singapore English refers to a variety that has little value, spoken only because of a lack of ability of the speaker, seen as a passive victim of social circumstances, while CSE refers to a variety valued as a functional variant, spoken as a consequence of active choice. In the glocalization model, we clearly present the speaker as an active agent. However, in doing so, we should not confine the scope to educated speakers, but extend it to ‘a whole wide spectrum of people in society’ (Kandiah 1998: 100) for whom English may not be a L1 in terms of acquisition, but a primary language of communication because of school or work. In developing variation as style, rather than code-switching, and in including educational attainment as well as register as part of the social variables that define the global-local variational space, the glocalization model is able to accommodate both the type of speech associated with the basilect as well as the colloquial variety, since these are able to exist in the global-local variational space. The idea that variation of English in Singapore is dynamic and fluid, and more fittingly described in terms of style-shifting rather than codeswitching is congruent with the idea that Singlish is less aptly characterized as a well-defined variety with clearly delineated boundaries, and more appropriately conceptualized as ways or styles of speaking that serve to foreground the localist orientations of the speaker. The lack of agreement among researchers investigating the structure of Singapore English as to what counts as acceptable or grammatical in Singlish lends support to this view. In the research literature on Singapore English, there have been a number of disagreements about grammaticality judgements and data, e.g., Wee’s (2002) critique of Gupta (1992) in discussing the acceptability and interpretations of the discourse particle lor; Wee and Ansaldo’s (2004) critique of data in Alsagoff and Ho (1998a) on the non-marking of number in bare noun phrases. There have also been differences in opinion in identifying the range of grammaticality of certain constructions, such as the relative clause (see Alsagoff and Ho 1998b; Newbrook 2003; Wee and Ansaldo 2004). Clearly, the dispute centres on what Singlish actually is. Before continuing, I necessarily digress on the following point. While I have no doubt that the data I have collected is valid, I am even more aware of the ways in which recent research relying on corpora has somehow assumed that what exists in the corpora presents a complete

120 Lubna Alsagoff picture of Singapore English. We must bear in mind that corpora are not incidental collections of data, but are designed and focused collections of data based on assumptions as to what counts as representative of Singlish, and who counts as representative speakers of Singlish. The corpora of Singapore English currently available focus primarily on the educated speaker, very much in line with the diglossia model. For example, both the NIE Corpus of Spoken Singapore English (Deterding and Low 2001) and the Grammar of Spoken Singapore English Corpus (Lim and Foley 2004) focus on the language of university student teachers and (largely) university undergraduates respectively. Evidence of the narrowness of corpus-based descriptions can, in particular, be seen in Deterding (2007), which is an entire book describing Singapore English based almost exclusively on a ‘corpus’ comprising one single speaker.7 What these disputes show is that the assumption that Singlish is a single stable language variety should be re-examined. Rather than try (and fail) to draw definitive boundaries around Singlish as a clearly distinct and well-bounded variety, we should acknowledge it for what it actually is — hybridized, creative ways of speaking that embody the local by tapping into an expansive set of linguistic resources for the expression of sociocultural meanings, identities and practices. Instead of trying to delimit the boundaries around what Singlish is, it might perhaps be more fruitful for researchers to investigate the meanings behind such constructions, and focus not on just how something is said, but why. The different ways of saying the same thing crucially elaborate on the socio-cultural content of the message, as well as the identities of speakers and participants. Appropriating different ways of speaking, for example, allows speakers to consciously identify with certain social groups or to express their different roles in a communicative event.8 An example of the way in which Singapore English is a subtle and complex integration of the global and local, where, for example, there is a careful interplay of features which index authority with those signalling closeness between the interlocutors, can be found in teacher talk in Singapore classrooms (Kwek 2005; Rubdy 2007; Doyle 2009). In her analysis of teacher talk in two Singapore classrooms, Kwek (2005) shows how deftly the two teachers in her study weave standard and non-standard English features in their facilitation and delivery of lessons. The use of Singlish features in complex communicative performances such as teacher talk highlights the way in which language choice is a delicate balance that shifts and varies as speakers construct, on the fly, representations of their identity, negotiate their role and relation with their interlocutor and context, as well as respond to the fluid and changing demands of the communicative event (Li 2005).

Hybridity in ways of speaking 121 To give a more specific illustration as to how Singapore English can best be seen as style-shifting in which speakers select, from their linguistic resources, features from both the globalist as well as the localist, we examine teacher talk from Kwek’s (2005) study of two Singaporean English language teachers. Lisa teaches English in a neighbourhood secondary school to 13and 14-year-olds. According to Kwek, Lisa is a highly skilled user of English, and thus her use of non-standard features does not stem from a lack of fluency, ability or competence. Lisa’s unmarked socio-stylistic orientation in the classroom is a standard variety of English, i.e., the globalist. However, as we shall see, Lisa constantly shifts to and from the localist orientation, through the inclusion of non-standard forms of English, as she purposefully negotiates her lesson delivery and classroom management. The following transcript from Kwek (2005: 126–7) is a single contiguous turn at the beginning of Lisa’s lesson. For ease of readability and discussion, however, it has been broken up and presented here as separately numbered paragraphs (1) to (5). Singlish features are marked in italics to more easily indicate switches in the focus of the pedagogical discourse, which as we shall see, are congruent with the switches between the globalist and localist orientations in language use. Lisa begins the turn with what Kwek (2005) refers to as ‘organizing’, where she does some housekeeping, and spells out what the tasks are for the particular lesson. Other than the two instances of the pragmatic particle ah, Lisa keeps to a more or less standard variety of English: (1) No, guys, enough is enough, shsss. Ok, look here. Hurry up! Thank you very much. This is what we’re going to do ah. Today’s there’s going, it’s double period. There’s quite a bit to do. Guys, I am serious ah! If you are going to go into your own little conversation while I am teaching or whatever, I am going to … I know some of you have lost your file ah, your CMF file but I can still book you. That means that if I don’t have your CMF file, it goes direct to the yellow form or pink form. I am very sorry because I don’t have a file. And at this level, I shouldn’t have, so if I give you warning, once, twice, third time, I am very sorry. Because sometimes, you all go too much. Okay, rules for today, simple, we have two tasks. I am going to first recap, task one, task two, then we will sum up. Ok, so actually only two things to do today, task one, task two, then a group presentation.

Interestingly, however, as she transitions and starts a discussion of the content of the lesson, many more Singlish features appear in Lisa’s speech, marking her shift away from a globalist orientation towards a more localist orientation. The speech is less formal and contains a noticeably larger number of localist features such as bald noun phrases without plural inflections (text, a number of situation, certain word, adjective, adverb), the non-

122 Lubna Alsagoff inverted form of the embedded question (we went on to see what is narrative), verbs without tense inflections (talk), as well as the expression of a conditional clause without the subordinating conjunction if (Sometimes [ ] you want, you may have LITTLE excitement goes down). What is interesting to note is that this is in contrast to (1), where Lisa uses standard forms of many of these very same constructions, e.g., she clearly uses the subordinator if as well as because (… so if I give you warning, once, twice, third time, I am very sorry. Because sometimes, you all go too much) and she marks plural nouns with the -s suffix (rules, things, tasks): (2) Ok, recap, last lesson what we did was we began to, we went on to see what is narrative and then we saw some movies and videos and we talk about narrative text and I told you that part of it what we were doing of different types of narrative ah. Instead of action stories, we are going into suspense stories. What’s so different between action stories and suspense stories? The only thing is that, certain things that you need to, in your story, somewhere down the line, you need to build up tension in a particular situation, sometimes in a number of situation. That means you have to write so that you bring in certain suspense, tension ok, certain scenarios using certain word, adjective, adverb to describe so that it brings the reader up to a level. Ok, you don’t have to, you know, do many situations; you can do one. Sometimes [ ] you want, you may have little excitement goes down; ok you saw the roller coaster one right, they thought they went in; and they thought that part of it is just to go up and then the thing fell down and they thought it was the end; that was not the end ok. It was the ride.

In excerpts (3) and (4), this same sort of style shift is seen: when Lisa switches back to organizational discourse in (3), much of her discourse is clearly in standard English, apart from the use of the particle ya, which means ‘yes’. However, as she moves towards the end of her organizational talk in (3), to a subject-specific content focus in (4), the shift towards the localist orientation is quite apparent, with an increase in the occurrence of non-standard features. Thus, at the tail end of (3), Lisa again marks her localist orientation through the use of several conditional clauses without subordinating conjunctions ([ ] you want to act it out, [ ] you want to film it, film it, [ ] you want to make an audio tape out of it, make an audio tape out of it, [ ]you want to write it, script it, go ahead; [ ] You want to make a comic script out of it, go ahead): (3) So you can do things like this … ok ... [Long pause] Ya, I don’t want to keep mentioning your name here. Ya, Go blush. Ok, structure. This is about the same structure as we did for action story. Recap, you are going to be given a template later. A template will be given; I am not going to reprint the template. This template is given to you as a model

Hybridity in ways of speaking 123 and this template, you’ll be using for all your work from now on from all the other passage I’ll be giving, for all the writing that you are going to do, for even the production that you are coming. What is our end result of this production here, ya, Ian? That means our end result in two, three weeks’ time, instead of coming up with a commercial you will be given a choice to come up with a production of your suspense story. Ya, but this time it need not be; [ ] you want to act it out, [ ] you want to film it, film it, [ ] you want to make an audio tape out of it, make an audio tape out of it, [ ]you want to write it, script it, go ahead.[ ] You want to make a comic script out of it, go ahead, but we will lead on to it and we’ll show you different types, ok.

As Lisa moves into a focus on content in (4), more non-standard features appear in her speech, including the increased use of discourse markers (ah, eh, then how), expressions of clauses without overt subjects or objects (e.g., we mark [ ]; [ ] Very hard ok to have [ ]; [ ] Cannot leave the, the reader hanging up there ah; Can [ ] remember [ ] or not?), as well as the use of don’t have in place of the existential construction there isn’t: (4) Scaffold (…) always ah, in any writing. You it, eh, Band B, you need to know all this, you know, because when we mark [ ] ah, we’ll check ah. Intro, that one, you all know ah, introduction, how you start. Very important, when I pick up a story, the first thing that does is the intro. [ ] You go and buy a story book, the intro doesn’t always start, ‘long time ago, once upon a time, far, far away in a long lost land ter ra ra ra. Don’t have. Theme? What theme? Although it’s a suspense story, but what you doing, which theme are you going about? (Announcement on loudspeaker). Eh, 4G. Ok, the, whatever theme you set, the setting must match ok. If you want to do a theme of ok, maybe eh, where there are knights and castles or whatever [ ] ok, Peter’s favourite, then the setting must be there. You want to talk about the pirates and everything at that time, then the setting must be there. You want to talk about eh say the, what we did, we watch the roller coasters and it’s about a fair and the modern times, then the setting must be there. So it must come down. Then comes your characters ok. [ ] Very hard ok to have [ ], … you can have one character, but that means you must really write the person’s inner feelings, the surroundings, the atmosphere and everything. Normally, average two to three ok. Action. You need to add in a bit; that means you need to use a lot of verb, adverb because you need to describe [ ]. Settings, character is adjective to describe the person. Conflict, complication, what happened, what led to it. [ ]You always have little, little complication, then have ONE MAJOR complication that climaxed. Always have [ ] resolution; how it’s solved ok. Then you can conclude. [ ] Cannot leave the, the reader hanging up there ah. Climax, and then you put the all the dots there, dot, dot, dot, dot ok. Or some of you who wrote the ‘chatasahib’ the trap I finished, later I give it back to you and then you end up with ‘hah’ and then dot,

124 Lubna Alsagoff dot, dot. Then I go back, then how? Ok, so, know this, anytime you do narrative you need this. Can [ ] remember [ ] or not?

Finally, at the end of this very long turn, Lisa goes back to organizing the class activity, and her use of English once again becomes standard in grammatical form, as shown in (5): (5) No, because if you miss any of these, especially when you write and you plan, you’re going to have trouble trying to link the thing up. No need to write it down; I am giving you a scaffold. Ok, [Pause] The scaffold is this one ah (opening scaffold). I tell you, I told you all these ah guys, intro, theme, setting, characters, action, conflict, climax, resolution. You are going to get these. Group leader, come, take one. I need to take a pencil out first. 3, 4, 5. 1, 2, 3, 4. How many?

The analysis of this excerpt clearly shows the teacher using globalist, i.e., standard, features of English, when she is organizing the lesson, alongside localist features when she focuses on teaching content. What is even more interesting is that the same construction can appear, even within the same turn, both in the standard form as well as the non-standard form. As we have seen, in one part of the turn, verbs may show past tense marking (e.g., did, went, edited, climaxed) and subject-verb agreement (e.g., the only thing is, you have, it brings ...), while in another, they appear without such inflections (e.g., we talk). Nouns, too, appear both in the inflected plural form (e.g., knights, castles) as well as uninflected bare forms (e.g., a number of situation; certain word, adjective, adverb). There is also the use of sentence connectors in conditional clauses (e.g., because, if), and yet, further on in this same turn, similar structures appear without sentence connectors. This to-ing and fro-ing between localist and globalist is a feature of all the four lessons in Kwek’s (2005) transcripts. The co-presence of both standard and non-standard features is deliberate and purposeful, with the shifts between a more standard globalist style and a more colloquial localist style clearly aligned to the pedagogical objectives and the role the teacher wishes to construct and present. In a delicate weaving of language, Lisa moves back and forth, from a formal style which signals her authority, giving instructions about the organization to the lesson and attending to classroom management, to a more colloquial and casual style which purposefully marks her not as distant and superior in status, but as close and an equal. Lisa balances her authoritarian role and her facilitative roles in the classroom much through her use of language. In explaining concepts, Lisa’s style eases the students into a more participatory and collaborative mode of interaction, but at the same time, she ensures that discipline is kept, and the class is on task by demonstrating her authority and position through her use of a distinct global style, which she

Hybridity in ways of speaking 125 occasionally punctuates, nonetheless, with indexical markers of localism, such as the ubiquitous ah and even the occasional lah to soften the tone of her discourse. Teachers like Lisa use English in creative ways that allow them to be responsive to the needs of the classroom, to enable them to build rapport with their students so as to be able to teach more effectively. Their use of English is standard, and yet deliberately includes many features that are non-standard. The inclusion of such features is not a result of the teacher’s lack of awareness. Kwek (2005) also demonstrates how both the teachers in her study are skilled and deliberate in varying the kind of English they use; for example, they increase their use of Singlish features when there is small group teaching, and move towards a more standard form of English in whole class lectures or discussions. Explaining the interlacing of standard and non-standard grammatical features as the simultaneous presence of both globalist and localist structures enables us to understand how teachers use the linguistic resources of Singapore English in a complex play of identity construction and negotiation, which, for example, would allow both the presence of standard features as well as non-standard ones. While there is variability, such use should not be seen as inconsistent. If we think of variation as fluid, and of users as selecting resources in the form of linguistic features as the ‘need’ arises, we are better able to see such variation in the context of use. We should not need to see Singlish as a homogeneous and well-structured linguistic system. Rather than focus on the linguistic environment of the variable, the focus of research into Singapore English should centre on what the inclusion of such variables means. Such an approach will also be able to account for the co-presence of features of both globalist and localist orientations. The variation of English in Singapore is clearly shaped by forces intricately bound with negotiations in identity formation and presentation. These negotiations are too complex and subtle to be characterized simply in terms of the use of particular varieties. I concur with Pennycook’s (2007: 85) observation that use of language represents ‘sites of struggle’ in which ‘meanings are always in flux and in contention’. The focus on features such as pragmatic particles, or bare verbs or nouns is common in the Singapore English research literature. However, the argument that language and culture are inextricably linked suggests that global-local cultural orientations may also be linguistically realized in a multitude of ways, and not restricted to discrete phonological, grammatical or lexical features. Shifts in identity construction and presentation could, for example, instead be signalled by more subtle pragmatic aspects of talk. Thus, ways of speaking could also encompass how politeness is negotiated — in accepting compliments, for example, or in terms of address, or in asking a question of the teacher. Thus, a speaker may use grammatical features that are globally oriented, but

126 Lubna Alsagoff choose to signal his or her ethnic membership of the local community by enacting pragmatic acts grounded in the endogenous cultures (Ho 2001).

Conclusion Glocalization has been used here to offer an alternative means to understanding variation of English use in Singapore. It presents language and identity as intertwined, and as fluid — just as language is used to present, re-present and construct and re-construct identity, it, in turn, is similarly constructed by its speakers in the service of identity and cultural representations, communication and interactions. Glocalization has presented variation within Singapore English as style-shifting, suggesting a fluidity of movement within a dynamic multidimensional space, rather than as diglossic code-switching between two well-bounded varieties. Such a model also offers a more dynamic orientation in which to understand the ways in which people appropriate English for their own purposes, but are at the same time constrained by institutional discourses and policies favouring standardization and conformity (Bhatt 2005). In a move towards a more holistic understanding of the indigenization of English in a context such as Singapore, it is imperative that language be seen as a means of identity formation and representation, where local appropriations of global forms by speakers to construct and represent their thought, practices and culture are realized as fluid variations in a multidimensional discursive space. Singlish is thus more aptly presented in the glocalization model as a range of lingua-cultural resources speakers use in order to identify or mark a change in cultural orientation or style. More importantly, a model of variation of the use of an indigenized English in multilingual, multicultural speech communities such as Singapore must be developed from a perspective that allows us to see language as a meaningmaking and identity-creation resource in a culturally grounded manner.

Notes 1.

2.

Glocalization can, as a model, clearly be applied to other contexts in which New Englishes are spoken (e.g., the Philippines, India, Nigeria, etc.) since it offers an understanding of language variation in relation to the global and local interactions resulting from processes of globalization. See Pakir (2000) for a discussion of language teaching in relation to ‘glocal Englishes’. Evidence of this fast-evolving language situation is captured in MOE’s statistical information on the number of English speakers among Primary One students (age 7) entering school. In 1996, pupils who spoke English numbered only 1 in 3, whereas in 2006, they numbered 1 in 2 (MOE 2006: 4).

Hybridity in ways of speaking 127 3.

4. 5.

6. 7.

8.

The account here is necessarily simplified because of space constraints. The lectal continuum does in fact discuss variation along functional lines, though in a limited manner. See Lim and Foley (2004: 2–7) for a more detailed discussion of the changing profile of the Singaporean English speaker. I am aware that Gupta does discuss the leakiness of diglossia, though her reasons for suggesting this in fact lend support to the argument that variation should not be discussed as a code-switch but rather as a continuum of change. See Alsagoff (2007, forthcoming) for a more detailed discussion of the issues relating to Gupta’s description of Singapore English variation as diglossia. I am grateful to Anthea Fraser Gupta for her suggestions in relation to this point. Interestingly, Deterding’s reason for using a single speaker on which to base his description is to obviate the need to explain the variation of Singapore English (2007: 6). This does not detract from previous arguments in the literature that Singlish is autonomous and rule-governed. It clearly is a highly organized system. However, it may be the case that we need to look at this organization not just in terms of whether features appear consistently in a linguistic environment, but of organizing principles that enable speakers to exercise choice of linguistic features to index global-local orientations. An example of such principles is seen in Table 5.2 where I describe differences between the global and local in relation to systemic choices. For example, instead of speaking about whether Subject-Auxiliary Inversion is used in question formation, I present this as a contrast between the use of the interrogative mood to indicate questions in globally-oriented speech, versus the use of the declarative for locally-oriented speech.

References Alsagoff, Lubna (2007) Singlish: Negotiating culture, capital and identity. In Language, Capital and Culture. Edited by Viniti Vaish, S. Gopinathan and Yongbing Liu. Rotterdam: Sense Publishers, pp. 25–46. Alsagoff, Lubna (forthcoming) English in Singapore: Culture, capital and identity in linguistic variation. World Englishes. Alsagoff, Lubna and Ho, Chee Lick (1998a) The grammar of Singapore English. In English in New Cultural Contexts: Reflections from Singapore. By Joseph A. Foley, Thiru Kandiah, Bao Zhiming, Anthea Fraser Gupta, Lubna Alsagoff, Ho Chee Lick, Lionel Wee, Ismail S. Talib and Wendy D. Bokhorst-Heng. Singapore: Oxford University Press, pp. 127–51. Alsagoff, Lubna and Ho, Chee Lick (1998b) The relative clause construction in colloquial Singapore English. World Englishes, 17(2), 127–38. Bakhtin, Mikhail (1981) Unitary language. In The Dialogic Imagination: Four Essays. Edited by Michael Holquist. Translated by Caryl Emerson and Michael Holquist. Austin: University of Texas Press, pp. 269–95.

128 Lubna Alsagoff Bhatt, Rakesh M. (2005) Expert discourses, local practices, and hybridity: The case of Indian Englishes. In Reclaiming the Local in Language Policy and Practice. Edited by A. Suresh Canagarajah. Mahwah, NJ: Lawrence Erlbaum, pp. 25–54. Block, David and Cameron, Deborah (eds.) (2002) Globalization and Language Teaching. London: Routledge. Bokhorst-Heng, Wendy D. (2005) Debating Singlish. Multilingua, 24(3), 185–209. Bourdieu, Pierre (1986) The forms of capital. In Handbook of Theory and Research for the Sociology of Education. Edited by John G. Richardson. New York: Greenwood Press, pp. 241–58. Bourdieu, Pierre (1991) Language and Symbolic Power. Translated by Gino Raymond and Matthew Adamson. Cambridge, MA: Harvard University Press. Canagarajah, A. Suresh (ed.) (2005) Reclaiming the Local in Language Policy and Practice. Mahwah: Lawrence Erlbaum. Canagarajah, A. Suresh (2006) Negotiating the local in English as a Lingua Franca. Annual Review of Applied Linguistics, 26, 197–218. Chew, Phyllis G.L. (1995) Lectal power in Singapore English. World Englishes, 14(2), 163–80. Chew, Phyllis G.L. (2006) Remaking Singapore: Language, culture and identity in a globalized world. In Language Policy Culture and Identity in Asian Contexts. Edited by Amy B.M. Tsui and James W. Tollefson. Mahwah, NJ: Lawrence Erlbaum, pp. 73–93. Deterding, David (2007) Singapore English. Edinburgh: Edinburgh University Press. Deterding, David and Low, Ee Ling (2001) The NIE Corpus of Spoken Singapore English (NIECSSE). Singapore Association of Applied Linguistics Quarterly, November, 2–5. Doyle, Paul G. (2009) Language development in Singapore classrooms: A corpusbased description of the ‘school variety’. In Language Learning in New English Contexts. Edited by Rita E. Silver, Christine C.M. Goh and Lubna Alsagoff. London: Continuum, pp. 91–111. Geertz, Clifford (1973) The Interpretation of Culture. New York: Basic Books. Gupta, Anthea F. (1992) The pragmatic particles of Singapore Colloquial English. Journal of Pragmatics, 18, 31–57. Gupta, Anthea F. (1994a) A framework for the analysis of Singapore English. In Language, Society and Education in Singapore: Issues and Trends. 1st ed. Edited by S. Gopinathan, Anne Pakir, Ho Wah Kam and Vanithamani Saravanan. Singapore: Times Academic Press, pp. 119–32. Gupta, Anthea F. (1994b) The Step-Tongue: Children’s English in Singapore. Clevedon, UK: Multilingual Matters. Ho, Chee Lick (2001) The cultural grounding of Singapore English. In Evolving Identities: The English Language in Singapore and Malaysia. Edited by Vincent B.Y. Ooi. Singapore: Times Academic Press, pp. 102–11. Ho, Chee Lick and Alsagoff, Lubna (1998) English as the common language in multicultural Singapore. In English in New Cultural Contexts: Reflections from Singapore. By Joseph A. Foley, Thiru Kandiah, Bao Zhiming, Anthea F. Gupta, Lubna Alsagoff, Ho Chee Lick, Lionel Wee, Ismail S. Talib and Wendy D. Bokhorst-Heng. Singapore: Oxford University Press, pp. 201–17. Ho, Mian-Lian and Platt, John T. (1993) Dynamics of a Contact Continuum: Singaporean English. Oxford: Clarendon Press.

Hybridity in ways of speaking 129 Kachru, Braj B. (ed.) (1992) The Other Tongue: English Across Cultures. 2nd ed. Urbana: University of Illinois Press. Kachru, Braj B. (1997) World Englishes and English-using communities. Annual Review of Applied Linguistics, 17, 66–87. Kandiah, Thiru (1998) The emergence of New Englishes. In English in New Cultural Contexts: Reflections from Singapore. By Joseph A. Foley, Thiru Kandiah, Bao Zhiming, Anthea F. Gupta, Lubna Alsagoff, Ho Chee Lick, Lionel Wee, Ismail S. Talib and Wendy D. Bokhorst-Heng. Singapore: Oxford University Press, pp. 73–105. Kwek, Melody Y.P. (2005) English teachers using Singapore Colloquial English: An examination of two secondary school teachers’ lessons. Unpublished BA Academic Exercise, National Institute of Education, Nanyang Technological University, Singapore. Le Page, Robert B. and Tabouret-Keller, Andree (1985) Acts of Identity: Creole-Based Approaches to Language and Ethnicity. Cambridge: Cambridge University Press. Lee, Kuan Yew (2005) Speech by Minister Mentor Lee Kuan Yew at the Tanjong Pagar Chinese New Year Dinner at Radin Mas Community Club, 17 February 2005. http://www.sprinter.gov.sg. Accessed 10 July 2006. Li, Wei (2005) ‘How can you tell?’ Towards a common sense explanation of conversational code-switching. Journal of Pragmatics, 37, 375–89. Lim, Lisa and Foley, Joseph A. (2004) English in Singapore and Singapore English: Background and methodology. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Varieties of English Around the World. Amsterdam/ Philadelphia: John Benjamins, pp. 1–18. Ministry of Education (2006) Report of the English Language Curriculum and Pedagogy Review. Singapore: Curriculum Planning and Development Division, Ministry of Education. Newbrook, Mark (2003) Features of the relative clause in Singapore English. In English in Singapore: Research on Grammar. Edited by David Deterding, Low Ee Ling and Adam Brown. Singapore: McGraw-Hill, pp. 67–76. Pakir, Anne (1991) The range and depth of English-knowing bilinguals in Singapore. World Englishes,10(2), 167–79. Pakir, Anne (1994) English in Singapore: The codification of competing norms. In Language, Society and Education in Singapore: Issues and Trends. 1st ed. Edited by S. Gopinathan, Anne Pakir, Ho Wah Kam and Vanithamani Saravanan. Singapore: Times Academic Press, pp. 63–84. Pakir, Anne (1999) Bilingual education with English as an official language: Sociocultural implications. Georgetown Roundtable of Linguistics Archive. Georgetown University Press, pp. 341–49. http://digital.georgetown.edu/gurt/1999.cfm. Accessed 20 October 2006. Pakir, Anne (2000) The development of English as a ‘glocal’ language: New concerns in the old saga of language teaching. In Language in the Global Context. Edited by Ho Wah Kam and Christopher Ward. Anthology Series 41. Singapore: SEAMEO Regional Language Centre, pp. 14–31. Pakir, Anne (2001) The voices of English-knowing bilinguals and the emergence of new epicentres. In Evolving Identities: The English Language in Singapore and Malaysia. Edited by Vincent B.Y. Ooi. Singapore: Times Academic Press, pp. 1–11. Pennycook, Alastair (2007) Transcultural Flows. London: Routledge.

130 Lubna Alsagoff Phillipson, Robert (1992) Linguistic Imperialism. Oxford: Oxford University Press. Pieterse, Jan Nederveen (2006) Globalization and culture: Three paradigms. In McDonaldization: The Reader. 2nd ed. Edited by George Ritzer. Thousand Oaks: Pine Forge Press, pp. 278–83. Platt, John and Weber, Heidi (1980) English in Singapore and Malaysia: Status, Features, Functions. Kuala Lumpur: Oxford University Press. Robertson, Roland (1995) Glocalization: Time-space and heterogeneity-homogeneity. In Global Modernities. Edited by Mike Featherstone, Scott M. Lash and Roland Robertson. London: Sage, pp. 25–44. Robertson, Roland (1997) Comments on the ‘Global Triad’ and ‘Glocalization’. In Proceedings: Globalization and Indigenous Culture. Institute for Japanese Culture and Classics, Kokugakuin University. http://www2.kokugakuin.ac.jp/ijcc/wp/ global/15robertson.html. Accessed 20 October 2006. Rubdy, Rani (2001) Creative destruction: Singapore’s Speak Good English movement. World Englishes, 20(3), 341–55. Rubdy, Rani (2007) Singlish in the school: An impediment or a resource? Journal of Multilingual and Multicultural Development, 28(4), 308–24. Rubdy, Rani and Saraceni, Mario (eds.) (2006) English in the World: Global Rules, Global Roles. London: Continuum. Wee, Lionel (2002) ‘Lor’ in Colloquial Singapore English. Journal of Pragmatics, 34(6), 711–25. Wee, Lionel (2003) Linguistic instrumentalism in Singapore. Journal of Multilingual and Multicultural Development, 24(3), 211–24. Wee, Lionel and Ansaldo, Umberto (2004) Nouns and noun phrases. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam/Philadelphia: John Benjamins, pp. 57–74. Widdowson, Henry G. (1997) EIL, ESL, EFL: Global issues and local interests. World Englishes, 16(1), 135–46.

Part III

Ethnicity and Ownership

6

Whose English? Language ownership in Singapore’s English language debates Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff

‘Language ideological debates’ (Blommaert 1999) are a common feature of Singapore politicking, and are centrally implicated in much of the social, economic, and political constructedness of the nation. As articulated by Blommaert (1999: 1–2), language debates ‘are organized around issues of purity and impurity of languages, the social “value” of some language(s) as opposed to (an)other(s), the socio-political desirability of the use of one language or language variety over another, the symbolic “quality” of languages and varieties as emblems’ of socio-cultural notions of nationhood, cultural authenticity, progress, modernity, and so forth. At the core of such debates are struggles ‘over definitions of social realities’, or the ‘politics of representation’ (Blommaert 1999: 9). With respect to the ‘politics of representation’ within the context of Singapore’s English language ideological debates, the notion of ownership becomes relevant — how speakers of different varieties of English are positioned and represented. Such institutional positioning involves value judgements (often negative) about languages and speakers of languages, often through the idealization of the native speaker found within the native speaker(NS)/non-native speaker (NNS) dichotomy, as well as an unwillingness to recognize the different varieties of world Englishes as legitimate languages. What is interesting in the Singapore context is how government leaders have appropriated much of this NS/NNS discourse in their own English language ideological debates. In broad strokes, English has been ideologically constructed as a purely instrumental and functional language within the context of nation building and the global economy. In this context, the officially preferred model is British RP, and the Inner Circle speakers of English continue to be regarded as the true owners of English. At the same time, speakers of the local variety of English, especially the colloquial form Singlish, are portrayed as uneducated, uncouth, and unworldly (Bokhorst-Heng 2005; Alsagoff 2007). One of the consequences of this ideological construction is that there is no discursive

134 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff space in Singapore’s language ideological formation to label English as a mother tongue or to allow Singaporeans the acquisition of native speaker membership — even though more and more Singaporeans do in fact primarily use English (Wee 2002) and English is the medium of instruction in all schools. According to the 2000 Census (Statistics 2000), 71% of the population is literate in English, and 23% report using predominantly English at home. Lim and Foley (2004: 6) maintain that there is a ‘growing body of English users for whom English has gone beyond the lingua franca stage, who are native speakers of the language, following the simple definition that a native speaker is a fluent speaker of the language, typically after having learnt the language as a child’. While the Singapore census does not ask questions about bilingual language practices, data from an ongoing Sociolinguistic Survey of Singapore (see Vaish et al. in this volume) involving 10-year-old Singaporeans indicate that 14.2% report using predominantly English at home — and 53% report bilingual practices, saying they use both English and their (ethnic) ‘mother-tongue’ (the official language associated with one’s father’s ethnicity) habitually at home. According to a survey administered by the English Language Curriculum and Pedagogy Review Committee in 2006, only 12% of Primary 1 students indicated they hardly or never use English at home (Ministry of Education 2007: 4). English clearly has a significant place in the everyday lives of many Singaporeans, and, contrary to the official diglossic discourse that separates language use (English is for the purposes of meeting the nation’s functional and economic needs) and language ownership (‘mother tongue’ languages are for cultural and personal identity), operates in dynamic and interactive relation with the mother tongue languages. In the first part of our chapter we will unpack the key parameters of this debate, drawing on speeches given by various government officials and documents related to the Speak Good English Movement (SGEM). In the second half of this chapter, we propose a model that nuances the meanings of English language ownership in Singapore, taking us away from the native/non-native speaker binaries and replacing the idea of ‘native speaker’. Drawing on research by Alsagoff, Bokhorst-Heng, McKay, and Rubdy (Bokhorst-Heng et al. 2007; Rubdy et al. 2008), we consider speakers’ orientations towards English norms to foreground speakers’ degree of ownership of the English they speak. While the study involved Chinese, Malay, and Indian participants, our discussion here focuses on the Malay and Indian dyads. Malays and Indians are both ethnic minority groups in Singapore’s population pool. English is routinely used the most by members of the Indian community, where 36% report using English as their dominant home language. English is used the least by the Malays, with only 8% indicating using English routinely at home. Yet even among the Malays, the bilingual education policy, coupled with the socio-economic

Whose English? 135 forces that highlight the prestige and instrumental value of English, have led to an increase in the number of English users (Kamsiah and Bibi 1998). This is especially true among the educated, the well-off, and the younger school-going population across all ethnic groups. In our conclusion, we summarize our discussion, and note the contributions that a view of language ownership has to Singapore’s language ideological debates.

English language ideological debates

Some background to Singapore’s language ideological debates Singapore is a multi-ethnic and multilingual society, representing three main racial groups (about 76% Chinese, 14% Malays, and 9% Indians, and 1% ‘others’) and four official languages: three associated with the nation’s main racial groups (Mandarin, Malay, and Tamil), and English. The current education system supports these languages through what is commonly referred to as English-knowing bilingual education, with all children educated in English and taught their ethnic ‘mother tongue’ languages. However, bilingualism has very particular meanings in this multilingual context. The following excerpt from then prime minister Lee Kuan Yew’s 1972 speech to the Singapore Teachers’ Union captures the essence of language constructedness in Singapore. While he is talking about English and Mandarin bilingualism, it is important to note that the same arguments have been applied to the other ‘mother tongue’ languages as well. Here, he first presents his view of Mandarin: Please note that when I speak of bilingualism, I do not mean just the facility of speaking two languages. It is more basic than that, first, we understand ourselves, what we are, where we came from, what life is, or should be about, and what we want to do ... And it is not just learning the language. With the language goes the fables and proverbs. It is the learning of a whole value system, a whole philosophy of life, that can maintain the fabric of our society intact, in spite of exposure to all the current madnesses around the world.

He then goes on to talk about English: Then the facility of the English language gives us access to the science and technology of the West. It also provides a convenient common ground on which … everybody competes in a neutral medium.

There is a very strong constructed dichotomy between one’s ‘mother tongue’ language and English. One’s ethnic ‘mother tongue’ language

136 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff is the language of identity, culture, values, rootedness, and of self. In contrast, English is defined in pragmatic and utilitarian terms, to be used for the economic and global connectedness that English provides. English is something external, something to be accessed and tapped into for pragmatic and instrumental purposes. English is also the language of interethnic communication. However, Lee Kuan Yew significantly does not suggest that this means English is a language of pan-ethnic national identity. Rather, somehow, by each of the different ethnic groups rooted through their own languages, a national multicultural identity will emerge. Thus, while the ethnic languages are owned, English is used. This distinction becomes even more transparent later in his speech: Whilst we may speak English, whilst we may use the English language, whilst we may watch what the English-speaking world in America and Canada, in Britain, in Australia and New Zealand, are doing, … much of it is not us. Only when we first know our traditional values, can we be quite clear the Western world is a different system, a different voltage, structured for purposes different from ours.

And therefore, English cannot be regarded as a mother tongue. In his speech at the launch of the 1984 Speak Mandarin Campaign, Lee Kuan Yew asserted: One abiding reason why we have to persist in bilingualism is that English will not be emotionally acceptable as our mother tongue.

By not allowing English to be given the status of ‘mother tongue’ in the ideological constructedness of language, English remains a utilitarian language to be accessed and to access. While these speeches were given over two to three decades ago, the views expressed continue to inform Singapore’s language ideological debates today. In the next section, we examine the current English ideological debates more closely as they are articulated in the current SGEM, with a particular focus on the parameters of ownership.

Singapore’s English ideological debates Current English ideological debates are primarily concerned with the kind of English Singaporeans should be using. In broad strokes, they pivot on the tensions emerging from globalism — the competing needs of internationalism and identity (Crystal 1995: 110; see also Bokhorst-Heng 2005) — with internationalism being the primary concern of government

Whose English? 137 leaders, and the view of identity promoted primarily by the more educated and yuppie sector of society (who are able to speak both standard English and Singlish). The debate somewhat mirrors that of Quirk and Kachru, two key figures in the growing debate over standards in international English (Quirk 1985 and Kachru 1985, discussed in McKay and Bokhorst-Heng 2008: 140–1). Quirk argues for the need to uphold a common standard of English in all contexts of English language use. Kachru argues that there needs to be acceptance and understanding of the linguistic creativity that occurs in multilingual situations, and an adoption of a pluricentric model of standards. Quirk’s view maintains a ‘native-speaker’ model, with those in Inner-Circle countries as the true owners of English, whereas Kachru’s model suggests co-ownership or multiple-ownership. The government’s concern for internationalism has to do with the particular nationalist agenda that it has pursued. Claiming to be a small country with limited natural resources, the government has actively pursued a development model that positions Singapore firmly within the global community. Furthermore, all policy initiatives are guided by the logic of pragmatism which prioritizes continuous economic growth. As we have seen in the above discussion, such rationale lies at the core of the ideological constructedness of English. And it continues to be at the centre of the government’s rationale, as evident in the SGEM politicking. The SGEM was first launched in 2000 by then prime minister Goh Chok Tong ‘who was concerned that Singapore’s aim to become a first world economy would be affected by poor English standards among Singaporeans’ (Speak Good English Movement 2007). The movement’s initial message was ‘Speak Well. Be Understood.’ The current 2006/2007 slogan (originally taken from the popular Phua Chu Kang sitcom)1 expands this a bit: ‘Be Understood. Not only in Singapore, Malaysia and Batam.’ While the NS/NNS dichotic discoursing is evident in most government speeches, for the purposes of discussion, we will examine two SGEM speeches: Prime Minister Goh Chok Tong’s speech at the first launch of the SGEM in 2000 (Goh 2000) and Professor Koh Tai Ann’s (SGEM’s then chairman) speech given at the launch of the 2006 SGEM (Koh 2006). While Goh concentrates on denying ownership by invalidating Singlish, Koh focuses on the global purposes of English, thereby maintaining a NS/NNS divide.

Goh Chok Tong: Good English versus corrupt English In his speech at the launch of the 2000 SGEM, Goh Chok Tong immediately establishes the illegitimacy of Singlish as a language and as a marker of Singaporean identity. He defines Singlish as ‘ungrammatical and truncated’ and ‘incomprehensible’. Consider the following excerpt from his speech:

138 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff The ability to speak good English is a distinct advantage in terms of doing business and communicating with the world. This is especially important for a hub city and an open economy like ours. If we speak a corrupted form of English that is not understood by others, we will lose a key competitive advantage. My concern is that if we continue to speak Singlish, it will over time become Singapore’s common language. Poor English reflects badly on us and makes us seem less intelligent or competent … all this will affect our aim to be a first-world economy.

In this excerpt, Singlish is defined as a ‘corrupted form of English’, as a sign of less intelligence and less competence, as something that is a bad reflection of the nation. Singlish is presented as the lowest common denominator, which over time ‘will become Singapore’s common language’. And, bringing us back to the ideological constructedness of language discussed earlier, to speak Singlish has national consequence as it undermines the government’s economic objectives to be an economic hub and first-world economy. Later in his speech, he makes this last argument even stronger: They [younger Singaporeans] should not take the attitude that Singlish is cool or that speaking Singlish makes them more ‘Singaporean’. They have a responsibility to create a conducive environment for the speaking of good English … If they speak Singlish when they can speak good English, they are doing a disservice to Singapore.

Speaking ‘good English’ thus has national merit and is connected to notions of good citizenship. But note that the national value of English comes not through ownership, which the users of Singlish sought to establish, but rather, through use — through its economic utility. Throughout his speech, Goh continues to invalidate Singlish as a language of ownership through a dichotic structuring between Singlish and English. As noted by Bokhorst-Heng (2005: 203), Goh uses the word ‘English’ in his SGEM speech 40 times: almost half of these references (19) were used in the expression ‘good English’ (rather than standard English); an additional four references were used in the phrases ‘proper English’, ‘excellent English’ and ‘elegant English’. In all other instances, the phrase was used in contrast with Singlish with the obvious connotation of Singlish thus being ‘bad English’. Goh illustrates his argument with an example of a conversation he had with a Zimbabwean caddy: Some years ago, I played golf in Zimbabwe when I had some spare time before my meetings. I was impressed by the excellent English spoken by my African caddy. For example, when he found me putting badly ever so often, he asked me politely, ‘Would you permit me to test your putter?’ He

Whose English? 139 tried it and advised me that it was too heavy. In Singlish, we would have said, ‘Can try your putter or not?’ After the game, he asked me, ‘Would you have some used balls to spare me?’ I was so impressed by his elegant English that I gave him all my used golf balls, and some new ones too. In Singlish, it would be, ‘Got old balls give me can or not?’ My Zimbabwean caddy did not complete his secondary school education. He picked up his English from the white Zimbabwean golfers.

The example of course does not work (Fong, Lim and Wee 2002); the caddy’s speech was most likely influenced by the fact that he was in the company of a prime minister, as it would also be for most Singaporeans. However, his illustration does a number of things, in addition to demonstrating the perceived benefits of speaking ‘good English’ (i.e., you get free golf balls from a prime minister). First, note the use of his words to describe the caddy’s speech: ‘excellent English’, ‘politely’, and ‘elegant English’ — all of which he contrasted with samples of Singlish which, by implication and position, would be the opposite. Second, he demonstrates the benefits of speaking good English by suggesting his giving of the golf balls was a reward for the caddy’s standard of English.

Koh Tai Ann: Discourses of NS/NNS The themes raised in Goh’s speech continue with Koh. However, she places greater emphasis on the discourse of global instrumentality, and spends less time on invalidating Singlish. The global position of English frames her opening line: This year, the Speak Good English Movement continues with its mission to encourage Singaporeans to speak Good English — English that is intelligible to English speakers all over the world.

She then immediately clarifies what is meant by intelligible English, which is: It is not about accent. It is about speaking grammatically correct English, so as to be understood, as our tag line this year puts it, ‘not only in Singapore, Malaysia and Batam’.

Good English, global English and intelligibility are all intertwined. Intelligible English is defined firstly as an issue of grammatical correctness — which, in the context of Singapore’s exonormative practices, is determined outside of Singapore, primarily the United Kingdom. Accent is considered separate from issues of intelligibility — although some, like Bamgbose (1988), define intelligibility based on various linguistic features,

140 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff including phonology. Second, intelligible English is defined by what is globally understood. In keeping with the pragmatic legitimization for English as needed to access the global economy, if the English used by Singaporeans is not globally understood, then there is no reason for English to have a presence in Singapore. This stance comes through even stronger in her next comment: In an increasingly complex, knowledge-based global economy, the better our command of English, the greater will be our ability to comprehend and communicate in contexts where English matters. (our emphasis)

In her speech, Koh also makes it clear that the centre of English is outside of Singapore, and that Singaporeans are not native speakers of English. Consider the following two excerpts: Anyone who learns a second or foreign language will learn the standard form, not a dialect or a non-standard variety. We too must learn and use English in the standard form. … Indeed, native English speakers have said that in the region, the standard of English in Singapore is comparatively high.

Her first excerpt is curious. What she is suggesting is that everyone is learning English, making English a second language for all when in fact many Singaporeans would regard English as their first language. She is also suggesting that those who speak Singlish (which is what is implied here by the dialect/non-standard variety) are non-native speakers of English. The second excerpt is unequivocal in its statement that Singaporeans are non-native speakers of English in her reference to exonormative native speakers (speakers from Inner Circle countries) who are apparently the authority on the standard of English spoken in Singapore. The choice of an exonormative model perpetuates the division of language users into native and non-native speakers. What we see in the discourse of government leaders is thus a steadfast rejection of any possibility of Singaporeans to take up ownership of either standard English or Singlish. As we argue elsewhere (Bokhorst-Heng et al. 2007), one of the main criticisms against the NS/NNS dichotomy has been how it suggests negative assumptions regarding English speakers in the Outer Circle with respect to their general proficiency level, the credibility of their intuitive judgements, and their overall authority over the language in terms of their confidence and sophistication in using it. It does not take into account how the Outer Circle speakers themselves feel about their sense of ownership, belonging, or identity towards English. And it ignores the complexities emerging from the manner in which particular historical, socio-cultural, economic, and

Whose English? 141 political factors have shaped the realities of these speakers as reflected in their use of the language. Finally, such a dichotomy assumes a single uncontested linguistic norm within which English operates in Outer Circle countries. Yet, as the debates around Singlish and English indicate, there are competing standards and measures at play in Singapore English (Pakir 1998), whereby the government and some members of the population accept an exonormative standard of English, while others embrace the development of indigenous endonormative standards. These two speeches by Koh and Goh illustrate the dynamics of the English ideological debates, anchored on discourses of NS/NNS and ownership. The net result of such discourses is that English remains a language to be used and to provide access to the benefits that come with proficiency in English; however, it is not to be owned. And therein lies a key threat to the legitimacy of Singlish, for, as a home-grown local variety, it threatens to emerge as a language of identity. For many in Singapore who see a place for both standard English and Singlish, Singlish offers uniqueness in an English-dominant globalized world; while English is used by ‘everyone’, Singlish is owned by Singaporeans. As put by one contributor to The Straits Times forum page, ‘Singlish is a mark of how we have evolved as a nation and should surely have a place in our culture’ (Lee 1998). However, by denying a place for Singlish in the official discourse, English remains a language of the so-called native speaker, and a language to only be used — denying Singaporeans ownership (both in usage and positioning) of both standard English and Singlish. In the remaining section of this chapter, we examine ways to think about these issues of NS/NNS and nuance notions of ownership.

Language ownership

Background to the study Drawing on Higgins’ (2003) comparative study on speakers’ orientations towards English norms of members of Inner and Outer Circle countries, we examined Singaporeans’ degree of ownership of their norms of English. By ‘ownership’, we do not mean indigenization in which speakers appropriate English for their own needs (as in, for example, Widdowson 1994), although that is part of it. Rather, along with Norton (1997) and Higgins (2003), we take ownership to mean the degree to which speakers of English in Singapore ‘project themselves as legitimate speakers with authority over the language’ (Higgins 2003: 615). That is, in contrast to the government’s institutional stance, which focuses on the degree of English language use, we focus on speakers’ own subjective stand. Norton implicitly

142 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff makes this distinction in her conceptualization of ownership in reference to second language acquisition among immigrants in Canada, by establishing a relationship between ownership and language acquisition. Furthermore, through her research she demonstrates how speakers’ investment in the language they were learning ultimately leads to ownership through a sense of the ‘right to speak’ (that is, their legitimacy as a speaker): ‘[i]f learners of English cannot claim ownership of a language, they might not consider themselves legitimate speakers of that language’ (Norton 1997: 422). In her work, Norton saw legitimacy as being constructed and revealed through discourse. Higgins extends Norton’s argument to speakers of English in the Outer Circle: ‘If these speakers [in the Outer Circle] are invested in their local varieties and view them as forms of symbolic capital, it follows that their standard (i.e., target) variety is a local variety, and hence, that they view themselves as legitimate speakers of English’ (Higgins 2003: 621). She notes as well that the situation is often complicated by various varieties of English co-existing in a community (for example, a local variety and an Inner Circle variety), and that not all speakers of English have equal access to the claim of ownership. In her words, ‘[g]iven the inequitable social, economic, and political histories of certain groups in colonial and postcolonial contexts, relatively few populations have achieved full access to English via English-medium schooling, the primary setting for acquisition of English’. Furthermore, she notes (and as we have seen in our earlier discussion), governments may block claims to ownership. Building on Norton’s conceptualization of legitimacy as discursive, Higgins developed an Acceptability Judgement Task (AJT) to elicit conversation that would reveal variation in degree of ownership through participants’ responses to a series of English sentences. The AJT (see Appendix 6.1) consisted of twenty-four English sentences, some of which were taken from real data of Englishes around the world, and others that were deliberately made ungrammatical (Higgins 2003: 626–8). The AJT was administered to dyads. The task of eliciting their intuitive responses was accompanied by a recording of their conversation, reflecting the way these speakers positioned themselves in the process of articulating their orientations to English norms. In our view, this also reflected the way they were enacting their sense of ownership towards English. Using the tools of conversational analysis, Higgins measured the degree of ownership in terms of criteria such as the certainty, confidence, and selfreliance in their own linguistic intuitions about what was deemed correct usage with regard to the AJT sentences. Central to Higgins’ analysis is the concept of footing — a term from Goffman (1981), referring to the position or alignment an individual takes when using a given linguistic expression. Goffman distinguishes three senses of a speaker: animator, as someone

Whose English? 143 who is only involved in the production; author, as a speaker who creates and owns the production; and principal, ‘the party to whose position, stand and belief the words attest’ (Goffman 1981: 226). To further elaborate Goffman’s concept of footing, Higgins introduces Scollon’s (1998) ‘receptive roles’ — receptor, which mirrors animator; interpreter, the counterpart of author; and judge as the receptive role in relation to the speaker role of principal. The receptor receives the communication, but does not evaluate it; the interpreter construes a meaning from the communication; the judge evaluates and validates the communication. In replicating Higgins’ study, we introduced a further complexity to the analysis by focusing on degrees of ownership in terms of age, ethnicity, and socio-economic class. Twelve dyads were formed on the basis of (a) ethnicity: Malay, Chinese and Indian (Tamil) dyads; (b) age: old (around 50 years of age) and young (between 25 and 35 years of age), each having received their education during different language-in-education policy eras; and (c) socio-economic status: upper middle class and lower middle class. 2 Each of the twelve dyads consisted of persons of the same raceage-SES category. Paralleling the national census data, none of the Malay participants use only English at home; three indicated using Malay and English. And of the Indian participants, two use only English at home; four use English and Tamil; and two (the lower-middle class, older dyad) use only Tamil. While participants of different ages had received their education during different language-in-education policy eras, nonetheless, they all had received English language instruction — either in schools that had English as a ‘second language’, with maths and science taught in English, or in schools that used English as the medium of instruction and taught English as a ‘first language’. They all use English on a daily basis, as their primary language at work, and in other public domain interactions. In an unaccompanied recorded conversation, dyads were asked to read each sentence one at a time and then together decide if they thought the sentence was ‘acceptable’, ‘not acceptable’, or if they were ‘not sure’. If they thought the sentence was not acceptable, they were to revise the sentence and asked to indicate how they determined their decisions. In attempting to gauge the degree of English language ownership felt by the participants, we used three linguistic markers to determine their changes of footing: (a) references of legitimization; (b) the use of pronouns; and (c) modal use. In relation to the first marker, we focused on the use of external sources versus references to their own English usage. The participants’ use of ‘grammar’ was most often used in reference to an external authority, with no relational value to themselves or to other members of the speech community, and was used to legitimize their decisions in their role as judges or principals. The use of ‘grammar’ is thus seen as an impersonal concept as opposed to the use of concepts where the footing is clearly as

144 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff judge. This distinction is further made by the second linguistic marker, the use of pronouns, whereby we focused primarily on the differences between a choice of using personal pronouns such as ‘I’ and ‘you’ in contrast to the more general statements such as ‘(it) sounds wrong/correct’. Statements involving verbs of perception and belief such as ‘I think’ and ‘I hope’ were also seen as indicative of a strong sense of solidarity and community, and of a sense of ownership. The third linguistic marker used to indicate changes in footing are modality markers, which range from the use of a variety of modal verbs to adverbials such as ‘definitely’. Such usage reflects the position of the speaker, particularly in the sense of being judge. In our view, in a country like Singapore, where many people feel that they are first-language speakers, the issue of confidence is an important indicator of ownership (although this is not to suggest that all confident speakers necessarily claim ownership). There are of course various reasons that could explain the different uses of modals and pronouns and the references to grammar rules in the data. However, at the same time, this mode of analysis is in line with Goffman’s (1981: 126) stand that ‘change in footing is very commonly language-linked’. And as indicated by Higgins (2003: 629), these linguistic features can be related to shifts in footing from being receptor (if they read the sentence aloud) to interpreter (offering their understanding of the sentence) to being judge (when they determined whether the sentence was acceptable or not and then offered their basis for their judgement). The focus of our analysis was thus on changes of footing in the data and not directly about the actual words used. By weaving together these various linguistic strategies, a picture begins to emerge as to the different orientations the various participants held towards English and their comparative degree of ownership.

Findings Our discussion here focuses on the Malay and Indian dyads. Looking at the decision strategies these dyads reported using in making their decisions (Table 6.1), we find that overall, the Malay dyads were much more exonormative in their judgements than were the Indians. That is, apart from the lower-middle class, older (LM-O) dyad, the Malay dyads were more inclined to reference grammar rules to substantiate their decisions. And for both ethnic groups, the upper-middle class, older (UM-O) dyad relied on ‘grammar’ most frequently in justifying their judgements compared to the other dyads. Furthermore, for both ethnic groups, the lower-middle class, younger (LM-Y) dyad relied most on their own intuition, on whether or not it ‘sounded’ right or wrong — although the Indian dyads, apart from the lower-middle class, older (LM-O), were markedly

Whose English? 145 Table 6.1 Decision-making strategies of acceptability judgements (%)* Dyad**

‘It sounds right/ wrong’ Malay

Grammar rule

Guess

No answer

Indian

Malay

Indian

Malay

Indian

Malay

UM-O

8.3

45.8

91.7

54.2

0.0

0.0

0.0

Indian 0.0

LM-O

25.0

4.2

10.4

12.5

0.0

4.2

64.6

79.2

UM-Y

25.0

43.8

70.8

35.4

0.0

2.1

4.2

16.7

LM-Y

39.6

61.9

54.2

38.6

6.3

0.0

0.0

0.0

*

Where dyad members disagreed in their recorded strategies, the average of their responses is used. ** UM-O = Upper-middle class, older; LM-O = Lower-middle class, older; UM-Y = Uppermiddle class, younger; LM-Y = Lower-middle class, younger

more inclined to use this strategy than the Malays. This intuitive strategy suggests a stronger sense of ownership among young Singaporeans, as they position themselves as able to make judgements without any external authority. For both Malays and Indians, the LM-O dyad included far more ‘no answers’ in their written judgements than any other dyad. This suggests that, overall, this dyad demonstrated far less confidence in their judgements than did the other dyads. The actual judgements made by the dyads also suggest patterns of ownership. For both the Malay and Indian dyads, the UM-O dyad had the greatest number of ‘not OK’ judgements, which then required negotiation and discussion as to the basis for their decision and a revised version of the sentence. However, the Malay dyads across all groups had the greatest number of ‘not OK’ judgements compared with the Indian dyads. It appears that a lack of sense of ownership leads to more ‘not OK’ judgements. The Malay dyads also gave the greatest number of ‘not sure’ judgements. Table 6.2 Acceptability judgements of target sentences (%) Dyad

‘OK’ judgements

‘Not OK’ judgements

Malay

Indian

Malay

Indian

UM-O

4.2

25.0

87.5

LM-O

25.0

54.2

58.3

UM-Y

20.5

16.7

LM-Y

29.2

18.7

‘Not sure’ judgements Malay

Indian

75.0

8.3

0.0

41.7

16.7

4.2

70.8

4.2

8.3

4.2

66.7

4.2

4.2

0.0

Among the Indian respondents, the LM-O gave the greatest number of sentences ‘OK’ judgements, which suggests their overall hesitancy in making judgements as to the correctness of a sentence, finding it easier to accept a sentence than to find a basis for judging it correct.

146 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff According to Goffman (1981), pronouns index the source of authority, or judge. Higgins (2003: 635) takes up this view in her discussion of the syntactic frame ‘you + can say/use’, which, she argues, indexes ownership among speakers who use it in their judgements. One is reminded too of Fairclough’s (1989: 88) observation that the pronoun ‘you’ has relational value (in contrast to ‘one’), being used to register solidarity with the others presumably in the same community. The use of the pronoun ‘we’ has the same value. And the use of ‘I think’ indicates personal assurance of judgement. In this way, the participants’ use of pronouns reveals their degree of reliance on external authority to determine their footing as judge (see Table 6.3). What is striking here is the high percentage of the use of ‘I think’ by the Indian LM-Y dyad, in contrast to the others. And, overall, the Indian dyads made greater use of ‘I think’ than the Malays, again suggesting a pattern we are beginning to see of greater confidence by the Indian dyads in their judgements and ownership than the Malay. The greater use of ‘you’ and ‘we’ by the older Indian dyads than their Malay counterparts again suggests a greater awareness of the speech community to which they belong, and hence, a stronger sense of ownership. However, the younger Malay dyads showed slightly greater solidarity with the speech community than did their Indian counterparts. Table 6.3 Use of pronouns (% of turns) Dyad

you (generic)

I think

we (generic)

Malay

Indian

Malay

Indian

Malay

Indian

UM-O

4.7

7.9

0.5

2.2

1.3

3.2

LM-O

0.8

0.6

0.4

2.4

0.0

3.0

UM-Y

2.8

3.2

2.3

0.7

1.6

0.2

LM-Y

0.0

16.6

0.9

0.5

1.8

1.4

Finally, in Table 6.4, we see how the participants differ in their use of modals. The use of these terms signals someone who has authority over English and is able to assert what is possible and not possible to say in English. What is notable here again is the greater use of these modals by almost all of the Indian dyads compared with the Malay dyads, and, for both ethnic groups, the generally lower use of these modals by the LM-O dyads. Table 6.4 Use of modals (% of turns) Dyad

can/cannot (emphatic)

should

is (emphatic)

Malay

Indian

Malay

Indian

Malay

Indian

UM-O

2.1

0.2

2.1

4.0

6.1

22.3

LM-O

0.4

0.2

2.4

3.7

3.2

6.7

UM-Y

0.5

0.7

2.1

2.3

13.1

9.9

LM-Y

0.5

7.6

6.8

3.8

9.9

17.1

Whose English? 147 Taking these figures together, the general pattern that emerges is thus one that suggests the LM-O dyads as being the least confident of their judgements, and reflecting the weakest sense of ownership; and the younger dyads showing a greater inclination to draw upon their own intuition rather than follow exonormative rules. Finally, the Indian dyads overall indicate a greater sense of confidence and ownership compared with the Malay dyads. To nuance these patterns, we turn to a detailed discussion of the discourse patterns used by the dyads, demonstrating how their discourse positions their orientation towards their sense of ownership. Due to the limitations of space and for clarity, we focus here on their responses to just one of the sentences (more detailed analyses for each set of dyads can be found in Bokhorst-Heng et al. 2007 and Rubdy et al. 2008): Sentence 8: I picked up a rice from the floor and threw it away.

The first two excerpts (1 and 2) are from the UM-O dyads (speaker A and speaker B in each), both are concerned with grammar rules as they negotiate their decisions. Excerpt 1: UM-O-Malay A:

B: A: B: A: B: A: B: A:

I’m not talking about the rice, er, rice as uncountable noun. A grain of rice and threw it away? Or I pick up the rice on the floor? And threw it away? That could be collective noun.= =Um, um. But I think maybe, both can lah, huh? Um. A grain of rice? Pick up.= =A grain of rice. [I picked up THE grain of rice from the floor, [‘The grain’, because it’s er, SPECIFIC. Yah.

Excerpt 2: UM-O-Indian A: B: A: B: A: B:

Okay. I picked up a rice from the floor and threw it away. A rice, wrong. A rice is actually uncountable. Um. We don’t usually, erm, the article there. A grain of rice.

B: A: B:

No, no, I, a. We, we, I picked up rice from the floor and threw it away. A GRAIN of.

148 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff In both excerpts 1 and 2, the dyads make reference to grammatical terms such as countable nouns, articles, and collective nouns. And both identify ‘grammar’ as their means of justification for their judgements. For the Indian dyad, the fact that they were teachers may have played a role in their frequent references to grammar in justifying their judgement. In addition, their discourse is also reflective of the attention given to grammar in language teaching during their time in school. However, in addition to reliance on the exonormative principles of grammar, members of the Indian dyad also position themselves within the wider speech community through their pronoun use in ‘we don’t usually’ and ‘we don’t use’, and after beginning their deliberation with the firm assertion that ‘ “A rice” IS ACTUALLY …’. They also move very quickly from receptor to judge, as was evident in almost all of their exchanges (Rubdy et al. 2008). There were no similar assertions or rapid judgements in the Malay dyad’s deliberation. Instead, there is the use of ‘that COULD BE’ and ‘I think’ but then followed by ‘maybe’, with a rising tone suggesting uncertainty and the need for affirmation. In contrast to the UM-O dyads, the LM-O dyads in excerpts 3 and 4 indicate no strategy for their decisions made, and indicate no positioning in relation to ownership. Excerpt 3: LM-O-Malay B: A: B: A:

B:

Ok. I pick up a rice from the flour, floor, and threw it away. Hah. Lagi tak ok: ‘Even worse:’ Not ok eh? You pick abih you throw ni apa? Ok lah … peduli lah apa-apa ah: ‘You pick then you throw this what? Ok lah … anything lah don’t care:’ Not ok. (pause). Abih? ‘Not ok. (pause). Then?’ (‘Then’ in this instance is used to signal a move to the next sentence for discussion.)

Excerpt 4: LM-O-Indian A: B: A: B: A:

I picked up a rice and throw it. Straight fowarded thaana athu. I pick up a rice from the floor. ‘Straight foward that one. I pick up a rice from the floor.’ Very (.) clear, where did they pick it up= =Ah. From the floor not ___ Fine.

Whose English? 149 Throughout the task, the LM-O dyads provided very little discussion, and, as we noted earlier, tended to leave many decisions blank. They were the only dyads to use their mother tongue in their discussions (Malay and Tamil respectively above), and it was evident that the task was difficult for them. At one point, one of the Malay participants in this dyad said, ‘apa-apa lah. Mata naik kelabu asap, eh’, ‘Anything lah. My eyes are turning gray, eh.’ And very similarly, one of the Indian participants in this dyad expressed, ‘enakku appadi thala kanamaa irruku, kaacha vara poothu’, ‘My head is very heavy, going to get fever.’ In excerpts 3 and 4, while the Malay dyad felt the sentence was incorrect, there was a sense of just wanting to get through the exercise (‘Ok lah … peduli lah apa-apa ah …’,‘Ok lah … anything lah don’t care …’). The Indian dyad felt that the sentence was correct and was more concerned about the location of the rice. These dyads systematically showed the least confidence in their judgements and were the least willing to commit to a position, ultimately indicating little sense of ownership. Excerpts 5 and 6 come from discussions by the two UM-Y dyads. Both dyads throughout the task demonstrated confidence in their responses, making frequent use of the emphatic is, and often relying on their own intuitions and membership in the speech community that adheres to the conventions they were arguing for to support their answers. But the Malay dyad was also particularly concerned with ‘getting it right’. They were careful to revise a sentence so that it was correct in every aspect, and even while relying on their own intuitions, they would also wonder about the appropriate grammar rule associated with their decision as if seeking further validation. At one point, they even turned on the computer they had with them and used Microsoft Word’s grammar and spelling check to confirm their decision. This is evident as well in excerpt 5, when one of the partners comments on the need of Singaporeans to have things ‘right’. This exchange also suggests a view that there must be one correct form, a single set of norms. Excerpt 5: UM-Y-Malay A: B: A: B: A:

It’s a grain of rice right? Not the something else of rice I’ll hope not. (laughter) Can we know the correct answer _____. Like taking exam (laughter). Singaporeans has this mentality. Have to get it right.

Excerpt 6: UM-Y-Indian A: B:

I picked up= =A rice from the floor and throw it away.

150 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff A: B: A: B: A:

I picked up rice from the floor and throw it away.= =Yes. [So, it’s definitely grammar rule. [Like that lah. Okay.

Unlike the Malay UM-Y dyad, the Indian dyad demonstrated a very strong intuitive sense of ownership. They actively — and often at great length — engaged in the role of interpreter by discussing their interpretations of particular sentences. However, they also frequently, as in excerpt 6, moved quickly from receptor to judge, with very little discussion. They also displayed a great deal of authority in reaching their decision, as is evident in their frequent use of the emphatic ‘is’ (9.9%), and in their lexical choices such as, in this example, the use of ‘definitely’. Both the LM-Y dyads spent very little time discussing their answers or interpreting the various possible meanings of a sentence, and both very quickly moved from receptor to judge. For the Indian dyad, for example, in the course of eight turns, they reached a judgement on three target sentences. Both dyads also displayed a great deal of authority in reaching their decisions, relying on their own intuitions. In 40% of their answers, the Malay dyad determined their answers by ‘it sounds right/wrong’. They made frequent use of modals, as in excerpt 7, ‘cannot’, in addition to the emphatic phrases ‘it is either’, and ‘no need to’. In their discussion, there is no explicit reference point; they simply assume what is to be true. Neither was there any sense of belonging to a wider speech community; they made very little use of the relational pronouns ‘we’ and ‘you’ — the use of ‘we’ here being one of only two instances. Excerpt 7: LM-Y-Malay A:

B: A: B: A: B: A: B:

I picked up a rice from the floor and threw it away. I picked up the rice from the floor and threw it away. The rice. A rice. Let say, I picked up rice from the floor and threw it away because rice cannot be, is either one what, a ___ is one. :mmm We cannot count rice. So it’s the right? It’s either the a or none; no need to put the. __________ I picked up rice, ah? If I put the here. I picked up the rice from the:Mmm

Whose English? 151 Excerpt 8: LM-Y-Indian B: A:

B: A: B: A: B: A: B: A: B:

I picked up a rice from floor and threw it away. Usually rice. Yes. No count lah. So, I picked up rice from the floor unless I picked up a grain of rice from the floor and threw it away. :Mmm (0.5) I would just put a= =Picked up rice Ya. Picked rice from the floor and threw it away. Ah okay. [ ___ grain of rice? How is it if it’s not a grain? :Mmm If it’s not a grain of rice or lump of rice. If it’s a lump of rice of rice we use rice, lah. But if it’s like one single piece that’s then you use a grain. Ya. (0.5)

The Indian dyad’s discussion in excerpt 8 similarly shows strong confidence in their judgements. Their reference to ‘no count’ also indicates an awareness of the grammar rule at play. But more frequently evoked in this exchange and throughout the exercise is their own intuition and membership of the speech community: ‘usually’, indicating an awareness of what is the accepted norm among other users of English, and the use of relational pronouns in ‘if it’s a lump of rice WE use’ and if ‘one single piece’ then ‘YOU use’ indicating membership. One of them even evokes her own norms as valid rationale: ‘I would just put’.

Summary and discussion The purpose of our analysis was to examine speakers’ orientations towards their English norms, and the ways in which they positioned themselves in the process of articulating their orientations. What distinguished the UM-O dyads was their reliance on exonormative grammatical rules to justify their judgements. This may in part be accounted for by the pedagogical emphasis on grammar when they learned English. For both the Indian and the Malay dyads, the LM-O dyads demonstrated the least confidence in their judgements, suggesting little ownership with their norms of English usage. And what distinguished the Malay dyads overall was their greater reliance on exonormative rules (the UM-O showing the greatest tendency, but evident in all four dyads) compared to the Indian dyads, rather than relying on their own patterns of language use or that of the larger speech community.

152 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff The younger Malay dyads did show a slightly greater sense of ownership than did the older ones; however, there was a clear friction between the exonormative government-sanctioned and school-taught norms of ‘standard English’ and their acceptance of norms as they occur in everyday speech. That the younger dyads exuded a greater degree of certainty about their judgements on the correctness of particular sentences and the rapidity with which they reached their judgements demonstrates the degree to which Singapore English has undergone endonormative stabilization. The Indian dyads were particularly confident in this stabilization. Why the (younger) Indian dyads showed the greatest sense of ownership of their English norms might have to do with the degree to which they have appropriated English in their everyday speech, thus maximizing their sense of English ownership. Recall that only 8% of Malay Singaporeans speak English predominantly at home, and so it is very much a school-learned (and assessed) practice, whereas 36% of those in the Indian community habitually use English.

Conclusion We began our chapter with a discussion of language ideological debates, and Blommaert’s view that language ideological debates have to do with the ‘politics of representation’ (1999: 9). In the Singapore English language debates, one key aspect of the ‘politics of representation’ has to do with ownership, and the NS/NNS divide. The NS/NNS discourse followed by the Singapore government leaders serves to keep English at a level of use, rather than ownership. In their view, the wider use of English by Singaporeans is merely the confirmation of English as an ethnically neutral, interethnic lingua franca, and does not bear on the question of language ownership. Yet, as we examine Singaporeans’ own orientations towards their English language norms, we do see a growing, albeit uneven, sense of ownership, in the way they position themselves as judge and arbitrators of English usage, and their willingness to rely on their intuitions, and the way they identify themselves with a broader community of English speaker. Their subjective speaker ownership thus stands in direct contrast to the government’s position on the mother tongue (which, as we have noted, is not aligned with patterns of language use). This is particularly true of the younger dyads who come from a generation where being a native speaker of English is increasingly the norm. What this suggests is that, as Singaporeans increasingly use English habitually in their everyday home and work lives, this sense of ownership and intuitive authority over their English norms will increase. It also suggests a growing endonormative stabilization of English in Singapore, and with it a sense of there being a community of speakers with which to identify. We thus see an increasing tension between government policy and Singaporean’s own relationship with the English language.

Whose English? 153 This study is limited in its analysis of ownership because it involves only language identities expressed in an experimental setting. Participants may orient to English very differently in different contexts with different speakers. The study also only involved a small number of dyads, thus limiting the scope of analysis. However, this discussion adds another layer to the ‘English language debates’ in Singapore, making transparent the need to reconsider the parameters of the ideological constructedness of English, particularly in replacing the notion of native speaker with the concept of ownership.

Notes 1.

2.

Phua Chu Kang, the lead character in a local sitcom by the same name, has been at the heart of English debates. A Singlish-speaking contractor, Phua Chu Kang’s use of English prompted a heated debate in late 1999 about Singlish. At the prompting of then prime minister Goh Chok Tong, Phua Chu Kang was sent to English language school to improve his English — and to set an example for all Singaporeans on the importance of ‘good English’. SES was determined by housing type (small apartment versus house), income level (less than S$4,000 versus over S$5,000), and highest level of education attained (less than a university degree versus at least one degree). Full details of the participants and the methodology used can be found in Bokhorst-Heng et al. (2007) and Rubdy et al. (2008).

Transcription conventions: (.) (0.5) [ = : CAPS ?

___

micropause half-second pause talk in overlap cut off latched talk sound stretch loud volume rising contour slowly enunciated speech inaudible

References Alsagoff, Lubna (2007) Singlish: Negotiating culture, capital and identity. In Critical Studies and Education in Singapore. Edited by Viniti Vaish, S. Gopinathan and Y. B. Liu. Rotterdam: Sense Publishers, pp. 25–46. Bamgbose, Ayo (1988) Torn between the norms: Innovations in World Englishes. World Englishes, 17(1), 1–14.

154 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff Blommaert, Jan (1999) The debate is open. In Language Ideological Debates. Edited by Jan Blommaert. Berlin/New York: Mouton de Gruyter, pp. 1–38. Bokhorst-Heng, Wendy D. (2005) Debating Singlish. Multilingua, 24(33), 185–209. Bokhorst-Heng, Wendy D., Alsagoff, Lubna, McKay, Sandra and Rubdy, Rani (2007) English language ownership among Singaporean Malays: Going beyond the NSNSS dichotomy. World Englishes, 26(3), 424–45. Crystal, David (1995) The Cambridge Encyclopedia of the English Language. Cambridge: Cambridge University Press. Fairclough, Norman (1989) Language and Power. London/New York: Longman. Fong, Vivienne, Lim, Lisa and Wee, Lionel (2002) Singlish: Used and abused. Asian Englishes, 5(1), 18–39. Goffman, Erving (1981) Forms of Talk. Philadelphia: University of Pennsylvania Press. Goh Chok Tong (2000) Speech by Prime Minister Goh Chok Tong at the launch of the Speak Good English Movement on Saturday, 29 April 2000, at the Institute of Technical Education (ITE) headquarters auditorium, Dover Drive, at 10.30 a.m. Document number gct20000429j. Singapore: Ministry of Information, Communication and the Arts. http://stars.nhb.gov.sg/stars/public/. Accessed 26 May 2007. Higgins, Christina (2003) ‘Ownership’ of English in the outer circle: An alternative to the NS-NSS dichotomy. TESOL Quarterly, 37(4), 615–44. Kachru, Braj B. (1985) Standards, codification and sociolinguistic realism: The English language in the Outer Circle. In English in the World: Teaching and Learning the Language and Literatures. Edited by Randolph Quirk and Henry Widdowson. Cambridge: Cambridge University Press, pp. 11–30. Kamsiah Abdullah and Bibi Jan Mohd Ayyub (1998) Malay language issues and trends. In Language, Society and Education in Singapore: Issue and Trends. 2nd ed. Edited by S. Gopinathan, Anne Pakir, Ho Wah Kam and Vanithamani Saravanan. Singapore: Times Academic Press, pp. 179–90. Koh, Tai Ann (2006) Speech by Professor Koh Tai Ann, Chairman, Speak Good English Movement Committee, Launch of the Speak Good English Movement 11.00 a.m., The Plaza, National Library Building. http://www.goodenglish.org. sg/site/official-speeches/speech-by-prof-koh-tai-ann.html. Accessed 26 April 2007. Lee, M. (1998) No shame in using Singlish. The Straits Times, 3 November, 35. Lim, Lisa and Foley, Joseph A. (2004) English in Singapore and Singapore English: Background and methodology. In Singapore English: A Grammatical Description. Edited by Lisa Lim. Amsterdam/Philadelphia: John Benjamins, pp. 1–18. McKay, Sandra Lee and Bokhorst-Heng, Wendy D. (2008) International English in Its Sociolinguistic Contexts: Toward a Socially Sensitive EIL Pedagogy. Mahwah, NJ: Lawrence Erlbaum/ Routledge. Ministry of Education (2007) Executive Summary Report on the English Language Curriculum and Pedagogy Review. Singapore: Ministry of Education. Norton, Bonny (1997) Language, identity, and the ownership of English. TESOL Quarterly, 31(3), 409–29. Pakir, Anne (1998) English in Singapore: The codification of competing norms. In Language, Society and Education in Singapore: Issues and Trends. 2nd ed. Edited by S. Gopinathan, Anne Pakir, Ho Wah Kam and Vanithamani Saravanan. Singapore: Times Academic Press, pp. 64–84.

Whose English? 155 Quirk, Randolph (1985) The English language in a global context. In English in the World: Teaching and Learning the Language and Literatures. Edited by Randolph Quirk and Henry Widdowson. Cambridge: Cambridge University Press, pp. 1–6. Rubdy, Rani, McKay, Sandra Lee, Alsagoff, Lubna, and Bokhorst-Heng, Wendy D. (2008) Enacting English language ownership in the ‘Outer-Circle’: A study of Singaporean Indians’ orientations to English norms. World Englishes, 27(1), 40–67. Scollon, Ron (1998) Mediated Discourse as Social Interaction: A Study of News Discourse. London: Longman. Singapore Department of Statistics. (2000) Singapore Census of Population, 2000, Advance Data Release No. 3, Literacy and Language. http://www.singstat.gov.sg/ keystats/people.html. Accessed 8 May 2007. Speak Good English Movement (2000–2007) Speak Good English Movement. http:// www.goodenglish.org.sg. Vaish, Viniti, Tan, Teck Kiang, Bokhorst-Heng, Wendy D., Hogan, David and Kang, Trivina (2010) Language and social capital in Singapore. In English in Singapore: Modernity and Management. Edited by Lisa Lim, Anne Pakir and Lionel Wee. Hong Kong: Hong Kong University Press, pp. 159–80. Wee, Lionel (2002) When English is not a mother tongue: Linguistic ownership and the Eurasian community in Singapore. Journal of Multilingual and Multicultural Development, 23(4), 282–95. Widdowson, H.G. (1994) The ownership of English. TESOL Quarterly, 28(2), 377–89.

156 Wendy D. Bokhorst-Heng, Rani Rubdy, Sandra Lee McKay and Lubna Alsagoff

Appendix 6.1 Acceptability Judgement Task Sentences Type and sentence

Country where attested

Neologism 1. If a passenger on a preponed flight shows up at the time written on his ticket and finds that the plane has already left, he should be entitled to a refund. 2. I am sorry for the botheration I have caused you.

India

India, Malaysia, United Kingdom

3. The gloriosity of the sunset made us wish that we had our camera.

Invented

4. The perfectity of a new computer program can only by tested by running it.

Invented

Countability of nouns 5. The school was able to buy new computer equipments for the students last year. 6. Many researches have shown that smoking cigarettes is dangerous.

India, Malaysia, Philippines, Singapore *KDQD,QGLD Malaysia, Singapore, United Kingdom

7. The children fell in the muds near the swamp behind the house.

Invented

8. I picked up a rice from the floor and threw it away.

Invented

Topic-comment structure 9. English they have declared the official language of Kenya.

India, Nigeria

10. TV I don’t usually watch because I have too much homework.

Malaysia

11. Outside the boys they like to play even if it is extremely cold.

Invented

12. For research papers, the students, they use computers to type them.

Invented

Tense/aspect 13. She was having a headache and could not concentrate on the lecture. 14. I have read this book yesterday.

India, Nigeria, Kenya India

15. Jones breaks the record for the highest number of invented goals per game this season.

Invented

16. Every time we go to the movies, my father bought popcorn for us.

Invented

Whose English? 157 Type and sentence

Country where attested

Prepositions 17. It is difficult for me to cope up with all the work that my boss gives me. 18. The student requested about an extension for her research paper because she was sick for five days.

Malaysia, Singapore India, Nigeria, South Africa

19. After you have read the instructions, please fill the form so that your request can be processed.

Invented

20. After the meeting, the managers discussed about the possibility of raising their invented employees’ salaries by 10%.

Invented (possibly undergoing codification)

Distractors 21. Although many students have studied English for more than five years, many of them have not mastered punctuation skills. 22. One of my instructor told me that when a person learns a language, he or she also learns the culture of that language. 23. In the presidential election last year, he won by substantial majority. 24. Your daughter will attend the University of Wisconsin next year, isn’t it?

India, Kenya, Nigeria, Singapore, South Africa Malaysia

,QGLD0DOD\VLD Singapore, United States, United Kingdom ,QGLD0DOD\VLD Singapore, United Kingdom

Note: Except for the distractors, the first two sentences in each category are attested. The second two are invented based on productive morphological rules in English (neologisms and countability of nouns), misplacement of the topic in topic-comment structure, violation of punctual/nonpunctual distinction in tense and aspect, and unattested combinations of prepositions with verbs. Distractor sentences include variation in subject-verb agreement, articles, and tag question concord. The attested forms may occur in additional Englishes to those listed.

7

Language and social capital in Singapore Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang

Competence in EL [English Language] will be a source of competitive advantage for Singapore. Knowing our Mother Tongue Languages gives us confidence in our culture, roots and identity as a people. We have to stay an Asian society. Report of the Chinese Language Curriculum and Pedagogy Review Committee (2004) Malay is our national language and an important regional language. The learning of Malay is critical for cultural transmission in the Malay community. A sense of identity and the ability to access readily the values, history, literature and the arts of one’s own community are inextricably linked with a facility with one’s own language. The Malay language joins generations. Report of the Malay Language Curriculum and Pedagogy Review Committee (2005) The Committee’s vision is for Tamil to be a living language among future generations of Tamil Singaporeans, and vibrant part of Singapore’s identity as a multicultural, global city. Report of the Tamil Language Curriculum and Pedagogy Review Committee (2005)

The opening quotes above articulate Singapore’s commitment to bilingualism and to bilingual education, and make a link between the language and the culture of an ethnic group. The statements suggest that social capital is maintained through language — as seen in the phrases which state that ‘the Malay language joins generations’, and Tamil is to be ‘a living language among future generations’, and in the Chinese report where reference is made to ‘culture’, ‘roots’, and ‘a people’. Embedded in the significance of what we call the mother tongue policy is thus the notion of identity and connectedness to other members of the ethnic group through social ties.

160 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang Despite this strong commitment regarding bilingualism, it appears that the adoption of the ascribed mother tongue has not occurred equally across all ethnic groups. Table 7.1 Language trends in Singapore Language spoken at home

Ethnicity Chinese 1990

2000

Malay

Indian

2005

1990

2000

2005

1990

2000

2005

6.1

7.9

13.0

32.3

35.6

39.0

93.7

91.6

86.8

English

19.3

23.9

28.7

Mandarin

30.1

45.1

47.2

Malay Tamil Dialect Others Total

50.3

30.7

14.5

11.6

10.6

43.2

42.9

38.8

23.9

0.3

0.4

0.2

0.1

0.5

0.2

10.0

9.9

11.6

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Source: Singapore Department of Statistics (2001) Census of Population 2000, Advance Data Release, and Singapore Department of Statistics (2006) General Household Survey 2005, SocioDemographic and Economic Characteristics, Release 1.

Table 7.1 shows that from 1990 until 2005 English increased as the language spoken at home in all three ethnic groups. At the same time the use of mother tongue (Mandarin, Malay and Tamil) was going down. Furthermore in the case of the Chinese, the report from the Chinese Review committee states that ‘the number of Chinese students entering Primary 1 (P1) who speak predominantly EL at home has risen from 36% in 1994 to 50% in 2004’ (Ministry of Education 2004: 4). This discrepancy between the investment that the Singapore government has made in bilingual education to preserve the mother tongue and outcomes in terms of actual language use is a major challenge for a highly globalized Singapore. The uneven success of the mother tongue policy across ethnic groups also suggests that there is more at play than mere government policy. In fact, given the annual commitment to raise the status and awareness of Mandarin within the Chinese community through the annual Speak Mandarin Campaign, one could expect the acceptance and use of the mother tongue to be highest among the Chinese, when in fact it is not. Thus, clearly more is going on. In this chapter we use Coleman’s (1988) idea of social capital as a conceptual framework. Although this concept has been used to analyze many outcomes — like the drop-out rate of students and the level of industrialization in a country — it has not been used to analyze language use in a multilingual environment. We explore whether or not the use of the mother tongue, and thereby acceptance of the mother tongue policy, is a result of strong social ties, or social capital of an ethnic group. Specifically

Language and social capital in Singapore 161 we ask: What are the differences in language use on the basis of ethnicity? How are these differences materialized in the various socio-spacial domains? What is the relationship between funds of social capital and language across domains of language use? The term ‘language maintenance’ needs to be problematized at the outset. This term best describes language use in the Malay community as the government’s ascription of Malay as their mother tongue aligns with their actual patterns of language use. As Table 7.1 shows, in the Indian case there is a gradual language shift, though it is not as pronounced as in the case of the Chinese community. For the Indian ethnic group, we only present data on Tamil as we do not have any data on Non-Tamil-MotherTongues (NTMLs). Finally, the term ‘language maintenance’ is problematic for the Chinese, as embedded in their stories is a tale of language shift and the acceptance (or rejection) of government policy regarding an ethnically determined mother tongue — a policy to which we refer as the mother tongue policy. Thus, top-down institutional support and bottom-up language loyalty have not been for the maintenance of the Chinese ‘dialects’, which have, to some extent, been replaced by Mandarin and English. For the Chinese, when we speak of ‘language maintenance’, we are referring to Mandarin and not the Chinese ‘dialects’.

Social capital and language: A review of literature

Definitions ‘Social capital’ is a slippery concept. Though this concept has been known to scholars from the time of Durkheim and Marx, it is through the work of Bourdieu and Coleman that it gained popularity both in and outside the field of sociology. According to Portes (1998), the first clear definition of social capital was supplied by Bourdieu (1986 [1980]) who defined it as ‘the aggregate of the actual of potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition — or in other words, to membership in a group — which provides each of its members with the backing of the collectively owned capital, a “credential” which entitles them to credit, in the various senses of the word’ (Bourdieu 1983: 249). Portes points out that this definition actually did not catch on until Coleman’s (1988) influential paper on this topic. Coleman does not provide a statement that would clearly define social capital but elaborates on the nature of this concept in great detail. The closest he comes to a definition is when he writes ‘the consensus is growing in the literature that social capital stands for the ability of actors to

162 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang secure benefits by virtue of membership in social networks or other social structures’ (Coleman 1988: 6). Social capital is intangible because it inheres in the relationships of trust between community members. In this sense it is different from other forms of capital like economic capital (money), physical capital (machines and tools), and so forth. Thus, he argues for a definition of social capital based on its function. In his view, social capital is productive and has concrete outcomes that can be measured quantitatively. As an illustration of what is social capital, one of the examples that Coleman offers is that of Jewish diamond merchants in Brooklyn. He emphasizes the high level of trust amongst these merchants: a seller could leave a bag of diamonds with a colleague for evaluation without fear that his diamonds would be substituted for fakes. This level of trust is based on specific attributes of social structure like high frequency of interaction and degree of intermarriage, membership in the same synagogues in the Brooklyn area, and residence in specific neighbourhoods in Brooklyn: ‘these close ties, through family, community, and religious affiliation, provide the insurance that is necessary to facilitate the transactions in the market’ (Coleman 1988: 99). Coleman, like Bourdieu, sees social capital as an essentially positive concept that benefits all individuals between whom it exists. Like economic and physical capital, social capital is fungible in that it can be exchanged for other goods for the mutual benefit of parties. ‘The premise behind the notion of social capital is rather simple and straightforward: investment in social relations with expected returns in the marketplace’ (Lin 2001: 19). Portes (1998) extends this idea to also include negative consequences of social capital and emphasizes that these negative aspects of social capital must be acknowledged to give a more nuanced understanding of the concept. ‘Recent studies have’, according to Portes, ‘identified at least four negative consequences of social capital: exclusion of outsiders, excess claims on group membership, restrictions on individual freedoms, and downward levelling norms’ (Portes 1998: 15). Putnam (1995) and Fukuyama (1995) take social capital even further to be the characteristic of entire nations and talk about its positive outcomes, for instance, democracy or industrialization. Despite the fact that Portes (2000) has pointed to the problems inherent in measuring social capital at macro-social levels (for instance, the level of a nation), this concept is popular in current literature. The slipperiness of the term ‘social capital’ is also apparent in the different ways it is interpreted, for instance, by Putnam (1995) and Fukuyama (1995). In his renowned essay on the topic, Putnam (1995) explains declining social capital in the US on the basis of lack of trust amongst people, apathy with regards to the government (based on poor voter turnout), poor attendance at public meetings on town/school affairs or political rallies and lack of interest in serving on a committee of some

Language and social capital in Singapore 163 local organization like the Parent Teacher Associations (PTA). Putnam’s point, which created a stir in political circles, is that this lack of social capital undermines the very principle of democracy on which America is founded. He clearly sees social capital as a positive ideal that leads to ‘better schools, faster economic development, lower crime, and more effective government’ (Putnam 1995: 66–7). Whereas Putnam (1995) thinks that America’s social capital is declining, Fukuyama (1995) places both Japan and USA in the same basket of countries with high trust societies in which there are large funds of social capital. In contrast, China, according to Fukuyama, is a low trust society because kinship ties take precedence over ties amongst unrelated people. In his view, China is essentially a country of small businesses and cannot create large efficiently run corporations precisely because of this low social capital. However, both Putnam and Fukuyama are agreed over one aspect of social capital: that it can be depleted. We will see important parallels between the characteristics of social capital in China and those of the Chinese community in Singapore when we talk about the shift from the use of dialects to Mandarin. Castells (2004) disagrees with Putnam’s and Fukuyama’s notion of declining social capital. Though Putman’s (1995) focus is only on America and Castells’ (2000, 2004) is global, the latter’s view is that the era of globalization is marked by the rise of the local resulting in the strengthening of social ties at the level of community. He writes, ‘[p]eople resist the process of individualization and social atomization, and tend to cluster in community organizations that, over time, generate a feeling of belonging, and ultimately, in many cases, a communal, cultural identity’ (Castells 2004: 64). Castells also points out how religion and membership in religious organizations are the main form of cultural identity in an age of globalization. In this chapter we will explore how in the case of Singapore, religion, and its link with language, is an important identity marker and source of social capital for the Singaporean, though with differences across ethnic groups. In concluding this section we wish to acknowledge the other, and amongst applied linguists, more popular concept: linguistic capital. A comparison of social and linguistic capital is not the focus of this chapter, however; for an in-depth look at the way linguistic capital can be applied to the Sociolinguistic Survey of Singapore, see Vaish and Tan (2008).

Language as an outcome of social capital What is clear from Putnam’s and Fukuyama’s analyses is that there are outcomes of social capital, such as membership in voluntary organizations,

164 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang high voter turnout, and so forth. However, the idea of language use as an outcome of social capital has not been explored in depth. An exception is Milroy’s (2002) work whose social network theory has made important contributions to the field of sociolinguistics. Social network theory was developed in the 1960s and 1970s, based on the assumption that the relationships between speakers have important implications for language maintenance or loss. The key point in this theory is that ‘[n]etworks constituted chiefly of strong (dense and multiplex) ties support localized linguistic norms, resisting pressures to adopt competing external norms. By the same token if these ties weaken, conditions favorable to language change are produced’ (Milroy 2002: 550). We note the difficulty in discussing language maintenance in the pan-Singapore context where government policy has endeavoured to ascribe a ‘mother tongue’ to each of the three main ethnic communities, because the term ‘maintenance’ is not as appropriate for the Chinese ethnic group as it is for the Malay and Indian groups. However, we find Milroy’s social network theory useful as a way to theorize this relationship between language and social capital. Granovetter (1973) developed an argument about the strength of weak ties, which has had direct influence on the work of sociolinguists such as Milroy. The crux of Granovetter’s argument is that even weak social ties can have positive outcomes because they create bridges between diverse social groups and members thus have a larger pool of information to draw from. Granovetter’s thesis is based on the labour market, and how job seekers gather information that leads them to a new job. He found that in most cases it was through a friend of a friend, or through a weak tie, that the subject found a new job. More importantly, when a person changes jobs, he or she moves from one network of ties to another, thus creating a bridge between two weakly connected communities. Milroy’s social network analysis draws on both variationist sociolinguistics and the work of sociologists like Granovetter. Milroy defines a social network ‘as a boundless web of ties which reaches out through a whole society, linking people to one another, however remotely’ (2002: 550). The key finding of Milroy’s study of language shift/maintenance in Belfast was that a ‘close-knit network is interpreted as an important mechanism of dialect maintenance’ (Milroy 2002: 556). Milroy also suggests that weak social networks can create innovations in language which one does not find in dense and multiplex social ties. This is because members of weak social networks are mobile and might live in numerous locations becoming influenced by different dialects. Though Milroy’s work was on monolingual speakers in Belfast, social network theory has been applied to bilingual contexts by Zantella (1997) who studied the Puerto Rican Community in New York City, and Li (1994) who did the same with the Chinese immigrant community in Tyneside, UK.

Language and social capital in Singapore 165 Both these studies found that strong social ties support minority languages and resist the pressure to shift towards monolingualism in English. They also found that, when these networks weaken, language shift is likely to take place. Our chapter presents a different multilingual situation in that both Zantella and Li were looking at immigrant communities in a monolingual English environment. In contrast, the Sociolinguistic Survey of Singapore 2006 is located at a site where bilingualism is the norm and a relatively stable condition of the linguistic ecology.

The Singapore context Language planning in Singapore is both interventionist and egalitarian. Sociolinguists in Singapore acknowledge the fact that language planning in this country is interventionist in nature (Gopinathan et al. 1998), and see the promotion of Mandarin over dialects among the Chinese (to the point of even attempting to infiltrate the home) as an illustration of this. At the same time, there have been egalitarian moves towards allocation of resources to minority languages like the inclusion of five Non-Tamil-MotherTongues (NTMLs) into the school system in the 1990s through a bottomup demand from the Indian community (Tan 1998). Tan shows how it was a demand from the non-Tamil speaking communities of Singapore that led to the inclusion of Hindi, Gujerati, Urdu, Bengali and Punjabi in the Primary School Leaving Exam (PSLE). Even though there remains limited state funding for these NTMLs and they are largely community-funded, the heterogeneous Indian community values the move to include languages other than Tamil into the national school system. In these interventionist and egalitarian policies, we thus see efforts that lead to language loss, shift and maintenance. Sociolinguists in Singapore are concerned about mother tongue language loss (in the conventional sociolinguistic sense of the word) (Gupta and Siew 1995; Li et al. 1997; Saravanan 1999, 2001) and the implications this has for Asian culture and identity. For instance, Gupta and Siew (1995) point to the language gap between dialect-speaking grandparents and Mandarin- and English-speaking grandchildren in the Chinese community, with the result that cultural and family traditions are being lost. This illustration of language shift has the potential for weakening social ties amongst family members along with the concomitant loss of cultural practices linked with language. Government leaders have repeatedly defended their position that this language gap is a necessary sacrifice for the good of children’s education and for the successful implementation of the language policy, and have presented Mandarin as a more viable

166 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang

and desirable basis for the creation of social networks and social cohesion among the Chinese.

Methodology

The Sociolinguistic Survey of Singapore 2006 Part of the data for this chapter come from a project called the Sociolinguistic Survey of Singapore 2006 (SSS 2006), undertaken by the Centre for Research in Pedagogy and Practice (CRPP) in Singapore. The research question for SSS 2006 is: Who speaks what language to whom in what context with what attitude with what level of fluency and to what end? The survey has a sample size of 716 students from the Grade 5 cohort of Singaporean schools, who were randomly selected on the basis of race (Chinese, Malay and Indian) and socio-economic status (SES). The children are between 10 and 11 years of age. The sample of students includes only Singapore citizens. Due to small cell sizes, it does not include non-Tamilspeaking Indians and Singapore Permanent Residents. Though the Census of Singapore does not have a separate category called ‘socio-economic status’ (SES), it does have questions on income, housing type, occupation and highest qualification attained, all of which are items in SSS 2006. We take household income rather than housing type to be the main surrogate to classify SES since, in Singapore, families with extremely high household incomes live mostly in private housing and household earnings have a close tie with housing type in public housing. The SES factor structure shows that occupation is less related to SES and hence not considered for the classification. More importantly, it would be operationally convenient to rely on one single most important factor for the selection of students in the qualitative follow-up study. The following is the way SSS 2006 defines the four levels of income in Singapore dollars (S$): SES 1: more than S$5,001/month SES 2: S$3,501–5,000/month SES 3: S$1,501–3,500/month SES 4: less than S$1,500/month The survey examines language use in five domains: school, family and friends, religion, public space and media. It also includes sections on language attitudes, ideology and proficiency. The language attitudes section is based on a variation of the matched guise method, involving a set of three recordings or guises. In this part we play a set of three recordings for each child. In each recording a 10-year-old speaker is talking on the phone to a

Language and social capital in Singapore 167 friend asking him/her to come over to do homework together. Recording 1 is in English, 2 is in a mother tongue (Tamil, Malay or Mandarin), and 3 is in English and a mother tongue with code-switching. Participants listen to the set of recordings that corresponds to their mother tongue (thus each participant listens to a total of three recordings). The script is the same in all three recordings. After listening to each of the three recordings, respondents are asked a series of questions designed to reveal their attitudes towards the use of English, their mother tongue, and code-switching. The items on proficiency are all based on self-report and are divided into sections on literacy (reading and writing on a four-part scale) and fluency (speaking and understanding on a five-part scale) for both English and Mother Tongue. This section ends with questions on what marks the children received in the last language/s test they took in school. SSS 2006 has a funnel-shaped design, with the large-scale quantitative survey described above leading to small-scale follow-up studies that are planned as summarized in Table 7.2. Table 7.2 Design of follow-up studies in SSS 2006 Chinese

Malay

Indian

Total

SES 1

1

1

1

3

SES 2

1

1

1

3

SES 3

1

1

1

3

SES 4

1

1

1

3

Total

4

4

4

12

As shown in Table 7.2, there are twelve follow-up studies, with participants chosen on the basis of their race and SES. These twelve children have been observed for about two weeks in the five domains of the survey. Data collection includes audio-recording and video-taping of family and other domain interactions, observation and participant-observation, and unstructured interviews/conversations. Artefacts such as emails, photographs, journals and activity logs with a focus on language and literacy are also collected. The project emphasizes the child as participant in and leader of data collection. Thus one tape recorder is left with the children and they are asked to record what they think is typical language use in their lifeworlds.

The CRPP Panel 6 Survey The SSS 2006 data are subsequently merged with the Centre for Research in Pedagogy and Practice’s Panel 6 survey (henceforth CRPP Panel 6 2006).

168 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang CRPP Panel 6 2006 is a multi-cohort, cross-classified, longitudinal survey with an aim to study the class, school and student effects on various social outcomes. It is an ecological study of some 30,000 students over a five-year period, a longitudinal survey designed to capture and measure changes and growth of student’s demographics, life experience, patterns of social participation, social attainment and performance, social attributes, life goals and choices, pathways and beliefs. Three different cohorts were identified for the survey: the primary, secondary and the post-secondary cohort. The same Grade 5 cohort, the first wave of the CRPP Panel 6 carried out in the year 2003, is thus the source of participants for both SSS 2006 and Panel 6, and the data are merged. CRPP Panel 6 is currently at the third wave of the study.

Measures of social capital in Panel 6 Seven variables are identified as items of social capital characteristics. They are categorized under two main subheadings, namely family social capital and personal social capital. The former, bonding in nature, measures the family functioning and attachment, whereas the latter is of bridging characteristics including friendship skills, number of close friends, social and leadership skills, trust in persons and social participation in religious practices. Family climate measures the level of relationship between student and family members. Family attachment refers to the emotional attachment of respondents to parents, grandparents and siblings. Friendship skills and social leadership skills, shown under B 3 and 5, measure students’ socializing skills with friends, organizing skills and working relationships with friends. Number of close friends is the extent of a student’s social network. Trust in persons measures the trust level between students and family members/friends. Participation in religious practices is the frequency of visits to religious institutions. A. Family Social Capital 1. Family Climate 2. Attachment B. Personal Social Capital 3. Friendship Skills 4. Number of Close Friends 5. Social Leadership Skills [Application of Social Skill] 6. Trust in Persons 7. Participation in Religious Practices [Social Participation]

Language and social capital in Singapore 169

Measures of social capital in SSS 2006 On the following topics, there are items in the SSS 2006 instrument that can be used to measure social capital: 1. Household structure (single parent/multi-generation/nuclear family/ extended family/other) 2. Household size 3. Interaction with grandparents: • Does the child stay with his/her maternal/paternal grandparents? • How often does the child visit his/her maternal/paternal grandparents?

Results and discussion Table 7.3 shows the dominant language across the three ethnic groups for each of our five domains in SSS 2006 with a focus on English and Mother Tongue (MT).

Table 7.3 Dominant language across domains and ethnic groups in SSS 2006 Table 7.3a Family/friends Race

Chinese (%) (n=400)

Malay (%) (n=206)

Indian (%) (n=92)

English Mother Tongue (MT)

41.7 31.8

24.4 46.0

41.3 32.8

36.2 36.3

Language

Total (%) (n=716)

English-MT

15.9

18.3

13.9

16.4

Other N.A.

0.4 10.4

0.1 11.2

2.2 9.8

0.5 10.6

Total

100.0

100.0

100.0

100.0

Table 7.3b School Race

Chinese (%) (n=400)

Malay (%) (n=206)

Indian (%) (n=92)

Total (%) (n=716)

English Mother Tongue (MT)

68.5 19.9

60.4 30.1

72.4 19.4

66.5 23.0

English-MT N.A.

11.3 0.2

9.4 0.1

7.1 1.1

10.2 0.3

100.0

100.0

100.0

100.0

Language

Total

170 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang Table 7.3c Media Race Language English

Chinese (%) (n=400)

Malay (%) (n=206)

Indian (%) (n=92)

Total (%) (n=716)

69.2

57.7

66.5

65.2

2.7

4.6

1.1

3.1

English-MT

7.2

12.6

4.7

8.6

Other

0.7

0.5

0.3

0.6

Mother Tongue (MT)

N.A.

20.2

24.6

27.4

22.5

Total

100.0

100.0

100.0

100.0

Indian (%) (n=92) 69.4 20.8

Total (%) (n=716) 63.7 23.3

Table 7.3d Public space Race Language English Mother Tongue (MT) English-MT N.A. Total

Chinese (%) (n=400) 65.3 20.2

Malay (%) (n=206) 58.2 30.1

14.5 0.0

11.7 0.0

9.5 0.3

12.9 0.1

100.0

100.0

100.0

100.0

Indian (%) (n=92) 26.1 58.1

Total (%) (n=716) 26.1 52.2

Table 7.3e Religion Race Language English Mother Tongue (MT) English-MT

Chinese (%) (n=400) 38.7 35.5

Malay (%) (n=206) 3.7 79.5

4.8

2.6

5.6

4.2

Other N.A.

0.7 20.1

12.2 2.0

9.1 1.1

5.5 12.0

Total

100.0

100.0

100.0

100.0

In the domains of school, media and public space, English is clearly the dominant language for Singaporean children across all ethnic groups. In these three domains, 66.5%, 65.2% and 63.7% of the children respectively, reported English as their dominant language. However, the domains of family and friends and religion tell a different story. In the domains of family and friends, Malay is the dominant language for the Malay ethnic group. As many as 46% of the Malay children reported that Malay is their dominant language in this domain, whereas only about a quarter, or 24.4%, reported English. In the domain of religion, the Mother Tongue (MT) is dominant for both the Malay and Indian ethnic groups, though English is dominant for the Chinese ethnic group. Comparing across all three

Language and social capital in Singapore 171 ethnic groups, then, we can see that English has the least presence in the religious domain: amongst the Malays only 3.7% of the children reported that English is dominant for them in the religious domain whereas 79.5% reported that Malay is dominant. The Malay community has the highest levels of mother tongue language use, and in their case, maintenance, across all five domains. Table 7.3 implies use of code-switching between English and mother tongue in all cells, which is not the purpose of this chapter to explore. For information on the implications of code-switching, see Vaish (2007a and b) and Bokhorst-Heng (forthcoming). Does MT dominance for Malays in the domain of family and friends point to stronger social ties for Malay families leading to Malay language maintenance in this domain? The story here is confounded by Panel 6 data. In this database, ‘Family Climate’ is a construct that includes the following: • I feel loved by my family. • My parents listen carefully to what I have to say in important family discussions. • We show care and concern for our family. • In our family everyone listens to and respects each other. As such family climate is highest for the Indians: 21.3. It is roughly the same for the Malays and Chinese: 19.5 and 19.8 respectively. Thus family climate is not the highest for the Malays. Also, statistically, the number 21.3 for the Indians is not significantly higher than that for the Malays and Chinese. This key finding suggests that we cannot make a simple link between stronger family ties amongst the Malays with stronger mother tongue maintenance in this community. We will come back to this point in detail when we discuss language maintenance in the Indian community.

Measures of social capital in SSS 2006 On many measures of social capital in the SSS 2006, all the three ethnic groups show similar characteristics. For instance, in terms of household structure, 76.3% of Chinese children, 73.9% of Malay children and 80.6% of Indian children in our sample live in nuclear families, thus showing that the predominant household structure across all three ethnic groups is similar. On the basis of these numbers, it is not possible to hypothesize, for instance, that multi-generation families maintain language more than nuclear families do. Similarly, the majority of the children do not live either with paternal or maternal grandparents. Also, there are similarities across ethnic groups regarding how many times the children visit the grandparents. About 6% of the children in our sample visit their maternal grandparents every day, with most of them evenly spread out between the

172 Viniti Vaish, Teck Kiang Tan, Wendy D. Bokhorst-Heng, David Hogan and Trivina Kang ‘once a week’ or ‘once a month’ category. Again, the point to note here is that there are no glaring differences between the practices of the children across the three ethnic groups which could point to increased social capital leading to language maintenance. In terms of household size, the Malay families tend to be somewhat larger than the Chinese or Indian families. On average, 40% of the Chinese and 39% of Indian families are four-person families. However, 37% of Malay families are five-member families, and as many as 23% have six members. Though it may be possible to hypothesize that a larger number of siblings leads to increased language maintenance, again the data do not appear strong enough to warrant this claim.

Measures of social capital in Panel 6 As shown in Table 7.4, except for family climate and participation in religious practices, there are no statistical differences between the three ethnic groups in the other measures. Table 7.4 Measurement of social capital by race (means) Social Capital Measurement Family Social Capital Family Climate* Attachment

Chinese

Malay

Indian

19.8

19.5

21.3

20.6

21.1

22.0

Personal Social Capital Friendship Skills

17.3

17.9

18.1

Number of Close Friends

5.5

6.0

4.9

Social Leadership Skills

5.9

6.0

6.5

Friendship Attachment

4.6

4.2

4.4

18.3 3.9

21.5 6.0

18.8 5.1

Trust in Persons Participation in Religious Practices* * Significantly different p