Applied Linguistics Review

Applied Linguistics Review 1 2010

Applied Linguistics Review 1 2010 Editor

Li Wei

De Gruyter Mouton

ISBN 978-3-11-022264-7 e-ISBN 978-3-11-022265-4 ISSN 1868-6303 ISSN online 1868-6311 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de. ” 2010 Walter de Gruyter GmbH & Co. KG, Berlin/New York Printing: Hubert & Co. GmbH & Co. KG, Göttingen ⬁ Printed on acid-free paper Printed in Germany www.degruyter.com

Editor

Associate Editors

Li Wei Department of Applied Linguistics & Communication University of London, Birkbeck College 43 Gordon Square, Bloomsbury London WC1H 0PD UK E-mail: [email protected]

David Block University of London, Institute of Education Bencie Woll University College London Itesh Sachdev University of London, SOAS

International Advisory Board Keiko Abe Kyoritsu Women’s University, Japan

Alex Housen Vrije Universiteit Brussel, Belgium

Kingsley Bolton City University of Hong Kong, China

Andy Kirkpatrick Hong Kong Institute of Education, China

Vivian Cook Newcastle University, UK

Claire Kramsch University of California, Berkeley, USA

Donna Christian Center for Applied Linguistics, USA

Mayouf Ali Mayouf University of Sebha, Libya

Annick De Houwer University of Erfurt, Germany

Tim McNamara University of Melbourne, Australia

Patricia Duff University of British Columbia, Canada

Ben Rampton King’s College London, UK

Diana Eades University of New England, Australia

Elana Shohamy Tel Aviv University, Israel

Yihong Gao Beijing University, China

David Singleton Trinity College Dublin, Ireland

Ofelia Garcia City University of New York Graduate Center, USA

Anna Verschik Tallinn Pedagogical University, Estonia

Susan Gass University of Michigan, USA Fred Genesee McGill University, Canada Nancy Horberger University of Pennsylvania, USA

Terry Wiley Arizona State University, USA Lawrence Zhang National Institute of Education, Singapore

Contents

Li Wei Editorial

ix

Ben Rampton Social class and sociolinguistics

1

Tim McNamara Reading Derrida: Language, identity and violence

23

David Block Engaging with human sociality: Thoughts on communication and embodiment

45

Helen Kelly-Holmes Language trends: Reflexivity in commercial language policies and practices

67

Anna Verschik Contacts of Russian in the post-Soviet space

85

Francesco Cavallaro and Stefan Karl Serwe Language use and language shift among the Malays in Singapore

129

Mila Schwartz Family language policy: Core issues of an emerging f ield

171

Aurora Donzelli Is ergativity always a marker of agency? Toraja and Samoan grammar of action and the contribution of emancipatory pragmatics to social theory

193

viii

Contents

Amelia Church Opportunities for learning during storybook reading at preschool

221

Lawrence Jun Zhang Negotiating language, literacy and identity: A sociocultural perspective on children’s learning strategies in a multilingual ESL classroom in Singapore

247

Editorial LI WEI

The original idea for this new publication came out of the weekly seminar series which is jointly run between Birkbeck College and the Institute of Education, both of the University of London, and the annual Round Table whose organization also involves colleagues from other colleges of the University in the Bloomsbury area, especially the University College and the School of Oriental and African Studies. Given the quality and variety of the presentations at these events, we thought that it would be very nice to find an outlet for them to be published and read by a wider range of colleagues. Mouton de Gruyter enthusiastically supported the idea and we agreed to produce an annual publication under the title The Bloomsbury Review of Applied Linguistics and Communication. However, just as the papers for the inaugural issue were ready to be submitted to the publisher, the publishing giant, Bloomsbury Publishing Plc, sent us a letter through its lawyers insisting that the word “Bloomsbury” had been trademarked and therefore could not be used in any publication that they did not own. Whilst we were flattered by the fact that our new and rather modest venture was noticed by such a major player in the publishing world (and the thought that a handful of publishing houses are trying to dominate the world never crossed our mind. Honest!), we had to find an alternative title and amend all the documents and publicity material that had already been registered. It has turned out that Bloomsbury Publishing Plc’s intervention was a blessing in disguise. The papers we had received were of a more global appeal than the name Bloomsbury Review would indicate. And as the experts in Mouton’s marketing department suggest, the simpler the title the better – hence the current title Applied Linguistics Review. Applied Linguistics Review is a peer reviewed annual publication. It aims to serve as a testing ground for the articulation of original ideas and approaches in the study of real-world issues in which language plays a crucial role, by bringing together new empirical and theoretical research and critical reflections of current debates. As you can see in this inaugural issue, the issues dealt with in the

x

Li Wei

Review are wide ranging, covering aspects of the linguistic and communicative competence of the individual such as bilingualism and multilingualism, first or second language acquisition, literacy, language disorders, as well as language and communication related issues in and between societies such as linguistic discrimination, language conflict, communication in the workplace, language policy and language planning, and language ideology. One key difference between this publication and the other existing journals and annuals is that we encourage personal reflections on emerging issues and themes in applied linguistics as a broad, multidisciplinary field. We urge our contributors not to shy away from expressing their own opinions. And we make sure that the peer review process is thorough and constructive, and in the meantime allows the personality of the author to shine through the writing. We invite potential contributors to contact the Editors directly to discuss their ideas before submitting the papers. We are particularly interested in papers that not only report new research findings but also critically engage in current debates over theoretical and methodological issues in applied linguistics and communication generally and point to directions of future research. Finally, readers of this volume will want to know what we mean by applied linguistics in the title. Here is our answer: applied linguistics is any attempt to work with language in a critical and reflective way, with the ultimate goal of understanding the role of language in the construction and expression of human sociality. We will try to elaborate on this through the contributions in future issues. Birkbeck College, University of London December 2009

Social class and sociolinguistics1 BEN RAMPTON

Abstract Thisarticle makes the case for resuscitating social class as an issue in British applied and sociolinguistics. It begins with a sketch of the treatment of class in post-war social science in the UK, drawing out the implications for sociolinguistics. It then moves to a fuller review of how sociolinguistics has actually handled class, and considers Bernstein’s work and its relationship to classic US research in the ethnography of communication, as well as the reasons from the ‘retreat from social class’ in discourse-oriented UK sociolinguistics from mid-1980s onwards. After that it offers a class-oriented reinterpretation of my earlier work on ethnolinguistic crossing and stylization, and it concludes with some suggestions for further research, stressing the need to develop interactional and ethnographic perspectives on class processes. Keywords: Social class; sociolinguistics; interaction; crossing; ethnicity.

This article attempts to make the case for resuscitating social class as an issue in applied and sociolinguistics in Britain. Certainly, there is a great deal of contemporary work on discourse, culture, power and social inequality, but this generally focuses on gender, ethnicity and generation much more than class. And yes, in ordinary everyday activity, ‘transportable’ identities like class, ethnicity, gender, sexuality and age are all blurred and interwoven with a considerable range of institutional and interactional identities (Zimmerman 1998), so that it is very much an analytic act separating class out from everything else. Even so, many many people – and not just academics – engage in this kind of analytic differentiation, and separating out ‘class’ has distinct implications. So for example, when migrant ethnicities are discussed in education, students’ practices-anddispositions are generally treated independently of the inherently stratifying

2

Ben Rampton

processes of schooling, and there is talk of eradicating inequality by closing the gap between school and ethnic culture, either making schools more hospitable, or tuning home cultures more to education. Class analyses are potentially more pessimistic. Social and cultural identities are defined inside mainstream processes of stratification, not outside, and instead of being overcome by cultural bridge-building, inequality and discrimination are treated as central to schooling itself. So class isn’t an inconsequential concept, and there are four parts in my argument that contemporary linguists need to give it more attention. I shall start with a sketch of how class has been treated in post-war British social science. This will be brief and very second hand, but there are still some significant implications for sociolinguistics. In the second part, I shall comment on how sociolinguistics itself has dealt with class. Here, Bernstein looms large, though my main point is that there is still a great deal of scope for ethnographic and interactional sociolinguistic analyses of class processes. The third section tries to illustrate this by turning to my own interactional sociolinguistic research and looking at ethno-linguistic crossing and stylization through the lens of social class. And then to end, I offer some suggestions for further work.

1. Class in post-war British social science and some implications for sociolinguistics According to Mike Savage (2003, 2005, 2007), there have been three distinguishable ‘waves’ in the study of class in post-war Britain. The first dates from the mid-1950s to the mid-70s, and Savage associates it with figures like Halsey, Lockwood, Young, Hoggart, Williams, Willis, Thompson and Hobsbawm. The relationships between class consciousness, social stratification, community and family were of central interest in the first wave, and there was a lot of qualitative research, with, for example, in-depth interviews playing a major part in studies of community. In the second wave from the mid-1970s to the early 1990s, Goldthorpe and Marshall were pre-eminent in British research on class, and this focused much more on the structural aspects of inequality. Statistical methods were used to link class with life-chances in health, housing, crime, education, etc., and class was simplified and operationalized as occupational employment.2 Interest in class consciousness and solidarity declined (Savage 2003: 536), while over the same period, there was also a growing body of work asserting the ‘demise of class’ in late-modernity, a line of argument that Savage associates with Bauman, Lash and Giddens. In the third wave of work on British social class, starting in the early 1990s, researchers such as Skeggs,

Social class and sociolinguistics

3

Walkerdine, Lucy, Reay and Ball have contested these claims about the demise of class, and there has been a reaction against the reductive equation of class with occupation in survey work, turning from this to qualitative analysis of the cultural and psycho-social processes associated with class. The interest of the earlier generation in class identities and class identification has been revived with work on, for example, stigmatized working class identities and middleclass strategies in education markets, and the focus has also expanded beyond community, work, industrial relations to schooling, parenting, leisure and consumption. From the vantage point of sociolinguistics, there are at least three things that stand out in this work. First, even though it was pre-eminent in Savage’s second wave, there is no need to confine our notion of class to a single indicator like occupation – after all, in a country with a long history of stratification like Britain’s, ‘class’ can embrace a huge range of cultural and material processes, covering social differences in “family background, main source of income, place of residence, cultural tastes, . . . political affiliations etc” (Abercrombie and Warde 2000: 145–146). Next and much more suggestively, there are openings for sociolinguistics in the first and third wave interest in class identities and the ‘meanings’ of social class. In the first wave, both Raymond Williams and E. P. Thompson were concerned with how class consciousness and experience were shaped in social activity. Thompson insisted that class needs to be studied “in the medium of time – that is, in action and reaction, change and conflict” (1978: 295–296), focusing on the articulation of a sense of solidarity or opposition in the struggle for resources in particular locations. ‘Class’ here means a sensed social difference that people and groups produce in interaction, and there is struggle and negotiation around exactly who’s up, who’s down, who’s in, who’s out, and where the lines are drawn. Material and cultural inequalities matter a great deal, but human agency still plays a vital part in class processes. So when analysts see people in better and worse, higher and lower positions, they need to look for the cultural practices that accomplish this, which is something that sociolinguists are potentially quite adept at. Williams also attended closely to the subjective side of social class, and was interested in how the experience of living in a stratified society works its way into “the fibres of the self”, producing social instincts and tacit sensibilities that are shaped by “the lived dominance and subordination of particular classes” (1977: 110). This has been picked up in the third wave by feminist researchers such as Skeggs (1997) and Reay (1998) – Skeggs, for example, emphasises the role that “everyday negotiations of the mundane” play in the formation of classed subjectivities (1997: 167),3 and she explores the

4

Ben Rampton

implications in some long-term ethnography with working class women. So the second point to draw from Savage’s overview is that if sociolinguists want to investigate class, we don’t have to bind ourselves to large-scale comparisons of high- and low-placed social groups. In class societies, people carry class hierarchy around inside them, acting it out in the fine grain of ordinary life, and if we look closely enough, we may be able to pick it out in the conduct of just a few individuals. Third, alongside the perspectival shifts that Savage describes, there have obviously also been fundamental changes in the economy, culture and society since the 1950s – de-industrialisation, the decline of traditional collectivism and the emergence of gender, race and ethnicity as political issues, globalization, marketization and the ascendance of the individual as consumer (Abercrombie and Warde 2000: 148). “The working class”, Savage suggests, is now “no longer a central reference point in British culture”, and the middle class has “become the class around which an increasing range of practices are regarded as universally ‘normal’, ‘good’and ‘appropriate”’(2003: 536). But this doesn’t mean that class no longer matters – instead, “there has been a fundamental re-working of class relationships [and this] affects the mode by which class is articulated, imagined and thought” (2007). How far, we should ask, has sociolinguistics tuned into this kind of reworking? Indeed, how far has sociolinguistics contributed to its description? These are substantial questions, inviting consideration of some of the history of sociolinguistics itself, and I would like to try to respond, starting very briefly with Labovian variationism, turning to Bernstein and moving from there to ethnographic and interactional sociolinguistics.

2. Social class and sociolinguistics? 2.1.

Labovian variationism

Variationists have probably had a longer and more enduring interest in social class than any other branch of sociolinguistics, and they certainly got off to a very good start. According to Savage (2003: 536), Bourdieu has been the most important theoretical influence on the most recent, post-1990s third wave of class-focused British social science, but in fact Bourdieu was using Labov to develop his notion of classed habitus4 back in the mid-1970s. Variationists have repeatedly shown that in class-stratified societies, society-wide speech variation is ‘echoed’ in the style variation of individuals – the patterns of accent difference that you can see when you compare class-groups-distributed-across-society-as-

Social class and sociolinguistics

5

a-whole are mirrored (more weakly) within the speech repertoire of individuals – and Bourdieu took this as impressive testimony to class reproduction, largescale stratification being inscribed even into the apparently flexible conduct of individuals (Bourdieu 1991 [1977]: Part 1; Woolard 1985; Eckert 2000: 13). But has variationist sociolinguistics in general lived up to its early promise in the analysis of class? In recent years, the link between Bourdieu and language variation has been renewed in Eckert’s sociolinguistic ethnography of adolescents in the US. Taking up Bourdieurian practice theory, Eckert insists on seeing speakers as “agents in the continual construction and reproduction” of the language system, not just as its “incidental users” (2000: 43), and she also looks hard for the local meanings of social class, finding it in the polarization of school-enthusiastic ‘Jocks’ and disaffected ‘Burnouts’ and in the contrasting connections that adolescents make between on the one hand, local social activities and categories, and mainstream values on the other. Even so, Eckert’s linguistic analyses are primarily statistical, and there are no detailed descriptions of classed language being used or developed in situated interaction (compare, e.g. Coupland 2007). More generally, as late as 1996 Hymes took the view that in variationist research, “class has entered sociolinguistic analysis as an indispensable parameter of change and statistical difference, but as a lived reality [it] has hardly begun to appear as a focus of inquiry in its own right” (1996: 73). The central and abiding interest is in variability and change in language itself; class tends to be defined occupationally, as in Savage’s second wave; and classically anyway, variationists emphasize attitudes and prejudice in their explanations of linguistic inequality. In Hymes’ encapsulation of the Labovian view of class, “[d]ifferential access to resources there might be, but so far as ability was concerned, class had no cost” (1996: 188). Which brings us, of course, to Bernstein. 2.2.

Bernstein

Bernstein’s analysis of language and class starts with macro-social structure and the division of labor, moves into the institutional organization of family and education, homes in on interactional practices deemed critical in socialization, and from there looks for links to the communicative disposition of individuals and their impact on school achievement. In doing so, Bernstein spans most of the levels where researchers have located class processes – the economy, the community, occupations, families, activity, discourse, language, consciousness and school career. In fact, Savage identifies Bernstein alongside Bourdieu as a major influence on the current wave in the British sociology of class (2003: 537), and Bernstein’s

6

Ben Rampton

reputation has also substantially recovered among linguists over the last 10 to 15 years. With common ground in Halliday, critical linguistics and critical discourse analysis have always had a lot of time for Bernstein; Bernstein’s critique of ‘masking’ and the concealment of power relations in progressive pedagogy feeds into Fairclough’s account of the conversationalization of public discourse (e.g. 1995); and few can match Bernstein’s magisterial overviews of changing trends both in social science and in higher education generally (e.g. 1996: Chs. 3 and 7). Yet in the UK at least,5 subsequent to the critique by linguists like Labov and Trudgill (cf. Atkinson 1985: Ch. 6), there has been very little active interest in Bernstein’s analysis of class culture and language, in class “as a general position in society, which is also a tradition, with typical values [that have a bearing] on language use” (Collins 1988: 307). Bernstein’s work on class, family socialization and its educational effects generally gets bracketed off as ‘early Bernstein’ and it is worth considering the reasons for this. 2.3.

Bracketing off ‘early Bernstein’

There are many things that could be said to explain the sequestration of early Bernstein. First, early Bernstein is very much the product of its time, the 1950s and 1960s. Gender is taken for granted (Collins 2000: 73); there is nothing on ethnicity; and the main lines of the theory are formulated prior to the development of sophisticated discourse analysis, in the absence of detailed naturalistic observation of either homes or classrooms (compare Wells 1981; Tizard et al. 1988; Edwards and Westgate 1987: 35). Second, there is a neglect of all the ideology, politics, resistance and reflexive self-differentiation that shapes and develops in the interaction across classes (Rosen 1972; compare, e.g. McDermott and Gospodinoff 1981, Collins 1988). Third, within the highly ambitious, multi-leveled panorama that is covered in Bernstein’s argument, there is a great deal of scope for different specialists in this or that to zero in on particular details. And of course, fourth, from the late 1970s onwards, Bernstein himself shifted away from language and group culture to the social organization of educational discourse. But in trying to explain the bracketing off of Bernstein’s early work, I would like to dwell on his approach to knowledge production, and indeed his epistemology, it seems to me, also limits his contribution to our understanding of how class has changed in Britain.

Social class and sociolinguistics

2.4.

7

Bernstein’s non-ethnographic epistemology

In his paper on “Research and languages of description” (Bernstein 1996: 134– 146), Bernstein distinguishes two kinds of analysis, differing in their degrees of abstraction. On the one hand, there is what he calls an ‘external language of description’ – the frameworks and procedures for classifying, comparing, contrasting, sorting and organizing the recordings and observations from fieldwork. And on the other, there is the ‘internal language of description’, which involves more abstract models attempting to capture the fundamental principles that structure the patterns identified in field data. There is a strong resemblance in this account to research in syntax, and Bernstein is very explicit about the influence that linguistics had on the shape his theorizing took (1996: 127; also, e.g. 1971: 173). He refers to his own theory as a “generating grammar” (e.g. 1996: 127), and is open about being more interested in the abstract generative principles than in the more concrete specifics of class (1996: 126), which is also obvious in the proliferating grids of multi-directional arrows in his work, as well as in skimpy caricatures of class types like the ‘Millers’ (1971: 177). In consequence, Bernstein’s is very much a deductive, theory-driven account of class processes, and a great deal of effort goes into configuring data in ways that make it directly relevant to his theoretical models. Yes, the empirical descriptions can still exceed what the model anticipates (1996: 129, 138), but Bernstein doesn’t have the discursive disposition of the novelist, and his methods of work are emphatically non-ethnographic.6 ,7 This has several consequences for the account of class. First, one of ethnography’s key characteristics is its commitment to taking a long hard look at empirical processes that are hard to understand within established frameworks, and so when people start asking “What is the working class today? What gender is it? What colour is it” (Gilroy 1987: 19), ethnography is a potentially rather important resource. In the 1950s and 1960s when everyone thought they more or less knew what class meant, Bernstein’s thumb-nail character portraits and picture task elicitations might just about cover the lower-level empirical specifics of language and class, leaving him free to model the underlying dynamics, but this is far harder when the ground changes radically. When class identities become problematic, it is necessary to give much more attention to bottom-up description of the kind promoted in ethnography, and if that is what it would have taken – if what was needed was ethnography – then in the light of his stated methodological preferences, it is not surprising that Bernstein moved away from the work on language and group culture. Of course, Bernstein was just as aware as Savage of changes in how “class is articulated, imagined and thought”, and he himself discusses the formation of identities in reorganising

8

Ben Rampton

capitalism (1996: 75–80). But these insights are highly schematic, delivered in what Gemma Moss has described as Bernstein’s ‘marmite’ prose; there is nothing on identity and language; and it would be very hard overall to count this later work as a specifically socio-linguistic contribution to understanding the reworkings of class. Second – and focusing here on the bracketing off of early Bernstein – different positions on ethnography help to explain the ‘Bernstein / Brice Heath paradox’ – the extraordinary contrast between, on the one hand, the general condemnation of Bernstein’s work on language and class, as opposed to the almost universal celebration of Shirley Brice Heath’s on the other.8 2.5.

Bernstein and Heath

There is actually a great deal of overlap in the ground covered by Bernstein and Heath. They both start by contrasting communities and then follow the lines of influence from different kinds of family relationship to different kinds of interaction, producing different ways of taking meaning, resulting in different educational careers. Indeed, in their neglect of inter-group politics,9 they share a common weakness. And yet while the refutation of Bernstein helped to make Labov’s “Logic of Non-Standard English” (1969) one of the most anthologised sociolinguistic papers in the 1970s, Heath’s take on the issues that concern Bernstein – her ethnographic remix in “What no bedtime stories mean” (1982) – became one of the most famous papers of the 1980s. From what I have already said, it should be clear that the issue of being ethnographic or not is far from being just a trivial methodological matter. Rather than having a few social types quickly flashed at us, Heath’s work describes people in all their individual and contextual particularity, and whereas Bernstein presents ‘critical socialising contexts’ programmatically, as a necessary link in the chain between role-systems and language development, Heath’s equivalent idea, ‘literacy event’, works as a sensitising construct, opening up a huge body of descriptive work documenting hitherto unimagined complexities in reading and writing (the New Literacy Studies). Heath’s empirical three-way community comparison escapes the dichotomisation that grips Bernstein’s thinking, and the balance between theory and empirical description is almost wholly reversed, with theory emerging inductively from data and scholarly abstractions like ‘cognitive style’ given only very subsidiary walk-on parts. And whereas Bernstein’s work seeks to disclose the rules and mechanisms that generate the patterns of conduct and achievement that schools are all too familiar with, Heath’s work bathes familiar classroom behaviours in new light, opening new dimensions and possibilities.

Social class and sociolinguistics

9

Maybe if the ethnography of communication had been fully invented when Bernstein did his work on language and class (cf. Bernstein 1975: 154, 156), or if he had been less attracted to models of knowledge production propounded in formal linguistics, we would have had something more like Ways with Words in England in the 1970s and 1980s. Certainly, whereas Bernstein simply left it as a rather ominous black box, contemporary combinations of ethnography, conversation analysis, pragmatics and the semiotics of indexicality can now delve deep into ‘restricted code’, going far beyond the analysis of explicit propositions into robust and detailed description of eloquent silence, strategic indirectness, and allusive nuance (see, e.g. Clark 2003). But in spite of this, it is important to recognise that Heath and Bernstein were working with a very similar ontological map of the different macro-, meso- and micro-levels of social process that sociolinguistics should try to bring together. Yes, in his drive to fit all the jigsaw pieces together, Bernstein rushed past the image printed on each piece, but Heath and Bernstein were still working on very much the same puzzle, and in fact one can also see roughly the same pieces in John Gumperz’s work on Crosstalk (Gumperz et al. 1979). 2.6.

Bernstein, Heath and Gumperz united in a wider interdisciplinary field

Like Bernstein, Gumperz takes large scale social structural trends as the point of departure, pinpoints critical institutional encounters, and tries to show how unrecognised differences in the communicative dispositions of subordinate groups produce disadvantage in institutional environments, in spite of the participants’ good intentions. Within this, of course, there are substantial differences. Gumperz addresses racial discrimination in employment rather than class underachievement in education, and rather than mother-and-child interaction, the focus is on ‘gate-keeping’ interviews. In addition, the processes that Gumperz covers are much more limited, so his claims seem less speculative. Compared with the task of linking dialogue-at-home to achievement-at-school, it is relatively easy tying interactional practice to institutional outcome if all you’re trying to do is see how discourse in an interview affects the decision ten minutes after. Discussion of different communicative dispositions is also potentially more straightforward if these have developed in another language in another country. But despite such differences and the larger contrast between Bernstein’s epistemology and Gumperz’s affiliation with ethnography, there is a broad family resemblance in the sites, phenomena and processes that they are seeking to connect. So overall, although the differences between Bernstein, Heath and Gumperz loom large for specialists in applied and sociolinguistics, when they are placed

10

Ben Rampton

Table 1. Similarities between Bernstein, Heath and Gumperz Bernstein

Heath

Gumperz

‘critical socialising contexts’

≈

‘literacy events’

≈

‘gate-keeping’ interviews

‘coding orientation’

≈

‘ways with words’

≈

‘communicative style’

‘elaborated’ vs ‘restricted code’

≈

‘Mainstream’ vs ‘Roadville’ and ‘Tracktown’

≈

‘they-code’ vs ‘we-code’

in a wider perspective, there are a number of striking similarities in their central concerns and orientation points. Some of the broad parallels can be mapped as in Table 1. More generally, these formative figures in sociolinguistics focus on: – language and literacy but seek to situate these in: – – – –

historical process institutions and networks situated activity and practices individuals and their repertoires

and they share a driving interest in how these basic coordinates are linked together in different settings. The consequence is that in sociolinguistics and linguistic anthropology 20 to 30 years later, there are now rather a lot of rather well-developed, well-tried frameworks and procedures for exploring all this (even though the coverage of some of these coordinates may often be fuller than others). For quite a lot of the British social science research on class described by Savage in Section 1 above, this is news. As already mentioned, Skeggs emphasises the role that “everyday negotiations of the mundane” play in the formation of classed subjectivities, but very few sociologists working on class have the tools or procedures for analyzing this, and it is remarkable how much scholars still rely on the research interview.10 Inside applied and sociolinguistics, this kind of ontological map of society, culture and discourse may fade into the background in arguments over detail, but I am currently involved in a three-year training course designed for non-sociolinguistic social science researchers UK-wide,11 and it is remarkable how illuminating the participants seem to find it when they

Social class and sociolinguistics

11

are introduced to this combination of a basic orientational map and tools for connecting it to empirical social processes. So my claim is that there is ‘powerful medicine’ in sociolinguistics in the tradition of Bernstein, Heath and Gumperz. But if that is actually true, why, post-Bernstein, has there been so little British work on culture, identity and discourse as this relates to social class? 2.7. The retreat from class in British sociolinguistics Although class, gender and ethnicity flow together in practical experience, it is helpful to use Bradley (1996) to define class analytically as relations and identities lived within “the social arrangements of production, exchange, distribution and consumption”; to say that gender “refers to lived relationships between women and men”, including the production of “sexual divisions and definitions of masculinity and femininity”; and to identify race and ethnicity as “categories used to explain a . . . set of territorial relationships . . . involv[ing] conquests, . . . the historical development of nation states, and . . . migrations” (1996: 19–20). As mentioned in Section 2.1 above, social class is still important in variationist work, operating more or less in the survey mode that Savage associates with Goldthorpe, but from the mid-1980s onwards, non-variationist, discourse-oriented sociolinguistics in Britain has shown relatively little interest in social class, and instead it has focused much more on gender and ethnicity (though cf. Preece 2009). No doubt there are many different reasons for this loss of interest, many of them rooted in much wider cultural and economic changes (see Rampton et al. 2008). But closer to home, this sociolinguistic retreat from class analysis in Britain in the 1980s coincided with the growing influence of sociolinguistic ethnography and linguistic anthropology from North America. In the US, rather than being “spoken in its own right”, class tends to be “represented through other categories of social difference: gender, ethnicity, race, and so forth” (Ortner 1991: 164; see also, e.g. Bradley 1996: 75), and as it is set in very much its own continental history of conquest and slavery, North American work often shows an acute awareness of how the issues associated with class connect to race and ethnicity.12 But this can be lost in the transition to the UK, where class and ethnicity are more easily separated. In the UK, race and ethnicity get linked much more to post-war migrant groups than to indigenous populations, and so instead of seeing class and race as intricately linked, it becomes ethnicity for black and brown minorities and class for whites. With ethnicity and race de-classed in this way, it is quite easy to treat improved inter-group understanding as the key to full participation in mainstream society,

12

Ben Rampton

and that has been the dominant interpretation of Gumperz’s work on Crosstalk (Gumperz et al. 1979). Indeed, you can draw the same interpretation from a great deal of the work on multilingualism that has flourished in Britain from the 1980s. But as I have stressed, race and ethnicity are meta-analytic constructs, even though participants and researchers may agree on their reflexive relevance, and if you change your lens, you can see class (or gender) going on in exactly the same strips of activity where you first saw race and ethnicity. After all, the main speech event in Crosstalk is a job interview, and as well as illustrating inter-ethnic misunderstanding, it is a case-study of the socio-economic class demotion of migrant professionals from India, of their devaluation on the labor market. Now methodologically, ethnography not only helps us pick up on both the multiple, unpredictable and changing ways in which something like social class manifests itself. It also allows one to revisit earlier work with new questions, partly because it generates much more data than one can ever use, and partly because the data-collection isn’t tightly bound to the research questions that one first set out with. In the 1980s and 1990s, I did a lot of ethnographic work on ‘language crossing’, working across the different levels of social process that Gumperz, Heath and Bernstein study, looking in particular at how adolescents in multi-ethnic adolescent peer groups used each others’ ethnic languages (Rampton 2005). I interpreted this both as a locally generated, vernacular anti-racism, and as a central sociolinguistic process in the development of what Hall (1992) describes as new, hybrid ethnicities. I certainly recognized that social class was an important background influence, but I didn’t appreciate just how important it was, and so in the penultimate part of my presentation, I would like to present a summary reanalysis of crossing in terms of class (see Rampton [forthcoming] for a fuller exposition).

3. Crossing into class My research focused on Ashmead, a working class neighbourhood in the south Midlands of England, and I discovered that there was rather a sharp symbolic opposition between Creole and Asian English. There is a glimpse of this in Extract 1, and this comes from a playback interview: Extract 1: Participants: Asif (15 years old, male, Pakistani descent), Kazim (15, male, Pakistani descent), Alan (15, male, Anglo descent), Ben (the researcher/author, 30+, male, Anglo

Social class and sociolinguistics

13

descent). Setting: An interview, in which Ben is struggling to elicit some retrospective participant commentary on extracts of recorded data, and is on the point of giving up . . . 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

26 27 28 29 30 31 32 33 34 35 36

Ben: right shall I- shall we shall we stop there Kazim: no Alan: no come [ on carry on Asif: [ do another extract Ben: le- lets have (.) [ then you have to give me more= Alan: [ carry on Ben: =attention gents Asif: ((quieter )): yeh [ alright Alan: ((quieter )): [ alright Asif: ((quieter )): [ yeh Ben: I need more attention Kazim ((in Asian English)): I AM VERY SORRY BEN JAAD [AI AI æm veRi s6Ri ben dZA:d dZA:d] Asif ((in Asian English )): ATTENTION BENJAMIN [@thenSA:n @thenSA:n bendZemIn bendZemIn] : [ ((laughter)) Ben: [ right well you can- we cnAlan: [ BENJAADEMIN Ben: we can continue but we er must concentrate a bit [ more Asif: [ yeh Alan: alright [ (go on) then Asif: ((in Asian English )): [ concentrating very hard [ k6ns@stRetIï veRi AR AR] Ben: okay right : ((giggles dying down )) Kazim ((in Asian English)): what a stupid ( ) [v2d v2d @ stupId ] Ben ((returning the microphone to what he considers to be a better position to catch all the speakers)): concentrate a little bitAlan: alright then Kazim ((in Creole )): stop movin dat ting aroun [dæt dæt tIN eôAUn eôAUn] Ben: WELL YOU stop moving it around and then I’ll won’t need to (.) r[ight Kazim ((in Creole )): [stop moving dat ting aroun [dæP dæP tIN @ôAUn @ôAUn] Ben: right okay [ Kazim: [ BEN JAAD Alan: ((laughs)) Ben: what are you doing Alan: ben jaa[ad Ben: [ well leave ( ) alone

14

Ben Rampton

37 Kazim: IT’S HIM that ben jaad over there 38 Ben: right ((Ben continues his efforts to reinstitute the listening activity))

In this episode, things aren’t going quite as I had planned, and at the point where I threaten to stop the interview, Asif and Kazim switch into exaggerated Indian English in a sequence of mock apologies. Then a moment later in line 27, just as I seem to be signalling ’back-to-business’ by repositioning the microphone, the boot moves to the other foot, Kazim switches into Creole and directs a ‘prime’ at me, this time constructing my activity as an impropriety. This difference in the way Asian English and Creole are used fitted with a very general pattern in my data. Admittedly, away from stylized practices like these, Ashmead youngsters encountered many different uses of Asian and Creole English, and inside minority ethnic networks, the forms, functions and associations of Creole and Asian English were obviously a great deal more complex and extensive. But in spite of this, in crossing and stylization the images evoked were quite specific, and there was a sharp polarisation across a wide range of instances. Creole indexed an excess of demeanour over deference, displaying qualities like assertiveness, verbal resourcefulness, and opposition to authority, while Asian English stood for a surfeit of deference and dysfluency, typified in polite and uncomprehending phrases like ‘jolly good’, ‘excuse me please’, ‘I no understanding English’. Creole was clearly much more attractive to youngsters of all ethnic backgrounds, and it was often reported as part of the general local linguistic inheritance, particularly among Asian boys. So in the interpretation in Rampton (2005), I situated this socio-symbolic polarization in the larger context of migration. On the one hand, I said, Creole indexed an excitement and an excellence in youth culture that many adolescents aspired to, and it was even described as ‘future language’. On the other, Asian English represented distance from the main currents of adolescent life, and it stood for a stage of historical transition that a lot of youngsters, both male and female, felt they were leaving behind. Actually, though, this symbolization of a large scale historical trajectory went deeper. There was also a class dimension to the path indexed in the binary opposition of Creole and Asian English, and I shall mention four of the ways in which this showed up. First and most notably among boys, crossing and stylization themselves figured as something of a local class emblem, signifying the difference between Ashmead’s mixed adolescent community and the wider Stoneford population. When they described the kinds of people who wouldn’t do crossing and stylization, my informants referred to groups who were vertically placed at either end

Social class and sociolinguistics

15

of a bipolar hierarchy of wealth and status – a hierarchy that matched the demographic facts quite closely. Up above, there were the ‘posh wimpies’ living in wealthier districts outside Ashmead, and down below, there were Bangladeshis living in the very poorest parts of town. Second, there was significant overlap in the evaluation of Creole and local non-standard working class English. So when Asian and Anglo youngsters of both sexes described the efforts of their mums and dads to get them to speak properly, it was often the intrusion of swear-words, question tags and tense forms in Creole that were targeted. Third, the Creole and Asian English contrast itself actually mapped into a set of high/low, mind/body, reason-and-emotion oppositions that have been central to British class culture for several hundred years. According to Phil Cohen, this dualistic high/low idiom was generated “from within certain strategic discourses in British class society” from the 17th Century onwards, and “from the very outset [it was] applied across a range of sites of domination, both to the indigenous lower orders and ethnic minority settlers as well as to colonial populations overseas” (1988: 63). Given what I have just said about the evaluation of Creole overlapping with the evaluation of working class English, it is not hard to see Creole linked to the low side of the traditional British class semiotic. But just as important, the high side of the class binary was linked to Asian English. English is a prestige variety in the Indian sub-continent, and when my informants compared themselves with relatives there, they saw their own varieties as ‘slangish’, ‘not really proper’, inferior to the ‘strict’, ‘clear’, ‘good solid English they teach ‘em over there’. Transposed to the UK, however, Ashmead kids depicted this Indian English orientation to the high, proper and polite as absurd, its aspirations hopelessly marred by foreignness. Lastly, there was very little indication of a commitment to education in ethno-linguistic crossing and stylization in Ashmead.13 Of course, schools were a vital meeting point for kids from different ethnic backgrounds, and the general pastoral and extra-curricular ethos played a very significant part in promoting good interethnic relations. But Creole, which many admired, hardly featured at all on the curriculum, and rather than being tolerant of learners of English as a second language, or respecting them for their progress (as the teaching staff might hope), adolescents generally stigmatized pupils who hadn’t yet been fully socialized into the vernacular ways of ordinary youth. Overall, instead of curriculum learning, it was the activities and codes of conduct characteristic of popular culture and playground recreation that were central to the cross-ethnic spread of minority languages, and if anything, this was easier when a style was used in opposition to school authority.14

16

Ben Rampton

Drawing all this together, it looks as though the Creole / Asian English contrast oriented Ashmead adolescents to two major social processes. Not only did ethnolinguistic crossing and stylization situate these young people at an endpoint in the migrant transition from outside into Britain, but then also once inside, the binary lined them up with values much more associated with the lower than the higher classes. Yes, iconically, Creole was first and foremost associated with Caribbeans, Asian English with Asians. But in the problems, pleasures and expectations of working class adolescent life together, these kids experienced enough common ground to open up ethnolinguistic speech styles, respecifying their significance in crossing-and-stylization practices which recognized and cultivated the shared social space that labor migration had now created.

4. Conclusion I would like to conclude with two suggestions for further research and a general injunction. First, we need to understand how national and ethnic languages and speech styles contribute to the construction of globalized class identities. Under globalization, standardized minority language multilingualism is becoming more respectable and cosmopolitan – a new posh – and there is a strong case for seeing the kinds of ethnically-marked mixing, crossing and stylization described in the previous Section as an emergent counterpart to this, as a key feature in more globalized urban working class vernaculars. Second, in the words of one of Collins’ recent papers, we need to ask: “Where’s class in second language learning?” (2006). In a working class school I researched in the 1990s, it was quite common for multi-ethnic kids born or raised in Britain to direct stylized English foreigner talk at migrant and refugee newcomers, projecting a very non-cool, school-enthusiastic persona oriented to the high and polite. They were engaged in roughly the same kind of sociolinguistic status deflation that I had seen with stylized Asian English in Ashmead. Abroad, the global prestige of English might have encouraged you to see your acquisition of the language as an upward move, but for your classmates here and now, this was strictly comical (cf. Blommaert 2005). So how do newcomers navigate this kind of peer group pressure, what kinds of class origins and aspirations do they have, and how do they manage them in the relatively run-down urban environments where they often find themselves? Last and most obviously, we need to watch out for class much more generally, looking beyond occupation to the full range of processes and sites where central systems of production, distribution and consumption generate inequality, class

Social class and sociolinguistics

17

relations and class identities. The pioneers of sociolinguistics have identified a really rather robust set of multi-level coordinates for examining change in the workings and meaning of class, ethnicity, gender and generation; Bourdieu, a familiar figure for sociolinguists, has been a major influence in contemporary sociological research on class identification, culture and experience; and according to Savage (pers. comm.), one of the first people ever to write about social class in the British Journal of Sociology was Erving Goffman, one of the pillars of interactional sociolinguistics (cf. Goffman 1951). So there is a strong base for the sociolinguistic analysis of social class, and of course finally, my own recommendation is that we make full use of the perspectives and procedures of linguistic ethnography in the pursuit of this enquiry. King’s College London

Notes 1. This paper is based on a lecture delivered at the 2008 Annual Conference of the British Association of Applied Linguistics, and although I cannot claim that he agrees with every part of its argument, I would like to acknowledge a considerable debt to the work and influence of my friend Jim Collins, who has been working these seams far longer and deeper than me. 2. Savage characterises the equation of class with occupation as a retreat from the complexity of people’s perceptions of class. See also, e.g. Marshall (1997: 53). 3. “Representations [of class] . . . are not straightforwardly reproduced but are resisted and transfigured in their daily enactment. Categories of class operate not only as an organising principle which enable access to and limitations on social movement and interaction but are also reproduced at the intimate level of ‘structures of feeling’ (cf. Williams 1961, 1977) in which doubt, anxiety and fear inform the production of subjectivity” (Skeggs 1997: 6) 4. In a linguistic context, ‘habitus’ can be defined as a pre-conscious disposition to hear and speak in class and gender specific ways, inculcated into the individual through long-term experience of the purchase that their language resources provide in different kinds of setting. 5. The situation is different in the US and Australia (see Collins 2000). 6. Contrast Bourdieu, as discussed by Blommaert (2006). 7. “The perspective focuses upon the construction of rules which generate what may be called official pedagogic discourse/practices, whether these be in the school or the family. The perspective does not include the study of the full choreography of interaction in the context of the classroom or the family. Nor does the perspective offer the possibility of a delicate description of the full repertoire of arabesques of interaction within any classroom, staffroom or family. . . However, the principles of

18

8. 9. 10.

11.

12.

13.

14.

Ben Rampton description, whilst not capable of describing the full repertoire of classroom interactions (for which they were not designed), are well capable of describing features relevant to the theory of classroom interaction, its organisational contexts, and their relation to external agencies (e.g. family and work).” (Bernstein 1990: 7) This can be seen in the responses of, e.g. Rosen (1972, 1985) and Pratt (1987). See Rosen’s otherwise very positive (1985) review of Heath (1983). Admittedly, Part III of Raymond Williams’ (1977) Marxist and Literature develops an extraordinarily subtle pragmatics, prefiguring a lot of contemporary work in sociolinguistics and linguistic anthropology, but I’m not aware of this being picked up and used to any significant degree in, for example, Cultural Studies. “Ethnography, Language & Communication”, funded from 2007 to 2010 under the Economic & Social Research Council’s Researcher Development Initiative. See www.rdi-elc.org.uk. See, e.g. Collins (1988, 2000); McDermott and Gospodinoff (1981); Hymes (1996); Rampton et al. (2008: 73–74). Not that this critical integration of class and race perspectives is universal in NorthAmerica. Hymes speaks critically of an ethnolinguistic tradition which construes difference as “difference between separate, autonomous groups” (1996: 187), and this is central in Pratt’s (1987) critique of the ‘linguistics of community’. Delamont and Atkinson (1995) also suggest that this differential attention to ethnicity and class also has disciplinary underpinnings, not just national ones (1995: 34). At the same time, it is important to emphasize that for the most part, my research in Ashmead focused on a rather particular set of collective perceptions and conventional practices, and it would be a mistake to extrapolate from these to the more general attitudes, aspirations and trajectories of individuals. As a group, their crossing and stylization were embedded in practices and pastimes that were popular rather than elite, and looked towards (lower) class solidarity. But these kids also engaged in lots of other types of practice which I haven’t considered, and in fact in my datacollection, there was an explicit bias towards recreational sites. So it is very possible – in fact highly likely – that when, say, they settled down to work in lessons, a lot of youngsters could put aside the stances and alignments displayed in crossing – the attraction to Creole’s street credibility, the pleasures of Punjabi abuse, the mockery of Asian English deference. All of that might well be sectioned off quite easily as just having a laugh, messing around with friends. As well as Creole and Asian English, Punjabi had a very significant place in the local interethnic repertoire, and it was also beginning to be taught as a subject at school. But nobody white or black attended these classes, and instead of being book-learned, the active lexicon of cross-ethnic Punjabi among boys consisted of nouns referring to parts of the body, bodily functions, animals, ethnic groups and kin, as well as verbs to do with sex, violence and ingestion, while among a smaller number of primarily white girls, there was a degree of familiarity with bhangra. And in the three situations where most crossers got interested in Punjabi, the ethos was broadly non- or antischool. First, Punjabi was quite useful as a language for excluding teachers and other

Social class and sociolinguistics

19

white adults; second, it had entered the kind of traditional playground language and lore of school-kids described by the Opies, figuring among boys in incrimination traps, jocular abuse and chasing games; and third, white girls’ cross-ethnic interest in bhangra music and dance was generally embedded in heterosexual relations, with learning much more a matter of legitimate peripheral participation than classroom study (cf. Rampton 2005: Part III; Lave and Wenger 1991).

References Abercrombie, N. & A. Warde, with R. Deem, S. Penna, K. Soothill, J. Urry, A. Sayer & S. Walby. 2000. Contemporary British society. 3rd edn. Oxford: Polity. Atkinson, P. 1985. Language, structure and reproduction: An introduction to the sociology of Basil Bernstein. London: Methuen. Bernstein, B. 1971. Class, codes and control, volume 1: Theoretical studies towards a sociology of language. London: Routledge & Kegan Paul. Bernstein, B. 1975. Class, codes and control, volume 3: Towards a theory of educational transmissions. London: Routledge & Kegan Paul. Bernstein, B. 1990. The structuring of pedagogic discourse. London: Routledge Bernstein, B. 1996. Pedagogy, symbolic control and identity. London: Taylor & Francis Blommaert, J. 2005. Discourse: A critical introduction. Cambridge: Cambridge University Press. Blommaert, J. 2006. Bourdieu the ethnographer: On the ethnographic grounding of habitus and voice. The Translator 11. 219–236. Bourdieu, P. 1977. Outline of a theory of practice. Cambridge: Cambridge University Press. Bourdieu, P. 1991. Language and symbolic power. Oxford: Polity. Bradley, H. 1996. Fractured identities: Changing patterns of inequality. Oxford: Polity. Clark, J. T. 2003. Abstract inquiry and the patrolling of black/white borders through linguistic stylisation. In R. Harris & B. Rampton (eds.), Language, ethnicity and race: A reader, 303–313. London: Routledge. Cohen, P. 1988. The perversions of inheritance: Studies in the making of multi-racist Britain. In P. Cohen & H. S. Bains (eds.), Multi-racist Britain, 9–120. Basingstoke: Macmillan. Collins, J. 1988. Language and class in minority education. Anthropology and Education Quarterly 19. 303–319. Collins, J. 2000. Bernstein, Bourdieu and the New Literacy Studies. Linguistics and Education 11(1). 65–78. Collins, J. 2006. Where’s class in second language learning? Working Papers in Urban Language and Literacies 41. http://www.kcl.ac.uk/ldc. Coupland, N. 2007. Aneurin Bevan, class wars and the styling of political antagonism. In P. Auer (ed.), Style and social identities. Berlin & New York: Mouton de Gruyter. Delamont, S. & P. Atkinson. 1995. Fighting familiarity: Essays on education and ethnography. Kresskill, NJ: Hampton Press.

20

Ben Rampton

Eckert, P. 2000. Linguistic variation as social practice. Oxford: Blackwell. Edwards, A. & D. Westgate. 1987. Investigating classroom talk. Lewes: Falmer Press. Fairclough, N. 1995. Critical discourse analysis. London: Longman. Gilroy, P. 1987. There ain’t no black in the Union Jack. London: Hutchinson. Goffman, E. 1951. Symbols of class status. British Journal of Sociology 2(4). 294–304. Gumperz, J., C. Roberts & T. Jupp. 1979. Crosstalk: A study of cross-cultural communication. Southall, Middlesex: National Centre for Industrial Language Training. Hall, S. 1992. New ethnicities. In A. Rattansi & J. Donald (eds.), ‘Race’, culture and difference. London: Sage / The Open University. Harris, R. & B. Rampton (eds.). 2003. The language, ethnicity and race reader. London: Routledge. Heath, S. 1982. What no bedtime story means: Narrative skills at home and school. Language in Society 11(1). 49–76. Heath, S. 1983. Ways with words. Cambridge: Cambridge University Press. Hymes, D. 1996. Ethnography, linguistics, narrative inequality. London: Taylor and Francis. Labov, W. 1969. The logic of non-standard English. In J. Alatis (ed.), Georgetown monographs on language and linguistics 22, 1–22, 26–31. Washington: Georgetown Press. Lave, J. & E. Wenger. 1991. Situated learning: legitimate peripheral participation. Cambridge: Cambridge University Press. Marshall, G.. 1997. Repositioning class: Social inequality in industrial societies. London: Sage. McDermott, R. & K. Gospodinoff. 1981. Social contexts for ethnic borders and school failure. In H. Trueba, G. Guthrie and G. Au (eds.), Culture and the bilingual classroom, 212–236. Rowley: Newbury House. (Also in Harris & Rampton [eds.] 2003: 276–290.) Ortner, S. 1991. Reading America: Preliminary notes on class and culture. In R. Fox (ed.), Recapturing anthropology: Working in the present, 164–189. Santa Fe: School of American Research Press. Pratt, M. L. 1987. Linguistic utopias. In N. Fabb, D. Attridge, A. Durant & C. McCabe (eds.), The linguistics of writing, 48–66. Manchester: Manchester University Press. Preece, S. 2009. Post talk: language and identity in higher education. Basingstoke: Palgrave Macmillan. Rampton, B. 2005. Crossing: Language and ethnicity among adolescents, 2nd edn. Manchester: St Jerome Press. Rampton, B. 2006. Language in late modernity: Interaction in an urban school. Cambridge: Cambridge University Press. Rampton, B. Forthcoming. Style contrasts, migration and social class. Journal of Pragmatics. Rampton, B., R. Harris, J. Collins & J. Blommaert. 2008. Language, class and education. In S. May & N. Hornberger (eds.), Encyclopedia of language and education, Volume 1, 71–82. New York: Springer.

Social class and sociolinguistics

21

Reay, D. 1998. Rethinking social class: Qualitative perspectives on class and gender. Sociology 32(2). 259–275. Rosen, H. 1972. Language and class: A critical look at the theories of Basil Bernstein. Bristol: Falling Wall Press. Rosen, H. 1985. The voices of communities and language in classrooms: A review of Ways with Words. Harvard Educational Review 55(4). 448–456. Savage, M. 2003. A new class paradigm? British Journal of Sociology of Education 24(4). 535–541. Savage, M. 2005. Working-class identities in the 1960s. Sociology 39(5). 929–946. Savage, M. 2007. An alternative history of class identities in post-war Britain. ESRC Identities and Social Action Programme Public Lecture, London Metropolitan University, 7 March. Skeggs, B. 1997. Formations of class and gender. London: Sage. Thompson, E. P. 1978. The poverty of theory and other essays. London: Merlin. Tizard, B., M. Hughes, H. Carmichael & G. Pinkerton. 1988. Language and social class: Is verbal deprivation a myth? In N. Mercer (ed.), Language and literacy from an educational perspective, 4–15. Milton Keynes: Open University Press. Trudgill, P. 1975. Accent, dialect and the school. London: Arnold. Wells, G. 1981. Learning through interaction. Cambridge: Cambridge University Press. Williams, R. 1977. Marxism and literature. Oxford: Oxford University Press. Willis, P. 1977. Learning to labour: How working class kids get working class jobs. Farnborough: Saxon House. Woolard, K. 1985. Language variation and cultural hegemony: Toward an integration of sociolinguistic and social theory. American Ethnologist 12. 738–748. Zimmerman, D. 1998. Discourse identities and social identities. In C. Antaki & S. Widdicombe (eds.), Identities in talk, 87–106. London: Sage. Ben Rampton is Professor of Applied and Sociolinguistics and Director of the Centre for Language Discourse and Communication at King’s College London. He does interactional sociolinguistics, and his interests cover urban multilingualism, ethnicity, class, youth and education. He is the author of Crossing: Language and ethnicity among adolescents (Longman 1995 / St Jerome 2005) and Language in late modernity: Interaction in an urban school (Cambridge University Press 2006), and a co-author of Researching language: Issues of power and method (Routledge 1992). He has also edited or co-edited The language, ethnicity and race reader (Routledge 2003), Working papers in urban language and literacy (www.kcl.ac.uk/ldc), and special issues of Journal of Sociolinguistics (3[4] 1999; 11[5] 2007), International Journal of Applied Linguistics (7[1] 1997), Critique of Anthropology (21[1] and [2] 2001) and Pragmatics (13[1] 2003). He is currently Director of ’Ethnography Language and Communication’, a three-year research training program designed for social scientists UK-wide (www.rdielc.org.uk), and he is convener of the UK Linguistic Ethnography Forum (www.lingethnog.org.uk).

Reading Derrida: Language, identity and violence1 TIM MCNAMARA

Abstract This article offers a reading of a major text on language by the French philosopher Jacques Derrida, Monolingualism of the Other, and considers its relevance, and Derrida’s work more generally, to the concerns of applied linguistics. Derrida’s argument about the necessarily conflicted nature of identity and its relationship to language has implications for work in many areas of applied linguistics, especially bilingualism, language attitudes, language maintenance and shift, the role of English as a global language, theories of motivation in language learning and current debates over nationality and citizenship. Keywords: language and identity; Derrida; poststructuralist theory; applied linguistics; mother tongue.

1. Introduction: Reading Derrida This paper argues for the relevance to applied linguistics of the work of the French philosopher Jacques Derrida (1930–2004) by offering a reading of a text of his on language and identity, Monolingualism of the Other (Derrida 1998), and considering its relevance to a range of issues in applied linguistics. Derrida’s work is relatively little cited in applied linguistics,2 which is on the face of it surprising given that language is a persistent focus of his work. His discussion of spoken and written language in his definitive early text, Of Grammatology (Derrida 1976), together with the work of his mid-career on the linguistic pragmatics of the philosophers Austin and Searle (Derrida 1988), would alone provide a basis for claiming that Derrida was one of the most important theorists of language of the 20th and early 21st centuries. He addresses the specific issue of the relation of language and identity in several places in his work, particularly, as

24

Tim McNamara

we shall see in this paper, in his writing about himself as a speaker of French, but also for example in his discussion of the linguistic and cultural identity of the German-Jewish poet Paul Celan as a speaker of German writing in German about the Holocaust (Derrida 2005). There are a number of reasons for the relative unfamiliarity of Derrida’s work, perhaps largely a question of its difficulty for the applied linguistics reader. First is the fact that Derrida is of course not a linguist but a philosopher, and draws heavily on the philosophical tradition of both the Greeks and the philosophers of the Enlightenment. That should itself not be a problem, as the work of philosophers writing in English such asAustin and Searle have been important in applied linguistics. As a philosopher, he is not of course part of the Anglo-American analytic tradition, and must be read in relation to the philosophical tradition known in the English-speaking world as Continental Philosophy; he draws in particular on the work of Nietzsche, Husserl, Heidegger and Levinas. Secondly, Derrida engages not only with the work of philosophers but with a wide range of nonphilosophers: Marx in political theory, de Saussure in linguistics, Levi-Strauss in anthropology, Freud and Lacan in psychoanalysis, Mallarm´e, Joyce, Jab`es and Celan in literature (in several languages), to name just a few. Thirdly, there is his style – poetic, meticulous, allusive, full of extended metaphors, rapidly shifting, and characterized by puns and paradoxes. And finally, his own texts consist of exacting readings of the texts of other thinkers, and assume in the reader a close familiarity with the original text – which may of course be lacking. In Of Grammatology, for example, a series of brilliant discussions of language, he offers careful readings of texts by Saussure, Rousseau and Levi-Strauss. Sometimes, the work of more than one thinker is the subject of commentary at the same time. The breadth and depth of Derrida’s erudition and familiarity with work in many linguistic and cultural traditions can be daunting. Perhaps because of the demands that his texts place on the reader, Derrida’s relation to the tradition of earlier thought is often misunderstood. The term deconstruction, the philosophy with which his name is synonymous, is associated in the popular imagination with a debunking or negative pulling apart, but in fact it is better captured by the term close reading. Such a reading shows at the same time as a spirit of interrogation a deep appreciation of and respect for the source text. The philosopher Judith Butler, in her published remarks at the time of Derrida’s death in October 2004, comments on this aspect of his work: These are authors that he could not do without, ones with whom and through whom he thinks. He writes only because he reads, and he reads only because there are these authors to read time and again. He “owes” them something or, perhaps, everything, if only because he could not write without them; their writing exists

Reading Derrida: Language, identity and violence

25

as the precondition of his own; their writing constitutes the means through which his own writing voice is animated and secured. (Butler 2004: 32)

A further misunderstanding of deconstruction is that it is at bottom nihilistic or at least in a limiting way relativist, or worse, apolitical. In fact Derrida, as I will show, is a profoundly ethical thinker, whose larger project can be seen as a sustained interrogation and revealing of the conceptual and intellectual underpinnings of violence in modernity. What is also clear is Derrida’s sustained interest in language as part of this problem of violence. The frequent (mis)conception of deconstruction as being about playfulness and irresponsibility could not be farther from the truth. Given then the relevance, but also the difficulty, of Derrida’s texts, how can we begin to approach them? Derrida is at his most accessible in interviews, but also and in particular when he reflects on the facts of his own autobiography, which he did in a number of texts, particularly towards the end of his career.3 These essays are not mere reminiscences but close meditations on the facts of his experience: his life becomes, as it were, the text he is deconstructing. This paper, then, offers a reading of a text on language and identity, Monolingualism of the Other, published in French in 1996 and in English in 1998. I begin with an overview of the argument of the book and its historical context, before discussing its themes in detail, and conclude by considering its specific relevance to applied linguistics.4 2. Derrida’s Monolingualism of the Other In Monolingualism of the Other, Derrida uses an account of the violent and contradictory social forces central to the formation of his own identity as the occasion for a general reflection on linguistic and cultural identity and its relationship to violence. Born into a Sephardi Jewish family in colonial Algeria in 1930, he and his family suffered under the Vichy administration of Algeria during the war; his experience of colonialism and anti-Semitism in the formation of his identity as a speaker of French is the subject of the book. But the significance of the discussion goes well beyond this setting. His argument applies to all of us: our mother tongue is not our own, but is the language of the Other. Further, this means that to the extent that one’s mother tongue is an aspect of identity, this identity is always of the Other. While this becomes unavoidably clear for linguistic and cultural groups who find themselves the target of overt violence, it has a universal application. The title of the book needs explanation in relation to the political and cultural history of Algeria, the society into which Derrida was born. Algeria forms

26

Tim McNamara

part of the larger area of North Africa known as the Maghreb. Berber people were among the early settlers of the Maghreb (certainly by 3000 bc); Berbers still make up approximately 20% of the population of the country, and speak a range of Berber languages including Kabyle. The majority of Algerians (around 80%) are Arabs, who first settled the Maghreb in the 600s ad and developed their own Maghrebian variety of Arabic. The Arab settlement of Algeria also saw the development of a small Jewish community5 who spoke a Jewish ethnolect of Arabic known as Judaeo-Arabic; this community was supplemented after 1492 with Jews expelled from Spain, who brought with them a Jewish ethnolect of Spanish known as Ladino. Combined, the various Jewish ethnic communities constituted by 1931, the year after Derrida was born, less than 2% of the population of Algeria (the proportion at colonization may have been less than this). The linguistic environment in which Derrida found himself as a child was formed within the context of colonialism. Algeria had begun to be colonized by the French in 1830. The explicit ideology of this colonization saw itself as bringing the fruits of the French revolution to other peoples, and thus as liberal and welcoming. This was expressed in various ways, including the encouragement to learn and use French, and the extension of certain civil liberties, specific products of the French revolution, to the colonial setting. As a result, French became the sole official language, the sole medium of education and the sole language of administration. The colonial language policy also meant that languages other than French were effectively, although not technically, forbidden in the education system and in public life. In this sense, according to Derrida, the “unilateral imposition of [a] ‘politics’ of language” was “disguised under alibis of ‘universal’ humanism, and . . . of the most generous hospitality” (1998: 39). The policy favoring French led, for example, to the absurdity that Arabic, the language of the vast majority of the population, was not recognized in public life, and was taught as if it were a foreign language like English or German; other minority languages, such as Berber or the Jewish ethnolects (the vernaculars Judaeo-Arabic, Judaeo-Berber and Ladino, and Hebrew, the written language of religion) had no place in public life. Derrida comments on the damaging impact of the colonial language policy on the indigenous communities of Algeria (and I think we can see many parallels to this in other situations of liberal domination): When access to a language is forbidden, nothing – no gesture, no act – is forbidden. One forbids access to speech, that is all, a certain kind of speech. But that is precisely the fundamental interdiction . . . . This limit was never set down, enacted either as an act of law – an official decree, a sentence – or like a physical, natural or organic barrier . . . . We had the

Reading Derrida: Language, identity and violence

27

choice, the formal right, to learn or not learn Arabic or Berber. Or Hebrew. It was not illegal, or a crime . . . . The interdict worked therefore through other ways. More subtle, peaceful, silent, and liberal ways . . . . In the manner of permitting and giving, for, in principle, everything was given, or at any rate permitted. (1998: 32)

Derrida grew up as a specific kind of colonial subject in Algeria: as a member of the indigenous Sephardi Jewish community in Algiers. The word Sephardi comes from the Hebrew word for Spain (Sfarad), and represents one of the two broad cultural traditions of Jewish life (the other, Ashkenazi, deriving from central and eastern Europe). All Jews and Muslims who refused to convert to Christianity were expelled from Spain in 1492 and spread throughout the Ottoman Empire, including in North Africa, the Jews bringing Ladino with them. The term Sephardi was then extended to refer to all Jews living in the Muslim world, even if they were not originally from Spain; i.e. it was extended to members of the ancient Jewish communities in Arab countries who spoke Jewish ethnolects of Arabic known collectively as Judaeo-Arabic. The tiny indigenous Jewish community of Algeria thus had a tradition of speaking Ladino or JudaeoArabic, although this had been lost in the generation preceding Derrida’s birth. This community had benefited from one of the fruits of the Revolution extended to colonized peoples, the granting of French citizenship to the Jews of Algeria in 1870. This Republican tradition in favor of civil rights for Jews was not matched in other spheres of French public life, which remained very divided in its attitudes to Jews, as the anti-Semitism of the Dreyfus affair of the 1890s made painfully evident. Despite these ominous signs, the Algerian Jews responded to the gesture of citizenship by enthusiastically assimilating in a way which Derrida characterizes as “profound, rapid, zealous, and spectacular. In two generations” (1998: 17). One of the casualties of this assimilation was the loss of the traditional Jewish vernaculars. This, and the effective interdict against Arabic, meant that the indigenous Jews of Derrida’s generation were monolinguals – speakers of French. And then in 1940, following the occupation of France by Nazi forces, the Jews of Algeria were abruptly deprived of their French citizenship and many other rights. Derrida and his family experienced this directly, as he explains in a later text: When state anti-Semitism was unleashed in Algeria, in 1940–42, my father was grateful to his employers for protecting us, for keeping him in their service, whereas they could simply have fired this Jewish employee, as some people were urging them to do, and as they had the legal right to do. I felt humiliated to see him overflowing with respectful gratitude to these people for whom he had worked for forty years and who generously ‘consented’ to ‘keep him on’. (Derrida and Roudinesco 2004: 108)

28

Tim McNamara

Derrida, who was in early adolescence during this period, was expelled from school as a result of these measures: I was expelled from the Ben Aknoun high school in 1942, and beyond any anonymous ‘administrative’ measure, which I didn’t understand at all and which no one explained to me, the wound was of another order, and it never healed: the daily insults from the children, my classmates, the kids in the street, and sometimes threats or blows aimed at the ‘dirty Jew’, which, I might say, I came to see in myself . . . . (Derrida and Roudinesco 2004: 109)

The result was the painful realization that Derrida’s Self was constituted in the discourse and language of the Other. He expresses this idea as a paradox: ‘I only have one language; it is not mine.’ I am monolingual. My monolingualism . . . feels like one to me, and I remain in it and inhabit it. It inhabits me. . . . This monolingualism is me . . . I would not be myself outside it. It constitutes me . . . Yet it will never be mine, this language . . . . (Derrida 1998: 1–2)

However, in a later commentary on this text, he argues again for its significance beyond the immediate context of his autobiographical narrative. When I say “I only have one language: it is not mine”, such a statement defies common sense and is self-contradictory. Such a contradiction is not someone’s personal heartrending contradiction but a contradiction inscribed within the possibility of language. Without this contradiction there would be no language. I believe, therefore, that one must endure it . . . . (Derrida 2005: 104)

The complex and paradoxical argument of the book introduces a number of concepts and themes which, I will argue, are highly relevant and productive for issues central to the discussion of the relationship of language and identity in applied linguistics. In the next section, I will attempt to explicate how the text supports the following propositions, which have important implications for such discussions: 1. Subjectivity involves a relation to the Other (and its language). 2. Subjectivity involves desire in relation to the Other (and its language). 3. Deconstruction can be understood as the subject rewriting the language of the Other.

Reading Derrida: Language, identity and violence

29

3. Subjectivity and the relation to the other (and its language) Derrida does not use the term subjectivity, a term more familiar in the work of Foucault (e.g. Foucault 1977), but rather continues to use the terms identity and identification. For Derrida, identity is ‘disordered’: To be a Franco-Maghrebian, one ‘like myself’, is not . . . a surfeit or richness of identities, attributes, or names. In the first place, it would rather betray a disorder of identity [trouble d’identit´e ]. (Derrida 1998: 14)

What are the features of this disorder? First, and crucially, identity is dependent on identification by the Other, that is, on the terms in which one is recognized or identified. The input of the Other into the formation of consciousness is the subject of a famous passage of Hegel, the discussion of the Lord and Bondsman in Phenomenology of the Spirit (Hegel 1977). Judith Butler (1999) traces the powerful role of this passage in French intellectual life following Koj`eve’s influential lectures on Hegel in Paris in the 1930s, which were attended by Sartre, De Beauvoir, Lacan, and those who were some years later to teach philosophy to Foucault and Derrida.6 The relevance of the Hegelian tradition here is that the Algerian Jewish case illustrates how subjectivity is constructed through the terms in which one is recognized by the Other. Recognition or identification of the indigenous Jews of Algeria by the French Other occurred at two quite distinct moments, in 1870 and again in 1940, and their impact was contradictory. Even separately, however, in their different ways each of these recognitions created a ‘disorder of identity’. First, the granting of citizenship to the indigenous Jews of Algeria in 1870 extended the recognition represented by French citizenship to this previously ‘unrecognized’ group. This led to a transformation in their identity, involving an enthusiastic assimilation, which Derrida describes critically in the following terms: Since the end of the last century, with the granting of French citizenship, assimilation, as we say, and acculturation – the feverish bid for a ‘Frenchifying’ which was also an embourgoisification – were so frantic and so careless that the inspiration of Jewish culture seemed to succumb to an asphyxia: a state of apparent death, a ceasing of respiration, a fainting fit, a cessation of the pulse . . . . (Derrida 1998: 53)

The assimilation even affected the outward appearance of religious practices: Social and religious behaviour, even Jewish rituals themselves, were tainted . . . . The churches were being mimicked, the rabbi would wear a black cassock, and the verger [chemasch] a Napoleonic cocked hat; the ‘bar mitzvah’ was called communion, and circumcision was named ‘baptism’. (Derrida 1998: 54)

30

Tim McNamara

Worse, the younger generation of Jews such as Derrida even began to see Jewish religious practices negatively through Christian eyes. The influence on his attitudes he describes as: . . . an insidious Christian contamination: the respectful belief in inwardness, the preference for intention, the heart, the mind, mistrust with respect to literalness . . . in short, a denunciation, so conventional, of Pharasaism. (Derrida 1998: 54)

This assimilation led to these younger Jews of Derrida’s generation living in a sort of no-man’s-land of identity: They could not properly identify themselves, in the double sense of ‘identifying oneself’ and ‘identifying oneself with’ the other. They could not identify themselves in terms of the models, norms, or values whose development was to them alien because French, metropolitan, Christian, and Catholic . . . . But being already strangers to the roots of French culture, even if that was their only acquired culture, their only educational instruction, and, especially, their only language, being strangers, still more radically, for the most part, to Arab or Berber culture, the greater majority of these young ‘indigenous Jews’ remained, in addition, strangers to Jewish culture; a strangely bottomless alienation of the soul: a catastrophe; others will also say a paradoxical opportunity. (Derrida 1998: 52)

This disordered identity was also reflected in language affiliation in terms of a mother tongue. For the indigenous Jews, monolingualism (in French) was the norm. Because of the ‘interdict’ represented by French language policy, Arabic was not available: These young indigenous Jews could easily identify neither with the ‘Catholics’ [the term used for all Christians], the Arabs, nor the Berbers, whose language they did not generally speak in that generation. Two generations before them, some of their grandparents still spoke Arabic, at least a certain form of Arabic. (Derrida 1998: 52)

Jewish languages were no longer available, either: As for language in the strict sense, we could not even resort to some familiar substitute, to some idiom internal to the Jewish community, to any sort of language of refuge that, like Yiddish, would have ensured an element of intimacy, the protection of a ‘home-of-one’s-own’[un ‘chez-soi’] against the language of official culture, a second auxiliary in different socio-semiotic situations. ‘Ladino’was not spoken in the Algeria I knew, especially not in the big cities like Algiers, where the Jewish population happened to be concentrated. (Derrida 1998: 54–55)

Abruptly, however, in 1940, the indigenous Jews of Algeria were re-recognized or re-identified, this time in very different terms. Suddenly, they were no longer recognized as French citizens, but were identified exclusively in terms of their

Reading Derrida: Language, identity and violence

31

(violently stigmatized) Jewishness. Derrida is careful to establish that this was the expression of a specifically French anti-Semitism:7 The withdrawal of French citizenship from the Jews of Algeria, with everything that followed, was the deed of the French alone. They decided that all by themselves, in their heads; they must have been dreaming about it all along; they implemented it all by themselves . . . . It was a Franco-French operation, one ought to say an act of French Algeria in the absence of any German occupation. One never saw a German uniform in Algeria. No alibi, denial, or illusion is possible: it was impossible to transfer the responsibility of that exclusion upon an occupying alien. (Derrida 1998: 16–17)

Derrida links his experience at that time to the issue of his identity as a French speaker: I was very young at the time, and I certainly did not understand very well . . . what citizenship and loss of citizenship meant to say. But I do not doubt that exclusion – from the school reserved for young French citizens – could have a relationship to the disorder of identity of which I was speaking to you a moment ago. I do not doubt either that such ‘exclusions’ come to leave their mark upon this belonging or non-belonging of language, this affiliation to language, this assignation to what is peacefully called a language. (Derrida 1998: 16–17)

In summary, processes of recognition by the Other were critical in shaping the identity, the language behavior and language attitudes of the community to which Derrida belonged. Moreover, this identity was shaped in a contradictory fashion. Recognition came first as ‘recognition as like’ (inclusion, leading to cultural and linguistic assimilation and alienation from traditional and local cultures and languages) and then without warning it became ‘recognition as unlike’ (rejection and exclusion, leading to another more obvious kind of alienation). In each case, the process of subjection and subjectification of the indigenous Jews of Algeria in these very different terms of recognition – ‘silent, peaceful’ assimilation on the one hand, violent anti-Semitism on the other – created a subject that was vulnerable in different ways. Each process contributed to the ‘disorder of identity’ which is the theme of Derrida’s text.

4. Subjectivity and desire in relation to the other (and its language) The account so far, however, would tend to suggest the essential passivity and receptiveness of the subject in these processes of recognition. But this is a misrepresentation, as there is another, compounding factor: the Other which identifies or recognizes the subject is itself the object of the subject’s desire.

32

Tim McNamara

Desire here refers to a range of complex, even contradictory feelings: on the one hand passionate attachment, even physical desire, and on the other shame, pain, anger, and the desire to hurt. French, the only language that the community knows, as the embodiment of the Other, becomes the focus of desire in this sense. Thus, Derrida speaks of “an impulse of love and aggression toward the body of [French]” (Derrida 1998: 66), “one of these impulses of revenge out of which I here shape my theme” (Derrida 1998: 51), scandalously claiming that he both suffers and takes pleasure in8 his paradoxical relationship to the language: You at once appreciate the source of my sufferings, the place of my passions, my desires, my prayers, the vocation of my hopes, since this language runs right across them all . . . . I wonder if one can love, enjoy oneself [jouir], pray, die from pain, or just die, plain and simple, in another language or without telling anyone about it, even without speaking at all. But above all, and this is the double edge of a sharp sword that I wished to confide to you almost without saying a word: I suffer and take pleasure in [jouis de] what I am telling you in our aforementioned common language: ‘Yes, I only have one language; yet it is not mine.’ (Derrida 1998: 2)

He describes his discovery as a young man of philosophy and literature in French in terms of love and pain, of desire experienced as penetration: Several years later . . . I seemed to be harpooned by French philosophy and literature . . . wooden or metallic darts [fl`eches], a penetrating body of enviable, formidable and inaccessible words even when they were entering me . . . . (Derrida 1998: 50)

In an extraordinary passage Derrida talks of making deconstructive love to French: What dream? Not that of harming the language (there is nothing I respect and love as much), not that of endangering it or injuring it . . . But the dream . . . was perhaps to make something happen to this language. The desire to make it arrive here, by making something happen to it . . . something so intimate that it would no longer even be in the position to protest . . . so intimate that it cannot oppose it . . . something so intimate that it comes to take pleasure in it as in itself, at the time it loses itself by finding itself . . . at the time when an incomprehensible guest, a newcomer without assignable origin, would make the said language come to him, forcing the language then to speak itself by itself, in another way, in his language. To speak by itself. But for him, and on his terms, keeping in her body the ineffaceable archive of this event: not necessarily an infant but a tattoo . . . . (Derrida 1998: 51–52)

Reading Derrida: Language, identity and violence

33

It is not possible to do justice here to the theme of desire in subjectivity. It is a central theme in the work of Judith Butler (her thesis on the reception of Hegel in twentieth century France is entitled Subjects of Desire [Butler 1999]). In her books on gender and sexuality (Butler 1990, 1993), and in her later work on subjectivity (for example, Butler 1997, 2005) Butler’s project is to explore, using psychoanalytic theory, the psychic processes necessarily involved if the power represented by discourses (Foucault) or ideologies (Althusser) is to create the subject. The subject cannot exist prior to the process of subjectification; it is not just a tabula rasa on which is inscribed the subject positions determined by the operation of power. Suffice it to say here that Derrida is close to this view; the subject constituted through the power of the Other is involved in a relationship of desire with the Other which in this case complicates the relationship to the Other’s language. The option for Derrida is not simply to reject French (and remember he had no other language to turn to) but to critically re-work its significance for him by revealing that significance to itself and in the process in a sense re-inventing it – or deconstructing it.

5. Deconstruction as the subject rewriting the language of the other Derrida’s project of ‘deconstruction’ – the questioning of the linguistic, cultural and philosophical traditions of modernity in the context of the violence of Western culture as exemplified by colonialism and the Holocaust – can thus be understood as Derrida’s response to the dilemma, the disorder of identity, that I have outlined in the previous two sections. Derrida explicitly links the argument about language in Monolingualism of the Other to the project of deconstruction, which is at once a philosophical project and a reworking of Derrida’s problematic relationship with French: This debate with monolingualism will have been nothing other than a piece of deconstructive writing [´ecriture]. Such writing always attacks the body of this language, my only language, and . . . the philosophical tradition that supplies us with the reservoir of concepts that I definitely have to use . . . . (Derrida 1998: 59) Yes, by this word e´ criture we would indicate, among other things, a certain mode of loving9 and desperate appropriation of language and through it of a forbidding as well as a forbidden speech (for me, the French language was both) . . . the loving and jealous vengeance . . . which attempts to restore the language and believes it is at the same time reinventing it, finally giving it a form (deforming, reforming and transforming it), making it pay the price of the interdict . . . . (Derrida 1998: 32–33)

34

Tim McNamara

Having no other language, he will show the Other what its language means, what it is saying, by critically examining its very vocabulary – ‘fraternity’, ‘brotherhood’, the ‘mother tongue’ – and in a sense turning the language on itself. . . . ‘My mother tongue’ is what they say, is what they speak; as for me, I cite and question them. I ask them in their own language . . . if they indeed know what they are saying and what they are talking about. Especially when, so lightly, they celebrate ‘fraternity’. At bottom, brothers, the mother tongue, and so forth pose the same problem. It is a bit as if I was awakening them to tell them: ‘Listen, pay attention, now that is enough, you must wake up . . . . One day, you will see that what you are calling your mother tongue will no longer even respond to you.’ (Derrida 1998: 33)

What is the general significance of this autobiographical text? In certain respects, his singular experience raises the general issue, discussed in parts of this text, and in other work, of the complex dilemmas of other assimilated European Jewish intellectuals prior to and following the Nazi period, including Rozenzweig, Sholem, Arendt, Celan and Kafka, and their relationships to Yiddish, Hebrew and German. One of the strongest impulses behind Derrida’s work as a whole, including his concerns in relation to forgiveness and hospitality (Derrida 2001; Derrida and Dufourmantelle 2000), seems to me to be his position as an assimilated, secular Jewish intellectual in Europe in the immediate aftermath of the Holocaust. Its violence, a violence of identity, is a fundamental challenge for all of us who wish to remain working within the European intellectual tradition. More generally, however, the paradox of Derrida’s relationship to French is characteristic of the way in which, throughout his writing, he draws out the simultaneous presence of apparent opposites, often in order to challenge what are seen as binary distinctions (for example, between speech and writing in Of Grammatology). One such binary is the distinction between inside and outside (e.g. in the discussion of the pharmakon in Plato’s Pharmacy [Derrida 1981]). This critique perhaps becomes clearer when we understand it in the context of Derrida’s life, where he is both insider and outsider, writing on the inside of language from outside, bringing the outside of language inside language. Here, the paradox of Derrida’s relationship to French carries with it an emotional burden (of alienation, of an undermining on one’s personhood, of the inward psychological impact of cultural violence). One’s innermost sense of linguistic and cultural identity is experienced as a contradiction – more, as an experience of violence. This is why Derrida’s narrative has implications quite beyond the particular context of which he is speaking. In his words:

Reading Derrida: Language, identity and violence

35

I knew that what I was saying in The Monolingualism of the Other was valid to a certain extent for my individual case, to wit, a generation of Algerian Jews before the Independence. But it also had the value of a universal exemplarity, even for those who are not in such historically strange and dramatic situations as . . . mine. I would venture to claim that the analysis is valid even for someone whose experience of his own mother tongue is sedentary, peaceful, and without any historical drama. (Derrida 2005: 101)

Thus the point here is not about a particular language (French) as such, nor about colonialism, nor even about anti-Semitism and the Holocaust, but about the violence within cultural discourses constituted through language, including within monolingual contexts. In summary, then, Derrida’s project in relation to French is both singular and universal, a theme he elaborates in different parts of the text. In interpreting his disorder of identity in terms of this broader deconstructive project, Derrida is providing the basis for his assertion that all of us have a relation to the mother tongue as problematic as his, even when we are not aware of it; language, as he says, is always of the Other. In spite of appearances, this exceptional situation is, at the same time, certainly exemplary of a universal structure; it represents or reflects a type of originary ‘alienation’ that institutes every language as the language of the other. (Derrida 1998: 63)

6. Derrida and applied linguistics I have tried in this reading of the text to give a feeling for the richness, the complexity and the provocative quality of Derrida’s writing in relation to language. But how relevant is this work and the issues it raises to our concerns within applied linguistics? First, Derrida’s work contributes to the ongoing work of re-thinking our assumptions about ‘mother tongue’ within linguistics and applied linguistics. While the privileging of the native speaker is under considerable challenge, particularly in work on English as a lingua franca (Seidlhofer 2005, 2006, 2009; Seidlhofer et al. 2006), the ‘mother tongue’largely retains the romantic and positively charged aura given it within 19th century writings on German nationalism, whose toxic legacy has so clearly been demonstrated by Chris Hutton in his Linguistics and the Third Reich: Mother tongue fascism, race and the science of language (Hutton 1999). Mother tongue is also a key and relatively unexamined concept in work on language maintenance and shift, and is ripe for critique, a process which has now commenced (Cameron 2007; Block 2008). Wiley’s trou-

36

Tim McNamara

bling analysis of the ideology of research on mother tongue maintenance in the work of Heinz Kloss (Wiley 2002), central to so much that has followed in the sociology of language, is a case in point. Derrida provides a general theoretical basis for critique of the notion of the mother tongue. He contributes a general troubling of the assumed security and privileging that the mother tongue appears to offer by insisting that our relationship to the mother tongue is necessarily an experience of colonization; and that the resulting allegiance is open to question and indeed resistance if the experience of such colonization is recognized as psychologically or literally violent. The embrace of another language may be seen as a form of resistance, although it too will necessarily involve the experience of a secondary colonization. Thus even bilingualism or multilingualism offer no escape from the monolingualism of the Other – they involve simply a proliferation of monolingualisms. Second, Derrida’s account of his relationship to French is clearly relevant to the debate about the spread of English as an international language.10 Certainly there is material in this book to support the perspective that sees the spread of English in terms of linguistic imperialism (Phillipson 1992). But by extending the notion of colonialism to include all culture, and all language, including the mother tongue, Derrida makes a radical extension of the notion of linguistic imperialism in such a way as to complicate the notion of resistance to it: I would not like to make too easy use of the word ‘colonialism’. All culture is originarily colonial. Every culture institutes itself through the unilateral imposition of some ‘politics’ of language. Mastery begins, we know, through the power of naming, of imposing and legitimating appellations. (Derrida 1998: 39)

Moreover, the role of desire in the relation to the colonising Other complicates the situation further, as we have seen. Derrida thus provides a theoretical basis for those views of the cultural politics of the spread of English which are informed by poststructuralism and which are critical of simplistic accounts of linguistic imperialism (Pennycook 1995, 2001). He does so by removing the possibility of simple opposition between colonizing and colonized languages: there are no ‘good’ and ‘bad’ languages. Similarly, the notion of attachment to the colonizing language cannot simply be rejected as perverse. We are reminded of Achebe’s often-cited remark: “Let no one be fooled by the fact that we may write in English for we intend to do unheard of things with it.” (Achebe 1988: 50) Third, in much work in applied linguistics and sociolinguistics on multilingualism, there is a celebratory emphasis on hybridity and hybrid identities. Derrida’s account of the hybridity of Jewish culture in Algeria, however, emphasises its vulnerability.

Reading Derrida: Language, identity and violence

37

To be a Franco-Maghrebian, one ‘like myself’, is not . . . a surfeit or richness of identities, attributes, or names. In the first place, it would rather betray a disorder of identity [trouble d’identit´e ]. (Derrida 1998: 14)

In other words, for Derrida being located at the intersection of two cultures does not represent a place of rest or abundance; rather it is a site at which the violence of the conflict between the cultures is experienced, resulting in an inner conflict which is irresolvable. We can surely learn from this perspective, so that the psychological and literal violence of the social, political and cultural context is fully countenanced in discussions of multiple linguistic identity. Fourth, the centrality of desire in Derrida’s account of his relationship to French strongly suggests that accounts of identity in language learning need to find a place for desire. While this is beginning to be explored within work on emotions and language learning (Pavlenko 2005, 2006) and in other areas of applied linguistics, particularly language and sexuality (Cameron and Kulick 2003), we need as a field to engage more deeply with the challenging and complex theories of desire that Derrida and Butler (e.g. Butler 1997) offer us. This will provide a more appropriate theoretical basis for work within a variety of areas in applied linguistics, including an opportunity to re-think the old chestnut of motivation in second language learning. Motivation as conceptualized within second language acquisition is socially theoretical in only a shallow way; the classic formulation for example of the distinction between instrumental and integrative motivation (Gardner and Lambert 1959) does not properly theorize the painfully conflicted intergroup context within which it was formulated, and discussions within that tradition have continued to be relatively asocial (McNamara 1997). Derrida reminds us how complex and conflicted the desire to be considered part of the polity of the Other may be. Furthermore, studies of integrative motivation do not typically conceptualise motivation as passionate, and as touching the deepest unconscious feelings of the individual, a perspective that is now being opened up in the work of Pavlenko, as stated above. Fifth, in recent work on the social and political functions of language tests (McNamara and Roever 2006) I have explored the notion of language tests as being tests of identity (McNamara 2005). The emblematic example here is the biblical story of the shibboleth, in which a particular ethnically marked pronunciation of a word was used as identifying the boundary between insider and outsider, with violent consequences for those identified as outsiders. The shibboleth is frequently used as a metaphor for the social/political function of tests as instruments of social control in situations of intergroup conflict and competition. Derrida’s own discussion of the shibboleth in texts on the contradictory relationship of the poet Paul Celan to his native German in the aftermath of

38

Tim McNamara

the Holocaust (Derrida 2005) stresses its paradoxical character: he speaks of “The terrifying ambiguity of the shibboleth, sign of belonging and threat of discrimination . . . ” (Derrida 2005: 27). This is helpful as a way of re-thinking the necessarily complex and contradictory impact of language tests, a priori neither necessarily good nor necessarily bad. Recent studies of the impact in different national settings of the OECD’s program of language testing of reading skills in the main national language (PISA) (McNamara in press; McNamara 2009) demonstrate the powerful impact of this test on the education systems in the countries concerned; nevertheless the impact is not always negative, and in some cases actually positive, and thus escapes or exceeds the intentions of the test’s architects. Language tests, we might argue, are Derridean undecidables. Finally, Derrida’s complex notions of identity as inherently contested and divided are relevant to debates on the politics of language, identity and citizenship in a time of globalization and transnational flows, particularly as research has shown how ideologies of national belonging are enforced through language tests for residence and citizenship (Extra et al. 2009; Hogan-Brun et al. 2009; McNamara and Shohamy 2008; Shohamy and McNamara 2009). In Derrida’s words: Our question is still identity. What is identity, this concept of which the transparent identity to itself is always dogmatically presupposed by so many debates on monoculturalism or multiculturalism, nationality, citizenship, and in general, belonging? (Derrida 1998: 14)

7. Conclusion This paper has given a reading of Derrida’s Monolingualism of the Other in order to argue for the relevance of his thought and in particular his work on language to concerns within applied linguistics. These concerns include studies of language learning, language maintenance and shift, language attitudes, the spread of English as a global language, language testing, and debates on language and citizenship, to name just a few. If applied linguistics is a field which “mediates between theory and practice” and “makes abstract ideas and research findings accessible and relevant to the real world” (Kaplan and Widdowson 1992: 76), then it will constantly discover new sources of theory which are relevant to its concerns. But the work of Derrida as one such potential source is not to be appropriated easily. I am referring here not simply to its great difficulty, or the fact that it lies not within social science, but the humanities. Rather, the challenge is

Reading Derrida: Language, identity and violence

39

that its notions are profoundly and deliberately destabilizing and disruptive, requiring of us an imagination and a tolerance of paradox that our very practically oriented field may find difficult to sustain – to its own great cost. University of Melbourne

Notes 1. I wish to acknowledge the contribution of my colleague in French at Melbourne, Anne Freadman, to the development of the argument of this paper. 2. One recent exception is Ramanathan (2008). 3. Particularly in Monolingualism of the Other (1998), “Circumfession” in Bennington and Derrida 1993, For WhatTomorrow (Derrida and Roudinesco 2004), Counterpath (Derrida and Malabou 2004), and in interviews (Derrida 1995, 2002). Further biographical information is presented, with Derrida’s cooperation, in Bennington and Derrida (1993) and Cixous (2004). 4. Accessing Derrida is particularly difficult for a non-philosopher. I offer the following account of my own personal experience as a person without any philosophical training, in the hope that it might prompt others to consider similar approaches. I became interested in reading Derrida during a sabbatical leave I had in 2003. I was keen to return to work on language and identity, and particularly language and racism, after a gap of a few years, and knew that I needed to become better informed about contemporary social theory and its view of the subject and of subjectivity. During some time spent at UC Berkeley in April 2003, I learned of a small, mostly invitational conference on racism at UC Irvine at which Derrida would be present, and saw and heard him there. The conference, which was attended only by about 40 people, and in which all sessions were in plenary, was intended to bring together those working in postcolonial theory and those working in deconstruction to explore their sometimes sharp differences. Derrida was discussant to a paper by the French philosopher Etienne Balibar, and he used the occasion to reflect on many of the papers that had been presented that day. He was clear and very constructive, though his remarks on his own work proved controversial. On my return to my university I was invited by some postgraduate students, with whom I had been sitting in on a seminar on postcolonialism and postmodernism, to join a reading group they had set up. We initially read Frantz Fanon’s beautiful and moving Black Skins, White Masks (Fanon 1967), and then decided we were most interested in reading Derrida. A group of five of us met for two hours each week in the five years following. We read Of Grammatology (Derrida 1976), Writing and Difference (Derrida 1978), some individual essays on Walter Benjamin and Lacan, Spectres of Marx (Derrida 1994) and other texts. We would read about half a dozen pages at a time, or less – all we could manage, none of us being philosophers; we would read it more or less line by line. Altogether, it was one of the most invigorating intellectual experiences of my whole career.

40

Tim McNamara

5. There is some evidence of a Judaeo-Berber community predating the arrival of the Arabs. 6. They were also influential for Althusser, who wrote a PhD on Hegel. 7. The extent of French anti-Semitism and cooperation with the Nazi authorities in Vichy France has been widely documented (Todorov 1996), most recently for example in Callil (2006). 8. The term Derrida uses, jouir de, suggests sexual pleasure. 9. This phrase reminds us that restoring the language is a work of desire, and this is reinforced later when he writes of “an impulse of love and aggression toward the body of any given language . . . ” (Derrida 1998: 66). 10. The book originated as a paper given at a conference in Louisiana on la Francophonie.

References Achebe, C. 1988. Hopes and impediments: Selected essays 1965–1987. London: Heinemann Bennington, G. & J. Derrida. 1993. Jacques Derrida. Geoffrey Bennington (trans.). Chicago: University of Chicago Press. Block, D. 2008. On the appropriateness of the metaphor of LOSS. In R. Rubdy & P. Tan (eds.), Trading language: Of global structures and local market places, 187–203. London: Continuum. Butler, J. 1990. Gender trouble: Feminism and the subversion of identity. New York & London: Routledge. Butler, J. 1993. Bodies that matter: On the discursive limits of sex. New York & London: Routledge. Butler, J. 1997. The psychic life of power: Theories in subjection. Stanford, CA: Stanford University Press. Butler, J. 1999. Subjects of desire: Hegelian reflections in twentieth-century France. New York: Columbia University Press. Butler, J. 2004. Jacques Derrida. London Review of Books 26(21). http://www.lrb.co. uk/v26/n21/butl02 .html (accessed 23 October 2006). Butler, J. 2005. Giving an account of oneself. New York: Fordham University Press. Callil, C. 2006. Bad faith: A forgotten history of family and fatherland. London: Jonathan Cape. Cameron, D. 2007. Language endangerment and verbal hygiene: History, morality and politics. In A. Duchˆene & M. Heller (eds.), Discourses of Endangerment, 268–285. London: Continuum. Cameron, D. & D. Kulick. 2003. Language and sexuality. Cambridge: Cambridge University Press. Cixous, H. 2004. Portrait of Jacques Derrida as a young Jewish saint, trans. Beverley Bie Brahic. New York: Columbia University Press.

Reading Derrida: Language, identity and violence

41

Derrida, J. 1976. Of grammatology. G. C. Spivak (trans.). Baltimore: Johns Hopkins University Press. Derrida, J. 1978. Writing and difference. Alan Bass (trans.). London & New York: Routledge. Derrida, J. 1981. Dissemination. Barbara Johnson (trans.). London & New York: Routledge. Derrida, J. 1988. Limited Inc. Evanston, IL: Northwestern University Press. Derrida, J. 1994. Specters of Marx: The state of the debt, the work of mourning, and the new International. Peggy Kamuf (trans.). New York & London: Routledge. Derrida, J. 1995. Points: Interviews 1974–1994. E. Weber (ed.). Peggy Kamuf (trans.). Stanford, CA: Stanford University Press. Derrida, J. 1998. Monolingualism of the Other; or, the prosthesis of origin. P. Mensah (trans.). Stanford, CA: Stanford University Press. Derrida, J. 2001. On cosmopolitanism and forgiveness. M. Dooley & M. Hughes (trans.). London: Routledge. Derrida, J. 2002. Negotiations: Interventions and interviews, 1971–2001. Elizabeth N. Rottenberg (ed.). Stanford, CA: Stanford University Press. Derrida, J. 2005. Sovereignties in question: The poetics of Paul Celan. T. Dutoit & O. Pasanen (eds.). New York: Fordham University Press. Derrida, J. & A. Dufourmantelle. 2000. Of hospitality. R. Bowlby (ed.). Stanford, CA: Stanford University Press. Derrida, J. & C. Malabou. 2004. Counterpath: Travelling with Jacques Derrida. David Wills (trans.). Stanford, CA: Stanford University Press. Derrida, J. & E. Roudinesco. 2004. For what tomorrow . . . : A dialogue. J. Fort (trans.). Stanford, CA: Stanford University Press. Extra, G., M. Spotti & P. Van Avermaet (eds.). 2009. Language testing, migration and citizenship: Cross-national perspectives. London: Continuum. Fanon, F. 1967. Black skin, white masks. C. L. Markmann (trans.). New York: Grove Press. Foucault, M. 1977. Discipline and punish: The birth of the prison. A. Sheridan (trans.). London: Allen Lane. Gardner, R. C. & W. E. Lambert. 1959. Motivational variables in second-language acquisition. Canadian Journal of Psychology 13. 266–272. Hegel, G. W. F. 1977. Phenomenology of spirit. A. V. Miller (trans.). Oxford: Clarendon Press. Hogan-Brun, G., C. Mar-Molinero & P. Stevenson (eds.). 2009. Discourses on language and integration: Critical perspectives on language testing regimes in Europe. Amsterdam: John Benjamins. Hutton, C. 1999. Linguistics and the Third Reich. Mother-tongue fascism, race and the science of language. London & New York: Routledge. Kaplan, R. B. & H. G. Widdowson. 1992. Applied linguistics. In W. Bright (ed.), International encyclopedia of linguistics, vol. 1, 76–80. New York & Oxford: Oxford University Press.

42

Tim McNamara

McNamara, T. 1997. What do we mean by social identity? Competing frameworks, competing discourses. TESOL Quarterly 31(3). 561–567. McNamara, T. 2005. 21st century Shibboleth: Language tests, identity and intergroup conflict. Language Policy 4(4). 1–20. McNamara, T. 2009. Are cognitive and social perspectives on assessment incommensurable? The case of comparative assessment frameworks and language education. Plenary address, 1st Combined Conference, The Applied Linguistics Associations of New Zealand and Australia, Auckland, December. McNamara, T. In press . Measuring deficit. In C. N. Candlin & J. Crichton (eds.), Discourses of deficit. London: Palgrave Macmillan. McNamara, T. & C. Roever. 2006. Language testing: The social dimension. Malden, MA & Oxford: Blackwell. McNamara, T. & E. Shohamy. 2008. Language tests and human rights. International Journal of Applied Linguistics 18(1). 89–95. Pavlenko, A. 2005. Emotions and multilingualism. Cambridge: Cambridge University Press. Pavlenko, A. (ed.). 2006. Bilingual minds: Emotional experience, expression, and representation. Clevedon, UK: Multilingual Matters. Pennycook, A. 1995. The cultural politics of English as an international language. London: Longman. Pennycook, A. 2001. Critical applied linguistics: A critical introduction. London: Routledge. Phillipson, R. 1992. Linguistic imperialism. Oxford: Oxford University Press. Ramanathan, V. 2008. Applied linguistics redux: A Derridean analysis of Alzheimer life histories. Applied Linguistics 29(1). 1–23. Seidlhofer, B. 2005. Standard future or half-baked quackery? Descriptive and pedagogic bearings on the globalisation of English. In C. Gnutzmann & F. Intemann (eds.), Globalisation and the English language classroom, 155–169. Tübingen: Narr. Seidlhofer, B. 2006. English as a lingua franca in the expanding circle: What it isn’t. In R. Rubdy & M. Saraceni (eds.), English in the world: Global rules, global roles, 40–50. London: Continuum. Seidlhofer, B. 2009. Common ground and different realities: World Englishes and English as a lingua franca. World Englishes 28(2) [Special issue: Symposium, “Perspectives on Lingua Franca”]. 236–245. Seidlhofer, B., A. Breiteneder & M-L. Pitzl. 2006. English as a lingua franca in Europe. Annual Review of Applied Linguistics 26. 1–34. Shohamy, E. & T. McNamara. 2009. Language tests for citizenship, immigration and asylum. Language Assessment Quarterly 6(1). 1–5. Todorov, T. 1996. A French tragedy: Scenes of civil war, summer 1944, trans. Mary Byrd Kelly, edited and annotated by Richard J. Golsan. Hanover, NH: University Press of New England.

Reading Derrida: Language, identity and violence

43

Wiley, T. G. 2002. Heinz Kloss revisited: National Socialist ideologue or champion of language-minority rights? International Journal of Sociology of Language 154. 83– 97. Tim McNamara is Professor in the School of Languages and Linguistics at the University of Melbourne. His interests include language testing, language and identity, poststructuralism, the history of applied linguistics, and languages for specific purposes. His recent publications include Language testing: The social dimension (Blackwell, 2006) and Language testing (OUP, 2000).

Engaging with human sociality: Thoughts on communication and embodiment* DAVID BLOCK

Abstract In introducing the term ‘human sociality’into the social sciences lexicon, scholars such as Levinson and Enfield attempt to bring together several emergent tendencies. First, what they propose is part of a general move in the study of communication away from a dominant focus on language (phonology, syntax, morphology and lexis). For some time now, the study of communication in terms of language has been complemented by a growing interest in pragmatics, i.e. the study of how individuals accomplish goals drawing on the communicative resources at their disposal in ways deemed legitimate and appropriate in sociocultural contexts. More recently, further additions to language have come to the fore in the study of communication, most notably the growing interest in the communicative resources deployed in different communicative contexts. In particular, there has been a good deal of work on how semiotic resources, such as facial expression, gaze, stance, position and so on, come together as semantic assemblages in ongoing communication. In addition, human sociality is about how human activity, and indeed the very being of human beings, is biologically grounded and cognitively mediated. Thus, in any understanding of the social, it is worth bearing in mind (1) how communicative resources are embodied, i.e. they are the evolving products of the human body’s interactions in physical and social spaces over a lifetime, and (2) how a theory of mind, i.e. individuals’ theories about how other minds work, is essential to the ongoing cooperative activity which affords the continued existence and survival of human beings. In this paper, I engage with issues like the ones just listed, situated under the general umbrella of human sociality. Keywords: Human sociality; communication; multimodality; embodiment; neurobiology.

46

David Block

1. Introduction: Considering human sociality At the heart of the uniquely human way of life is our peculiarly intense, mentally mediated, and highly structured way of interacting with one another. This rests on participation in a common mental world, a world in which we have detailed expectations about what we share and do not share in the way of knowledge, intentions, and motivations. That itself relies both on communication (linguistic and otherwise) and on a level of cooperation unique in the animal world. This mode of cooperative, mentally-mediated interaction enables the accumulation of cultural capital and historical emergence of cultures. By inheriting a world of social organizations and values, individuals are released from reinventing the wheel. In turn, cultural capital shapes the style of interaction in local social groups, hiding shared commonalities behind the veil of distinct languages, cultural styles and different, social organizations. (Enfield and Levinson 2006: 1)

Thus begins Nick Enfield and Stephen Levinson’s introduction to their edited volume Roots of Human Sociality, an ambitious book published in 2006. I say ambitious for two primary reasons: first, because of its broadly humanistic and universalist approach to the study of human beings and their activity at a time when a generalized devotion to the local and particular has become prevalent in the social sciences (even if recently it is showing signs of intellectual and epistemological fatigue); and second, because it is an attempt to link what in the past have been fairly disparate and indeed at times antagonistic disciplines in the human sciences, such as developmental psychology and sociology, or biological anthropology and sociocultural anthropology. I come to this ambitious project as a newcomer and therefore in this paper my aim is to engage with what I think I understand about it, which is, for the moment, still relatively inchoate and far from being fully developed (but see Enfield [2009] for a further development of the overall concept). I see human sociality as above all, the bringing together of a good number of human sciences disciplines in an attempt to develop richer and more nuanced understandings of human behavior and activity. In the book introduction previously cited, Enfield and Levinson discuss several well developed areas of inquiry that they believe contribute to understandings of human sociality. First, they see as foundational to the enterprise the notion that human beings develop from early in life a Theory of Mind, which, put succinctly, is an understanding of how other minds work. Were it not for this Theory of Mind, individuals would not know where to start in their attempts to communicate with others. For example, small children must think that pointing at an object will, in some sense, be interpreted by an interlocutor as indicating the location of the object pointed at or even that the child desires to hold the object. For Enfield and Levinson, there is a need to accept Theory of Mind as foundational to communication; otherwise,

Engaging with human sociality

47

all communication would have to be seen to occur against an empty backdrop of tabula rasa. Emerging from the Theory of Mind is an acceptance of the notion that communication norms exist, such as those captured in Gricean (Grice 1975) and Gricean-inspired pragmatics (e.g. Levinson 2000) and maxims of communication around the principle of cooperation. As with the Theory of Mind, the basic idea here is that human beings could get very little communicative work done if there were not shared socio-histories and repeated experiences serving as a backdrop to this communicative activity. With interaction at its core, it is not surprizing that micro-analysis is a big part of the study of human sociality. And here scholars ranging from Emmanuel Schegloff (e.g. 2006) in conversation analysis to Erving Goffman (e.g. 1981) and his more socially informed dissection of interaction find a home in a broad church. Conversation analysis examines the architecture of conversation at the surface level – openings and closings, turn taking, repair and so on – but not in terms of deeper cognition. And this leads Enfield and Levinson to acknowledge that its consequent incorporation into the study of human sociality might be seen to be in conflict with the more psychologically grounded notions of Theory of Mind. However, the common ground here is an attention to detail and micro-analysis. At the same time, by drawing on scholars such as Goffman, and notions such as footing and the animator/ author/principle triad, human sociality researchers manifest a more sociologically grounded approach to interaction, which, in turn, may be seen to be in conflict with both Theory of Mind and conversation analysis. However, once again the combination can be made to work if one manages to see different disciplines making up for analytical gaps in others. I have always wondered about the maxim in conversation analysis of sticking to the data at hand and only the data hand, while eschewing speculations about more abstract notions such as social class, unless such abstract notions are explicitly mentioned by participants in an interaction being examined. Equally, I have noted how sociologists all too often make reference to psychological notions, such as motives and intentions, without really explaining what their theory of cognition is or exactly how they see the workings of the human mind. Perhaps human sociality is a move toward recognizing and working through such fissures and gaps. In any case, the fact that human sociality is also an attempt to bring biological and sociocultural schools of anthropology together under one tent is certainly enough to get anyone’s attention. This, of course, is not an easy task. On the one hand, we have a consideration of Robin Dunbar’s (1996) evolutionary argument that language practices such as gossiping parallel the grooming activities displayed by primates. On the other hand, we have work in language socialization in which communication practices are deemed to be embedded in

48

David Block

and emergent from the particularities of local cultures (e.g. Brice-Heath 1983). In conclusion, the human sociality project is broad, too broad in fact for any one author to go through every single constituent part of its edifice as matters stand at present. Indeed, my musings thus far have been partial to an extreme. However, they serve as a backdrop to this paper, which is to present my incursion into human sociality given my background and interests. I aim to proceed as follows. First, I align myself with a general move in the study of communication, away from an exclusive (or in any case dominant) focus on language, where language has traditionally been understood to be composed of phonology, syntax, morphology and lexis. For some time now, the study of communication in terms of language has been complemented by a growing interest in pragmatics; i.e. the study of how individuals accomplish goals drawing on the communicative resources at their disposal in ways deemed legitimate and appropriate in sociocultural contexts. More recently, further additions to language have come to the fore in the study of communication, most notably the growing interest in what communicative resources might be in different communicative contexts. In particular, there has been a good deal of work on how semiotic resources, such as facial expression, gaze, stance, position and so on, come together as semantic assemblages in ongoing communication and how this communication is deemed to be embodied in individuals. This move from communication as exclusively about language to communication as multimodal and embodied is one pillar of what I will discuss in this paper. At the same time, I am intrigued in my readings about human sociality by the idea that human activity, and indeed the very being of human beings, is biologically grounded. Thus, in any understanding of the communicative resources and embodiment as social, it is worth bearing in mind how communicative resources are physically embodied; i.e. they are the evolving products of the human body’s interactions in physical and social spaces over a lifetime. In this paper, I start with this former notion of communicative resources and social embodiment before moving to consider the issue of physical embodiment. In each case I can only work from where I have been and where I am in intellectual terms, which of course narrows my particular way of dealing with these issues.

2. The move from communication as exclusively about language to communication as multimodal and embodied In his oft-cited work on what he termed a ‘socially constituted sociolinguistics’ and the development of an ‘ethnography of communication’, Dell Hymes (1974)

Engaging with human sociality

49

includes some suggestive ideas regarding the limits of what we mean by language and linguistics. With reference to the ethnography of communication, he argues that apart from addressing instances of communication as events with various kinds of participants, settings, forms or genres of communication, attitudes and contexts, ethnographers need to be attentive to: . . . the various available channels, and their modes of use, speaking, writing, printing, drumming, blowing, whistling, singing, face and body motion as visually perceived, smelling, tasting,, and tactile sensation . . . [and] the various codes shared by various participants, linguistic, paralinguistic, kinesic, musical, interpretive, interactional, and other . . . . (Hymes 1974: 10)

By including channels and codes, and defining them in this way, Hymes showed himself to be prescient as regards how view of language communication would evolve from the 1980s onwards. Specifically, it is possible to establish a link between Hymes’s ideas in this regard and the rise of multimodality in the work of Gunther Kress, Leo van Leeuwen, Carey Jewitt and many others over the past two decades. Jewitt defines multimodality in manner consistent with Hymes’s views on communication some four decades ago as follows: Broadly speaking multimodality describes approaches that understand communication and representation to be more than about language, and in which language is seen as one form of communication (or a Mode – this refers to an organized set of semiotic resources for making meaning with) among other modes such as image, gesture, gaze, posture, and so on. These modes are understood as intimately connected, enmeshed through the complexity of interaction, representation and communication. (Jewitt 2009: 14)

However, Hynes was not alone in presaging the rise of multimodality as a generalized way of understanding communication. There were at least two other scholars – neither of whom had a background in linguistics and neither of whom worked in linguistics – who nonetheless had what were regarded by many linguists at the time as even more unorthodox views about language than Hymes, Erving Goffman and Pierre Bourdieu. Like Hymes, they incorporated into their comments about language the importance of other aspects of communication besides syntax, morphology and lexis. In the introduction of his oft-cited book Forms of Talk, Goffman writes the following about face-to-face communication: Everyone knows that when individuals in the presence of others respond to events their glances, looks, and postural shifts carry all kinds of implication and meaning. When in these settings words are spoken, then tone of voice, manner of uptake, restarts, and the variously positioned pauses similarly qualify. As does manner of listening. Every adult is wonderfully accomplished in producing all of these

50

David Block effects, and wonderfully perceptive in catching their significance when performed by accessible others. (Goffman 1981: 1–2)

On the traditional approaches to observed speech, Goffman (1981) states that it is primarily about analysing how the floor passes from one interlocutor to the other and thus how speaker-hearer roles are interchanged. About the notions of speaker and hearer, he has the following to say: . . . the terms ‘speaker’and ‘hearer’imply that sound alone is at issue, when, in fact, it is obvious that sight is organizationally very significant too, sometimes even touch. In the management of turn-taking, in the assessment of reception through visual back-channel cues, in paralinguistic function of gesticulation, in the synchrony of gaze shift, in provision of evidence of attention (as in middle-distance look), in the assessment of engrossment through evidence of side-involvements and facial expression – in all of these ways it is apparent that sight is crucial, both for the speaker, and for the hearer. For the effective conduct of talk, speaker and hearer had best be in a position to watch each other. (Goffman 1981: 129–130)

Elsewhere, Bourdieu has expressed his views about language along similar lines. A starting point is his continuous discussion of habitus, defined as: . . . systems of durable, transposable dispositions, structured structures predisposed to function as structuring structures, that is, as principles of the generation and structuring of practices and representations which can be objectively ‘regulated’ and ‘regular’ without in any way being the product of obedience to rules, objectively adapted to their goals without presupposing a conscious aiming at ends or an express mastery of the operations necessary to attain them, and, being all this, collectively orchestrated without being the product of the orchestrating action of a conductor. (Bourdieu 1977: 72)

One point that Bourdieu makes is that habitus, as predisposed ‘dispositions’ and ‘structured structures’, is embodied in what he calls ‘body hexis’, briefly defined as “a political mythology realized, em-bodied, turned into a permanent disposition, a durable way of standing, speaking, walking and thereby of feeling and thinking” (Bourdieu 1990: 69–70). This embodied disposition may be understood not only as a phenomenon that is deep, but also one that is embedded temporally in personal history, emergent in the individual trajectories of the lifecourse. Thus, Bourdieu writes the following with reference to the interrelationship between body hexis and childhood: The child imitates not ‘models’ but other people’s actions. Body hexis speaks directly to the motor function, in the form of a pattern of postures that is both individual and systematic, because linked to a whole system of techniques involving the body and tools, and charged with a host of social meanings and values: in all societies, children are particularly attentive to gestures and postures which,

Engaging with human sociality

51

in their eyes, express everything that goes to make an accomplished adult – a way of walking, a tilt of the head, facial expressions, ways of sitting and of using implements, always associated with a tone of voice, a style of speech, and . . . a certain subjective experience. (Bourdieu 1977: 87)

In his work, Bourdieu writes not of appropriacy but a ‘sense of acceptability’, the domain of the powerful in society (and indeed, the powerful in different fields, or spaces of social activity), those with the right symbolic capital, which allows them to move within the realms of valued practices and appropriate states of being. Relevant to my interest in communication as involving language, but not being exclusively about language, is Bourdieu’s observation about ‘the sense of acceptability which orients linguistic practices’. This ‘sense’ is said to be “inscribed in the most deep-rooted of bodily dispositions” (Bourdieu 1991: 86) in that it is “the whole body which responds by its posture, but also by its inner reactions, or more specifically, the articulatory ones, to the tension of the market” (Bourdieu 1991: 86). In short, for Bourdieu, “[l]anguage is a body technique . . . ” (Bourdieu 1991: 86).

3. The somatic appraisal system as embodied base for multimodal communication: Two examples Some years later, Charles Goodwin and Marjorie Goodwin (e.g. 2004) adopt Goffman’s notions of footing and frame analysis and echo Bourdieu’s views on body hexis and embodiment in their work on the analysis of participation patterns in a variety of communication contexts. Most importantly, they seem to have advanced on the notion that all face-to-face communication is multisemiotic or multimodal in nature, as the following comment suggests: The interplay between the semiotic resources provides in language on the one hand, and tools, documents, and artefacts on the other constitute a most important future direction for the analysis of participation. However, their multi-modal framework should not be seen as something new but instead recognition of the rich contextual configurations created by the availability of multiple semiotic resources that has always characterized human interaction. (Goodwin and Goodwin 2004: 239)

Drawing on this view of communication, Charles Goodwin (2007) looks at what he terms ‘embodied participation frameworks’; that is, how stance and affect are factors in ongoing communication in which language is a central feature. He examines how a father helping his daughter with her maths homework, makes use of his physical position vis-`a-vis her, points to particular parts of the prob-

52

David Block

lem being discussed and uses his gaze back and forth between her face and the notebook to guide talk about the problem that they are both focusing on. All three of these spatially shaped and shaping actions carry an affective load, as proximity, along with tone of voice serve to show emotions ranging from sympathy to anger. And of course, everything I have just said about the father is applicable to the daughter and her role in the interaction, that is, stance, proximity, tone of voice and so on all come together to configure her contribution to the effect of interaction. Goodwin makes the point that this use of combinations of multimodal resources in activity to create the effect of interaction occurs across a range of day-to-day interactions. As we see in Goodwin’s work, real life examples of how ‘language as a body technique’ are not hard to come by, examples that show how communication is neither just about language nor even most centrally about language, and how an array of semiotic resources help to configure, shape and indeed provide for language the essential backdrop by which it actually makes sense. Another example, in a very different domain of human activity, is the video of the 2004 hit single “Over and over” by Hip Hop artist Nelly and Country Western artist Tim McGraw. The video begins with two men waking up in the morning in their respective bedrooms. Nelly’s room has a modern urban feel against white, black and light blue tones (not only the walls are painted these colours but Nelly is wearing a white tank top and blue bandana). Meanwhile, McGraw’s room is decorated in a country motif against a backdrop of brown tones with exposed wooden logs and breams and canopy bed also done in wood. He is wearing a brown t-shirt which matches the colour of his bed sheets and walls. As other parts of the respective homes are revealed, even the exteriors are notably different as Nelly stands singing in front of what looks to be a block of flats in a city environment while McGraw has a country home as a backdrop. This split between urban hip hop and county styles continues throughout the video as contrasts are drawn between the way that the two men dress, down to their belt buckles and footwear: Nelly wears an urban bling over-sized belt buckle while McGraw’s is country bling and over-sized; Nelly wears fashionable black and white trainers while McGraw wear cowboy books. The vehicles that the two men drive are very different: Nelly’s pimped ride vs. McGraw’s oversized SUV. In addition, the two men move in different ways, including their walks (Nelly’s variation on a pimp roll vs. McGraw’s stiff country walk), facial expressions whilst singing (most notably McGraw’s very country wince) and their movements (Nelly’s hip-hop style hand movements vs. McGraw’s relatively motionless style). Their respective inflections on English while singing are also very different with Nelly producing an African American inflection and McGraw producing what might be termed a Southern White inflection. Finally,

Engaging with human sociality

53

different demographic and geographical backdrops are provided for the two men engaging in the same activities, such as driving, buying gas and singing autographs: for Nelly, African American and urban; for McGraw, White and rural. Thus, in the video the viewer is presented with two lives organized around what appear to be two separate ensembles of semiotic resources, which serve to frame the exact same words in English sung by the two men. Both thoroughly American, Nelly and McGraw live semiotically separate lives. In my view, the video can be seen alternatively as an indictment on the historical and current racial and cultural segregation of American society, or it can be seen as celebration of the notion that ‘deep down we are all the same’, although on the surface different. I would err towards the former interpretation, although this requires a consideration of how multimodality is historically situated. Though pursuing this theme in detail is interesting, more relevant to this paper is an examination of how multimodality of the type displayed by Nelly and Tim McGraw is deeply embodied as opposed to just existing superficially on show in day-to-day activities. Another example of embodiment, and one more language centred than the previous one, is how a significant part of second language learning is always set against backdrop of multimodal ensembles in which the language being learned is embedded. This is perhaps an obvious point to make, but it seems to be one lost on many second language learning researchers. To illustrate what I mean, I consider two expressions in Catalan which I first encountered naturalistically (in fact, it was probably a good deal of time before I actually saw them written anywhere) over 30 years ago. They are: D´eu n’hi do! Me cago en D´eu! The first expression is translatable into English as something along the lines of ‘I’ll say!’ or as a way of expressing agreement. My understanding of its use has always been that it shows the speaker’s agreement with someone describing a state of affairs, a situation or events and conveys emphasis and/or intensity. This view is supported by some linguists (e.g. Cuenca 2002; Sancho 2003), but contested by others (Mayol 2007; Mayol and Castroviejo 2008), who see it as often redundant and therefore lacking as a conveyer of emphasis and intensity. Translated literally, the expression is a grammaticalization of ‘May God give it to him/her’ (Mayol 2007). In my experience, the expression is sometimes accompanied by a gesture, which involves lifting the forearm slightly and then wagging the hand in a snapping manner as one simultaneously pronounces the words. It is as if the speaker were throwing the words out in parallel to the

54

David Block

wagging hand. The words, therefore, are very overtly embodied in gesture and inflection on speech. The other expression is one that I first experienced in a small bar in Barcelona, again some 30 years ago. I refer here to a small smoke-filled establishment where people either spend as little time as possible over a caf´e (an expresso) or spend hours talking as they play cards or watch a football match. In the latter case the people in question are generally male and quite often of retirement age. In any case, my experience was one of overhearing a conversation taking place among a group of men who were playing cards around a wooden table cluttered with beer bottles and glasses, ashtrays and burning cigarettes. In the midst of a cloud of smoke, I spotted a man, cigarette hanging from his mouth, empathically throwing a card down on the table, as he uttered Me cago en D´eu! (literally, ‘I shit on God!’). Now I might wonder what shitting on God has to do with playing a trump card, but the scene makes this clear. It is embodied emotion spat out with the inflection on words and the throwing of the card on the table. Meanwhile, the expression utterer’s spatial situation vis-`a-vis others at the table is also part of the ensemble of semiotic resources making up the communicative event. In my brief descriptions of these two examples, and in particular the latter case, I am drawing on Goffman’s understandings of face to face interaction, Bourdieu’s notions of habitus and body hexis and finally, Goodwin and Goodwin’s ideas about ‘[t]he interplay between the semiotic resources . . . language on the one hand, and tools, documents, and artefacts on the other . . . that has always characterized human interaction’ (Goodwin and Goodwin 2004: 239). There is no prospect of a well-developed understanding of the two expressions without a consideration of how they form part of ongoing interactions, involving embodied subjects and mediated multimodally. However, as with the previous examples discussed in this section, we are speculating from a social perspective about observed phenomena which might be framed more person-internal, more affective, more psychologically and even more biologically. In the next section, I explore an additional way of framing embodiment in discussions of communication and practice.

4. Affect and its neurobiological bases I begin my consideration of this issue by going back in time over a decade ago, when I was reading the work of John Schumann, in particular his (1997) monograph on affect in second language learning, Neurobiology of affect in language. This reading was completed with forays into some of the source material for Schumann’s speculations as well as other work which was coming

Engaging with human sociality

55

out at the time (e.g. Dalgleish and Power 1999; Damasio 1994, 1999; Lamb 1999; LeDoux 1996; Ortony, Clore and Collins 1999; Scherer 1988). More recently Schumann and his associates (Schumann, Crowell, Jones and Lee 2004) have published a volume examining the interrelationships between neurobiology and language aptitude, motivation, memory and attention. Through all of this work over the years, Schumann has rooted his discussion of learning in the ‘biological notion of value’, defined as follows: . . . a bias that leads an organism to certain preferences and enables it to choose among alternatives . . . Some aspects of value are so important that evolution has selected for them and they have become innate. Value is the basis for all activity; we perceive, move, cognize, and feel on the basis of value. (Schumann and Wood 2004: 24)

A value system is subsequently dependent on the development of three fundamental motivational systems, that is, systems that impel individuals to act in particular ways. The first of these systems is the homeostatic system which regulates the human organism, maintaining stable and consistent functioning. This system allows the human organism to sustain itself by monitoring and controlling such basic biological functions as breathing, eating, keeping warm or cool, maintaining the heart rate, procreating and caring for kin. The second system is called the sociostatic system which motivates human beings to associate with fellow human beings in their quest for feelings of attachment and group affiliation. As noted by Enfield and Levinson (2006), there is something intrinsically human about the need to associate with and engage in activities with fellow human beings. In the life of human beings, the same individuals – parents, family members, family friends – generally guarantee the provision of stimuli which cater to both sets of general needs and the relevant systems, homeostatic and sociostatic. For this reason, human beings develop them interdependently. In addition, through interaction with the environment, driven by these two systems, individuals acquire their individualized and idiosyncratic somatic system of preferences and aversions and likes and dislikes, the key element in the overall value system discussed by Schumann. This somatic system acts as an appraisal system for stimuli encountered as life unfolds. For example, if an individual initiates the study of an additional language in a formal classroom setting, the somatic system will act as a prism through which experiences are lived: the language being studied, learning activities engaged in, teachers, classmates, pedagogical materials and other features of formal learning environments. The idea here is that any instance of human behavior, taking place at a given point and time, should be understood as a set of locally and temporarily contin-

56

David Block

gent conditions interacting with traces of memories of previous experiences – the goals, interests and beliefs of the individual – which come together for the occasion. The reality of cognition then is a constant back and forth interaction of the past with the present, of the remembered with the understood, and ultimately of the accommodation of the old in order for assimilation of the new to take place. Over the years, Klaus Scherer has developed a model that sees this process in terms of dimensions of appraisal which may be seen to make up the somatic system. These are listed and glossed in Table 1. Table 1. Dimensions of appraisal in a value system based on Scherer (1988, 2007) Appraisal of 1. NOVELTY 2. INTRINSIC PLEASANTNESS

Gloss The extent to which the internal or external stimulus event contains novel or unexpected patterns. Whether the stimulus event is pleasant or unpleasant. Whether approach or avoidance is in order.

3. GOAL/NEED SIGNIFICANCE

The extent to which the stimulus event is relevant to the individual’s needs and goals. How expected the stimulus event is at this stage and time in the individual’s goal/plan sequence. The extent to which the stimulus event is conducive to satisfying my needs and goals. How urgently a response is required.

4. COPING POTENTIAL

The cause of the stimulus event. If the individual can control the stimulus event or its consequences. If the individual has the power to avoid the outcome of the stimulus event through fight or flight. If the individual can adjust to the outcome through internal restructuring.

5. NORM/SELF COMPATIBILITY

How the stimulus event conforms to external socially-determined cultural norms and the expectations of others. How the stimulus event conforms internally with cultural norms and expectations.

Scherer is not saying that individuals actually go through these dimensions as a checklist each time they encounter a new situation; rather, these are meant to be aspects of any new stimulus against which the individual might elaborate an evaluation. The first dimension, novelty, is in simple terms about how new and unexpected a particular stimulus is to an individual, based on his/her past experience. The second dimension is more clearly an affective evaluation (is

Engaging with human sociality

57

this stimulus pleasant or unpleasant for the individual?) along with a response to either engage with it or not. The third and fourth dimensions relate to Scherer’s and others’ theories of motivation, respectively, as revolving around goal and need based activity, as mediated by the individual’s awareness of his/her ability to control any progression of activities towards the satisfaction of a goal or a need. The fifth and final dimension is about the interaction between a need to conform to social norms and the expectations of others and the need to be true to oneself, to do only those things which are consistent with one’s self image. Discussing value and appraisal in terms of the homeostatic system, the sociostatic system and the somatic system could easily remain at the level of abstract modelling of speculation about what goes on in the minds of individuals as they encounter new experiences and develop over time. To put the neurobiological into the equation requires a theory of the hardware of human appraisal therefore making the case that the source of social regulation is at least in part biologically based. Goleman (1995) makes the following statement about the relationship between parts of the brain – the amygdala and the neo-cortex – and the interaction of what he calls emotional intelligence and more standard rational intelligence: . . . the pre-frontal amygdala circuit is a crucial doorway to the repository of the likes and dislikes we acquire over the course of a lifetime. Cut off from emotional memory in the amygdala, whatever the neocortex mulls over no longer triggers the emotional reactions that have been associated with it in the past – everything takes on gray neutrality. A stimulus, be it a favorite pet or a detested acquaintance, no longer triggers either attraction or aversion; these patients have ‘forgotten’ all such emotional lessons because they no longer have access to where they are stored in the amygdala. (Goleman 1995: 28)

The brain matter discussed by Goleman is in Figure 1.

5. Affect and its neurobiological bases: An example A well known vivid example of how the neurobiological links with the social behavior comes from Damasio (1994). Damasio describes in detail the intriguing case of Phineas Gage, a railroad construction foreman in New England in the mid-19th century. Gage was a highly qualified and respected specialist in engineering explosions to destroy rock encountered in the path of planned railroad lines. A part of this job involved the use of a tamping iron, an iron rod one meter in length, to pound the sand poured over gunpowder in order to ensure that an explosion moved inwards to destroy the rock as opposed to scattering away from

58

David Block

Figure 1. The brain (This figure is taken from: http://upload.wikimedia.org/wikipedia/ commons/b/bc/Constudoverbrain.png [accessed 1 June 2009])

the rock. One day in 1948, Gage was distracted in the middle of this process and accidently tamped the gunpowder before sand had been poured over it. The subsequent explosion drove his tamping iron up through his left cheek and behind his eye cavity, before exiting at the top of his head just behind his forehead. Gage survived this traumatic event and his wounds eventually healed. However, his personality seemed to change after the accident. Whereas previously Gage was considered a cool-headed competent professional and a shrewd businessman, after the accident he manifested extremely inconsistent and instable behavior, including capricious changes of opinion, fits of anger and an overall lack of discipline. He eventually alienated all of those with whom he had worked as a construction foreman and spent most of what remained of his life travelling and working in jobs which ranged from horseman to circus sideshow freak, before dying some thirteen years after the accident of what Damasio believes was a

Engaging with human sociality

59

form of epilepsy. For Damasio, the lesson to be learned from Gage’s tragic story is as follows: Gage’s example indicated that something in the brain was concerned specifically with unique human properties, among them the ability to anticipate the future and plan accordingly within a complex social environment; the sense of responsibility toward the self and others; and the ability to orchestrate one’s survival deliberately, at the command of one’s free will. (Damasio 1994: 10)

It is not my intention here to explore in depth the physical location and basis for the appraisal system which Schumann describes. Indeed, I have only gone into detail here about the evidence for a neurobiological base for affect, value and appraisal to make two important points. First, such detail is necessary if I am to portray in a fair way the approach adopted by Schumann, whereby human behavior, specifically language learning, is seen to be affectively mediated and the affect that mediates is seen to be neurobiologically based and indeed, physically situated in particular part of the brain. Second, I offer this discussion as a probe of the limits of the human sociality, of what might be allowed into the broad church, which aims to explore human interaction and communication in as much detail as possible. Is there space for neurobiology in the project? I ask because I suspect that to pursue this angle is perhaps to move a little too far away from what is at the centre of human sociality; i.e. interactions with fellow human beings and the physical and social environment. So, rather than continue this direction, I propose to focus on what I do think is most interesting and relevant to human sociality in Schumann’s model, the embodied somatic appraisal system which I mentioned previously.

6. The embodied somatic appraisal system: An example Given its resonances with Bourdieu’s notions of habitus, body hexis and a “sense [which is] inscribed in the most deep-rooted of bodily dispositions” (Bourdieu 1991: 86), the embodied somatic appraisal system shows a degee of explanatory potential as regards how individuals engage with and make sense of the various stimuli which they come into and out of contact in the course a typical day. To make this point, I apply this model to a small part of an interview I carried out with two Spanish women, Elena and Almudena, some ten years ago. At the time of the interview, the two women who were doing a PGCE in MFL (Post Graduate Certificate of Education in Modern Foreign Languages) at a university in London while participating in an exploratory study that I was carrying out on how foreign nationals on PGCE MFL courses adapted to various aspects of

60

David Block

MFL teaching in Britain (see Block [2001] for an account). In this part of the interview, Elena and Almudena are talking about motivational posters that they have encountered in the schools where they have done teaching practice. The exchange looks as follows (see Appendix for transcription symbols): Original in Spanish E: ahora hay posters! / {said with false earnestness} D: ¿de que clase? / A: motivational posters/ {said in dramatic tone} E: posters como (.5) the man hanging off a cliff = {extending arms upwards as if hanging} A: =no dos delfines!= {making a sweeping motion with her right hand} E: =y una frase que dice you can do it= {speaking hurriedly with false earnestness} A: =no group work leads to success / o algo as´ı / {speaking hurriedly with false earnestness} E: los hacen empresas [que A: [me guard´e uno para enviar a mis amigos [en Espa˜na] /

English translation E: now there are posters! / {said with false earnestness} D: what kind? / A: motivational posters/ {said in dramatic tone} E: posters like (.5) the man hanging off a cliff = {extending arms upwards as if hanging} A: = no two dolphins!= {making a sweeping motion with her right hand} E: =and a sentence that says you can do it= {speaking hurriedly with false earnestness} A: =no group work leads to success / or something like that / {speaking hurriedly with false earnestness} E: they’re done by companies [that A: [I kept one to send to my friends [in Spain] /

(A and E, 28 September 1999)

In this transcription, I have attempted to reproduce not only what was said but how it was said and inflected. I hope to have conveyed the high level of sarcasm and indeed cynicism coming from the two women as they almost gleefully ridicule the use of motivational posters in school settings. Motivational posters have become popular in recent years in the business world and progressively in school settings. They generally present a picture of someone overcoming adversity (such as the man hanging from a cliff, which Elena refers to) or a pleasant scene in nature (such as dolphins swimming in the sea, which Almudena refers to). However, they have also become the object of some derision as witnessed by the growing number of websites which are devoted to ridiculing them as naive, corny and silly. An example of one of the motivational posters mentioned in the exchange above – a man hanging from a cliff – is reproduced in Figure 2. As obvious ‘doubters’ as regards this practice, Elena and Almudena not only express in words their disdain but also they embody it with their voice

Engaging with human sociality

61

Figure 2. A man hanging from a cliff (http://www.allposters.com/-sp/PerseveranceCliffhanger-Posters i113373 .htm [accessed 1 June 2009])

inflections and body language, via dramatic false earnest tones and exaggerated body movements (Elena imitating the man hanging from a cliff and Almudena’s sweeping hand movement when talking about dolphins). The code switching into English might be interpreted as further distancing from the posters as the two women do not bother to translate into Spanish the term ‘motivational’ or, with the exception of Almudena’s dos delfines the slogans on the posters (‘the man hanging off a cliff’; ‘you can do it’; ‘group work leads to success’). However, it should be noted that in all of their interviews there was a tendency to code switch to English when what was being said was a word or phrase encountered through the PGCE course and other teaching-related activity. In order to understand why Elena and Almudena adopted this sarcastic and cynical position when talking about the motivational posters, it might be useful to speculate that the kind of value/appraisal systems and professionals and personal habituses that the two women have developed in their lifetimes do not predispose them to engage with such practices in a positive way. The posters are presented as something of a novelty for them, as regards what they would expect to find in the world of teachers, although they do seem familiar with the genre. The issue is, then, that their encounter with these posters in schools is a stimulus event that it is both novel and unexpected. Elena and Almudena make clear that they expect something less ‘feel good’ and trite in communications with and among teachers. Indeed, encountering these posters in schools would appear to be for

62

David Block

the two women a good example of an unpleasant stimulus, in Scherer’s terms, and one that is to be avoided if at all possible. As regards the dimension of goal/need significance, Elena and Almudena position the posters as irrelevant to their needs and goals as trainee teachers, something perhaps silly which crosses their path as they get on with more serious matters. As regards coping potential, Elena and Almudena clearly wish to avoid these posters, as regards any role they might have in their development as teachers and their somewhat aggressive treatment of them in this short stretch of conversation occurs as if to make two key points with regard to coping potential: (1) that they refuse to engage with the posters and therefore they resist any potential impact they might have on their way of viewing their professional activity as teachers; and (2) they distance themselves from the posters almost pre-empting any potential attempts by others to persuade them to engage with them. Finally, there is the issue of norm-self compatibility and lining up the cultural norms and the expectations of others with Elena’s and Almudena’s individual and socially situated cultural habituses and culturally based norms and expectations. From my interviews with these two women, I glean a common educational habitus forged via their formal education in Spain, one which was manifested despite the different geographical regions and types of schools which formed part of the their respective backgrounds and education: while Elena was from a city in the Basque country and had received part of her education in a French school, Almudena was from a small village in the central part of Spain and had received part of her education in a Catholic single-sex boarding school. This common educational habitus is one consistent with an educational system, which might be termed ‘rationalist’, as it “is associated with a systematic view of the physical world . . . [in which c]apacities for logic, deduction and abstraction together with systematization and synthesis should be developed . . . [and t]he private and irrational are rigorously excluded” (McLean 1995: 30). This system contrasts with the humanist tradition, predominant tin England, which emphasises “human character and its potential rather than the structure of the physical universe” (McLean 1995: 22). For these two women, the encounter with a humanistic approach to teacher development seems to have brought about a conflict. It is a conflict between their norm-self educational habitus, as “systems of durable, transposable dispositions, structured structures predisposed to function as structuring structures” (Bourdieu 1977: 72) and the new stimuli in the form of humanistic educational norms and expectations. The conflict is indexed in the multimodal way in which the encounter is portrayed in the interview excerpt cited.

Engaging with human sociality

63

7. Conclusion Scholars should always be careful about incursions into new epistemological territory and thus I have written this paper with some trepidation. As noted in the introduction, I have engaged with the emerging human society project exclusively on my own terms as regards both my portrayal of it and how I have related my work to it. Or, worded differently, I have only been able to engage with the new stimulus called human sociality as framed by my own socioculturally situated habitus. Those who see their work more centrally inside human sociality might well take exception to this paper and indeed might see it as under-informed or even irrelevant. However, the aim, again as stated at the outset, has been to situate human sociality as a mediator of my current thinking about communication and embodiment. I can now see avenues for further thinking about these matters and ultimately, this thinking may end up more clearly within a human sociality framework. University of London

Appendix Transcription symbols for the exchange in Section 6: E= Elena / (.5) (1) ? [xxx [zzz = italics ! {xxx}

D = David Block

A = Almudena

indicates the minimal but clear pause between phrases/sentences in normally paced speech. indicates pause of half a second indicates pause of one second indicates rising intonation (including questions) indicates overlapping speech indicates that the utterance latches with (i.e. occurs seamlessly after) the next utterance indicates words that were produced in English in the original version indicates raised intonation, exclamation comments describing aspects of body language, such as gaze, hand movements and facial expressions as well as voice inflection.

64

David Block

Note * This paper is based on a keynote address given at the Bloomsbury Round Table on Communication, Cognition & Culture, held 4–5 June 2009 at Birkbeck College, University of London. It retains many features of a spoken paper.

References Block, David. 2001. Foreign nationals on PGCE in Modern Languages course: Issues in national identity construction. European Journal of Teacher Education 24(3). 291– 312. Bourdieu, Pierre. 1977. Outline of a theory of practice. Cambridge: Cambridge University Press. Bourdieu, Pierre. 1990. The logic of practice. Cambridge: Polity. Bourdieu, Pierre. 1991. Language and symbolic power. Cambridge: Polity. Brice-Heath, Shirley. 1983. Ways with words. Cambridge: Cambridge University Press. Cuenca, Maria Josep. 2002. Els connectors textuals i les interjeccions. In Joan Sol`a, Joan Mascar´o, Maria Rosa Lloret & Manuel P´erez Saldanya (eds.), Gram´atica del Catal`a Contemporani, 3173–3237. Barcelona: Emp´uries. Dalgleish, Tim & Mick Power (eds.). 1999. Handbook of cognition and emotion. New York: John Wiley and Sons. Damasio, Antonio. 1994. Descartes’ error: Emotion, reason, and the human brain. New York: G. P. Putnam’s Sons. Damasio Antonio. 1999. The feeling of what happens: Body and emotion in the making of consciousness. New York: Harcourt Brace. Dunbar, Robin. 1996. Grooming, gossip and the evolution of language. London: Faber and Faber. Enfield, Nick. 2009. The anatomy of meaning: Speech, gesture and composite utterances. Cambridge: Cambridge University Press. Enfield, Nick & Stephen Levinson (eds.). 2006. Roots of human sociality. Oxford: Berg. Goffman, Erving. 1981. Forms of talk. Oxford: Blackwell. Goleman, Daniel. 1995. Emotional intelligence: Why it can matter more than IQ. New York: Bantum. Goodwin, Charles. 2007. Participation, stance and affect in the organization of activities. Discourse and Society 18(1). 53–73. http://www.sscnet.ucla.edu/clic/ win/ParticipationStanceAffect.pdf. Goodwin, Charles & Marjorie Goodwin. 2004. Participation. In Alessandro Duranti (ed.), A companion to linguistic anthropology, 222–244. Oxford: Blackwell. Grice, Paul. 1975. Logic and conversation. In Peter Cole & Jerry L. Morgan (eds.), Syntax and semantics, Vol. 3: Speech Acts, 41–58. New York: Academic Press. Hymes, Dell. 1974. Foundations in sociolinguistics:An ethnographic approach. Philadelphia: University of Pennsylvania Press.

Engaging with human sociality

65

Jewitt, Carey. 2009. Introduction: what is multimodality? In Carey Jewitt (ed.), The Routledge handbook of multimodal analysis, 14–28. London: Routledge. Kress, Gunther & L. van Leeuwen. 2001. Multimodal discourse: The modes and media of contemporary communication. London: Arnold. Lamb, Sydney M. 1999. Pathways to the brain: the neurobiological basis of language. Amsterdam: John Benjamins. LeDoux, Joseph. 1996. The emotional brain. New York: Simon and Schuster. Levinson, Stephen C. 2000. Presumptive meanings. Cambridge: Cambridge University Press. Mayol, Laia. 2007. Catalan ‘D´eu n’hi do’ and levels of meaning in exclamatives. In Charles B. Chang and Hannah J. Haynie (eds.), Proceedings of the 26th West Coast Conference on formal linguistics, 375–383. Somerville, MA: Cascadilla Proceedings Project. Mayol, Laia & Elena Castroviejo. 2008. Asserted and implicated meanings in Catalan D´eu n’hi do. In Arndt Riester & Torgrim Solstad (eds.), Proceedings of the 13th Sinn und Bedeutung Conference, Stuttgart, 371–384. Stuttgart: University of Stuttgart. McLean, M. 1995. Educational traditions compared. London: David Fulton. Ortony, Anthony, Gerald L. Clore & Allan Collins. 1999. The cognitive structure of emotions. New York: Cambridge University Press. Sancho, Pelegr´ı. 2003. An`alisi contrastiva interdialectal de la fraseologia: El cas de d´eun’hi-do, ausad.es/a gosad.es i espaiet. Catalan Review 172(2). 151–171. Schegloff, Emmanuel. 2006. Sequence organization in interaction: A primer in conversation analysis. Cambridge: Cambridge University Press. Scherer, Klaus. 1988. Facets of emotion: Recent research. Mahwah, NJ: Lawrence Erlbaum. Scherer, Klaus. 2007. Component models of emotion can inform the quest for emotional competence. In Gerald Matthews, Moshe Zeidner & Richard D. Roberts (eds.), The science of emotional intelligence: Knowns and unknowns, 101–126. New York: Oxford University Press. Schumann, John H. 1997. The neurobiology of affect in language. Oxford: Blackwell. Schumann, John H., Sheila E. Crowell, Nancy E. Jones & Namhee Lee (eds.). 2004. Neurobiology of learning: Perspectives from second language acquisition. Mahwah, NJ: Lawrence Elrbaum. Schumann, John H. & Lee Alexandra Wood. 2004. The neurobiology of motivation. In John H. Schumann, Sheila E. Crowell, Nancy E. Jones & Namhee Lee (eds.), Neurobiology of learning: Perspectives from second language acquisition, 21–38. Mahwah, NJ: Lawrence Elrbaum. David Block is Professor of Languages in Education at the Institute of Education, University of London. His main interests are globalization, migration and the interface between language and identity. He is author of Multilingual Identities in a Global City: London Stories (Palgrave, 2006) and Second Language Identities (Continuum, 2007).

Language trends: Reflexivity in commercial language policies and practices HELEN KELLY-HOLMES

Abstract Financier, philanthropist and economist George Soros is not the first to have used or derived the notion of reflexivity (for example, Anthony Giddens, Ulrich Beck and others have found it useful in their work). What is different about the approach of Soros is that he has used the concept of reflexivity to explain the workings of global markets, and for this reason his work seems to me to have particular potential for explaining language trends in the linguistic policies and practices of global marketers. Soros has challenged ruling economic orthodoxy by arguing that demand and supply are not given or objective, and far from determining how markets develop. Participants in economic processes inevitably and unavoidably shape and determine the outcome of those economic processes and also create the conditions for future development of those market processes. This particular insight, I would argue, could help us to understand and develop further the notion of normativity in‘global’linguistic practices and policies, particularly when linked to economic processes since “the self-validating capacity [of such processes] encourages trend-following speculation” (Soros 1994). We can equate the decisions of many global marketers to that of stock brokers, who largely act on their intuitions (linguistic intuitions), but who in turn create conditions for those intuitions to be successful now and in the future, in effect by creating language trends. Keywords: Reflexivity; language trends; language policy; normativity.

1. Introduction The notion of commercial language policies and practices can be seen to encompass all decisions made in the context of commercial activities (i.e. buying

68

Helen Kelly-Holmes

and selling) that have to do with language. Some of these decisions may be codified into explicit (Schiffmann 1996) or overt (Shohamy 2006) policies or documents; or they may have become established, implicit (Schiffmann 1996) or covert (Shohamy 2006) policies through long-term use, which become accepted as the status quo, the way things have always been done in a particular company; or, they may simply be one-off practices. Whatever the case, they involve decisions which relate to managing languages – or more precisely speakers (cf. Heller and Duchene 2007) – for commercial purposes. Languages/speakers can be managed either as a problem or as a resource from a commercial perspective. On the one hand, linguistic diversity can be seen as a problem for many businesses; it is an external, uncontrollable factor (cf. Kelly-Holmes [forthcoming a], for in-depth discussion of this; cf. also da Silva et al. [2006] in relation to Canada), which needs to be managed by categorizing markets (i.e. groups of people / speakers of languages) in terms of particular languages with which they can be then targeted. On the other hand, linguistic diversity can also be seen as a resource which can be managed and exploited in order to differentiate a particular product or brand from other products or brands by appealing to primarily ethnocentric perceptions of other cultures, or of some essential qualities inherent in the language, which are essentialized in the language-product link and which the targeted consumer is deemed to share (cf. Martin 2006; Piller 2003; Ustinova and Bhatia 2005; Lee 2006; Kelly-Holmes 2005). Advertisers thus use languages for differentiation, in order to differentiate their product from other products, and also for segmentation, in order to segment markets and break them down into manageable units that can be targeted by specific strategies (cf. Kelly-Holmes [forthcoming b] for an overview of this). However, there is a sameness – and to a certain extent a predictability – in the ways in which language is used in these differentiation and segmentation practices: global brands which originate in non-English-speaking countries of origin use English language slogans globally and use English on their corporate or dot.com pages, where they seek to address a ‘global’ audience (e.g. Japanese Toyota; Finnish Nokia); where global brands want to exploit a particular attribute associated with their country of origin, they may indulge in decorative use of a small amount of their ‘national’ language combined with global English (e.g. Spanish Seat’s use of auto emoci´on; German Volkswagen’s use of Das Auto). McDonald’s has been using multilingual graphic since 2004, and Burger King and Espirit de Corps also followed suit. In terms of segmentation, we can observe similar trends, for example, the provision by international brands of strictly bilingual (i.e. parallel monolingual in Heller’s [1999] terms) sites for Belgium, Switzerland and Canada, with the use of MENA (Middle East and North Africa) sites in English, French and Arabic, by a growing number of

Reflexivity in commercial language

69

brands, and also the growing use of Spanish on the USA sites of global and US American brands. All of these practices represent language trends and norms of a sort in advertising and market discourses. They are styles which may start with an individual, but which are then followed and come to be generally accepted as ways of ‘doing language’ in commercial contexts. Trends are by their nature transient, but while some inevitably fade away, others become established as norms. But, how might these trends or norms come about? As someone who has been observing such trends over a number of years, this question has come to concern and fascinate me, and in this article I would like to propose a framework that might enable us to track and understand these trends and to develop a sociolinguistics of the market (for an overview of approaches to this to date, cf. Kelly-Holmes [forthcoming c]). The framework that I am proposing is based on the economic writings of George Soros, and I want to begin by putting Soros’ work in context and arguing for the use of his work and insights from the field of economics in understanding commercial language trends, before going on to explain Soros’ most relevant theory, namely that of reflexivity. I then attempt to show how this could be useful for explaining how trends emerge in the use of languages for differentiation and segmentation in global, national and local advertising, and point to what I see as useful avenues for exploration in order to take these ideas forward.

2. Putting Soros in context George Soros is a controversial figure in all of the domains in which he is active – financial markets, academia and politics. He was born in Hungary, but his grew up in the USA. His brief encounter with the Nazi occupation and with the communist regime together with his discovery of the work of Karl Popper when he studied at the LSE have made him a lifelong advocate of the ‘open society’. The open society is, as one might expect, concerned with aspiring towards genuine freedom of information, thought etc., but – and this is relevant in terms of understanding reflexivity – also with an acknowledgement that there is bias at work and that we need to attempt to overcome or limit this bias, or work with it, rather than ignoring it or pretending it does not exist. Of interest to linguists may be the fact that Soros is a fluent Esperanto speaker, having been brought up with it as his first language. Soros is also known as the man who bankrupted the British pound on ‘Black Wednesday’ in October 1992. Much of the profit made by Soros in currency trading has gone towards philanthropic activities in former Eastern Bloc countries, particularly in the area of education (Soros established the Central European University in Budapest in 1991).

70

Helen Kelly-Holmes

What is compelling about Soros is his combination of theory and practice. There are few eminent economists or philosophers who actually try out their ideas on the financial markets or, in a grounded theory approach, formulate theories on the basis of their experience in such markets. Economic theory is notoriously and unapologetically abstract and generally not data-driven, and prides itself on ignoring real-life situations in favor of perfect market conditions and on the assumption of rational human behavior. Soros has argued, and his argument has added credibility because of his experience as a practitioner, that this is nonsensical. He is also anti-positivistic in his approach to studying economics, advocating, like Popper, that social sciences should not try to imitate natural or hard sciences and should instead work within and around the flawed, partial and unpredictable nature of human beings. Soros has published a number of works on his economic ideas (e.g. Soros 1987, 1998). However, to his disappointment, which he has spoken publicly about, the ideas have yet to be taken on board seriously (although as we shall see below there are points of comparison between him and the ideas of John Maynard Keynes), and he is generally seen as a maverick by orthodox economists. He has also been branded an ‘anti-capitalist capitalist’ and a hypocrite for being happy on the one had to make money from international financial markets, whilst at the same time criticizing them. Having put Soros in context, I will now turn to an explanation of reflexivity.

3. Reflexivity versus orthodox economic thinking In general terms, reflexivity can be seen to occur in social systems and processes “when an actor observes and thinks about his or her actions and their consequences and then modifies his/her behavior” (Umpleby 2006: 1). Soros did not of course invent the concept of reflexivity, nor is he the first person to use it as a way of explaining social processes and to “. . . refer both to the fact that and to the way in which agents are confronted with the results of their earlier actions” (Mügge 2007: 10). Soros and other proponents of the reflexivity approach would argue that these results are often different from what they may originally have intended, and the reason for this is that “. . . societies are extremely complex, bringing together myriad agents with different agendas, capability and perspectives” (Mügge 2007: 10). Consequently, these actions and their results, from a reflexive perspective, can “generate new agency and thus inject dynamism into social change” (Mügge 2007: 10). Furthermore, reflexivity means that the way agency affects overall structures is unpredictable, both for analysts and agents themselves (Mügge 2007: 10).

Reflexivity in commercial language

71

Change, therefore, should be understood as “the combination of intended and unintended consequence of purposeful agency” (Mügge 2007: 10). Reflexivity, thus, gives us a way of engaging with and studying systems that include thinking and acting participants (Birshtein and Borsevici 2001: 74), and this is reflected in the work of sociologists of globalization such as Anthony Giddens and Ulrich Beck (cf. Bryant [2002] for a comparison between Giddens, Beck and Soros in relation to reflexivity). As mentioned above, Soros claims to have used reflexivity to explain and understand the workings of international financial markets. He sees market developments as dependent on “the beliefs of market practitioners about them. Therefore, a clear separation between markets and the beliefs observers hold about them is impossible” (Mügge 2007: 12). Soros argues that “many market participants are perfectly aware of this recursiveness. And they try to use it to their own advantage” (Mügge 2007:12). Soros has a more holistic approach to finance and economics: he sees finance as a ‘human player game’ with himself as a participant. One of his key criticisms of economic orthodoxy is that it negates or disregards the human dimension to economics and finance. In such classical – or more correctly neo-classical – economic thinking, the individual is deemed to act rationally and to make decisions on the basis of having access to perfect information. According to mainstream or neo-classical theory, economic man or woman is rational, being rational means that s/he seeks to maximize his/her utility and that s/he does so in a consistent way, using the perfect information available to make these rational decisions. Neo-classical economic theory considers utility to be that which is of value to (or in the best interests of) the individual – something that is different to the classical or Marxist view of utility. Thus, individuals try to maximize the value of goods to them. There are of course many problems with this theory, which Soros and others have pointed out, and which, commonsensically, we can see ourselves: individual human beings are highly complex and have multiple and often contradictory motivations. As economist John Kay points out: “the economic imperialism which seeks to ‘explain’ all behavior by reference to rational choice is absurd” (Kay 2004: 194), and “. . . often this is achieved by stretching the meaning of rationality” (Kay 2004: 201). Likewise, we can clearly see that ‘perfect information’ is hardly ever available in all situations, hence the famous caveat, ‘buyer beware’. According to the mainstream view, however, left to its own devices, with a minimum of intervention by governments or other players, and with everyone behaving rationally, demand should stimulate supply in order to achieve equilibrium in the best possible way for all agents and all participants. Another major criticism, and one levelled by Soros, is that mainstream economic theory is not based on reality – in fact it ignores realities – or on any data or experience; instead it proposes a

72

Helen Kelly-Holmes

Utopian view of economics. This might perhaps be acceptable in and of itself – the need for abstraction in all theory being important – were mainstream economics not used as the way in which major decisions are taken by national and international economic players such as governments, companies, international organizations etc. who operate in a far from Utopian world. Soros has turned mainstream economic thinking (which is largely based on this neo-classical or Chicagoean model) on its head – causing “constant annoyance among advocates of the orthodox economic and political theory” (Birshtein and Borsevici 2001: 74) – and tried to move away from this abstract concept of economics, by arguing that “economic reality is actively shaped by the perceptions of market participants” (Cymbalista 2003: 1). Thus, rather than demand and supply determining what happens in markets, it is instead what happens in markets that determines demand and supply; and what happens in markets is what individual human beings do on markets, with all their flaws, unpredictable and irrational behaviour and complex ideologies. According to Soros, participants in economic processes inevitably and unavoidably shape and determine the outcome of these economic processes, rather than these processes being shaped by some abstract notion of demand and supply. Soros sees individuals as participating at a fairly unsophisticated level – they use guesswork, they formulate hypotheses and they act on these hunches or hypotheses. The key aspect, according to Soros, is that by acting on these hypotheses or guesses, they actually create the conditions for their hypotheses to be successful. Central to this understanding is an acceptance of, perhaps even an embracing of, human fallibility, “meaning both the belief in [our] own fallibility and the belief that the misconceptions and misunderstandings that go into our decisions help shape the events in which we participate” (Cymbalista 2003: 1). Taking the example of English as a global language of business, it could fruitfully be argued that a process of reflexivity is at work here. The received idea within sociolinguistics and applied linguistics could perhaps be formulated as follows: processes of globalization have made English a global lingua franca in business, and thus there is a demand for English globally, which is being met by the supply of English. This status of English as global lingua franca leads in turn to a greater demand for English, which stimulates supply, and thus leads to a greater supply of English. This type of argument is based on ‘classical’economic theory, which would predict that some kind of equilibrium would eventually be reached in terms of the demand for English and the supply of English internationally. However, demand appears to continue to grow, and so, correspondingly does supply. Soros would argue that, what such an understanding lacks is the human component. Looking at the global spread of English from the reflexivity point of view, it might be possible to formulate a contradictory point of view:

Reflexivity in commercial language

73

individual agents (e.g. companies) formulate their own linguistic hypotheses about using English in particular commercial contexts. These hypotheses are of course not just individual, but are also influenced by social, cultural and sociolinguistic factors; these are language ideologies (cf. Schieffelin et al. 1998; Kroskrity 2001; Bauman and Briggs 2003). But, basically, a decision is made somewhere, by someone, to use English in a particular advertisement for example, without there being any particular demand for English to be used in that advertisement. Thus, supply might precede demand. However, because the decision is made to supply or use English, English comes to be accepted in this particular context, and then more English is supplied. A reflexivity approach would argue that by using English, companies and agents are creating the conditions whereby more English can be used or is even expected to be used (e.g. corporate websites discussed above), the idea that ‘everyone is doing it, so we should be too’. Another important difference is that rather than simply focusing on variables, as is the practice in mainstream economics, Soros also looks at ideas, groups and events – factors that have either been deemed unimportant or have been neutralized or blanded out in mainstream economic theory (e.g. in the notion that there is perfect information, and that all individuals act rationally all the time). In his mind, it is particularly important that we take ideas into account – these can be seen as the bias which various actors bring to their decisions, their guesswork and hypothesizing, and, ultimately, their actions. From our point of view, this could be their own linguistic ideologies. The ability to incorporate individual language ideologies into a model of commercial language practices makes reflexivity particularly attractive to me, since too often in sociolinguistics and applied linguistics, actions are deemed ‘corporate’, thus hiding or masking the societal, educational, familial and individual inputs in these. Often, too, it is forgotten that copywriters, marketing managers, brand and image consultants are more than just ‘economic man’ or rational human beings too. They have their own complex linguistic identities and these may impact on their behavior, which may be far from rational in a mainstream sense. All too often, perhaps, linguists are seen on one side and marketers on the other, whereas the reality is much more complex. For example, Atkinson and Kelly-Holmes (2006) and Kelly-Holmes and Atkinson (2007) argue that advertisers as members of particular language communities do not always act rationally as the Chicagoean school would predict. For example, some may advertise in a minority language primarily to lend financial support to a minority language newspaper and in this way support revitalization or maintenance in a language community rather than to reach particular customers and even where they acknowledge that it will probably not attract business for them. Equally, advertisers possibly influenced

74

Helen Kelly-Holmes

by majority language ideologies may decide to keep their advertisement in the national language (in this case Castilian Spanish) in a newspaper that is entirely in the ‘minority’ language, and which is translated free of charge into the minority language (in this case Catalan). Here, advertisers actively and explicitly opt out of speaking the customer’s language. The idea then is disarmingly simple: individual ways of thinking influence events on markets – according to Soros, reflexivity equals perception plus action. In other words, our perceptions plus our actions creates events which have consequences on markets. A recent example which came to my attention may help to illustrate this: international drinks brand Carlsberg recently produced an advertisement for the Irish market in which Irish is used (cf. Kelly-Holmes [2010] for a full discussion of the advertisement). This was the first time that Irish had been used in this way (see discussion below) in such a context by a global advertiser (it has been used nationally, and there is growing use of commercial Irish in a number of domains in Ireland). According to a source in Diageo, the global drinks company that owns Carlsberg, the marketing team, a very small group of individuals, felt that Irish is becoming ‘trendy’ and that an advertisement in Irish would work well. So, we can see here how these individuals formulated a linguistic hypothesis on the basis of their own language ideologies – in turn fed by societal language ideologies in Ireland. By using Irish in an advertisement for a global beer brand, these social actors have created a new scale (Blommaert et al. 2005) for the language and have also created the conditions whereby more Irish can be used in future advertisements – they have perhaps started a trend. In Soros’ terms, their perceptions (changing language ideologies about the ‘trendiness’ of Irish) together with their action (using Irish in this particular action) create the reflexivity, whereby they help to create the conditions for their use of Irish (and the future use of Irish in such commercial contexts) to be successful. We can equate this decision to that of stock brokers, who, Soros claims, largely act on their intuitions and biases, but who in turn create conditions for those intuitions to be successful now and in the future. Mügge argues in relation to global financial markets that “producers in the financial sector increasingly act as rule-setters in the transnational domain” (Mügge 2007: 7) and Yeung (2002) also points out in relation to globalization, that it is the social actors who exploit global scales who also and at the same time by doing so create and reinforce these global scales. For instance the ‘scale’ or ideology that enables English to work globally does not just exist; it is created, reinforced by people and their everyday language decisions. As Cymbalista (2003) argues in relation to price trajectories on international stock currency markets, “valuation does not represent economic value but creates economic value . . . markets and their

Reflexivity in commercial language

75

participants are more than information-processors. They create information” (Cymbalista 2003: 20). So, we can see that rather than having perfect information about the sociolinguistic situation on a particular market or in relation to a particular customer or group, advertisers and marketers may have some – by definition – imperfect information, which they combine with their own perceptions and ideologies. Just as it is, Soros claims, impossible to have perfect information about market conditions, so too is it impossible to have perfect information about a sociolinguistic situation. Reflexivity also provides a way of explaining the ‘copycat’behavior which we can observe in commercial language policies and practices, the homogenizing trend we can observe around us (cf. Kelly-Holmes forthcoming b). Linguistic hypotheses, as argued above, become self-fulfilling: the presence of a language in a particular ‘event’ (see below) leads to an expectation of its presence. Consumers thus come to expect to see the particular language used in a particular way. Trends emerge, practices are copied and they become norms. For example, multilingualism has been used as an index of universalism or globalism by international organizations such as the UN and the EU. Promotional material for the EU tends to feature multilingual slogans and UNICEF’s Christmas cards have long featured multilingual greetings, a practice taken up by OXFAM and other international charities. Recently, brands which are seeking a similar type of global or universal image have also started/followed a trend in using multilingualism as a way of indexing this. Most notable has been the McDonald’s ‘i’m lovin’ it’ campaign, launched in 2003, which features a multilingual slogan with Arabic, Chinese, French, German, Filipino, Turkish, Ukrainian versions of ‘i’m lovin’ it’ alongside the English version. These versions together formed a multilingual graphic which was used on packaging in McDonald’s restaurants. When fast food rival Burger King relaunched its 30-year-old ‘Have it your way’ slogan in 2004, it too created versions in different languages and merged them into a multilingual graphic that appeared on packaging, cartons etc., and clothes brand Esprit de Corps (EDC) uses different language versions of its slogan ‘The world is our culture’ on its bags to reinforce its positioning as an “international youthful lifestyle company” (www.edc.com). The combinations of languages used, while displaying some differences (e.g. in McDonald’s use of Ukrainian and Filipino), also display remarkable similarities, with languages such as Arabic, Chinese, French, English acting as primary indexes of internationalization and globalization.

76

Helen Kelly-Holmes

4. A reflexive model of language trends In this section, I would like to attempt to model how commercial language policy and practice might operate as a social system, so that trends come about. The model, following Soros, has four key components. The first is ‘variables’, and in this model, the variables would be the language or languages used (including also varieties, accents, repertoires), as well as their modes of use (e.g. genres, discourses) which are also subject to reflexivity. In other words, which language(s) is/are chosen, and which are not, and how are they used? The second component is ideas, and these would be language ideologies – individual, group, societal – perceptions and preconceptions about language or languages or varieties etc. as well as about modes of use of language and about speakers. The third component Soros focuses on is ‘events’, and these would then be the particular texts and speech acts which take place, since these represent the action or fundamental intervention on the market (e.g. advertisements, websites, brochures etc.). Soros’ final component, ‘groups’ would in such a model refer to all actors at micro, meso and macro levels, who are involved in these processes, who can take part in these processes and who are affected by these process. So, the marketers themselves, companies, consumers, individuals, educational institutions etc. The proposed model is presented in Figure 1.

Figure 1. Reflexive model of language trends

The reflexivity model of language trends (see figure 1) shows the reflexive nature of the process, whereby action and participation are constantly being described and understood by the individual actors (groups), and whereby variables, ideas, events and groups are constantly interacting and affecting each other: . . . the psychology of participants in any historical process is its integral component and by continuously interacting with reality, it forms a reflexive process: a real situation affects the minds and behavior of its participants, while their thoughts and behavior affect the development of the situation. The participants’ ideas, evaluations, expectations and prevailing preferences, which are imperfect

Reflexivity in commercial language

77

by nature, greatly determine the natural course of events and their basic uncertainty. (Birshtein and Borsevici 2001: 75)

Another appealing aspect of Soros’ theory of reflexivity from the point of view of applying it to the language policies and practices of marketers is his focus on the interdependence of the various factors which other economists tend to treat as discrete; namely, variables, ideas, events and groups. According to Soros, interaction comes first, and the model is predicated on the interdependence of variables, ideas, events, and groups. It is in fact impossible to think of these variables as independent or existing on their own. This is nothing new to sociolinguists, but it is fairly radical in economics. Reflexivity in economic theory recognizes that “. . . a factor never acts alone, but only as it changes and is instantly changed by being in interaction with all other factors” (Cymbalista 2003: 19). Thus, rather than a static and independent set of variables, a reflexive model should instead be seen as an “unseparated multiplicity” (Soros 1987), in which “interaction comes first” (Cymbalista 2003: 19). The model proposed above should therefore not be seen as something static or measurable. It is instead an attempt to model or gain a snapshot of a process that is continuous. The variables (languages, varieties, accents) are not static and measurable, since they are always interacting with ideas (ideologies) and groups (speakers, hearers) in ever-changing, continuous conversations and narratives (events) – the speech act we encounter at this moment cannot be seen or measured or encountered in isolation from the speech act immediately before or after it. Our ideologies are inseparable from our experience of events and groups and vice versa. It is this state of constant diachronic and synchronic change that leads to uncertainty, which in turn leads to (and actually necessitates) guesswork on the part of individual actors in the market. This state of uncertainty is due to the ‘unseparated multiplicity’ of variables, events, ideas and groups, which are constantly participating and reflecting on and understanding this participation, and which in turn leads to a state of constant change. Thus, “uncertainty is therefore not just the absence of certainty about an outcome: it is also the presence of the ongoing process of events” (Cymbalista 2003: 19). Market processes are “not a series of discrete events, a chain of completed occurrences, of finished happenings. The future is uncertain because it’s always in the process of being formed” (Cymbalista 2003: 19), and all the components of the reflexive model have a hand in how it is being formed. This concept of an ‘unseparated multiplicity’ also shows how ideologies and norms come about. Just as ‘the market’ does not consist of a series of discrete events, information too is “a continuous process which is created over time as the market participant engages with the events” (Cymbalist 2003: 20).

78

Helen Kelly-Holmes

Trends emerge when consensus emerges in this process among a range of market participants, who “not only perceive outside events in a similar way but also share perceptions of transition or stability, thus creating, reinforcing of reversing trends” (Cymbalista 2003: 20). Reflexivity can also be seen to have something in common with the process of linguistic normalization, in particular as exemplified by Strubell’s Catherine Wheel (cf. for example, Strubell 1998, 2001) – the case of commercial Irish being a good example. Although not explicitly economic in orientation, Strubell’s Catherine Wheel does show the reflexive impact of intervention in processes of language revitalization. Strubell shows how supply and consumption of goods and services in a particular language can lead to perception of a greater need for the language, leading to more motivation to learn and use the language, leading to more learning, more informal use in social situations, more demand for goods and services in the language and back to even greater supply and consumption of goods and services in the particular language. Interventions set the wheel in motion. However, the Catherine Wheel is predicated on the same assumption as mainstream economic thinking, namely that demand stimulates supply or supply stimulates demand. A reflexivity modeling of the processes of language revitalization or language normalization would instead focus on expectations and acceptability (in Bourdieu’s [1991] terms), rather than seeing a straightforward cause and effect or input-output relationship – again keeping in mind the contradictory and paradoxical behavior and ideologies which individuals have and which they are constantly revisiting. Taking the example of the Irish language, the introduction of Irish language television station TG4 (originally TnaG when it was founded) in 1996 can be seen to represent an intervention that would correspond to an increase in the supply of goods and services in Irish that would lead to greater demand and greater supply, according to Strubel’s Catherine Wheel. If we take the example of advertising in Irish, however, the situation is not quite a straightforward as this – there is an effect, but it happens in a much more indirect, unpredictable and long-term way. When TG4 was set up by the Irish government, many national and international brands initially started to advertise on the channel in Irish and paid for the production of Irish language advertising. However, soon after its launch, the channel began to lose its viewership and it consequently also lost advertisers. When the station began to recover and relaunched itself, it did win back advertisers, but this time the majority of the advertisers kept their advertising in English (cf. Kelly-Holmes 2005). So, in the ad breaks in the programs (which are in Irish with English subtitles during prime time) the bulk of commercial advertisements for national and international products are in English.

Reflexivity in commercial language

79

While the intervention into the supply side (providing a television station in Irish) might not have led to a growth in the demand side (for advertising in Irish), the conclusion is too simple and points to the inadequacy of the mainstream economic view. In parallel with TG4’s growth and success, a number of interesting initiatives have happened – we could describe these as trends. An increasing number of businesses in particular domains are using Irish company and brand names, particularly in the areas of food (cf. Kelly-Holmes 2005), tourism (cf. Kelly-Holmes et al. forthcoming), but also in newer domains such as beauty (a local beauty parlor in Limerick uses the name Mise or ‘me’ to brand itself) and child care (a day care-centre uses the Irish word for care, Aire to mark itself out). The young producers on TG4 and the trendy programming have contributed to a phenomenon identified as ‘Sexy Irish’, and Irish has started to feature in comedy acts (Kelly-Holmes and Atkinson 2007b). Recently, as highlighted above, Carlsberg, a global brand, used Irish in an advertisement for lager. The last time Irish was used in a commercial for beer, it was for the iconic Irish brand of Guinness, and featured the inhabitants of the Irish-speaking Aran Islands, off the Irish west coast, watching the clock and the sea to await their supply of Guinness due to arrive by the traditional currach boat (1977). The Carlsberg ad on the other hand featured three young Irish men on holidays in Rio de Janeiro, and the message of the ad, that speaking Irish makes one attractive in international tourist destinations, was very much in tune with the ‘Sexy Irish’ discourse (cf. Kelly-Holmes [2010] for full discussion of this). What I think we can see happening here is something that cannot really be explained in terms of demand and supply or even demand/supply and consumption. It is instead a process of taking chances and making guesses about what is credible, acceptable (in Bourdieu’s [1991] terms), what is possible to say; and in the process of stretching the boundaries of what is currently acceptable, possible, credible, creating new levels of acceptability, credibility and possibility, while challenging prevailing ideologies. It involves very subtle guesswork based on knowledge of prevailing language ideologies, and knowledge of how far these can be stretched and challenged. In this way, trends are created, and reflexivity, by creating new trends, norms and expectations, ultimately leads to predictability and sameness.

5. Conclusion As stated in the introduction, this article grew out of my own curiosity about patterns and trends that I have been observing over more than ten years of research around the area of commercial language policies and practices. In particular, my engagement with reflexivity is driven by a desire to deconstruct

80

Helen Kelly-Holmes

the discourse around ‘the market’ and globalization, which is reflected in the ‘uneasy relationship’between sociolinguistics and economics (cf. Kelly-Holmes [forthcoming c] for a discussion of this). One appealing thing about reflexivity is that it enables us to stand back from vast nebulous processes like globalization or the global spread of English and break down these processes into the microlevel decisions of companies, teams, and sometimes individuals, sometimes working on their own, with their own particular language ideologies. The attempt to use reflexivity is also motivated by a conviction that orthodox economics, with its reliance on ‘perfect’ information, rational human beings and a notional equilibrium created by supply responding to demand and vice versa has little to offer us in understanding the complex and multi-layered interactions between languages and markets. More than this, it may even obscure what is actually taking place, masking the agency involved. What would be needed in order to build a credible and useful reflexive theory of language trends would be longitudinal, historical studies and tracking of institutional practices and of the development of such trends, in combination with ethnographic studies of the ‘groups’ identified in our model, in particular of course the actors who create the events and who make the language choices. Further fruitful research could look at the real effects of these choices, the ripple effects, in terms of speakers, ideologies, equalities and categorizations (cf. Blommaert 2003). Just as “. . . the evolution of prices on financial markets can be regarded as a reflexive historical process” (Birshtein and Borsevici 2001: 75), so too might we usefully research the evolution of language trends as ‘a reflexive historical process’. By acknowledging the ‘unseparated multiplicity’ of ideologies, languages, speakers and events, and their constantly modifying interactions, reflexivity allows us to understand a dynamic ‘future is now’, ‘past is now’ model of these interactions and their ongoing effects. Of course, reflexivity does not and cannot provide all of the answers; it may simply help to illuminate one part of the picture. As Cymbalist highlights, reflexivity in economics and international financial markets remains “mysterious, both at the theoretical and at the practical level: neither his [Soros’] conceptual framework nor the manner in which it gets translated into investment decisions is fully understandable” (Cymbalista 2003: 1). This mystery is, however, due to the fact that reflexivity allows us to acknowledge that we are dealing with human beings who are not rational and who are unpredictable, who behave in complex and contradictory ways – orthodox economics which eliminates these messy factors is much easier to comprehend and less mysterious, but not very useful, I would argue. Messiness is what sociolinguists and applied linguists deal in – the messy, mixed-up language practices that again and again defy categoriza-

Reflexivity in commercial language

81

tion and labeling – and a useful, workable theory that helps us to understand language trends and their implications needs to have this messiness built in. University of Limerick

References Atkinson, David & Helen Kelly-Holmes. 2006. Linguistic normalization and the market: Advertising and linguistic choice in El Peri´odico de Catalunya. Journal of Language Problems and Language Planning 30(3). 239–260. Bauman, Richard & Charles L. Briggs. 2003. Voices of modernity: Language ideologies and the politics of inequality. Cambridge: Cambridge University Press. Birshtein, Boris &Viktor Borsevici. 2001. Theory of reflexivity by George Soros:Attempt of critical analysis. Reflexive Processes and Control 1. 73–85. http://www.reflexion.ru/EJour2002 1.html (accessed 30 September 2009). Blommaert, Jan. 2003. Commentary: A sociolinguistics of globalization. Journal of Sociolinguistics 7(4). 607–623. Blommaert, Jan, James Collins & Stef Slembrouck. 2005. Spaces of multilingualism. Language & Communication 25. 107–216. Bourdieu, Pierre. 1991. Language and symbolic power. Cambridge, MA: Harvard University Press. Bryant, Christopher G. A.. 2002. George Soros’s theory of reflexivity: A comparison with the theories of Giddens and Beck and a consideration of its practical value. Economy and Society 31(1). 112–131. Cymbalista, Flavia. 2003. George Soros: How he knows what he knows, Part 1: The belief in fallibility. Stocks, Futures, and Options. July 2003 2(7). www.marketfocusing.com/downloads/soros1.pdf (accessed 30 September 2009). Da Silva, Emanuel, Mireille Mclaughlin & Michelle Richards. 2006. Bilingualism and the globalized new economy: The commodification of language and identity. In Monica Heller (ed.), Bilingualism: A social approach, 183–206. Basingstoke and New York: Palgrave-MacMillan. Heller, Monica & Alexandre Duchene. 2007. Discourses of endangerment: Sociolinguistics, globalization and social order. In Monica Heller and Alexandre Duchene (eds.), Discourses of endangerment, 1–13. London and New York: Continuum. Heller, Monica. 1999. Linguistic minorities and modernity: A sociolinguistic ethnography. New York and London: Longman. Kay, John. 2004. The truth about markets. London: Penguin. Kelly-Holmes, Helen & David Atkinson. 2007a. Minority language advertising: A profile of two Irish language newspapers. Journal of Multilingual and Multicultural Development 28(1). 34–50.

82

Helen Kelly-Holmes

Kelly-Holmes, Helen & David Atkinson. 2007b. When Hector met Tom Cruise: Attitudes to Irish in a radio satire. In Sally Johnson & Astrid Ensslin (eds.), Language in the media, 173–187. London & New York: Continuum. Kelly-Holmes, Helen, Sari Pietik¨ainen & M´air´ead Moriarty. Forthcoming. Minority languages, tourism and the Web. Tourism Culture and Communication. Kelly-Holmes, Helen. 2005. Advertising as multilingual communication. Basingstoke & New York: Palgrave-MacMillan. Kelly-Holmes, Helen. 2010. Rethinking the macro–micro relationship: Some insights from the marketing domain. International Journal of the Sociology of Language 202. 25–40. Kelly-Holmes, Helen. Forthcoming a. Raising language awareness or reinforcing monolingual norms? In Helen Kelly-Holmes & Gerlinde Mautner (eds.), Language and the market. Basingstoke and New York: Palgrave-MacMillan. Kelly-Holmes, Helen. Forthcoming b. Language and global marketing. In Nikolas Coupland (ed.), Handbook of language and globalization. Oxford & New York: WileyBlackwell. Kelly-Holmes, Helen. Forthcoming c. Markets: The linguists’ perspective. In Helen Kelly-Holmes & Gerlinde Mautner (eds.), Language and the market. Basingstoke & New York: Palgrave-MacMillan. Kroskrity, Paul V. (ed.). 2000. Regimes of language: Ideologies, polities and identities. Santa Fe, NM: School for Advanced Research Press. Lee, Jamie S.. 2006. Linguistic constructions of modernity: English mixing in Korean television commercials. Language in Society 35(1). 59–91. Martin, Elizabeth. 2006. Marketing identities through language: English and global imagery in French advertising. Basingstoke & New York: Palgrave-MacMillan. Mügge, Daniel. 2007. Reflexivity in global finance: How agency matters to market change. Garnet Working Paper 14(07). http://www.garnet-eu.org/fileadmin/documents/working-papers/1407.pdf (accessed 30 September 2009). Piller, Ingrid. 2003. Advertising as a site of language contact. Annual Review of Applied Linguistics 23. 170–183. Schieffelin, Bambi B., Kathryn A. Woolard & Paul V. Kroskrity (eds.). 1998. Language ideologies: practice and theory. New York & Oxford: Oxford University Press. Schiffman, Harold. 1996. Linguistic culture and language policy. New York & London: Routledge. Shohamy, Elana. 2006. Language policy: Hidden agendas and new approaches. London & New York: Routledge. Soros, George. 1987. The alchemy of finance: Reading the mind of the market. New York: John Wiley. Soros, George. 1998. The crisis of global capitalism: Open society endangered. Jackson, TN: PublicAffairs-Perseus. Soros, George. 1994. The theory of reflexivity. Lecture delivered to the MIT Department of Economics World Economy Laboratory Conference, Washington DC. http://www. geocities.com/ecocorner/intelarea/gs1.html (accessed 30 September 2009).

Reflexivity in commercial language

83

Strubell, Miguel. 1998. Language, democracy and devolution in Catalonia. In Sue Wright (ed.), Language, democracy and devolution in Catalonia, 4–38. Clevedon: Multilingual Matters. Strubell, Miquel. 2001. Catalan a decade later. In JoshuaA. Fishman (ed.), Can threatened languages be saved? Reversing language shift, revisited: A 21st century perspective, 260–283. Clevedon UK: Multilingual Matters. Umpleby, Stuart A. 2006. Reflexivity in social systems: The theories of George Soros. Proceedings of the 50th Annual Meeting of the International Society for the Systems Sciences. www.gwu.edu/∼umpleby/recent papers/2007 SRBS Reflexivity Theory.pdf (accessed 30 September 2009). Ustinova, Irina & Tej K. Bhatia. 2005. Convergence of English in Russian TV commercials. World Englishes 24(4). 495–508. Yeung, Henry W. 2002. The limits to globalization theory: A geographic perspective on global economic change. Economic Geography 78(3). 285–305. Helen Kelly-Holmes is a lecturer in sociolinguistics and new media at the University of Limerick, Ireland. Her research interests focus on the sociolinguistics of market processes. She has published widely in a range of international journals and is author of Advertising as Multilingual Communication (2005, PalgraveMacMillan).

Contacts of Russian in the post-Soviet space* ANNA VERSCHIK

Abstract The article discusses the importance of contact linguistic research for a general understanding of the language dynamics and linguistic processes in the post-Soviet space. Unlike political and macrosociolinguistic studies, contact linguistic case studies on language contacts in FSU are still scarce; some reasons being the heterogeneity of data and isolation of Eastern scholars. The review is concerned with contacts of Russian language with Ukrainian, Kazakh and Estonian. It is demonstrated that the particular cases are relevant for a general discussion on the directionality of contact-induced change, grammar of code-switching, and the emergence of contact varieties. Keywords: language contacts; post-Soviet studies; Russian; Ukrainian; Kazakh; Estonian.

1. Introduction The collapse of the USSR and the subsequent fundamental changes that affected politics, language policies, sociocultural settings, minority-majority relations, language choice patterns, and the character and distribution of bi- and multilingualism, immediately attracted attention of scholars in social and political sciences but, surprisingly, not so much in contact linguists. Political scientists, most notably Laitin (1998) and Kolstø (1995, 1999) from the West and Lebedeva (1995) and Savoskul (2001) from the East with their teams conducted necessary foundational research. Later, linguistic anthropologists and sociolinguists started their research in the post-Soviet countries (for instance, Ciscel [2007] on Moldova; Korth [2005] on Kyrgyzstan; see details and references in Pavlenko [2008a: 276–277]). However, among the researchers we do not find contact

86

Anna Verschik

linguists. Such little interest can be explained by a variety of reasons, one of them certainly being the heterogeneity of the subject: fifteen successor states, different language policies, different degrees of proficiency in Russian/titular languages, different ethnic makeup of the population and the proportion of Russian-speakers, myriad of language contact situations, inherited and new, and different history of the incorporation into the Soviet Union. Yet, there is at least one common denominator that allows comparison between different post-Soviet contact situations and further theorizing: namely, the contacts between Russian and other languages. This circumstance turns the whole post-Soviet space into a kind of language laboratory (Verschik forthcoming). So far that most of the literature on contact linguistic addresses Western contact situations where a sociolinguistically dominant language is one of the Indo-European languages (English, German, Dutch, etc). The studies focus on contacts of immigrant languages (Turkish in Europe) or minority languages (for instance, Swedish in Finland). In this situation, post-Soviet language contacts will certainly add new empirical evidence from a novel sociolinguistic context. As far as macrosociolinguistics and language policy of post-Soviet space are concerned, some commonalities, differences and challenges have been outlined and a preliminary basis for further research has been formulated, thanks to the efforts of Aneta Pavlenko who has published several comprehensive reviews (2006, 2008a, 2008b). From the point of view of microsociolinguistics and modern contact linguistics, that is, the study of spoken/written language data collected from individual speakers, it is too early for large-scale generalizations. The first attempt of bringing together scholars in contact linguistics who work on post-Soviet language contact data was the colloquium “Language contacts in the post-Soviet space” held at the 6th International Symposium on Bilingualism in 2007 in Hamburg by the current author. Several papers from the colloquium are to appear in the special issue of International Journal of Bilingualism (Verschik forthcoming). Still, all the articles in the named collection have different focus and address a wide range of topics, from pidgin formation (Stern forthcoming) to a variety that has emerged as the result of contacts between closely related languages, Russian and Belarusian (Liskovets forthcoming), from the continuing impact of Russian on Komi (Leinonen forthcoming) to the growing impact of Estonian on Russian (Zabrodskaja forthcoming a) and Russian conjunctions in the speech of urban Kazakhs (Muhamedova forthcoming). The special issue may be viewed as the first step towards defining the field. Of course, the topic of contacts between Russian and other languages is not new in itself (see for instance Comrie [1981] on the languages of the USSR). However, the character, focus and scope of such research are most heterogeneous and the quality of writings on this topic is rather uneven. Many Eastern scholars

Contacts of Russian in the post-Soviet space

87

are not familiar with modern theoretical frameworks of code-switching, contactinduced language change, bilingual speech, etc., and their studies are often rather descriptive without any reference to current discussions on contact linguistics in the West. In this respect, contact linguistic and sociolinguistic research in some post-Soviet countries is doing better than in others. The question why this happens is without doubts fascinating but cannot be addressed here. It has to be noted as well that many studies are published in languages other than English or German and, thus, remain inaccessible to Western scholars. It is difficult to present a concise review of such a new and constantly growing field of research. An all-inclusive comprehensive review on all language contact situations of FSU would be impossible and some delimitation is inevitable. The departure points in the current review are as follows. (1) Developing the metaphor of language laboratory, the focus will be on contacts between Russian with other languages, that is, the same language (once the dominant language of USSR) in different sociolinguistic situations. Thus, Latvian-Estonian or Tajik-Uzbek language contacts remain out of the scope of the present article. A view on contacts between Russian and typologically different languages or between Russian and other East-Slavic languages, Ukrainian and Belarusian, but in different sociolinguistic circumstances can in the long run provide relevant empirical data, which would hopefully improve the understanding of the interplay between typological and sociolinguistic factors in language contacts. (2) Directionality of impact and contact-induced language change. The question of directionality of impact and contact-induced language change appears a very obvious one, yet it has not been separately addressed in case-studies. Sometimes Eastern scholars write about ‘influence’ of other languages on Russian but most often it is not treated in any systematic manner within any modern contact linguistics theoretical framework (e.g., Külmoja 1999; Mechkovskaya 2005). As for directionality of impact, there appears to be a certain division of labor between disciplines. Traditionally, SLA research is concerned with the impact of L1 > L2 and contact linguistics deals more often with the opposite L2 > L1 impact. Thomason and Kaufman (1988) proposed a now classical model where the fundamental difference between two different sociolinguistic processes, language maintenance and language shift is highlighted. In language maintenance, the direction of impact is L2 > L1 (generally referred to as ‘borrowing’) and L1 > L2 in language shift (‘interference through shift’). In these two processes, different subsystems of language are affected. Synchronically, however, it is difficult if not impossible to say which process will eventually prevail. Communities are not homogenous and, especially in new sociolinguistic situations,

88

Anna Verschik

some segments of a community or individuals may be oriented at language maintenance and others at language shift. It has to be noted though that change starts at the level of individual speaker. Weinreich (1953) was right when he stated that the locus of language contacts is bilingual brain. Not only that a community may be fragmented as far as language maintenance and shift are concerned but also in an individual contact induced language change may occur in both directions, L2 > L1 and L1 > L2. Pavlenko and Jarvis (2002) emphasize, as they call it, bidirectionality of transfer in bilingual individuals. For the post-Soviet language contacts, the directionality of impact is especially relevant. What was the impact of Russian on other languages? Is it continuing or has the direction changed in post-Soviet countries? What can be said about directionality in different speakers? At times, speaking of directionality, it appears difficult to avoid politicization and ideological connotations. For example, Leinonen (forthcoming) describes the state of research on KomiRussian contacts and observes that mostly Komi linguists discuss ‘purity’ of Komi and how Russian influence can be resisted. In not a very distant past, Hint (1991) was justly concerned with the Soviet ethnic and language policy in Estonia but dramatized the impact of Russian on Estonian. As studies by Ehala (1994) and Hasselblatt (2000) demonstrated, Hint’s (1991) claims are not substantiated. Research on the directionality of change, based on data from postSoviet contact situations, may have implication for theory as well because in some countries previously monolingual Russians are gradually becoming bilingual and researchers can observe the dynamics of contacts from the point zero, so to say. The issue of directionality may become complicated in the case of multiple contact situation where, on the one hand, titulars have some competence of Russian as L2 and, on the other, Russians start using the titular language as L2 (for Estonia see Verschik [2007a]). Probably, this type of contact situation is especially characteristic of some post-Soviet countries because of the reverse of language status, power relations and language policy after the disintegration of USSR. I am not aware of any similar studies from the Western context, although cases of spontaneous acquisition of minority language by majority language speakers have been reported (Auer and Dirim 2003; Dirim and Hieronymus [2003] on Germans acquiring Turkish). (3) Whether and how the collapse of USSR has affected the process and the outcome of language contacts? It is clear that many contact situations originate from pre-Soviet times. However, the whole post-Soviet landscape, so to say, presents a sociolinguistic constellation previously unknown in the Western sociolinguistics. While Russian minorities in the classical sense existed and do

Contacts of Russian in the post-Soviet space

89

exist in the post-Soviet space, for instance, Old Believers in the Baltic countries, by and large Russians cannot be equalled with minorities and immigrants in the conventional sense (see Pavlenko [2008a] on terminology problems; also see Ozolins [2003] on the Baltic countries). In the East, mostly scholars with background in Russian linguistics consider Russians in the post-Soviet countries and, in general, abroad as a diaspora (Külmoja 1999, 2000; Mustajoki and Protassova 2004). However, considering all varieties of Russian in the post-Soviet context and elsewhere outside Russia as diaspora varieties may be counterintuitive in the view of sociocultural heterogeneity of Russian-speakers: can descendents of Russian post-revolution émigrés in France, Russian-speaking Jews in Germany, Kazakhs who have shifted to Russian during the Soviet era and Russian-speakers who viewed the Baltic region as a part of their country and settled there without any attention to the local languages, be characterized as speakers with the same profile? During the Soviet era, Russians were mostly monolingual in certain republics (Estonia, Latvia) but not necessarily everywhere. In Ukraine, due to the close relatedness of the varieties and exposure to Ukrainian, many Russian-speakers had at least a passive command of Ukrainian (Bilaniuk and Melnyk 2008: 345, 346– 347; Vakhtin et al. 2003). In Kazakhstan, Ukraine and some other republics, Russian was de facto considered more prestigious than the indigenous languages, so that urban elites tended to shift to Russian and send their children to Russian-medium schools; however, this was not the case in the Baltic republics where a silent but strong resistance to Russian as an ‘intruder’ language existed throughout the years of the Soviet rule (Raun [2001: 210] on the reported decline in Russian proficiency among ethnic Estonians in the census of 1979). Now the situation has changed in the terms of language policy and language status (on various versions of de-Russification in the post-Soviet countries see Pavlenko [2008a, 2008b]) but the question remains: what has changed in the linguistic behavior and repertoire of individual speakers? In Section 2, three case studies will be addressed: contacts of Russian and Ukrainian, Kazakh and Estonian. The reasons for choice of these particular cases will be presented and general points of comparison outlined. Section 3 of the current review comprises conclusions and possible topics for future research.

2. Review on case studies In what follows, I am going to focus on the contacts of Russian in three countries: Ukraine, Kazakhstan, and Estonia. First, some general considerations will be presented and then the rationale for the choice will be explained.

90

Anna Verschik

As stated before, the field of post-Soviet contact linguistics is new and so far there have been isolated attempts, albeit often successful and meaningful, to deal with aspects of contact-induced change. Apparently, it is impossible to give a systematic account on all existing literature because: (a) it is written in many different languages (Russian, but also Estonian, Lithuanian, Ukrainian, etc.) and often published in peripheral periodicals rather than mainstream linguistic journals, which makes it difficultly accessible; (b) many researchers in the post-Soviet countries are not aware of modern contact linguistic theories and their research is either too descriptive (i.e. lots of isolated examples like lists of loanwords or instances of ‘interference’ presented without any connection to theories); or they attempt too high a level of generalization, for instance, judging on the processes in Russian in the whole post-Soviet territory without ample empirical data (see reference to Mechkovskaya [2005] in the next paragraph); (c) purist focus of some research, i.e., the main concern is either un-doing Russification or the necessity of following the norms of Standard Russian outside Russia (Zharkova [2006: 52] on “violation of the norm” in the speech of Russians in Lithuania). To render it in a slightly different manner, diaspora approach (see more details in Verschik [forthcoming]) displays some interest towards the regionalization of Russian in the new circumstances, i.e., what is in the terms of contact linguistics sometimes labeled as divergence. Some authors make rather categorical claims about all varieties of Russian in all post-Soviet countries (“Regional changes in the literary Russian language so far do not contradict the norms of Russian” [Mechkovskaya 2005: 60]). However, given the fact that still there is a serious lack of systematically collected data and empirical research, it is too early for any such generalizations. Moreover, empirical data from Estonia’s Russian (Verschik 2004, 2008; Zabrodskaja 2009, forthcoming a, forthcoming b) show that there are indeed constructions conventionalized in local Russian that are at odds with monolingual grammar rules. Divergence approach has its strong sides but, if not balanced with case studies on particular contact situations, there is a risk of overlooking relevant evidence, processes and results. Regionalization, i.e., emergence of regional varieties of Russian, cannot be fully explained exclusively by the dispersion of Russianspeakers, emergence of new state borders and new language policies. Contacts with different languages and, in some cases, convergence towards these languages (see especially Verschik [2006]; Zabrodskaja [2009] on Estonia’s Russian), different patterns of bilingualization of Russian-speakers/titulars and

Contacts of Russian in the post-Soviet space

91

complexity of language contacts though the time span (like in Russian-Ukrainian case) are ‘responsible’ for regionalization not in the least measure than divergence by the means of population dispersion and establishment of new state borders. Significantly, purism, whether understood as de-Russification and stigmatization of ‘mixed’ varieties like Surzhik (literally, ‘a mixture of wheat and rye flour’ that was considered somewhat of low quality, see discussion below) or as strive towards the maintenance of Standard Russian in the post-Soviet countries in the face of impact of other languages, largely determines research climate. An essentialist view on monolingual ‘un-contaminated’ pure varieties (a case of monolingual bias, see Auer 2007; Ellis 2008) interferes with research goals and sometimes even the legitimacy of research on ‘inappropriate’ language phenomena is questioned (see discussion in Stavic’ka 2001). Despite the limitations that do not yet allow for a comprehensive, all-inclusive review on the post-Soviet language contacts, some recent research results can and should be compared and generalized. Three cases will be considered: contacts between Russian and (1) Ukrainian, (2) Kazakh, (3) Estonian. Below I will provide an explanation for such a choice. First, quite interestingly, David Laitin (1998) chose the same countries plus Latvia for his much quoted monograph. One of his reasons for picking exactly these cases was a considerable number of Russian-speakers in these countries, both ethnic Russians (like in Estonia and Latvia) and a high number of Russified titulars (Kazakhstan, Ukraine). Nauruzbayeva (2003) in her short review on language revival policies refers to the same cases (minus Latvia). I am not including Latvia in the current review article because, unlike in Ukraine, Kazakhstan and Estonia, no modern microsociolinguistic research on language contacts has been produced there and nothing has been written in any systematic way on say, Latvian-Russian code-switching or contact-induced change in Russian of Latvia. Instead, researchers there are preoccupied with language policy and macrosociolinguistic studies. There are similarities and differences between these countries as far as their Russian-speaking populations are concerned. In all the three countries, these populations are socially and culturally heterogeneous (in that respect social scientists emphasize the Estonian case; see Vihalemm & Masso [2002: 85]; other references in Verschik [2005a]); thus, the label ‘Russian-speakers’ is abstract and general. For instance, Ukraine and Kazakhstan have a considerable number of Russified titulars who clearly are Russian-speakers (Pavlenko 2008a: 283, 291). This happened because Russian was associated with modern education, high culture and urban life there. Ukrainian had had necessary registers but these fell into disuse for various reasons (Masenko [2004] discusses Russification in

92

Anna Verschik

corpus planning in the 1930s and the decreasing prestige of Ukrainian; see also Bilaniuk [2005: 90–91]). Kazakh, however, lacked important modern registers (academic writing, higher education, science, administration, etc.) because there had been too little time for urban modern culture in the national language to develop before Russian became omnipresent. Thus, Soviet language hierarchies have been internalized by certain titular elites, and often covert but consequent promotion of Russian at the expense of other languages lead to the loss of domains to Russian. This did not happen in Estonia and in the Baltic countries in general: the response to Soviet language policies, demographic Russification and pressure was a voluntary separation and segregation of Estonian-speakers, coupled with some functional EstonianRussian bilingualism for those who needed it because of the character of their occupation. The three cases illustrate that the terms ‘diaspora’ or ‘beached diaspora’ (Laitin 1998: 29) is too broad: some Russian-speakers have never left their homeland but shifted the language instead (for instance, urban Kazakhs in Kazakhstan) and thus differ from ethnic Russians who settled outside Russian Federation and felt no obligation to study the local languages (Russians in the Baltic countries). Probably, what is common now to all three countries is differentiation of Russian-speakers as far as their varieties of Russian, bilingualization, and civic identity are concerned. The three countries demonstrate that the ethno-demographic makeup of the population, while important, is not the only crucial factor that determines the outcome of language contacts. Ethnic and linguistic demarcation lines do not necessarily co-inside, as it is especially visible in Ukraine (see discussion in Section 2.1). Second, contacts of Russian with Ukrainian, Kazakh and Estonian have enjoyed in-depth monographic treatment, whereas the discussion is solidly grounded in modern sociolinguistic/contact linguistics theories. The existence of such scholarship provides a reliable basis for an overview and discussion. The monographs are all based on field work and real spoken language data and present a list of potentially relevant questions for further research and theory development: contacts between closely related languages, language ideologies and conversational practices in Ukraine, as well as a typology of Surzhik (Bilaniuk 2005); grammatical aspects of Kazakh-Russian code-switching and challenges to Matrix Frame Language model (Muhamedowa 2006); pragmatics of RussianEstonian code-switching among Russian-speaking schoolchildren (Zabrodskaja 2005) and contact-induced language change in Estonia’s Russian with a focus on non-monolingual constructions and code-copying framework (Verschik 2008). Third, contacts of Russian in these countries provide insightful data for comparison. Unfortunately, comparative glance at language contact processes in the

Contacts of Russian in the post-Soviet space

93

post-Soviet space is still limited because of the focus on language policy and challenges for education systems. In any case, post-Soviet language contacts are in need of a comparative research as much as other aspects, such as language policies, language management, etc., do. The countries have a different history of incorporation into Russian Empire/Soviet Union (see more details in Pavlenko [2008a]). Kazakhstan was a colony that became independent in 1991 for the first time in its history. Ukraine presents a less clear case but is viewed by some scholars, notably, Masenko (1999, 2002, 2004) in the terms of post-colonialism (see overview on Ukrainian history in Bilaniuk and Melnyk [2008: 347–349]). Eastern Ukraine voluntary submitted to the Russian crown because of unwillingness to remain under the Polish rule (hence the union treaty of 1654), whereas Western Ukraine belonged at first to Lithuanian-Polish commonwealth and after the partition of Poland in the late 18th century became a part of Austro-Hungarian Empire. Noteworthy, the two empires had different language policies vis-à-vis Ukrainian (toleration and even promotion in Austria, belittling and limitation of domains in the Russian Empire). After World War I, Western Ukraine was mostly under the Polish rule and some minor territories belonged to Romania, Hungary and Czechoslovakia. Western Ukraine was annexed by USSR in 1939 after Molotov-Ribbentrop pact. Eastern Ukraine enjoyed a short-lived independence in 1918–1920, to be soon incorporated into the Soviet Union. While in the 1920s a policy of indigenization (Ukrainization) was implemented, it was effectively undone in the 1930s (Bilaniuk 2005: 84–93; Taranenko 2007: 121–125).1 The differences in political history of different regions in Ukraine have far-fetched consequences for sociolinguistic profiles of the population and regionalization that is significant today. Estonia was conquered and ruled by different powers in its early history, ending up as the result of Northern War as a province within the Russian Empire in 1709. It was a fully independent state between 1918–1940 with Estonian as the official language functioning in all institutions and domains. It was a country with a fairly homogeneous population (before 1940, ethnic Estonians made up 88.8% of the population). Estonia was occupied and annexed by USSR in 1940 and after World War II demographically Russified by central authorities’ policy that facilitated migration of Russian-speakers (for a review of the sociolinguistic history of Estonia see Rannut [2004, 2008]; Verschik 2005a). Given all this, one should ask how different socio-political history has affected language contact processes and outcomes. Fourth, theorizing on correlation between sociolinguistic factors and structural characteristics of languages in contact has remained as relevant in contact linguistics as it was two decades ago when Thomason and Kaufman (1988) pre-

94

Anna Verschik

sented their first systematic generalization based on a solid body of case studies. While sociolinguistic factors themselves do not directly cause contact-induced language change, they set into motion other forces, such as spread of individual multilingualism with a wide range of consequences: purely linguistic, cognitive, pragmatic and social (Backus 2005). At the same time, structural and typological similarities are known to facilitate contact-induced language change (Clyne [2003: 162–175] on facilitation). For instance, Surzhik is a cluster of contact varieties that are a result of Russian-Ukrainian contacts. One may ask what role the proximity of varieties in question plays in the emergence of contact varieties (see further Section 2.2). Fifth, the contacts of Russian in the three countries provide a fruitful ground for the inquiry into the afore-mentioned directionality of contact-induced change. Is the impact unidirectional? Do pre-1991 trends prevail or not? Does attitudinal delay, i.e., following earlier practices and adhering to earlier language ideologies and hierarchies etc. play a role? The three situations will be reviewed in the following aspects: – overall sociolinguistic setting, status of languages; – distribution of Russian-non-Russian bilingualism: who is bilingual, who recently became bilingual and for what reasons; – regional differentiation and its significance for bilingualism and language contacts; – directionality of contact-induced change; – contact varieties and purity discourses (if applicable); – practices of bilingual communication; – relevance of a particular case study on contact-induced language change for general theory; – other languages, for instance, English, entering the scene.

2.1.

Ukraine

Ukraine presents a fascinating case where two closely related varieties, Russian and Ukrainian, are in a long-term contact, while sociolinguistic circumstances keep changing. Ukrainian, together with Belarusian and Russian, forms the EastSlavic branch of Slavic languages. As mentioned above, due to the differences in political and cultural history between East and West, the language situation of Ukraine is far from homogeneous. Restrictions against the use of Ukrainian in the East began already in the 17th century and under various pretexts continued till the collapse of the Soviet Union (Bilaniuk 2005: 74–76, 79–80, 84–93). Ukrainian was overtly banned

Contacts of Russian in the post-Soviet space

95

and excluded from education and public sphere under the Tsarist rule. Under the Soviet regime, after a short-lived policy of indigenization, Russification was systematically implemented in corpus planning in the 1930s as a means of fighting against ‘bourgeois nationalism’.2 The Russification became even more intense with the influx of Russian-speakers into industrial areas of South-East during the Soviet era (Donbass region) and introduction of compulsory instruction in Russian at school since 1938 (Bilaniuk 2005: 90). As of 1987, more than half of Ukrainian pupils attended Russian-medium schools (Bilaniuk and Melnyk 2008: 348). Gradually, Ukrainian acquired low prestige even among ethnic Ukrainians and was associated with backward, rural, uneducated speakers. For Western Ukraine, it was a different story. Ukrainian was never banned under the Austro-Hungarian empire, although in some regions such as Transkarpathia its usage was more restricted compared to Bukovina and Galicia (Bilaniuk 2005: 75). In the interwar period, some restrictions and relatively low prestige notwithstanding, Ukrainian was able to develop relatively freely (Bilaniuk and Melnyk 2008: 348). All this created a very different type of language attitudes and even under the Soviet regime Ukrainian enjoyed a covert prestige in the West, making the subsequent post-Soviet Ukrainization of public sphere smooth. Statistics provided by Bilaniuk and Melnyk (2008: 342) demonstrate the uneven distribution of Ukrainian and Russian population throughout the country, Ternopil oblast being the most Ukrainian region (98% Ukrainians, 1% Russians) as opposed to Donets’k oblast (38% Ukrainians, 57% Russians). The ethnic composition of the population in Ukraine is as follows. According to 2001 census, the whole population is 48.5 million. 78% identify themselves as Ukrainians and 17% as Russians. However, 68% indicated Ukrainian as their mother tongue and 30% Russian (http://www.ukrcensus.gov.ua/eng/results/ [accessed 10 December 2009]). While in comparison to the last Soviet census of 1989 the proportion of ethnic Ukrainians and the number of Ukrainian-speakers has increased by 5% and 3% respectively, this development may be accounted for by changes in civic identity but also in cultural and linguistic identity (Bilaniuk and Melnyk 2008: 342). The present debate on Ukrainian is complicated by the fact that the promotion of Ukrainian is accompanied by purist ideology, aimed at de-Russification (both in status and in corpus planning) and undoing the consequence of the historical subjugation to Russia.The stigmatization of ‘impure’language and non-standard speech, however, creates linguistic insecurity in many speakers (Bilaniuk and Melnyk 2008: 346). A rather useful overall description of Ukrainian-Russian contacts has been provided by Taranenko (2007), whose article belongs to the special issue of In-

96

Anna Verschik

ternational Journal of the Sociology of Language on contacts between Slavic languages. From a historical point of view, Taranenko (2007: 121) mentions bidirectional influence up to the middle of the 18th century and the later unidirectional impact of Russian on Ukrainian. Thus, Taranenko (2007) touches upon directionality, albeit implicitly, from a historical perspective. Starting from the 1930s, Russification of Ukrainian corpus became a dominant tendency in language planning (Bilaniuk 2005: 86–87); however, the result is more complex than just a Russified version of Ukrainian. Indeed, directionality in Ukrainian-Russian language contacts is a highly entangled issue. In addition to Standard Russian and Standard Ukrainian, there are regional varieties of both languages that bend more in one or the other direction and contact varieties exist as well (as we shall see, there are different kinds of Surzhik). The borders between varieties cannot be easily drawn. For instance, Bilaniuk (2005: 118) refers to a case where a Ukrainian linguist admired beautiful Ukrainian of his co-traveller and the latter answered that she was speaking Belarusian and her native village was in Belarus (although her variety was very close to Standard Ukrainian, some regional traits notwithstanding). Pavlenko and Melnyk (2009) found in their experimental research on cross-linguistic influence among students in Kyiv that in certain instances it was impossible to determine what element belongs to what language. Taranenko (2007: 126–127) describes the state of dynamic balance in border regions between Ukraine and Russia (see also Chorny [2006] and Zhurzhenko [2006] on post-Soviet language preferences and self-identification in border regions). Historically, Ukrainisms have permeated South Russian dialects on all levels (lexicon, pronunciation, word formation, etc.), and it is often impossible to assign an element to either language (see the digest of an early and rather exceptional study by Chizhikova [1968] in Bilaniuk [2005: 115–117]). Ukrainisms have been conventionalized in varieties of Russian outside Ukraine as well, mostly being a feature of casual speech, sometimes communicating irony and sometimes signaling low style. Interestingly enough, the ethnic coloring may have disappeared for the speakers so that they do not identify the source of the items they are using. Lexical and semantic Ukrainisms are also present in the ethnolectal speech of some Russian Jews who unknowingly introduce Ukrainisms that are present in also in Yiddish (Verschik 2007 b: 227). Even today when Ukrainian is the only official language (with the exception of Crimea, which is an autonomy within Ukraine and where, for various reasons, Russian is the second official language, see Bilaniuk and Menlyk [2008: 342]), various authors point out that the situation of Russian-Ukrainian bilingualism is far from stable (Taranenko 2007: 119). Bilaniuk and Melnyk (2008: 346) claim that ethnic Ukrainians are 2.7 times more likely to practice bilingualism than

Contacts of Russian in the post-Soviet space

97

ethnic Russians. Although the majority of the population is bilingual (including those who at least passively understand the other variety) and only a minority is completely monolingual Ukrainophones/Russophones (Vakhtin et al. 2003), the asymmetrical character of bilingualism has been still retained. Still, one cannot say that the promotion of Ukrainian and attempts to raise it prestige since 1991 have not brought about any changes in addition to the census statistics. While some authors, especially Masenko (2004, 2005, 2008) are inclined to speak about inequality of languages and de facto domination of Russian in mass media; others, like Vakhtin et al. (2006) have registered important changes in language attitudes even in areas with a prominent number of Russian-speakers (Kyiv and Kharkiv). They found that correspondence between language proficiency and support of the current language policy is not straightforward: the informants who are not fluent in Ukrainian and for that reason continue using Russian or versions of Surzhik, nevertheless, support the language policy and feel that Ukrainian should be the sole official language of the country. Field work data show that such a discrepancy is not conflicting: the same speaker can harmoniously use languages other than Ukrainian and yet favor the present Ukrainization policy. According to Vakhtin et al. (2003: http://old.eu.spb.ru/ethno/projects/project3/ukraine/004.htm), the present sociolinguistic situation of Ukraine can be characterized as follows: (1) almost equal segments of population speaking Ukrainian and Russian, whereas linguistic and ethnic borders are not identical; (2) territorial and social polarization of linguistic preferences; (3) declarative and often ambiguous statements in legislative acts. This leads us to the importance of empirical data on spoken language. While the concerns about the consequences of the Soviet Russification, certain retention of low prestige of Ukrainian among some ethnic Ukrainians, and problems of acquisition planning (especially in Crimea where only recently Ukrainianmedium schools have been established, Bilaniuk and Melnyk (2008: 342) are understandable and legitimate, these should not obscure the fact that, in reality, linguistic repertoire includes much more than just Russian, Ukrainian and Surzhik, the stigmatized contact variety. Although recently the interest towards Surzhik has increased, among Ukrainian scholars this was mostly dictated by the considerations of ‘purity’, ‘beautiful standard language’and the fear that Surzhik is a transitional stage on the way to ‘assimilation’. As Vakhtin et al. (2006: 200– 201) note, the question is often over-politicized and the fact that many speak Surzhik as L1 is disregarded or such speakers are viewed as deficient. In this respect, the project “New languages of the new states” of European University at St. Petersburg (Vakhtin et al. 2003), dedicated to field research on contact varieties (Surzhik in Ukraine and Trasyanka in Belarus), clearly dif-

98

Anna Verschik

fers from the mainstream research done by scholars in Ukraine. The researchers concur with Orlov ([1997: 80] quoted from Vakhtin et al. [2003]) who, speaking of the situation in Ukraine, arrived at the conclusion that the set of identities legitimized by the society is often in contradiction with the existing latent self-identification. This explains the mentioned fact that while many continue using Russian and Surzhik, they strongly support the policy of promotion of (Standard) Ukrainian. Vakhtin et al. (2003) argue that in the situation of close contacts between the two groups self-identification is complex and non-linear. In ethnic auto-stereotypes of Ukrainians, the importance of common customs, religion, tradition, and way of life prevail, whereas Russians tend to emphasize the importance of their language, culture and features of character. Ethnic Ukrainians form a heterogeneous assemblage of bi- and monolinguals, Ukrainian- and Russian-speakers and also Surzhik-speakers. The researchers justly emphasize that sociological surveys cannot show what languages a respondent really speaks: for instance, surveys do not identify groups whose linguistic repertoire is Ukrainian + Russian + Surzhik, Ukrainian + Surzhik, Russian + Surzhik, etc. The close relatedness of the varieties enables a relatively easy acquisition of passive competence that is often disregarded by researchers but which, according to Thomason (1997) is one of the mechanisms of contact-induced language change. Surzhik is a passionately discussed and debated subject (see Tarasenko [2006] for a summary of different views). Several ‘good language’ guides aimed at the eradication of ‘Surzhikisms’ have appeared since 1991 (Hnatkevich 2000; Serbens’ka 1994). Surzhik is highly stigmatized for a variety of reasons, for instance, because it is usually perceived as a hallmark of Soviet Russification and a step away from Ukrainian towards Russian, and because it is ‘neither one, nor the other’, that is, its legitimacy is disputed precisely because it is a contact variety and its relation to the ‘parent varieties’ is transparent to a layperson.3 Sometimes it is stated that code-switching between Ukrainian and Russian is legitimate because different languages are used in different situations for different functions but Surzhik does not fulfil these criteria and, thus, is not viewed as code-switching (Stavic’ka 2001). Elements of Surzhik are not ‘well-formed’ in the terms of two monolingual grammars. Such an attitude, when forms that cannot be attributed to monolingual varieties are not tolerated, exemplifies monolingual bias. The emotional load and negative connotations of the very term Surzhik become explicit if one looks at metaphors and comparisons applied to it. Clearly, Surzhik is a result of bilingualism, but in post-Soviet countries with the strong memory of the Soviet language policy that promoted bilingualism among nonRussians the very concept of bilingualism has acquired negative overtones in

Contacts of Russian in the post-Soviet space

99

public discourse (and also among some Ukrainian scholars). Sometimes the whole polemics reminds of old discussions on bilingualism where bilinguals were typically portrayed as people with split personalities and loyalty/identity conflict. Loaded metaphors and keywords appear both in popular writings and in scholarly literature. Surzhik is described as deformed, maimed, chaotic, poor language, a sign of the speakers’ low culture and ‘spiritual plebeianism’ (Tarasenko [2006: 64] for a review). It is interpreted as a sign of national schizophrenia, intellectual handicap and identity crisis (people who are ‘neither one nor the other’; see Vakhtin et al. 2003: http://old.eu.spb.ru/ethno/projects/project3/ ukraine/007.htm). For instance, Masenko (2002: 13) refers to Surhzik as to a “chaotic juxtaposition of superficially acquired Russian elements onto deformed Ukrainian structure”. However, in my view, the most marked metaphor has been coined by Yuryii Andrukhovich, a prominent Ukrainian writer: he calls Surzhik an “incestuous child of bilingualism” (кровозмiсне дитя двомовностi, quoted from Stavick’ka [2001: 20]). Thus, in this anthropomorphic metaphor, not only Surzhik is impure and contaminated by definition but, in view of Andrukhovich, this is a particularly ugly type of hybrid because of the relatedness of the two ‘parent’ varieties. Thus, emotional load, politicization of the subject and purism rather obscure the research agenda. Vakhtin et al. (2006: 200) describe even how at a conference in Ukraine they were viewed as ‘outsiders’, although, in fact, they are not (all have spent a considerable part of their life in Ukraine or continue living there; two are ethnic Ukrainians who graduated from Ukrainian universities, etc.). Although the scholars do not refer to Bilaniuk’s (2004, 2005) research and her typology of Surzhik, they also argue that, contrary to Stavic’ka (2001) and Masenko (1999, 2004), Surzhik is not necessarily a transitional stage on the road towards Russian because there are empirical data of the opposite linguistic behavior (see also Kuznetsova [1999: 9–10] who found in her study on bilingual families that speakers of Surzhik as L1 were capable of mastering Standard Ukrainian). Those whose work obligations presuppose the use of Ukrainian (teachers, workers in the public sectors, TV hosts and journalists, etc.) do their best, even if the final result is not necessarily monolingual Standard Ukrainian. Thus, the researchers confirm Bilaniuk’s findings on Russian-based Surzhik type (urban bilinguals’ Surzhik and post-Soviet Surzhik). In fact, Masenko (2002, 2004) never explains why Surzhik is a transitional stage on the way to Russian and not vice versa. Stavic’ka (2001) admits that theoretically a shift from Surzhik to Ukrainian is possible but not highly probable in the view of power and prestige of Russian. At the same time, during the field work in Kyiv and Kharkiv, summarized in Vakhtin et al. (2003), no

100

Anna Verschik

interdependence between acquisition of Russian and knowledge of Surzhik was confirmed. It was established that the informants proficient in Surzhik often learned it as L1. Numerous Ukrainian scholars are concerned with the linguistic classification of Surzhik, asking whether it is a jargon, sociolect (Flier 2000), hybrid sociolect (Taranenko 2007: 125), mixed language, regional dialect, pidgin (Masenko 2004), a continuum of Russified dialects (Mechkovskaya 2005: 52), etc. Masenko (2002: 13) and Stavic’ka (2001) argue against classifying Surzhik as a regional dialect or a sociolect. Masenko (2002) argues that a sociolect emerges within one language and jargons appear because of the speakers’ wish to be different from the rest of population. It is not clear what ‘one language’ exactly means and whether slang/jargon is always conditioned by the wish to stand out. Both Masenko and Stavic’ka draw attention to the lack of urban colloquial registers in Ukrainian. To a degree, Surzhik compensates for this gap (also Kuznetsova [1999] found that many speakers emphasize the expressiveness of Surzhik). Stavic’ka (2001) claims that the explanation lies in the fact that urban subculture of Ukraine has emerged as a Russian-language subculture and, to some extent, elements of casual urban register come from Russian and, being added to Ukrainian, form Surzhik. Similarly, Trub (2000) points at the lack of an everyday urban register in Ukrainian analogous to Russian prostorechiye (‘casual uneducated / non-standard speech’). However, even this does not legitimize Surzhik for Stavic’ka (2001). According to her, Surzhik is a dead end of linguistic development because, unlike urban slang, it lacks creativity. This claim is not substantiated; after all, linguistic creativity is a characteristic of speakers, not varieties. What’s more, Bilaniuk (2005: 159–165) dedicated an entire section of her book to the subject of creative use of Surzhik. For Stavic’ka (2001), Surzhik exemplifies linguistic deficiency (неповномовность; literally, ‘incomplete language’) which produces ‘simplified thinking’. In this discussion one can see that researchers, literati and public figures are worried about the consequences of asymmetrical bilingualism; still, the argumentation appears to be essentialist (only ‘pure’ varieties are legitimate) and monolingually biased (the assumption that varieties can be fullfledged and deficient, etc.). From a scholarly point of view, it is not the question of the legitimacy of Surzhik nor its classification in terms of sociolect, slang or casual register that is relevant. For contact linguistics, if a variety exists, it is legitimate by definition and calls for a detailed study. As for the attempts of classification, comparison with other known contact varieties would be a productive research perspective, but in order to do that more field data and more relating to similar studies elsewhere is needed. As Vakhtin et

Contacts of Russian in the post-Soviet space

101

al. (2003: http://old.eu.spb.ru/ethno/projects/project3/ukraine/007/011.htm) describe, many researchers analyze various texts (including fiction, e.g. Flier 2000) in Surzhik but few analyze real language data. Regional variation in Surzhik has been reported by various scholars (a short overview can be found in Vakhtin et al. 2003: http://old.eu.spb.ru/ethno/projects/ project3/ukraine/007/010.htm). The detailed account by Bilaniuk (2004, 2005: 121–135), based on ethnographic field work in Ukraine, presents different types of Surzhik, encompassing the following dimensions: (1) directionality (Ukrainian-based or Russian-based); (2) the division between urban and rural; (3) chronology (from the days of Russian Empire up to now); (4) position in Auer’s (1999) continuum from code-switching via language mixing towards fused lects. The types of Surzhik are as follows: urbanized peasant Surzhik, village dialect Surzhik, Sovietized-Ukrainian Surzhik, urban bilingual Surzhik and post-Soviet Surzhik. Such a multidimensional approach is probably the most productive one because linear approaches have failed to deal with the complexity of Surzhik. For theory building the discussion on various degrees of conventionalization of ‘switched’ elements in different types of Surzhik (i.e. in the terms of Auer [1999]) are most promising. Sometimes Surzhik is classified in the terms such as ‘mixed language’, ‘mixed type varieties’ (смешаные языковые формы [Vakhtin et al. 2003: http://old.eu.spb.ru/ethno/projects/project3/list.htm]), “pidgin” (Masenko 1999: 28). Masenko (1999, 2004) argues that it is the situation of post-colonialism in the contemporary Ukraine that is similar to other post-colonial situations elsewhere and Surzhik may thus be compared to pidgins. Although post-colonialism is actively discussed in connection with the post-Soviet context and this approach may prove productive in many respects (see Ozolins 2003; Pavlenko 2008a: 300), an automatic copying of post-colonial realities onto the Ukrainian case is not justified. One of the conditions for pidginization, namely, the lack of a common language at the first moment of the contact, is not fulfilled here. Components and patterns originating from different languages in a contact variety are not a sufficient basis for postulating a pidgin or a creole. Since the varieties in question are closely related, the origin of constructions/items/units is not always easily detectible. As for mixed languages, it seems that the label ‘mixture’ is used impressionistically in the literature on Surzhik without any references to the ever expanding literature on mixed languages (Matras and Bakker 2003). Generally speaking, whether and how continuum proposed by Auer (1999) accounts for the emergence of mixed language in the strict sense of the term, remains to be seen. As Backus (2003) shows, alternational code-switching is unlikely to be conventionalized in the way that leads to lexicon-grammar split that is observed in many

102

Anna Verschik

mixed languages. Because of the close relatedness and high material/structural similarity it is not easy to track whether there is a prototypical grammar-lexicon split (like in Media Lengua) or verb phrase vs. noun phrase originate from different varieties (as in Medny Aleut, Golovko 1994, 2003). In any case, a comparison between contact varieties of non-related languages and those of related languages would be useful to the general discussion on the emergence of contact varieties. A comparison of Surzhik to Trasyanka, a cluster of contact varieties in Belarus (see Giger and Sloboda 2008; Liskovets forthcoming; Woolhieser 2001; Zaprudski 2007), would be a step towards a better understanding of contacts between closely related languages. Similar linguistic features notwithstanding, it appears that in the post-Soviet period the dynamics of Surzhik and Trasyanka differ to some extent. Both are related phenomena, since both are (clusters of) contact varieties, both are stigmatized as ‘deviant’, ‘neither one nor the other’. Like Surzhik, Trasyanka varies regionally. Liskovets (forthcoming) has investigated Trasyanka in Minsk, the capital of Belarus. She refers to the lack of serious field work and empirical data on Trasyanka. In her opinion, Trasyanka can be qualified as a ‘mixed code’ but not a ‘mixed language’ because what is called grammar-lexicon split is not the case here (Liskovets forthcoming). Overall, the informants’ attitudes towards Trasyanka affect attitudes towards Standard Belarusian; for some speakers Belarusian has an extremely low prestige and many informants confuse Trasyanka and Belarusian, erroneously identifying Trasyanka speech as Belarusian. This may be true of Surzhik as well in the context of stereotyped presentation of Ukrainian in the guise of Surzhik for Russian-speaking audience in Russia (see Bilaniuk [2005: 165–172] on the comic persona of Verka Serduchka). There are voices for legitimization of Surzhik (Doni¨ı [2003] argues that attacks on Surzhik would push its speakers towards Russian; also see the summary in Tarasenko [2006: 62]), but, to the best of my knowledge, nothing of the kind is true of Trasyanka. It appears that, at least in Minsk, there exists only Belarusianbased Trasyanka of rural migrants but not a Russian-based version (i.e. a move towards Belarusian). This differs from the situation of Surhzik which, as we have seen may also be a step towards Ukrainian. Such a state of affairs is probably connected to the socio-political differences between the two countries: low status of Belarusian, marginalization of Belarusian-speaking elite, limited usage of and lack of practical necessity to speak the language, which is unlike the post-Soviet Ukraine. To date, Bilaniuk (2005: 175–177, forthcoming) is probably the only scholar who has investigated Ukrainian-Russian multilingual conversation. She discovered a practice of non-reciprocal, or non-accommodating, bilingualism, where

Contacts of Russian in the post-Soviet space

103

a Russian-dominant and a Ukrainian-dominant speaker interact in his/her respective languages (see also description in Bilaniuk and Melnyk 2008: 358). This conversation strategy is usual for TV shows; according to Bilaniuk, it helps avoiding the stigmatized mixing and thus diminishes tension between ideology of purity and official requirements on the one hand and, on the other, language proficiency. However, the avoidance of ‘mixing’ is a way of preservation of essentialist view on identity, i.e., people should not be ashamed of who they are and what they speak, at the same time rejecting the ‘imperfect’ non-native language use (Bilaniuk and Melnyk 2008: 358). In the terms of Auer’s (1995, 1998) pattern of multilingual conversation this practice corresponds to A1 B2 A1 B2, where the letters refer to languages and digits to the speakers. The practice of communication where each speaker adheres to his/her first language (with possible accommodation for the better understanding) has recently become known under the label of receptive multilingualism (Zeevaert and ten Thije 2007; Rehbein et al. forthcoming). Receptive multilingualism is not reserved exclusively to closely related languages because structural and material similarity is but one component necessary for understanding. Potentially, Ukrainian-Russian multilingual conversation can provide exiting empirical data for a study on receptive multilingualism. Interestingly, the same communication practice has been registered in Estonian-to-Russian conversation as well and, among other things, is characteristic of bilingual or monolingual (at least in theory) Russia-language TV shows but also of interaction over the counter. However, the sociopragmatic meaning of the same pattern is different in Estonia, as it is not a strategy of avoiding ‘mixing’ (which is not stigmatized). The Estonian version of ‘non-accommodating bilingualism’ will be discussed in the Section 2.3 on Estonia. New languages, such as English, appear on the sociolinguistic scene of Ukraine. It is ambiguous whether English as a subject in schools of Ukraine is called to replace Russian as a language of wider communication; whether it stands for Western values as opposed to the Soviet ones, or whether considerations are utilitarian (better jobs, better mobility, ease of international communication), or everything together. In any case, Bilaniuk (2005: 181–185) and Bilaniuk and Melnyk (2008: 363–365) point out that English is very much present in the linguistic landscape of Kyiv where “Ukrainian and Russian are often joined by English in commercial labeling and signage”. 2.2.

Kazakhstan

Kazakh is a Turkic language belonging to the Quypchaq (or Kipchak) branch. Typologically, it is an agglutinating SOV language. Russian and Kazakh differ ty-

104

Anna Verschik

pologically to a great extent but both languages are rich in inflectional morphology. This is one among several reasons why an investigation of Kazakh-Russian language contact can provide a substantial pool of empirical data relevant for theory development. Unlike Ukraine and Estonia, Kazakhstan never was an independent country prior to the collapse of USSR in 1991. Formation of the Kazakh people started in the 15th century when several tribal groups were assembled under the rule of a khan. In the beginning of the 18th century some of the tribes were not able to cope with the military threat from the tribe of Zhungars and had to seek military protection of the Russian Empire. As Smagulova (2008: 441) renders it, “this was the beginning of a long process of inclusion of the Kazakh tribes into the Russian Empire.” Administrative-politic reforms introduced by the Russian administration gradually lead to the transfer of power from Kazakh traditional elite to authorities imposed by the Empire. Demographic Russification of Kazakhstan started already in the 19th century (Smagulova 2008: 442), so that the share of Slavs increased from 14.7% in 1897 to 28.5% in 1911, reaching 30% in some regions. Russification continued throughout the Soviet era. Also Kazakhstan was the final destination for some deported ethnic groups such as Volga Germans, Koreans and others. According to the last Soviet census of 1989, Kazakhs made up merely 39.7% of the population, whereas the share of Russians was 38.8% and the rest comprised other ethnicities (Smagulova 2008: 446). Pavlenko (2008a: 283–284) mentions multiethnic and multilingual composition of the population as an additional factor contributing to Russification, since Russian would be typically used as a lingua franca for interethnic communication. As of 2004, the population of Kazakhstan was 14,951,200, of whom ethnic Kazakhs constituted 57.9%, ethnic Russians 27.24% and other ethnic groups 15.57% (Smailov 2004, quoted from Smagulova 2008: 440). After 1991, the domains of Kazakh expanded considerably compared to the Soviet era. However, Russian maintains its strong position: from 1997 on, Russian can be officially used in administration and other public spheres. As Fireman (2006) shows, Kazakhization of education and public life has created mixed feelings and scepticism among some Russified urban Kazakhs. Currently Russian is an official language beside Kazakh and has also assumed the role of lingua franca in the communication between Central Asian countries (Pavlenko 2008a: 298). Looking at these figures, one can think of other examples of Soviet republics with substantial Russian (Russian-speaking) population. For instance, in the late Soviet period in Latvia, the share of Russians in the population was comparable to that in Kazakhstan (34% according to the census of 1989). However, statistics by itself can be misleading because in the two cases language attitudes of the

Contacts of Russian in the post-Soviet space

105

titulars significantly differed. In Latvia, like in the Baltic republics in general, the ‘cultural superiority’ of Russian was not recognized, the period of independence was highly valued and the response to the Sovietization and Russification was a silent but very real protest. In Kazakhstan, Russian-speakers were not only numerically powerful, but also dominant politically, economically and culturally. Russian-speakers held more prestigious jobs, while Kazakh became associated with rural and unsophisticated way of life. Thus, a high proficiency in Russian became a means of social mobility in Kazakhstan (Fireman 2006: 101–102; Laitin 1998: 71; Smagulova 2008: 444–445). Unlike in the Baltic republics, higher education in Kazakhstan was available in Russian only. Graduates from Russian-medium school had better chances for finding a prestigious employment; many parents preferred Russian-medium schools because it was believed that the quality of education is higher there. In this way, the number of Russified titulars steadily increased and urban Kazakhs were shifting to Russian. As of 1989, only 1% of Russians claimed proficiency in Kazakh (Smagulova 2008: 445). According to Yessenova (2005: 678) and Smagulova (2008: 445), such language policy and the situation of asymmetrical bilingualism reinforced and reproduced social differences among urban and rural Kazakhs. Smagulova (2008: 444) stresses that Russian schools have longer traditions, better trained teachers and a wealth of teaching materials. Fireman (2006) substantiates this claim (see further in the current section). As mentioned, Russified titulars are a part of the sociolinguistic situation of Ukraine as well; however, one can say that this group of speakers is more visible in the Kazakh context, since the languages are typologically different, and cultural distance is significant, too. In Ukraine, the cultural and linguistic proximity makes borders less marked. Indeed, during the Soviet Russian had more prestige and was considered as a greater asset than the national language. One of the reasons for this course of events was the fact Kazakh had not had sufficient time to develop registers for the domains of higher education, academia, public administration and other important spheres before Russian firmly took the leading position in these domains. Still, in this respect the Kazakh language situation differs from that of Ukrainian because the latter has all necessary ‘high’ registers that at some point fell into disuse and are being restored now. Regional differences in ethnolinguistic composition of the population are prominent in Kazakhstan. North (Uralsk, Akmola, Kokshetau oblasts) and North-East (Pavlodar, Semipalatinsk) are more Russian-dominated, while the Kazakh population is concentrated in the South (Kyzylorda, Shimkent). Probably, the division line between urban and rural population is more significant because ethnic Kazakh urban dwellers are more likely to be Russified titu-

106

Anna Verschik

lars than rural Kazakhs. After 1991, many Germans, Russians and other Slavs left Kazahstan (Fireman 2006: 110). Kazakhs from rural areas started moving to cities (Fireman 2006; Nauruzbayeva 2003; Smagulova 2008; Yessenova 2005) and the repatriation of diaspora Kazakhs (oralman) is encouraged by the state. Speaking of language acquisition, multilingualism and identity, migration should be included as a weighty factor (Block 2007). At present, migration from rural to urban localities may trigger changes in the sociolingusitc setting. Official statistics says little about bilingualism. Muhamedova (forthcoming) refers to Suleimenova et al. (2007: 40) who believe that statistical data are contradictory and often ethnic groups are equaled to speech communities, that is, ethnic Kazakhs are considered as Kazakh-speakers. Although the phenomenon of Russified titulars is well known in the literature, little can be said about the number of monolingual Kazakhs (Muhamedova, forthcoming). In a similar vein, Fireman (2006) discusses complexities of school dynamics (in the terms of language of instruction). In Kazakhstan, there exist Russianmedium schools vs. Kazakh-medium schools vs. mixed schools (not to be mistaken for bilingual or immersion schools: Russian-medium and Kazakh-medium classes are housed in the same building).Although the number of ethnic Kazakhs attending Kazakh-medium schools has increased, nevertheless, in 2003 approximately 20% of Kazakhs studied in Russian medium schools (Fiereman 2006: 107). However, the number of pupils in various schools or universities is not the whole story. Indeed, how can one unambiguously name the language of instruction, if lectures are in Kazakh but the textbooks, especially in technical subjects, are in Russian (as Fireman [2006: 113] reports)? As for mixed school in urban environment, even Kazakh pupils tend to communicate with each other in Russian during the breaks (Fireman 2006: 103). An educated guess would be that in such circumstances one should expect code-switching and other phenomena of non-monolingual speech. In sum, the interest in macrosociolinguistics, language policy and management, language in education has dominated research on Kazakhstan sociolinguistic situation. To date, several comprehensive studies on the language situation are available (Baskakov 1992; Fireman 2006; Pavlenko 2008a, 2008b; Smagulova 2008). At the same time, contact linguistic research based on spoken language data is scarce. Muhamedowa (2006: 2–3) describes the studies on Kazakh-Russian contacts as belonging to two categories. First, there exists a strong tradition of language contact research in Turkology (for instance, Baskakov 1960; Johanson 1997), that is, a diachronic perspective on language contacts at a macro level. The second one is the contrastive approach, mostly focusing on ‘interference’ from Kazakh into Russian of bilingual Kazakhs (Kopylenko 1988; Kopylenko and Saina 1982). If more studies based on field data

Contacts of Russian in the post-Soviet space

107

exist, they are not easily accessible (i.e. they appear in peripheral local periodicals and are unknown to wider scholarly community). Probably, Muhamedova’s/Muhamedowa’s (2005, 2006, forthcoming) is the only microsociolingusitically oriented research on Kazakh-Russian contacts. She has collected ample data on Kazakh-Russian code-switching and bilingual speech of urban Kazakh. So far the direction of influence has been from Russian to Kazakh, which is not surprising in the situation of asymmetrical bilingualism where non-Kazakhs are very unlikely to know any Kazakh. Shaybakova (2005: 193, 195) describes the lexical impact of Kazakh on Russian as an increasing trend and suggests that a regional variety of Russian is gradually emerging (Shaybakova 2005: 205). Probably, the impact of Kazakh as L2 on Russian as L1 is gaining momentum but has not been investigated apart from occasional lists of lexical borrowing (Muhamedova, personal communication 2009). It may be assumed that this is a multiple contact situation where some Russified Kazakhs and Russian-speakers are (re-)learning Kazakh on the one hand and Kazakh-speakers have Russian as L2. Such a multiple contact situation exists in Estonia (Verschik 2007a) and, due to the bilingualization of Russians in the new post-Soviet reality, may be something especially characteristic of post-Soviet language contacts. Muhamedowa/Muhamedova (2006, forthcoming) took the research on Kazakh-Russian language contacts at a completely new level, discussing the grammar of code-switching in the terms of highly influential Matrix Language Frame (MLF) and 4M models (Myers-Scotton [2002], among many other publications). She shows that code-switching is an everyday practice among urban Kazakhs. According to the models, there is a strict division between two grammars: matrix language (ML) provides the morphosyntactic structure of the clause and the inserted, or embedded, language (EL), provides the lexicon, whereas ML and EL constituents are grammatically well-formed (Auer and Muhamedova 2005). Together with Auer, based on Kazakh-Russian code-switching corpus, she demonstrates that the proposed principle for determination of dominant, or matrix language (ML) cannot be upheld: Kazakh-Russian code-switching data show that EL insertions do not always take on the morphology of ML and can even influence ML (Auer and Muhamedova 2005). Also complex insertions are not necessarily well-formed from the point of view of monolingual grammars. One example of this is gender in inserted Russian NP/AP within the Kazakh matrix. Russian and Kazakh have different nominal systems; in the former, nouns have gender and an adjectival modifier agrees with the noun in gender, case and number; in the latter, the gender category is absent. Thus, in monolingual Russian, one would expect the nominative высшая школа [vysˇs-aja sˇkol-a] (higher-ADJ.FEM.NOM school NOM) ‘higher education establishment’

108

Anna Verschik

and the accusative vysˇs-uju sˇkol-u (higher-ADJ.FEM.ACC school-ACC). This is not what happens in code-switched NPs: instead of expected case, gender and number agreement the adjective often has masculine nominative singular ending, as follows (modified excerpt from Auer and Muhamedowa [2005: 43], Russian items in bold script): Высш-ий школ-d¨ı bıtır-d os¨ında. Highest-?? school-ACC finish-3PAST here ‘graduated from a university here’ Probably, masculine adjective suffix has been reanalyzed as a general adjective marker, whereas gender becomes irrelevant. Muhamedowa (2005) shows that the large majority (76%) of Russian ‘non-masculine’EL constituents in Kazakh ML are ungrammatical according to Russian monolingual grammar. This should not be attributed to an ‘incomplete’acquisition of Russian because the same speakers are highly fluent in Russian and only 7% of the adjective gender agreement in the monolingual corpus deviates from the rule. A similar tendency has been reported for other Turkic languages as well: Chamidova (1985) for Uzbek-Russian; Krippes (1994a) for Kyrgyz-Russian and Kazakh-Russian; and Menz (1999) for Gagauz-Russian contact situations. All these cases show that the structrure of ML influences the structure of EL and vice versa, contrary to the predictions of MLF and 4M models. Kazakh-Russian code-switching data have been used by Auer and Muhamedova (2005) for a general discussion on monolingual bias. Noteworthy, an analysis of Russian-Estonian codes-switching data has provided ground for a similar critique of MLF and 4M model. There are instances where ML cannot be assigned unambiguously (Zabrodskaja forthcoming a) because of innovative mixed constructions or because Russian constituents in some cases follow Estonian word order (genitive NP). Again, ‘well-formedness’ of EL and ML items is not always the case. Muhamedova (2006, forthcoming) discusses also code-switched Russian conjunctions and discourse pragmatic words (henceforth DPW). Previously, Johanson (1997) wrote on tendencies of copying of Russian conjunctions into the Turkic languages of FSU. Albeit marginally, he addresses issues of Soviet corpus planning: the increased use of Russian elements, conjunctions among them, was as ‘enrichment’becauseTurkic languages originally had very few conjunctions (Johanson 1997: 117; also Baskakov 1960: 30, quoted from Muhamedova forthcoming). However, Johanson’s is a macro approach and his interest is more diachronic and general than synchronic and concrete. It has been noted in the contact linguistic literature that code-switched conjunctions may trigger

Contacts of Russian in the post-Soviet space

109

a change in the structure (Salmons 1990; Wertheim 2003). The appearance of Russian subordinate conjunctions in Turkic languages has a deep impact on Turkic structure because Russian subordinate clause has a finite predicate and Turkic one contains verbal nouns, participles and converbs. The following example (Muhamedova forthcoming) illustrates how Turkic non-finite predicate functions as a conjunction: Saltanat [erteˇn biz-ge kel-etin-i-n] Saltanat tomorrow we-DIR come-PART.FUT-POS3PSG-ACC ayt-tı. say-PAST3P ‘Saltanat said that she would come to us tomorrow.’ Russian conjunction потому что ‘because’ is one of the most inserted conjunctions in the speech of Kazakh-Russian bilinguals. In some instances, word order in the constituent following the inserted conjunction is at odds with the rules of monolingual Kazakh syntax. This contradicts the principle of MLF, according to which ML must both provide late system morphemes and constituent order (Myers-Scotton 2002: 59). Muhamedova (forthcoming) concludes that ML cannot be clearly defined here and that under the impact of Russian, the genuine Turkic non-finite subordinate clause pattern is the least preferred. Borrowing of DPW has been widely discussed in the literature (Maschler 1994, 1998; Matras 1998; see references and discussion in Wertheim [2003]). Wertheim (2003) has done a thorough research on Russian DPW in urban Tatar, providing also comparative evidence from other Turkic languages. In addition to morphosyntactic restructuring caused by Russian DPW in Tatar, Wertheim (2003: 214–217) claims that borrowing of DPW may be representative of a certain stage of contact-induced language change. This may be characteristic, among other things, of so-called contracting languages like Tatar (i.e. when intergenerational transmission is disrupted). In addition, Wertheim mentions extensive bilingualism, loss of domains to sociolinguistically dominant language (Russian in this case) and notable purity discourse (Wertheim 2003: 84–98). Probably, Kazakh in the urban environment during the Soviet era can be considered as a contracting language; it remains to be seen, whether it has remained so. Again, Russian-Estonian language contact case (Verschik 2008: 151–170) demonstrates that borrowing of DPW may occur in languages like Russian in Estonia that do not satisfy all the criteria suggested by Wertheim. Thus, borrowing of DPW in the post-Soviet context calls for a comparative research.

110

Anna Verschik

Muhamedova (forthcoming) concurs with Shaybakova (2005: 195) that, unlike in Ukraine, Kazakh-Russian bilingualism has not resulted in a mixed variety. Indeed, nothing like Surzhik or Trasyanka has emerged. A relevant question to be asked is why it has not happened. Is typological proximity/genetic relatedness a factor here? What is the degree of habitualization and conventionalization necessary for a contact variety to emerge? The mentioned practices of code-switching are characteristic of bilingual urban Kazakhs. However, I do not know of any research on whether and how Russians are acquiring and using Kazakh and what kind of multilingual conversation patterns exist in their speech. The next issue to be discussed in this section is the connection between de-Russification and orthography. Orthography has recently become a topic of sociolinguistic and contact linguistic research (Angermeyer 2005; Sebba 2006). Orthography, choice and modifications thereof have been discussed in the postSoviet context, especially in the case of Turkic languages. Wertheim (2003: 106 and the rest of Chapter 2) describes in a greater detail the issues of Latin vs. Cyrillic orthography for Tatar. During the policy of indigenization, alphabets were designed for languages that lacked writing tradition. For languages that had used the Arabic script, Kazakh among them, new alphabets based on the Latin script were proposed. At first, a modified Arabic script was proposed for Kazakh in 1924, to be supplanted by the new alphabet based on the Latin script in 1929. In 1940, the script was abandoned in favor of the Cyrillic script, as it happened with other Latin-based alphabets in USSR. Although after the independence in 1991 there were voices in favor of replacement of the Cyrillic alphabet by one based on the Latin script; the opposition to this was rather strong and eventually the issue of alphabet was postponed for an indefinite period (Smagulova 2008: 443). However, the situation is more nuanced than a plain choice between the two script systems. Krippes (1994b) shows how de-Russification occurred in the rendering of the following items: older and newer loanwords from Russian, foreign phonemes and consonant clusters and Kazakh geographical names. Krippes (1994b: 110) discusses how corpus planning of Soviet Turkic languages went hand in hand with orthography planning. For instance, it was decided by Soviet language planners that Russian loanwords should retain their original Russian spelling, regardless of their pronunciation in Kazakh. Thus, nativized spelling of words like порум [porum] ‘form’, багон [bagon] ‘wagon, carriage’, тылыпун [tilipun] ‘telephone’ had to be abandoned in favor of форма, вагон, телефон respectively. After the independence, there is a minor trend of reintroduction of ‘colloquial spelling’ for Russian loanwords. While the public reaction to this trend was mixed, deRussified spelling of geographic names became generally accepted: for instance,

Contacts of Russian in the post-Soviet space

111

earlier Европа [yevrop@] ‘Europe’ was substituted with Еуропа [ewropa] and Россия [rasiy@] ‘Russia’ by Ресей [resey] (Krippes 1994b: 113–114). Finally, I will briefly stop on the role of English in the post-Soviet Kazakhstan. The earlier course of language policy was in favor of introducing English that would eventually replace Russian as a language of wider communication. However, such a replacement did not occur due to the lack of tradition, sufficient teaching materials and so on (Pavlenko 2008a: 229). The present language policy in Kazakhstan is more moderate: the government of Kazakhstan announced in 2007 a new cultural project, aimed at the development of societal trilingualism, including Kazakh, Russian and English (Smagulova 2008: 455). Pavlenko (2008a: 299) summarizes that the attitude towards Russian in Kazakhstan has changed since the proficiency in English has not been achieved as quickly as hoped earlier. As we shall see, in this respect Estonia differs from Kazakhstan and even from its Baltic neighbors. 2.3.

Estonia

Estonian belongs to the Finnic branch of Finno-Ugric languages. It has a highly developed inflectional morphology (14 grammatical cases). The present Russian-Estonian language contact situation can be better understood in comparison to other post-Soviet contact situations. A handful of scholars in Estonia deal with sociolinguistic topics but, unlike in Latvia or Lithuania, several branches of sociolinguistics are represented (from language policy and bilingual education to studies on contact-induced language change in Russian, from ethnolinguistic vitality research to sociophonetics, and so on). As described in the beginning of Section 2, Estonia differs from Ukraine and Kazakhstan sociohistorically. An interested reader will find an overview of history and ethnic dynamics in several papers by Rannut (2004, 2008). As of the population and housing census of 2000, the population of Estonia is 1,370,052, of whom ethnic Estonians constitute 67.9% and ethnic Russians 25.6% (respectively, 61.5% and 30.5% in 1989). Since migration from Estonia after 1991 was insignificant, the decrease of Russians is attributable to lower birth rates and also to the fact that representatives of some so-called third ethnicities who used to declare their ethnicity as Russian during the Soviet era now tend to identify themselves as Ukrainians, Belarusians, and so on. In 1989, only 15% of Russians claimed proficiency in Estonian (most probably, among them was the indigenous Russian minority who is bilingual and whose sociolinguistic profile is different from that of the Soviet-time newcomers; see Rannut 2008; Verschik 2005a). In 2000, already 44% of ethnic Russians declared proficiency in Estonian. Proficiency in Estonian was self-reported and the census statistics say nothing about

112

Anna Verschik

its degree; however, in any case this can be explained as an act of identity, i.e. the respondents value knowledge of Estonian (Le Page and Tabouret-Keller 1985; by the way, the denial of proficiency in Russian by Estonians in several Soviet censuses was definitely an act of identity as well, see Raun [2001: 210]). According to ethnodemographic criteria, there are three distinct regions in Estonia: pre-dominantly Russian-speaking North-East, bilingual Tallinn where Estonians constitute slightly over 50%, and pre-dominantly Estonian-speaking rest of the country. Thus, a potential research question would be whether and how contact-induced language change differs regionally. I will briefly return to this issue below. Macrosociolinguistic and societal factors, such as experience of own statehood, valuing Estonian, (tacit) opposition to Soviet language policy and Russification during the Soviet era, and, speaking of today, successful economy, language policy aimed at multilingualism and bilingualization of Russian-speakers, and many other similar factors have resulted in the significant change of language repertoires, emergence of bilingual speech and multilingual conversational practices. More particularly, this is notable in the directionality of change. The Russian (Soviet-time) impact on Estonian is to some extent visible on the lexical level. It would be better to label some of lexical Russianism as Sovietisms; those have become obsolete. During the Soviet era Estonian humanitarian intelligentsia tried to prevent Russianization of the corpus; this was discretely yet effectively done through the activity of the Republican Committee for Correct Language Usage (Vabariiklik o˜ igekeelsuse komisjon) (see Raag 1999: 33). Of course, prevention of Russification could never be declared openly as a goal, but strict language planning and high linguistic awareness helped to keep lexical, semantical and morphosyntactical Russianisms out of official registers. A handful of Russian DPW appeared in colloquial Estonian during the Soviet era but gradually fell in disuse during the recent decades. Only a couple of particles such as davai ‘let us’ and vot ‘so, well, here it is’ are still in active use. However, the latter is an ambiguous case because it is materially similar to Estonian vaat with the same meaning. As mentioned earlier, there is no significant morohpsyntactic or phonic impact of Russian on Estonian. Under the new circumstances, personal contacts between Estonian and Russian speakers became more usual and regular than under the Soviet regime when Estonians’ response to the incorporation into the USSR and language policy was withdrawal and creation of symbolic borders according to linguistic criteria. Now bilingualism is not asymmetrical (or, at least, not to such extent as before 1991). The question remains whether and how a relatively rapid acquisition of Estonian by a large number of Russian-speakers will influence Estonian. To answer this question more precisely, Ehala (2009) tries to detect

Contacts of Russian in the post-Soviet space

113

a possible impact of Russian on Estonian object case use. Estonian has fairly complicated rules that determine the choice of the object case. Unlike in Russian and many other languages that have a distinct object case accusative, Estonian object can be in the nominative, genitive or partitive. Ehala (2009) designed a questionnaire where he asked to judge on grammaticality of fabricated utterances.The questionnaire was distributed among Russian- and Estonian-speaking students in Estonian-medium gymnasia. Young people were asked to indicate whether they know Russian/Estonian and whether they have Russian/Estonian friends. Ehala (2009) assumes that the system of Estonian object cases may be undergoing change and speakers of Estonian as L2 may be a force behind it. There is a tendency of partitive overuse but, interestingly, it is more visible in male Russian-speakers than in female ones. The overuse of nominative among Estonian students correlates to their contact with Russian-speakers. However, at least two issues to be mentioned in this connection: first, there are no real spoken data to substantiate this claim and the grammaticality judgment is not triangulated; second, there is slight evidence in my own field-work data that in Russian as L1 there is a tendency of genitive overuse for functions that are covered by of Estonian partitive. Here the choice of object case is influenced by Estonian as L2. Thus, while there may be restructuring in individual speakers, it is difficult to say definitely whether L2 varieties would ever affect Estonian as a whole, whereas there is empirical evidence of overall impact of Estonian as L2 on Russian as L1 (see below). In-depth studies on multilingual communication in micro-communities are useful because they allow identifying contact-induced phenomena that are not visible otherwise. Ehala and Üprus (2008) have established that in some microcommunities where Russian-to-Estonian communication takes place, speakers of Estonian as L1 would sometimes copy L2 speakers’ accent (that is, Russian accent) as a part of accommodation strategy. What clearly renders the Russian-Estonian contact case different from Kazakh-Russian and, possibly, many others, is the impact of Estonian on Russian semantics and morphosyntax. Before the independence, Estonian influence on Russian was minimal if any (here I do not discuss Russian varieties in the lake Peipus region because it is a sociohistorically distinct group of population). Only a handful of lexical borrowings from Estonian occurred in colloquial Russian of the Soviet period. In less than 20 years since the re-establishment of the independence, the picture has changed considerably. Based on the fieldwork with Russian-speaking students (Zabrodskaja 2007), the field data on bilingual communication over the counter, and bilingual TV shows and experiments with Russian as L1/Estonian as L2 speakers (Verschik

114

Anna Verschik

2006, 2005b, 2008), it is safe to claim that (1) code-switching is an everyday practice among Russians and also in Estonian-to-Russian communication and (2) contact-induced change is happening in Russian, for instance, in semantics, combinability of items (word order in genitive constructions, Zabrodskaja [2009]; ‘unconventional’ equivalents of Estonian compound nouns, Verschik [2008: 118–135]; argument structure in certain VP, Verschik [2006]). Also Estonian DWP have been attested in Russian-to-Russian and Russian-to-Estonian communication, although DWP do not affect Russian morphosyntax. On the scale proposed by Thomason and Kaufman (1988), these changes correspond to a moderate impact of L2 (lexical influence, changes in semantics, new collocations, innovations in argument structure, etc.). According to Wertheim’s (2003) modification of the scale which incorporates borrowing of DPW as a separate indicator (i.e. different from borrowing of content words), the placement within the scale remains the same. Of course, it is too early to speak about a crystallized and conventionalized single contact variety of Russian in Estonia. Unlike in immigrant minority situation, in Estonia it is the third or even fourth generation of Russian-speakers who are the first generation of Estonian as L2 speakers. There are regional, generational, occupational, sociocultural and attitudinal differences. Some Russianspeakers, especially in the North-East, effectively create for themselves Russian monolingual environment (press from Russia is available and Russian TV channels may be more attractive to many than the local Russian-language TV, radio and press). Yet others prefer to work with Estonians and spend most of their time in the Estonian-language environment. Studies on micro-communities (for instance, students at Tallinn University and at Tartu University Narva College in Zabrodskaja [2007, forthcoming a, forthcoming b]) show that it is safe to speak of conventionalization of certain contact-induced phenomena. Research on communication patterns in the market place (Verschik 2008) demonstrates that there exist certain strategies of multilingual communication between Estonianspeaking clients and Russian-speaking salespersons, for instance, a compromise strategy of so-called market discourse where semantically specific nouns, numerals, relevant adjectives and some DPW are in Estonian and the rest is Russian. A further research on written texts (including public signage, press, written information issued by banks, local administration, etc.) would help to make an educated judgment on conventionalization. Of course, Russian-language press presupposes monolingual texts but some preliminary data show that oftentimes these are non-monolingual texts in a monolingual disguise, that is, there is no overt use of Estonian lexical items but semantics and morphosyntax can be traced to an Estonian model.

Contacts of Russian in the post-Soviet space

115

Russian-Estonian language contact data have been discussed in the frameworks of MLF and 4M model (Zabrodskaja forthcoming a) and in the codecopying framework (Verschik 2006, 2007b, 2008). Speaking of MLF and 4M model, Zabrodskaja (forthcoming a, forthcoming b) arrived to conclusions similar to those of Muhamedova (2006, forthcoming) and Auer and Muhamedova (2005). ML is a useful empirical notion, but the data show that in some cases it cannot be clearly determined. There are so-called amalgamated constructions where a conceptual unit and combination of items come from Estonian and one or all lexical items come from Russian (compound nouns, genitive constructions). Amalgamated constructions are not well-formed from the point of view of monolingual grammar. Code-copying framework, proposed by Johanson (1992, 2002 and numerous papers since then) is less known and much less influential than the mentioned MLF model. Differently from many theories and models, this model considers all non-monolingual phenomena within the same terminological system.The model is not based on constraints, is not concerned with well-formedness and does not draw a strict line between lexicon and morphosyntax. The latter proved to be an asset in explaining the emergence of non-monolingual multiple word items: idioms, constructions, fixed expressions, and compounds (compound nouns and analytic verbs), etc. For instance, Estonian analytic verbs may yield (1) selective combinational copies, that is, the collocation and overall meaning is copied into Russian, like endale saama ‘to receive, to end up with’ (literally, ‘to get to oneself’) > получить себе or (2) mixed combinational copies where one element in the multiple word item is overtly from Estonian, like pähe o˜ ppima ‘to learn by heart’ (literally ‘to learn into the head’) > учить pähe (Verschik 2008: 139–151). Whether there is a significant regional variation as far as contact-induced language change is concerned, remains unclear for now. The question definitely deserves a special investigation. On the basis of Zabrodskaja’s (2007, forthcoming a, forthcoming b) investigation on non-monolingual speech of students in Tallinn and Narva, it can be claimed that although linguistic environments differ considerably, the same phenomena appear in both localities. The reason for this may be that students are the segment of population who spends most of their time hearing and speaking Estonian. The quantitative aspect has not been studied: it is possible, that differences may be found in quantity of non-monolingual phenomena and in the pace of changes. Pragmatic aspect of Russian-Estonian code-switching has been considered as well. Zabrodskaja (2005) investigated pragmatic functions of code-switching among Russian-speaking schoolchildren in Kohtla-J¨arve, a town in the NorthEast with a dense Russian population. The schoolchildren study Estonian as a

116

Anna Verschik

subject at school and had some additional practice in summer Estonian language camps. It was demonstrated that most of pragmatic functions of code-switching mentioned in the literature were present in the case-study. This leads again to the question of the interplay between macrosociolinguistic factors (linguistic environment, ethnolinguistic makeup of the population, etc), pragmatic factors (conversational goals, participants, their background, language skills) and structural factors. Let us return to the notion of non-accommodating bilingualism, suggested by Bilaniuk (2005, forthcoming). In TV shows speakers of Russian and Ukrainian interact in their respective languages and understand their interlocutors. Her analysis shows that in Ukraine this practice allows avoiding much stigmatized ‘mixing’. Interestingly, in Estonia the interaction where each participant speaks his/her preferable language is characteristic of bilingual TV shows and communication over the counter. Probably, what makes it different from the Ukrainian case, the choice of languages is not dictated by the wish to avoid mixing: such a conversation can slip to code-switching mode within one participant’s utterances as well. In TV shows, when practiced by hosts who have different L1s, the choice of language is conditioned with the necessity of sticking to a certain role (‘Russian-language host’, ‘Estonian-language host’) but as the hosts are bilingual, they balance between A1 B2 A1 B2 pattern, insertional code-switching and occasional use of either language in several consequent utterances. Moreover, a symbolic exchange of languages when a speaker of Russian as L1 speaks Estonian and a speaker of Estonian as L1 speaks Russian has been attested in Russian-to-Estonian communication (‘paradoxical politeness’, Verschik [2005b]) but not in Ukraine (Bilaniuk, p. c., 2007). All this calls for a comparative research on multilingual conversation patterns. It becomes especially interesting in the view of ambiguity of A1 B2 A1 B2 pattern: the situation designated by the symbols and digits may refer to (1) non-cooperation and hostility (“I am sticking to my own language no matter what”); (2) avoidance of mixture (“better to speak one’s native language than to mix languages”); and (3) compromise (“everybody speaks his/her language and we understand each other”). If so, one can ask whether the seemingly same pattern of bilingual communication can be labeled as non-accommodation in the Estonian-Russian case. To conclude this section, it is necessary to mention the role of English. Of all Baltic countries, Estonia has the highest percent of people proficient in English: 25.2% of the population, compared to 16.9% in Latvia and the same for Lithuanian as of census of 2000/2001 (http://www.csb.lv/images/modules/items/item 2064 baltcen3.pdf: 37). Within the country, ethnic Estonians are more likely to know English than ethnic Russians but within the younger generation the difference is not as pronounced. The state does not prescribe what language should be

Contacts of Russian in the post-Soviet space

117

taught as the first foreign language at school; it is decided by schools and parents. Needless to say, English is most popular as the first foreign language, followed by Russian. The latter fact may be attributed both to a graduate de-politicization of Russian and to its instrumental value (it is an asset for many jobs, especially in Tallinn). As for English, there is some research on conventionalized English lexical borrowings into Estonian (Leemets 2002), while Estonian-English codeswitching, a common practice among young people, students, schoolchildren, some urban professionals, in online communication, in advertising, and so on, is waiting for research. English DPW are common in the speech of young Estonians; however, nothing firm can be said about it yet because we are in need of empirical data.

3. Conclusions and directions for further research The current review outlined and specified several points that make post-Soviet language contacts relevant for the general contact linguistic research. Looking at old and new contacts, especially at the fact that all languages of FSU have been and are in contact with Russian, one can think of linguistic laboratory. Under the change of socio-political and sociolinguistic circumstances, a closer look to directionality would be useful. We find a relatively rapid change in the linguistic behavior of Russians in Estonia and, subsequently, the increasing impact of Estonian at least on some varieties of Russian. Although KazakhRussian contacts present a different picture (the degree of cultural pressure, Russified titulars, etc.), the influence of Kazakh on Russian is probably a growing trend. In both cases code-switching between typologically different, genetically unrelated languages with a highly developed inflectional morphology provided some evidence that contradicts the predictions of MLF model. In Russian-Estonian and Kazakh-Russian contact situation nothing like Surzhik or Trasyanka has developed. Yet, this is not to exclude the emergence of contact varieties as such: case studies in Estonia prove that non-monolingual varieties of Russian are emerging, although different from Surzhik in the degree of crystallization, conventionalization, etc. This opens a new series of questions on the possible nature and typology of contact varieties and relevance of typological and genetic proximity between the languages in contact. In general, post-Soviet language contact research would gain if (1) more real spoken language data would be collected and (2) more data-driven case studies conducted in order to enable a comparative view. For instance, the comparison of Kazakh-Russian language contact data to other Turkic-Russian has proved that there are similar tendencies in code-switching patterns and borrowing of

118

Anna Verschik

Russian DPW that affect morphosyntactic structure. Even a superficial glance at pragmatics of multilingual communication in Ukraine and in Estonia suggests that, from a formal point of view, there are similar patterns when each speaker uses his/her language, but the driving forces behind the choices are different. Also a more systematic comparative approach to Trasyanka and Surzhik would be useful: first, it can contribute to the discussion on contact varieties and mixed languages (in the strict sense), second, it would illuminate similarities and differences between the two cases and assist to create a typology of Trasyanka (similar to that of Surzhik). Unfortunately, there is no basis for a comparative research on language contact processes in the Baltic countries. Cultural, linguistic and other differences notwithstanding, the Baltic countries were in the same boat in the 20th century and probably are now as well. One can make an educated guess on the directionality of influence (growing impact of titular languages on Russian) but more facts are needed. For instance, impressionistically, Russian DPW are frequent in spoken Lithuanian (at least in Vilnius), but, as mentioned earlier, not in Estonian where English DPW are becoming common. The underlying reasons are complex and probably have to do with subtle differences in sociolinguistic situation. It is not known whether typological factors play a role here: indeed, Lithuanian is a Baltic language and Baltic languages have a lot in common with Slavic languages; on the other hand, English DPW have been borrowed into many languages (Hlavac [2006] on Australian Croatian) that differ typologically (Estonian among them). What is more, despite of typological distances between Russian and Turkic languages, Russian DPWs are frequent there. Hopefully, in the nearest future more empirical data from post-Soviet language contact situations will become known to the students in contact linguistics. As demonstrated, post-Soviet situation has a potential of contributing to the general discussion as well as for illuminating new facts, which is necessary for the advancement of the theory. Tallinn University Notes *

I express my deep gratitude to the following colleagues (in alphabetical order) whose help was essential for the preparation of this article: Martin Ehala, Kapitolina Fedorova, Tetyana Kuznetsova, Raihan Muhamedova, Aneta Pavlenko, Li Wei, Anastassia Zabrodskaja. 1. The dynamics of the indigenization policy during the early Soviet period and subsequent Russification (both in corpus and in status planning) does not boil down to a somewhat naive formula ‘good Lenin and bad Stalin’. Indeginization was a prag-

Contacts of Russian in the post-Soviet space

119

matic step, partly following internationalist slogans of Marxism-Leninism, partly as a contrast to the politics of the Tsarist empire, but mostly in order to win over local populations for the Soviet cause. On the ambiguity of Ukrainization policy in the 1920s see Bilaniuk (2005: 80). 2. It is relevant that Soviet language policy was not identical in all republics.The Soviets erased, moved or created borders, ethnic groups, languages, etc. (Pavlenko 2008a: 280). The fluctuations in linguistic status of Karelian (i.e. the promotion of Karelian as opposed to Finnish or undoing planning of Karelian in favor of Finnish) before World War II reflects fluctuations in policy towards Finland. Before 1939, Finnish language was banned in Karelia and a new Karielian standard was created; when the plans to conquer and annex Finland in the late 1930s became more articulated, Karelian Autonomous Republic was promoted into Karelian-Finnish Union Republic and Finnish was legalized again. This was done either as a preparation for further annexation of Finland or, at least, for intimidation (Taagepera 2000: 126–127). 3. The question of perception and visibility of language change is relevant for the discussion on contact varieties. Often, purists wage wars against what is in their view contamination, i.e., foreign lexical items, constructions and patterns; yet they forget that what today looks as perfectly ‘authentic’ may be, in fact, a result of contact-induced language change.

References Angermeyer, Philipp S. 2005. Spelling bilingualism: Script choice in Russian American classified ads and signage. Language in Society 34(4). 493–531. Auer, Peter. 1995. The pragmatics of code-switching: A sequential approach. In Lesley Milroy & Pieter Muysken (eds.), One speaker, two languages, 115–135. Cambridge: Cambridge University Press. Auer, Peter. 1998. Introduction: ‘Bilingual conversation’ revisited. In Peter Auer (ed.), Code-switching in conversation, 1–24. London: Routledge. Auer, Peter. 1999. From code-switching via language mixing to fused lects: Toward a dynamic typology of bilingual speech. International Journal of Bilingualism 3(4). 309–332. Auer, Peter. 2007. Monolingual bias in bilingualism research – or: why bilingual talk is (still) a challenge for linguistics. In Monica Heller (ed.), Bilignualism: A social approach, 319–339. Houndmills: Palgrave. Auer, Peter & ˙Inci Dirim. 2003. Socio-cultural orientation, urban youth styles and the spontaneous acquisition of Turkish by non-Turkish adolescents in Germany. In Jannis K. Androutsopoulos & Alexandra Georgakopoulou (eds.), Discourse constructions of youth identities, 223–246. Amsterdam: Benjamins. Auer, Peter & Raihan Muhamedova. 2005. ‘Embedded language’ and ‘matrix language’ in insertional language mixing: Some problematic cases. Italian Journal of Linguis-

120

Anna Verschik

tics/ Rivista di linguistica 17(1). [Special Issue: “Commutazione di codice e teoria linguistica”]. 35–54. Backus, Ad. 2003. Can a mixed language be conventionalized alternational codeswitching? In Yaron Matras & Peter Bakker (eds), The mixed language debate. 237–270. Berlin & New York: Mouton de Gruyter. Backus, Ad. 2005. Codeswitching and language change: One thing leads to another? International Journal of Bilingualism 9(3/4). 307–340. Baskakov, Nikolai A.. 1960. The turkic languages of Central Asia: Problems of planned culture contact. The turkic peoples of the USSR: The development of their languages and writing. Oxford: Central Asian Research Centre. Baskakov, Aleksandr N. [Баскаков Александр Н.]. 1992. Социолингвистический анализ языковой ситуации в регионе Средней Азии и Казахстана [Sociolinguistic analysis of the linguistic situation in the Central Asian region and Kazakhstan]. Нукус: Билим. Bilaniuk, Laada. 2004. A typology of Surzhyk: Mixed Ukrainian-Russian language. International Journal of Bilingualism 8(4). 409–425. Bilaniuk, Laada. 2005. Contested tongues: Language politics and cultural correction in Ukraine. Ithaca, NY: Cornell University Press. Bilaniuk, Laada. Forthcoming. Language in the balance: The politics of nonaccommodation in bilingual Ukrainian-Russian television shows. International Journal of the Sociology of Language. http://www.ukrainianstudies.uottawa.ca/pdf/ P Bilaniuk Danyliw06.pdf (accessed 18 September 2009). Bilaniuk, Laada & Svitlana Melnyk. 2008.A tense and shifting balance: Bilingualism and education in Ukraine. International Journal of Bilingual Education and Bilingualism 11(3 & 4). 339–372. Block, David. 2007. Second language identities. London: Continuum. Chamidova, S. M. [Хамидова, С. М.]. 1985. Русские словарные единицы в узбекской разговорной речи (к постановке вопроса) [Russian lexical units in colloquial Uzbek (towards the formulation of the problem)]. Советская тюркология 1. 20–29. Chizhikova, L. N. [Чижикова Л. Н.]. 1968. Об этнических процессах в восточных районах Украины [On ethnic processes in the Eastern regions of Ukraine]. Советская этнография 1. 18–31. Chorny, Dmitro [Чорний, Дмитро]. 2006. „Анклави цивiлiзацiї“ чи „контактнi зони“: велике мiсто українсько-росiйського Порубiжжя [‘Enclaves of civilization’ or ‘contact zones’: big cities in Ukrainian-Russian borderland]. Схiд-Захiд: Iсторикокультурологiчний збiрник. Випуск 8. Харкiв; Київ: Критика. 69–81. Ciscel, Matthew. 2007. The Language of the Moldovans: Romania, Russia, and identity in an ex-Soviet republic. Lanham: Lexington Books. Clyne, Michael. 2003. Dynamics of language contacts. Cambridge: Cambridge University Press. Comrie, Bernard. 1981. The languages of the Soviet Union. Cambridge: Cambridge University Press.

Contacts of Russian in the post-Soviet space

121

Dirim, ˙Inci & Andreas Hieronymus. 2003. Cultural orientation and language use among multilingual youth groups: ‘For me it is like we all speak one language’. Journal of Multilingual and Mutlicultural Development 24(1 & 2). 42–55. Donii, Oles [Донiй, Олесь]. 2003. Хай живе суржик! [Long live Surzhik!]. http://www. molodaukraina.org/news.asp?Id=229 &IdType=12 (accessed 18 September 2009). Ehala, Martin. 1994. Russian influence and the change in progress in the Estonian adpositional system. Linguistica Uralica 39(2). 177–193. Ehala, Martin. 2009. Keelekontakti m˜oju eesti sihitisek¨aa¨ nete kasutamisele [The impact of language contacts on the use of Estonian object cases]. Keel ja Kirjandus 3. 182– 204. Ehala, Martin & Tene Üprus. 2008. The mechanism of substrate impact on superstrate: Assessing Uralic substrate in Germanic. Linguistica Uralica 54(2). 81–96. Ellis, Elisabeth. 2008. Defining and investigating monolingualism. Sociolinguistic Studies 2(3). 311–330. Fireman, William. 2006. Language and education in post-Soviet Kazakhstan: Kazakhmedium instruction in urban schools. The Russian Review 65(1). 98–116. Flier, Michael [Флар, Майкл]. 2000. Суржик: правила утворення безладу [Surzhik: rules for chaos formation]. Критика 6. 16–17. Giger, Mark and Marian Sloboda. 2008. Language management and language problems in Belarus: Education and beyond. International Journal of Bilingual Education and Bilingualism. 11(3 & 4). 315–339. Golovko, Evgenij. 1994. Mednyj Aleut or Copper Island Aleut: an Aleut-Russian mixed language. In Peter Bakker & Maarten Mous (eds.), Mixed languages: 15 case studies in language intertwining, 113–121. Amsterdam: IFOTT. Golovko, Evgeniy V. 2003. Language contact and group identity: The role of ‘folk’ linguistic engineering. In Yaron Matras & Peter Bakker (eds), The mixed language debate. 177–207. Berlin & New York: Mouton de Gruyter. Hasselblatt, Cornelius. 2000. Estonian between German and Russian: Facts and fiction about language interference. In Dicky Gilbers, John Nerbonne & Jos Schaeken (eds.), Languages in Contact, 135–144. Amsterdam & Atlanta: Rodopi. Hnatkevich, Yurii [Гнаткевич, Юрiй]. 2000. Уника ймо русизмiв в укра¨ıнскiй мовi! [Let’s avoid Russianisms in Ukrainian!]. Киiв: Просвiта. Hint, Mati. 1991. Russian influences in the Estonian language. In Lazslo Keresztesz and Sándor Maticsák (eds.), Congressus Septimus Internationalis Fenno-Ugristarum. 1A Sessiones plenares, 87–107. Debrecen, 27 August–2 September 1990. Debrecen: Debrecen University. Hlavac, Jim. 2006. Bilingual discourse markers: evidence from Croatian-English codeswitching. Journal of Pragmatics 38(11). 1870–1900. Johanson, Lars. 1992. Strukturelle Faktoren in türkischen Sprachkontakten (Sitzungsberichte der Wissenschaftlichen Gesellschaft an der J. W. Goethe-Universit¨at Frankfurt am Main, 29: 5). Stuttgart: Steiner. Johanson, Lars. 1997. Kopien russischer Konjunktionen in türkischen Sprachen. Geburtstag. In Dieter Huber & Erkia Worbs (eds.), Ars transferendi. Sprache, Übersetzung,

122

Anna Verschik

Kulturalit¨at. Festschrift für Nikolai Salnikow zum 65., 115–121. Frankfurt am Main: Peter Lang. Johanson, Lars. 2002. Structural factors in Turkic language Contacts. London: Curzon. Kolstø, Pål. 1995. Russians in the former Soviet republics. Bloomington, IN: Indiana University Press. Kolstø, Pål. 1999. Territorializing diasporas: The case of the Russians in the former Soviet republics. Millennium – Journal of International Studies 28(3). 607–631. Kopylenko, Moisey [Копыленко, Моисей]. 1988. Интерференция в русской речи казахов [Interference in the Russian speech of Kazakhs]. Алма-Ата: Наука. Kopylenko, M. & S. Saina [Копыленко, М. & C. Саина]. 1982. Функционирование русского языка в различных слоях казахского населения [Functioning of the Russian language in various strata of Kazakh population]. Алма-Ата: Наука. Korth, Brigitta. 2005. Language attitudes towards Kyrgyz and Russian. Discourse, education and policy in post-Soviet Kyrgyzstan. Bern et al.: Peter Lang. Krippes, Karl A. 1994a. Russian code-switching in colloquial Uzbek, Kazakh and Kyrgyz. General Linguistics 34(3–4). 139–144. Krippes, Karl A. 1994b. Russian loan-words and orthography in the post-Soviet Turkic Languages. General Linguistics 32(2). 107–116. Kuznetsova, Tetyana [Кузнєцова, Тетяна]. 1999. Двомовна сiм’я як об’єкт соцiолiнгвiстичного дослiдження [Bilingual family as an object of a sociolinguistic research]. Проблеми семантики слова, речення та тексту. Ки¨ıв: КДЛУ. Külmoja, Irina. 1999. Vene keel Eestis [The Russian language in Estonia]. In JüriViikberg (ed.), Eesti rahvaste raamat [The book of Estonia’s peoples], 516–524. Tallinn: Eesti Entsüklopeedia. Külmoja, Irina. 2000. Специфические черты языка русской диаспоры Эстонии [Specific features of the language of Russian diaspora in Estonia]. Труды по русской и славянской филологии. Лингвистика. Новая серия. (Язык диаспоры: проблемы и перспективы) 3. 84–93. Laitin, David. 1998. Identity in formation: the Russian speaking populations in the near abroad. Ithaca: Cornell University Press. Lebedeva, Nadezhda [Лебедева, Надежда]. 1995. Новая русская диаспора: социально-психологический анализ [New Russian diaspora: Socio-psychological analysis]. Moscow: Russian Academy of Sciences. Leemets, Tiina. 2002. Inglise laenud ja m˜ojutused 20. sajandi l˜opu eesti keele s˜onavaras [English loans and influence on the Estonian lexicon of the late 20th century]. Tallinn: Eesti Keele Instituut. Leinonen, Marja. Forthcoming. Russian influence on Iˇzma Komi dialect. International Journal of Bilingualism 13(3) [Special issue: Language contacts in the post-Soviet space]. Le Page, Robert & André Tabouret-Keller. 1985. Acts of identity: Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press.

Contacts of Russian in the post-Soviet space

123

Liskovets, Irina. Forthcoming. Trasjanka: a code of rural migrants in Minsk. International Journal of Bilingualism 13(3) [Special issue: Language contacts in the post-Soviet space]. Maschler, Yael. 1994. Metalanguaging in bilingual conversation. Language in Society 23. 325–366. Maschler, Yael. 1998. On the transition from codeswitching to a mixed code. In Peter Auer (ed.), Code-switching in conversation, 125–149. London: Routledge. Masenko, Larisa [Масенко, Лариса]. 1999. Мова i nолiтика [Language and policy]. Ки¨ıв: Соняшник. Masenko, Larisa [Масенко, Лариса]. 2002. Суржик як соцiолiнгвiстичний феномен [Surzhik as a sociolinguistic phenomenon]. Дивослово: Українська мова й лiтература в навчальних закладах 3. 11–13. Masenko, Larisa [Масенко, Лариса]. 2004. Мова i сусniльство: Постколонiальний вимiр [Language and society: Postcolonial dimension]. Kyiv: Kyiv-MohylaAcademy. Masenko, Larisa [Масенко, Лариса]. 2005. Українська мова у XX сторiччi: iсторiя лiнгвоциду [Ukrainian language in the 20th century: A history of linguicide]. Kyiv: Prosvita/Kyiv-Mohyla Academy. Masenko, Larisa [Масенко, Лариса]. 2008. Мовна ситуацiя України: Соцiолiнгвiстичний аналiз [Language situation in Ukraine: A sociolinguistic analysis]. In Juliane Besters-Dilger (ed.), Мовна иолiтика та мовна ситуацiя в Українi: Аналiз та рекомендацiї [Language policy and language situation in Ukraine: Analysis and recommendations], 6–131. Kyiv: Kyiv-Mohyla Academy. Matras, Yaron. 1998. Utterance modifiers and universals of grammatical borrowing. Linguistics 36. 281–331. Matras, Yaron & Peter Bakker. 2003. The study of mixed languages. In Yaron Matras & Peter Bakker (eds.), The mixed language debate. Theoretical and empirical advances, 1–20. Berlin & New York: Mouton de Gruyter. Mechkovskaya, Nina [Мечковская, Нина]. 2005. Постсоветский русский язык: новые черты в социолингвистическом статусе [Post-Soviet Russian: New features in the sociolinguistic status]. Russian Linguistics 29. 49–70. Menz, Astrid. 1999. Gagausische Syntax. Eine Studie zum Kontaktinduzierten Sprachwandel. Wiesbaden: Harassowitz. Myers-Scotton, Carol. 2002. Contact linguistics: Bilingual encounters and grammatical outcomes. Oxford: Oxford University Press. Muhamedova, Raihan. Forthcoming. The use of Russian conjunctions in the speech of bilingual Kazakhs. International Journal of Bilingualism 13(3) [Special issue: Language contacts in the post-Soviet space]. Muhamedowa, Raihan. 2005. Kasachisch-russisches Code-mixing: ein Fall von morphologischer Vereinflachung. Zeitschrift für Sprachwissenschaft 24. 263–309. Muhamedowa, Raihan. 2006. Untersuchung zum kasachisch-russischen Code-mixing (mit Ausblicken auf den uigurisch-russischen Sprachkontakt) (LINCOM studies in language typology 12). München: LINCOM Europa.

124

Anna Verschik

Mustajoki, Arto & Ekaterina Protassova [Мустайоки, Арто & Екатерина Протасова] (eds.). 2004. Русскоязычный человек в иноязычном окружении [Russian-speakers in foreign language environment]. Helsinki: Slavica Helsingiensia. Nauruzbayeva, Zhanara. 2003. Paradoxes of Kazakh language ‘revival’ in Kazakhstan. The New Hampshire Institute of Politics. https://www.anselm.edu/NR/rdonlyres FC460641-E210–4AC7-9401-E470E25B9FAC/1262/20024.pdf (accessed 17 September 2009). Orlov, A. V. [Орлов, А. В.]. 1997. Украинцы и русские: современные параметры этнокультурного взаимодействия. Диалог украинской и русской культур: Материалы международной конференции (24–25 окт. 1996 г.). [Dialog of the Ukrainian and Russian culture: Proceedings of international conference (24–25 October 1996)], 78–82. Kyiv. Ozolins, Uldis. 2003. The impact of European accession on language policy in the Baltic States. Language Policy 2(3). 217–238. Pavlenko, Aneta. 2006. Russian as lingua franca. Annual Review of Applied Linguistics 26. 78–99. Pavlenko, Aneta. 2008a. Multilingualism in post-Soviet countries: Language revival, language removal, and sociolinguistic theory. International Journal of Bilingual Education and Bilingualism 11(3 & 4). 275–314. Pavlenko, Aneta. 2008b. Russian in post-Soviet countries. Russian Linguistics 32. 59–80. Pavlenko, Aneta & Scott Jarvis. 2002. Bidirectional transfer. Applied Linguistics 23(2). 190–214. Pavlenko, Aneta & Svitlana Melnyk. 2009. Ukrainian-Russian language contact: A one-way affair? Paper presented at 7th International Symposium of Bilingualism, Utrecht, 8–11 July 2009. http://www.eventure-online.com/eventure/ publicAbstractView.do?id=86216 (accessed 18 September 2009). Raag, Raimo. 1999. One plus one equals one: The forging of Standard Estonian. International Journal of the Sociology of Language 139. 17–38. Rannut, Mart. 2004. Language policy in Estonia. Noves SL. Revista de Sociolinguistica. Spring-Summer 2004. http://www6.gencat.net/llengcat/noves/hm04primaveraestiu/docs/rannut.pdf (accessed 10 September 2009). Rannut, Mart. 2008. Estonianization efforts post-independence. International Journal of Bilingual Education and Bilingualism 11(3 & 4). 423–439. Raun, Toivo. 2001. Estonia and Estonians, 2nd edn. Stanford, CA: Stanford University, Hoover Institution Press. Rehbein, Jochen, Jan ten Thije & Anna Verschik (eds.). Forthcoming. Receptive multilingualism. [Special Issue.] International Journal of Bilingualism. Salmons, Joseph. 1990. Bilingual discourse marking: Code switching, borrowing and convergence in some German-American dialects. Linguistics 28. 453–480. Savoskul, Sergei [Савоскул, Сергей]. 2001. Русские нового зарубежья: выбор судьбы [Russians in the new abroad: The choice of destiny]. Moscow: Nauka. Sebba, Mark. 2006. Ideology and alphabets in the former USSR. Language Problems and Language Planning 30(2). 99–125.

Contacts of Russian in the post-Soviet space

125

Serebrens’ka, Oleksandra [Сербенська, Олександра] (ed.). 1994. Антисуржик [AntiSurzhik]. Лвiв: Свiт. Shaybakova, Damina [Шайбакова, Дамина]. 2005. Функционирование неорганического языка в полиэтническом обществе (на материале русского языка в Казахстане) [Functioning of a nonindigenous langauge in a multiethnic society (on the example of Russian in Kazakhstan)]. Àëìàòû PhD dissertation, draft version. Smagulova, Juldyz. 2008. Language policies of kazakhization and their influence on language attitudes and use. International Journal of Bilingual Education and Bilingualism 11(3 & 4), 440–475. Smailov, Alikhan A. [Смаилов, Алихан А.] (ed.). 2004. Статистический ежегодник 2004 [Satistical Yearbook 2004]. Almaty: Statistics Agency of the Republic of Kazakhstan. Stavic’ka, Lesya [Ставицька, Леся]. 2001. Кровозмiсне дитя двомовностi [Incestuous child of bilingualism]. Критика 10. 20–24. Stern, Dieter. Forthcoming. The Taimyr pidgin Russian morphology enigma. International Journal of Bilingualism 13(3) [Special issue: Language contacts in the postSoviet space]. Suleimenova, Eleonora, Nurilya Shaymerdenova & Dana Akanova [Сулейменова, Элеонора, Нуриля Шаймерденова & Дана Аканова]. 2007. Языки народов Казахстана. Социолингвистический справочник [Languages of the peoples of Kazakhstan. Sociolinguistic handbook]. Астана: Арман-ПВ. Taagepera, Rein. 2000. Soomeugri rahvadVenemaa Föderatsioonis [Finno-Ugric peoples in Russian Federation]. Tartu: Ilmamaa. Taranenko, Oleksandr. 2007. Ukrainian and Russian in contact: Attraction and estrangement. International Journal of the Sociology of Language 183. 119–140. Tarasenko, B. V. [Тарасенко, Б. В.]. 2006. До питання про дефiнiцiю мiшаного українсько-росiйського мовлення [On the question of defining a mixed UkrainianRussian speech]. Науковi записки 60. 62–66. Thomason, Sarah Grey & Terence Kaufman. 1988. Language contact, creolization, and genetic linguistics. Berkeley and Los Angeles: University of California. Thomason, Sarah Grey. 1997. On mechanisms of interference. In Stig Eliasson & Ernst H. Jahr (eds.), Language and its ecology: Essays in memory of Einar Haugen, 181– 207. Berlin & New York: Mouton de Gruyter. Trub, V. M. [Труб В. М.]. 2000. Явище ‘суржику’ як форма просторiччя в ситуацiї двомовностi [The phenomenon of ‘Surzhik’ as a form of casual speech in a bilingual situation]. Мовознавство 1. 46–58. Vakhtin, Nikolai, Oksana Zhironkina, Irina Liskovets & Yekaterina Romanova [Вахтин, Николай, Оксана Жиронкина, Ирина Лисковец & Екатерина Романова]. 2003. Новые языки новых государств: явления на стыке близкородственных языков на постсоветском пространстве [New languages of new states: Contact phenomena in genetically related languages in the post-Soviet space]. http://old.eu.spb.ru/ ethno/projects/project3/list.htm (accessed 18 September 2009).

126

Anna Verschik

Vakhtin, Nikolai, Oksana Zhironkina & Katerina Romanova [Вахтiн Нколаiй, Оксана Жиронкiна & Катерiна Романова]. 2006. Калампоцання над долею української мови [Pondering over the fate of the Ukrainian language]. Схiд-Захiд: Iсторикокультурологiчний збiрник. Випуск 8. Харкiв; Київ: Критика. 200–2008. Verschik, Anna. 2004. Estonian compound nouns and their equivalents in the local variety of Russian. Scando-Slavica 50. 93–109. Verschik, Anna. 2005a. The language situation in Estonia. Journal of Baltic Studies36(3). 283–317. Verschik, Anna. 2005b. Russian-Estonian language contact, linguistic creativity and convergence: new rules in the making. Multilingua 24(4). 413–429. Verschik, Anna. 2006. Convergence in Estonia’s Russian: Directional vs. static vs. separative verb. International Journal of Bilingualism 8(4). 383–404. Verschik, Anna. 2007a. Multiple contact situation in Tallinn: Transfer B2 > A1 or B1 > A2? International Journal of Bilingual Education and Bilingualism 10(1). 80–103. Verschik, Anna. 2007b. Jewish Russian and the field of ethnolect study. Language in Society 36(2). 213–232. Verschik, Anna. 2008. Emerging Bilingualism: From monolingualism to code-copying. London: Continuum. Verschik, Anna (ed.). Forthcoming. Language contacts in post-Soviet space. [Special issue.] International Journal of Bilingualism 13(3). Vihalemm, Triin & Anu Masso. 2002. Patterns of self-identification among the younger generation of Estonian Russians. In Marju Lauristin & Mati Heidmets (eds.), The challenge of the Russian minority. Emerging multicultural democracy in Estonia, 185–198. Tartu: Tartu University Press. Yessenova, Saulesh. 2005. ‘Routes and roots’ of Kazakh identity: Urban migration in postcolonial Kazakhstan. The Russian Review 64(4). 661–679. Weinreich, Uriel. 1953. Languages in contacts. The Hague: Mouton. Wertheim, Suzanne. 2003. Linguistic purism, language shift and contact-induced change in Tatar. Berkeley: University of California, Berkeley dissertation. Woolhieser, Curt. 2001. Language ideology and language conflict in post-Soviet Belarus. In Camille C. O’Reilly (ed.), Language, ethnicity and the state, Vol. 2, 91–122. New York: Palgrave. Zabrodskaja, Anastassia. 2005. Vene-eesti koodivahetus Kohtla-J¨arve vene emakeelega algkoolilastel [Russian-Estonian code-switching among Russian-speaking schoolchildren in Kohtla-J¨arve]. Tallinn: Tallinn University Press. Zabrodskaja, Anastassia. 2007. Code-switching and contact-induced language change in Estonia’s Russian. In Pille Eslon (ed.), Tallinna Ülikooli keelekorpuste optimaalsus, töötlemine ja kasutamine [Tallinn University linguistic corpora optimization, elaboration and use]. (Tallinna Ülikooli eesti filoloogia osakonna toimetised 9 [Publications of the Department of the Estonian Philology ofTallinn University 9], 13–64.) Tallinn: Tallinn University Press. Zabrodskaja, Anastassia. 2009. Towards establishing the matrix language in RussianEstonian code-switching:A corpus-based approach. In Stavroula Tsiplakou, Marilena

Contacts of Russian in the post-Soviet space

127

Karyolemou & Pavlos Pavlou (eds.), Language Variation – European Perspectives II. Selected papers from the Fourth International Conference on Language Variation in Europe (ICLaVE 4), Nicosia, Cyprus, 17–19 June 2007, 205–240. Amsterdam: John Benjamins. Zabrodskaja, Anastassia. Forthcoming a. Evaluating the Matrix Language Frame model on the basis of a Russian-Estonian code-switching corpus. International Journal of Bilingualism 13(3) [Special issue: Language contacts in the post-Soviet space]. Zabrodskaja, Anastassia. Forthcoming b. Morphosyntactic contact-induced language change in young Estonia’s Russian speakers variety. In Isabelle Léglise and Claudine Chamoreau (eds.), The interplay of variation and change in contact settings. Morphosyntactic studies in language variation. Amsterdam: John Benjamins. Zaprudski, Siarhej. 2007. In the grip of replacive bilingualism: The Belarusian language in contact with Russian. International Journal of the Sociology of Language 183. 97–118. Zeevaert, Ludger & Jan ten Thije (eds.). 2007. Receptive multilingualism. Amsterdam: Benjamins. Zharkova, A. [Жаркова, А.]. 2006. Родная речь русских в Литве: проблемы нормативности [Native speech of Russians in Lithuanian: problems of norms]. In Ирина Кюльмоя (ed.), Взаимодействие языков и языковых единиц [Interaction of languages and linguistic units], 47–53. Tartu: Tartu University Press. Zhurzhenko, Tetyana [Тетяна Журженко]. 2006. Нове пострадянське прикордоння: ностальгiя, опiрзмiнам, адаптацiя (на прикладi трьох сiл Харкiвської областi) [New post-Soviet borderlands: Nostalgia, resistance to changes, adaptation (on the example of three villages of Kharkiv oblast]. Схiд-Захiд: Iсторико-культурологiчний збiрник. Випуск 8. Харкiв; Київ: Критика. 185–99. Anna Verschik is professor of sociolinguistics and Estonian as a foreign language at Tallinn University and docent in Yiddish studies at Helsinki University. Her scholarly interests include Russian-Estonian language contacts, sociolinguistics of the Baltic countries and varieties of Yiddish in the Baltic region. Her publications include Emerging bilingualism: From monolingualism to code-copying (Continuum, 2008) and “Language contacts in the post-Soviet space” (special issue of the International Journal of Bilingualism, in press).

Language use and language shift among the Malays in Singapore* FRANCESCO CAVALLARO and STEFAN KARL SERWE

Abstract With active language planning policies in force since its independence as a nation, the linguistic situation in Singapore has received a substantial amount of scholarly attention. Yet, the focus has traditionally been on Singapore English, with issues regarding maintenance and shift of the other official languages of the republic attracting much less interest. Malay Singaporeans have often been enviously described as guardians of their ethnic language, apparently resisting the push and pull factors of English more successfully. This study aims to investigate to which degree the Malays are indeed still maintaining their community language. In this study a total of 233 participants from 12 to 72 years of age were asked to report on their language use across different domains, topics and interlocutors in semi-structured interviews. The results indicate that for Singaporean Malays the age of interlocutor is the most important factor when deciding on the language(s) of interaction. While Malay is still unrivaled in interactions with senior members of the community, English is making inroads everywhere else. The influence of English is particularly strong for young adults (18–24 years), young women and people of high socio-economic and educational status. This leads to the conclusion that domains that were traditionally considered safe havens for Malay in Singapore are slowly being eroded. Keywords: language maintenance; language shift; Malay language; bilingualism; language policy; Singapore.

1. Introduction Multilingual societies, such as Singapore, offer great opportunity for detailed sociolinguistic studies. However, getting a firm grasp on the issues involved

130

Francesco Cavallaro and Stefan Karl Serwe

in studying the use of two or more languages by the same social group or between social groups is never easy, all the more so when the sociolinguistic situation is as complicated and fluid as it is in Singapore. Issues focusing on language maintenance (from now on LM) and language shift (from now on LS) and their consequences are very significant aspects of situations where different languages come into contact. This is particularly so, because, very often through social and/or political processes, one or more language(s) become dominant at the expense of the others as is the case in Singapore. The last fifty years or so have seen a rapid growth in the study of LM and LS and more specifically, of the factors associated with shift or maintenance. The literature points to a number of clear-cut factors for or against LM and some that are ambivalent. That is, they can either lead to LS or support LM according to the circumstances of the group in question (see Kloss 1966; Clyne 1982, 1991, 2003; Fishman 1991; see also Cavallaro [forthcoming] for a summary; and Giles et al. [1977] on ‘ethnolinguistic vitality’). Some of the most important ambivalent factors are the numerical strength of a minority group in relation to the majority, the status of the language in a given society, generation and the institutional support. Tollefson (1991; and see discussion in Kamwangamalu [2006]) suggests that the older generation’s decision to pass down a language to subsequent generations depends, in particular, on the socio-political status of the language and the government’s policies and community support. However, Kamwangamalu (2006) relates the South African experience as a counterexample. Nine African languages are recognized alongside English and Afrikaans as official languages of the state, but none can compete against English and Afrikaans in terms of prestige and with English in terms of economic value, to the extent that many black South Africans are actively encouraging their children to speak only English in all facets of their community life. Moreover, Reagan (2001) points to similar signs of rapid shift to English for Afrikaans speakers, and thus one can imagine what the prognosis is for the fate of other minority languages. It is clear in these cases that social mobility strongly outweighs the values of the traditional languages of these communities. The gradual erosion of Afrikaans in South Africa bears some resemblance to the situation for Malay in Singapore. Malay is a language that for a long time has been maintained much better than the other minority languages spoken in Singapore. However, the rise of English as the global financial language worldwide and the emphasis on English as the language of instruction in Singapore has meant that even the Malay community is displaying strong signs of a shift away from their traditional language. Various authors have discussed language shift in various Singaporean communities, e.g., Li et al. (1997) for Teochew; Schiffmann (2007) for Tamil; and Gupta and Siew (1995) for Cantonese. Up

Language use and shift among Malays in Singapore

131

till now, LS in the Malay community in Singapore has not received a great deal of attention because the assumption has always been that it is not a language under threat. This paper presents an in-depth study into the language use of the Malay community with the specific aim to document whether the community is undergoing a shift to English. It will focus on language use with family members, since the family has been defined in the literature as the stronghold of Malay in Singapore (Chew 2006; Vaish 2008). The investigation aims to ascertain in more detail to which degree this is still true. In the first section a detailed demographic and linguistic analysis will be given using the latest census data. The second part of the article will present and discuss our data on the language use of Singaporean Malays of different ages, educational and socio-economic backgrounds. The findings will be compared to earlier studies on language use in Singapore. The main factors affecting LM or LS will then be discussed in the final section.

2. Singapore’s multicultural make-up Singapore is a multiethnic and multilingual society of 4.8 million people. Of these 4.8 million 75% are residents and 25% non-residents (Department of Statistics 2009a). The total population has been growing steadily over the years (Table 1), which is mainly due to an increased influx of non-residents, and is expected to reach 6.5 million in the next ten or so years according to government projections. Table 1. Singapore’s population (’000)

Total population Resident population

1980 2,413.9 2,282.1

1990 3,047.1 2,735.9

2000 4,027.9 3,273.4

2008 4,839.4 3,642.7

Source: Department of Statistics 2009a

Yet despite dynamic demographic changes, the ethnic composition of Singapore’s resident population has remained relatively stable throughout the last fifty years (Kuo 1980a; Department of Statistics 2001a). Recent figures (Department of Statistics 2006, 2009a) depict Singapore’s society as an ethnic mix of 75.6% Chinese, 13.6% Malays, 8.7% Indians and 2.1% so-called Others, most of whom are of Eurasian, European or Arab decent (Table 2). There has been a very small decline in the number of Chinese and Malays while the Indian population has increased slightly.

132

Francesco Cavallaro and Stefan Karl Serwe

Table 2. Ethnic composition of Singapore residents (%) 1990 2000 2005

Chinese 77.8 76.8 75.6

Malay 14.0 13.9 13.6

Indian 7.1 7.9 8.7

Others 1.1 1.4 2.1

Source: Department of Statistics 2001b, 2006

Singapore is truly a multilingual country. Table 3 shows that there are at least 23 living languages spoken in this country of only 692.7 square kilometers and this number does not include the languages spoken by non-residents living, studying and working in Singapore. Table 3. Most spoken (local) languages in Singapore (adapted from Gordon [2005]) Indian Tamil Bengali Gujarati Hindi Malayalam Panjabi, Eastern Sinhala

Chinese Mandarin Hakka Hainanese Min Nan (Hokkien) Teochew Yue (Cantonese)

Malay Malay Javanese Baba Malay Bazaar Malay Orang Seletar Madura (Boyanese)

Others English Malaccan Creole PortugueseCreole – Papia Kristang Singapore Sign Language

The 13.6% Malays account for 484,600 people according to the 2005 General Household Survey (Department of Statistics 2006). The Malay community in Singapore has also gone through significant changes, particularly in its linguistic make-up, as will be shown in more detail below.

3. Language policy Besides its colonial history and the politico-pragmatic circumstances before and after independence in 1965, the island nation’s language policy has been described as the result of ensuring the cohesion of its multi-ethnic fabric (e.g. Bokhorst-Heng 1998; Gupta 1998; Wee 2003). In rather stark contrast to Kuo’s (1980b) account of the natural diversity of languages spoken within and across ethnic lines before independence, the Republic of Singapore adopted four official languages, promoting English as the language of public administration, interethnic communication, education and commerce; and establishing Mandarin, Malay and Tamil as the home languages, officially referred to as ‘mother

Language use and shift among Malays in Singapore

133

tongues’, of the three major ethnic groups.1 Note that the term ‘mother tongue’ in the Singapore contexts refers to Mandarin Chinese, Malay and Tamil even if one’s mother tongue is English. Henceforth, the Singaporean usage of the term ‘mother tongue’ will be flagged by single quotation marks. While these languages may well be bona fide mother tongues for many, they are L2s for many others. In effect, for many Singaporeans who have shifted to English as a mother tongue, their ethnic languages are technically second languages. Singapore’s bilingual education policy was instituted in the late 1950s emphasizing equality for all the official languages. In its original form, it stated that the four official languages were also the media of instruction. In the time following independence, while most schools were English-medium, there were a number of Tamil, Malay and Mandarin medium schools. However, by 1987 all of them were closed because of falling student numbers (Tan 2007). This reduced Mandarin Chinese, Tamil and Malay to being taught as second languages in primary and secondary schools, and English has since dominated the country’s education system (Pakir 2004). Although the assigned ethnic language may come unnaturally to some, proficiency in it is considered very important socially, as the ‘mother tongues’ are deemed to function as “an anchor in their [students] ethnic and cultural traditions” in opposition to the Western values and world view supposedly imparted through the English language (Gopinathan 1998: 21). Regardless of how one judges such efforts at language planning, one can safely agree with Pakir’s (1991) prediction that post-independence-born Singaporeans today have become English-knowing bilinguals, confident in their use of the varieties of Singapore English plus their ethnic language. Yet, as Wee (2003) illustrates, the functional separation between English and the ‘mother tongues’ in Singapore has been shaken within the wake of economic globalization, so that now the utilitarian value traditionally assigned exclusively to English has also been extended to Mandarin Chinese. This shift in emphasis has been promoted by government policies and educational reforms. Public initiatives have been implemented to strengthen the position of Malay and Tamil, so as to preserve the equality between the ‘mother tongues’. Wee (2003), however, regards these efforts as futile, due to lack of practicality and bottom-up support. Since the 1980s, economic success has helped convince the majority of Singaporeans that a good knowledge of English is the basis for better career opportunities for themselves and their children. In actual fact, we can argue that due to the economic power of China, and Singaporeans’ innate sense of linguistic pragmatism, mastery of Mandarin is now also seen as an essential path to economic success, both at the individual level and for the country. The widespread use of English among all ethnic groups and the majority (in numbers)

134

Francesco Cavallaro and Stefan Karl Serwe

of the ethnically Chinese have essentially elevated English and Mandarin to the status of majority languages, and relegated Malay and Tamil to the rank of minority languages. This general perception and attitude has led significant numbers of Singaporeans to shift to using English as their home language. This is happening despite the government’s support of the teaching of the ‘mother tongues’ (David et al. 2009).

4. The special role of Malay among the other ethnic languages Malay has always had a special status among the languages spoken in the Straits region and retains this status even in modern day Singapore. As the language of the indigenous rulers, Malay used to be the administrative language as well as the language of trade and commerce in the region. Even after the establishment of the colonial rule in the Malaysian peninsula, starting with the Portuguese in the 16th century, Malay continued to function as the language of wider communication among the ethnically diverse population. Kuo (1980b) still ascribes this function to Malay for the 1950s in Singapore. Apart from Bahasa Melayu, Malay-lexified pidgin languages such as Bazaar Malay and Baba Malay were widespread. Bazaar Malay is reportedly still used by a significant number of older Singaporeans today (Aye 2005). Baba Malay is on the verge of disappearing as mostly only elderly people still speak it (Lee et al. 2009). Yet, apart from being the traditional lingua franca in Singapore, Malay is also the national language of the republic. The reasons can again be found in history. The status of Malay as the language of the people and the nation found its ultimate expression as the language of the anticolonial movement in post-World War II Malaysia. Speaking and learning Malay was institutionally supported in Singapore in the 1950s leading up to Singapore joining the Federation in 1963 (Abdullah and Ayyub 1998; Lowenberg 1988). This was politically motivated as it was then critical for Singapore to find a place within the Federation of Malaysia. Even though all efforts to promote it were stopped immediately after Singapore’s independence in 1965, Malay was kept as the national language, supposedly because of its widespread use among the people, as an expression of their identity, and its political value within the region (Alsagoff 2008). However, this value is largely symbolic and the only remnants of Malay’s national language status today can be found in Singapore’s national anthem, its coat of arms and in some military commands in parades (Lowenberg 1988; Llamzon 1978).

Language use and shift among Malays in Singapore

135

5. The Malay speech community in Singapore From these historical records it becomes evident that speaking Malay in Singapore has never been a purely ethnic Malay affair. These historical circumstances have their contemporary consequences. As in the past, in contemporary Singapore, speaking Malay is not necessarily indexical to ethnic Malay community membership. Data from the 2005 census (Department of Statistics 2006) prove that 10.5% of Indian Singaporeans and 12.2% of the category ‘Others’ also regard Malay as their home language. This paints a slightly more heterogeneous picture of the Malay speech community in terms of its ethnic composition as compared to the other language groups in Singapore. Data from the 2000 census verifies this. The Malay speech community is comprised largely of Malays (91.2%) with a significant proportion of Indians (6.0%), Chinese (1.3%) and Others (1.5%) (Department of Statistics 2001a). According to the latest census data, in 2005 13.2% of Singapore’s resident population 5 years and above speaks Malay at home most frequently (Department of Statistics 2006). This figure has remained relatively stable throughout the past 15 years (14.3% in 1990 and 14.1% in 2000). Interestingly the proportion of Malay home language speakers is highest for the younger cohorts. The 2005 census reports that 16.9% of 5 to 19-year olds and 16.3% of 20 to 24-year olds prefer Malay as compared to 11.5% of 25 to 44–year olds and 11.6% of 45 and above (Department of Statistics 2006). Since Malay is not only the politically assigned ethnic mother tongue of Malay Singaporeans, but as figures in Kuo (1980b) show, also historically the dominant variety among them (85% in 1957), it is not surprising that Malays have been described (e.g. Stroud 2007) as more resilient to language shift in contrast to the Chinese or Indian communities, whose ‘mother tongues’ have had proportionally fewer speakers. In 1990 93.7% of Malays indicated to use Malay at home with other varieties such as Javanese and Boyanese almost totally eradicated (Department of Statistics 2001a). Since then figures have dropped to 86.8% (Department of Statistics 2006). Census data suggest that this shift is due to the influence of English, as we will see in more detail below. An even more varied picture of the Malay speech community appears when literacy figures for Malay are considered. Of the total literate population above 15 years of age in Singapore, 16.4% can read and understand Malay, a number which is expectedly higher than the proportion of Malay home language speakers. Of those literate in Malay, more than three quarters claim to be at least bilingual in English and Malay (83.6%). Not surprisingly the bilingual literacy rates are particularly high for those under 45 years of age (15 to 24: 97%; 25 to 44: 89.6%), as compared to those 45 and above (61.4%), partly

136

Francesco Cavallaro and Stefan Karl Serwe

because the majority of the former group will have received their education in English with Malay as a language subject. Interesting is also the distribution of people literate in Malay across ethnic groups. Non-Malays make up a quarter of Malay literates (24.9%), with the Chinese accounting for the biggest portion (12.3%), followed by Indians (10.6%). This makes Malay by far the most widely understood language among the ‘mother tongues’ and thus the most ethnically inclusive language besides English. Nevertheless, like the other ‘mother tongues’ in Singapore, Malay seems to be on its way to becoming a more exclusive ethnic language. Unfortunately, no direct comparison with earlier data is possible due to changes in statistical sampling, but undoubtedly Malay literacy among Singaporeans has decreased sharply. Surveys discussed in Kuo (1980b) show that in 1957 almost half of all Singaporeans (48.0%) were able to speak Malay and in 1978 more than two thirds (67.3%) were still able to understand it.Additionally, contemporary language education does little to entice non-Malays to study the language. Although, Malay is offered as a third language option for non-Malays at Singapore’s schools, it has been the practice to allow only the very best students to enroll in a third language (top 10% of each cohort), thus limiting the number severely. Here, Malay also has to compete with other, traditionally more popular modern languages such as Japanese, French, German and, recently, Arabic. However, this regulation has now been somewhat relaxed. Kassim (2008) reports that Malay can now be taken up as an additional language by any student who would like to learn it.

6. Language maintenance and language shift in Singapore Pakir (1998) assigns three roles to the English language in contemporary Singaporean society: utilitarian, unifying and universal. In other words, English serves as a code for international transactions and for interethnic communication. Moreover, these roles mean that it is present in all domains of life, and thus has encroached on territory traditionally occupied by the ethnic ‘mother tongues’. The 28% of Singaporeans that indicated using English as a home language in 2005 (Department of Statistics 2006) might not seem significant, but it presents an enormous increase within only two generations from the 1.8% in 1957 (Kuo 1980b). Similarly, an overall literacy rate in English of 64.4% for the year 2000 (Department of Statistics 2001a) indicates that a third of the population is not able to read English, but a closer look at the data shows that this affects primarily the older generation. English literacy rates for the 15 to 24 year-olds (96.4%) and 25 to 44 year-olds (75.5%) confirm that English-knowing

Language use and shift among Malays in Singapore

137

bilingualism is becoming the norm, or in Riney’s (1998) terms, Singapore is fast moving towards linguistic homogeneity. One doesn’t need to venture very far to find the reasons for this shift. Models on variation in Singapore English (Platt and Weber 1980; Pakir 1991; Gupta 1994) identify two variables as crucial for acquiring and using the full spectrum of English, namely level of education and socio-economic status of the speaker. The models claim that the higher speakers’ education and professional status, and thus the variety of speech situations they are exposed to, the better their ability to master and use the different registers of Singapore English appropriately. Similarly, Bokhorst-Heng (1998) views English proficiency as a prerequisite to perform well academically and professionally in Singapore. As she pointed out, mastery of English “is directly associated with social mobility and socioeconomic status” (1998: 300). That Singaporeans are speaking more and more English is not a new phenomenon. Kong (1977) reported how, in the 1970s, parents were choosing to enroll their children in English-medium schools despite the possibility to enter Tamil, Malay or Mandarin Chinese-medium schools. For many Singaporeans, the struggle for economic success transcends their loyalty to their ethnic tongue. English has become the language of economic success for Singaporeans and with it come all the attractive material gains money and power can buy. However, there is a social cost to all gains. The cost borne by communities shifting to English is the loss of their traditional language. As shown earlier in this article and clearly seen in Figure 1, when compared to the other ethnic groups in Singapore the Malays display a better record of maintaining their language. But, as we will see later in this article, they are showing signs of greater alternation between Malay and English, and an accelerating rate of shift to English. Language maintenance and language shift have been little researched in Singapore. The few studies that have been carried out have ranged from detailed studies of the census data (Kuo 1980b, Kuo and Jernudd 2003) to only a few others that have analyzed the shift of particular ethnic groups in detail, such as Saravanan (1995, 1999) and Schiffman (1998, 2002) who have looked at the shift to English of Tamils; Vaish (2007) who studied the general Indian community; Li et al. (1997) the Teochew; the Cantonese were investigated by Gupta and Yeok (1995); and the Chinese community by Kwan-Terry (1989, 2000) and Xu et al. (1998). Very few have investigated the languages used by the Malays in Singapore in great detail. In fact past research has reduced the issue of language maintenance in the Malay community in Singapore to its association with Islam (Saravanan 1999; Stroud 2007).

138

Francesco Cavallaro and Stefan Karl Serwe

Figure 1. A comparison of English and mother tongue use in Singapore in 2000 and 2005 (Source: General Household Survey, Singapore Department of Statistics 2006)

7. English use among Malay Singaporeans increases significantly In line with the increased use of English in Singaporean homes nationally, there has been a steady increase in the use of English in the homes of Malay Singaporeans over the years. However, Malay Singaporeans show some of the most drastic developments in embracing English, as Table 4 shows. In comparison with the other ethnic groups, the proportion of Malay Singaporeans who use English as a home language has more than doubled within 15 years, whereas the increase among Chinese and Indians, while large, has not been as significant. Data from 2005 show that among the Malays the highest proportion of English users is to be found among the cohorts of 5 to 14 year-olds (17.2%) and 25 to 44 year-olds (16.1%) (Department of Statistics 2006). Table 4. English as home language (%) Year 1990 2000 2005

Total 18.8 23.0 28.0

Malay 6.1 7.9 13.0

Chinese 19.3 23.9 28.7

Indian 32.3 35.6 39.0

(Calculated from Department of Statistics 2001a, 2006)

Language use and shift among Malays in Singapore

139

On the other hand, these same figures have often been used to suggest that Malay Singaporeans still manage to maintain their mother tongue more effectively than other ethnic groups. However, we would argue otherwise. As outlined above there is no doubt that English is recognized among Singaporeans as the language of economic success and social mobility. Thus, a doubling of the number of English users among Malays is in practical terms a very dramatic increase. A correlation between English and socio-economic advancement in Singapore has been posited by some scholars. Analyzing data from the 1990 census, Bokhorst-Heng (1998) highlights that only one out of ten low-income families use English at home. Her analysis also reveals that the use of English as a home language has increased in higher income brackets. Data from the 2000 census (Department of Statistics 2001b) sees three quarters of English language households with incomes above $4000. In comparison, only one quarter of Malay language households has incomes of this level. If one then examines the type of professions the heads of Malay language households are engaged in, a similar picture emerges (Department of Statistics 2001b). Almost three quarters of English language household heads work in high income positions (managers, professionals, technicians), as compared to slightly more than one fifth of Malay-speaking household heads. Malay-speaking household heads are over-proportionally represented in occupations such as clerks and sales staff (28.7%), and almost half of them work in low-income jobs such as craftsmen, machine operators or cleaners. This has important bearings for the Malay community. The real and perceived economic values of English might motivate a shift to English, since it is recognized that not speaking English at home is a factor against upward social mobility. Statistical data (Department of Statistics 2001b) from the education sector point to similar discrepancies between speakers of English or Malay as a home language. The data show that 80% of Malay home language users aged 15 years and above have only graduated with secondary school qualifications and below. While almost two thirds of English home language users have obtained a college or post-secondary school degree. More equity between home language use and education exists only among the population 5 years and older that is still schooling. Here little difference between users of Malay and English as a home language exists. The only exceptions are institutions of higher learning: only 0.5% of Malay language users were attending a university in the year 2000, in comparison to 4.8% of English language users. If we look at this issue within the Malay and Indian communities, a similar pattern emerges. According to data from the General Household Survey 2005 (Department of Statistics 2006), 80% of Malays above the age of 15 who use Malay as a home language have not obtained a qualification higher than

140

Francesco Cavallaro and Stefan Karl Serwe

secondary school. In contrast, 50% of Malays who indicate using English as a home language have qualifications higher than secondary school. Among Indian Malay speakers the proportions are almost identical. If Malays and Indians who are still schooling above the age of 5 are considered, the difference between Malay and English home language users again dissolves. However, the figures for all Malays and the Malay-speaking Indians in institutions of higher learning are still significantly lower than the national total. Therefore, one may conclude that in education the home language has been and still is significant when it comes to high levels of educational achievement.

8. The study A total of 233 Malay Singaporeans took part in this study. The participants were divided into four age groups. Table 5 shows the demographic details of all the participants. Table 5. Participants – Descriptives (by age group)

N Male (N ) Female (N ) Min. Age Max. Age Mean Age / SD

12–17 years 45 22 23 12 17 14.76 / 1.75

18–24 years 65 25 40 18 24 20.22 / 1.64

25–45 years 64 27 37 25 45 33.97 / 7.25

> 45 years 59 28 31 46 72 52.49 / 5.94

A questionnaire was designed to map the language use of the participants in different domains, with different topics and with different interlocutors. Participants were asked to indicate their language use for a particular situation on a 7-point Likert scale with English only (1) and Malay only (7) at its extreme end points. The middle of the scale (4) represents an equal use of Malay and English. The domains were those of the home and in a crowded public place, for example a train or bus stop. The topics ranged from family affairs to the discussion of English and Malay TV or radio programs. The interlocutors ranged from the members of the immediate family, relatives, to close Malay friends. The questionnaire was divided into two sections. The first section contained 78 questions on language use and the second section elicited the demographic data of each participant. The questionnaires were administered to the participants in semi-structured interviews conducted by Malay research assistants in Malay or English accord-

141

Language use and shift among Malays in Singapore

ing to what the participants preferred. In the examples below, for Question 1 the setting is at home. The interlocutors are the participant’s parents and the topic is a family matter. For Question 24 instead, the discussion is with the participant’s younger sibling in a public place (in this case a shop) over a newspaper article (English Media). The participants were asked to circle or indicate to the interviewer whichever number they thought best reflected their use of English and Malay. 1

You are at home alone with your mother or your father; if you are talking about buying a present for a relative, what language do you use? Malay only

Mostly Malay

7 7

6 6

Mother Father

24

More Malay 50/50 More English Mostly English than English than Malay English only 5 5

4 4

3 3

2 2

1 1

You are in a shop with your younger sibling; if you are talking about an article that you have just read in the Straits Times or the New Paper, what language do you use? Malay only 7

Mostly More Malay 50/50 More English Mostly English Malay than English than Malay English only 6

5

4

3

2

1

9. Results The results from the questionnaires were coded into Excel and SPSS and then analyzed. All the results are plotted in the line graphs and tables below. Each line graph shows the average language use in the y-axis. The higher the number (maximum of 7) means the higher the Malay used. The x-axis shows the two domains and the topics of the conversations. The first three topics are presented in the home domain, and then these three topics are repeated in the public domain. 9.1. Age When taking a look at the different age groups’ language use with their immediate family members (Figure 2), the most striking features are the clear demarcations between the lines depicting the interlocutors and the lowering of the lines as the interlocutors’ age decreases in each subsequent chart.

Note: the symbol ‘>’ in the last column denotes a greater use of Malay with the interlocutor on the left of the symbol.

Table 6. Group statistics, family – total language use by age group (p < 0.05 = statistically significant)

142 Francesco Cavallaro and Stefan Karl Serwe

Language use and shift among Malays in Singapore

143

Figure 2. Language use within the family by age group

Results of the statistical analysis (Table 6) prove that Malay is still used very frequently with grandparents and parents, but significantly less with siblings. Although they use mostly Malay, the > 45 group also follows this trend. The greatest stratification can be found for the 25–45 and the 18–24 age groups. Both groups show a declining use of Malay along the three age levels of interlocutors, grandparents, parents and siblings. Members of the youngest age group (12– 17), interestingly, do not seem to significantly alter their language use when talking to parents and siblings, while sticking to a high use of Malay with their grandparents (see also Figure 3). Topic is mostly not a significant factor for the older age groups, but for the younger groups it is significant for some topics with certain interlocutors (see Table 7). When they talk to their parents in a public setting, 18 to 24-year olds use more Malay when discussing a topic which was triggered by a Malay media input as compared to an English language one. They react the same way in conversations with their older siblings in public and at home. Additionally,

144

Francesco Cavallaro and Stefan Karl Serwe

Figure 3. Language use within the family (All age groups)

this group uses more Malay with older siblings at home, if the topic is familyrelated. Similarly, 12 to 17-year olds’ use of Malay increases with their younger siblings in public and at home, when the topic is a Malay TV or radio program as compared to an English one. The charts in Figure 3 are used to compare the overall amount of Malay used by the different age groups with different members of their core family. While grandparents trigger a uniformly high use of Malay, Table 8 shows that the members of the youngest age group use noticeably less Malay as compared to their peers aged 25 and above. Interactions with parents show the greatest differentiation. Again, the 12 to 17 year old group speaks less Malay than the two oldest cohorts, while there seems to be no difference between them and the 18 to 24–year olds. The latter group also shows a significantly reduced amount of Malay with the two older cohorts. However, this clear demarcation is not present when we look at the charts of the language used with siblings. The statistical analysis (Table 8) shows that there appears to be no difference in the amount

Language use and shift among Malays in Singapore

145

Table 7. Group statistics, family – topics (p < .05 = statistically significant) Age group

Interlocutor

Oneway ANOVA Results

18–24

Parents

F(5, 384) = 3.82, p < .05

18–24

Older Sibling

F(5, 365) = 4.56, p < .05

12–17

Younger Sibling

F(5, 372) = 4.82, p < .05

Tukey HSD Post Hoc Test Results Topics with Significant multiple comparisons 1) Public Malay Media > Public English Media 1) Home Family > Home English Media; 2) Home Malay Media > Home English Media; 3) Public Malay Media > Public English Media 1) Home Malay Media > Home English Media; 2) Public Malay Media > Public English Media

Note: the symbol ‘>’ in the last column denotes a greater use of Malay on the topic to the left of the symbol.

of Malay used with siblings among those aged 45 and below. However, all of them use noticeably less Malay with their siblings as compared to the oldest age group. Figure 3 shows that it is members of the 18–24 group that use the least Malay with their siblings and that they are statistically on a par with the 12–17 age group in the use of Malay with their parents. Language use with relatives (Figure 4) bears some resemblance with the interactions with core family members discussed above, but the graphs also highlight interesting differences. The graphs visualize the expected assumption that generally less Malay is used by the younger cohorts. It was also expected that old relatives trigger the highest use of Malay. However, it is interesting to note that the older relatives do not elicit the same high amount of Malay as grandparents. Moreover, age of interlocutor effects are only significant for the > 45 group and the 25–45 group, while the two youngest cohorts do not adjust their patterns of language use to the age of their relative (Table 9). Unlike in conversations with family members, topic does not alter the language of interaction with relatives. When one compares the amount of Malay used by the different age groups with relatives, an interesting result emerges. We see in Figure 5 that the 12– 17 group shows a consistently higher use of Malay than the 18–24 group and even of the 25–45 group when speaking with younger relatives. Table 8 proves that this trend is statistically significant. This high use of Malay by the 12–17 age group is consistent with the census data reported above and is very likely due to these teenagers’ social networks being still more family- and (ethnic) peer group-oriented. Moreover, many students in this group are still taking

Note: The symbol ‘>’ in the last column denotes a significantly greater use of Malay and ‘<’ denotes significantly less Malay.

Table 8. Group statistics, family and relatives – total language use by age (p < 0.05 = statistically significant; ns = not significant)

146 Francesco Cavallaro and Stefan Karl Serwe

43.02

34.14

3 8.62

25–45

18–24

12–17

11.24

11.19

9.21

8.35

SD

33 .6

3 0 .0 3

34.39

45.74

1 2.9 2

11.13

11.73

11.21

36.87

3 0 .2

28.75

38.41

13 .9 2

11.53

10.59

12.47

Younger Relative Mean SD

F(2, 132) = 1.80, ns

F(2, 190) = 2.72, ns

F(2, 189) = 29.63, p < .05

F(2, 169) = 16.72, p < .05

Oneway ANOVA Results

Not applicable

Not applicable

1) Older Relative > Same Age Relative; 2) Older Relative > Younger Relative; 3) Same Age Relative > Younger Relative

1) Older Relative > Younger Relative; 2) Same Age Relative > Younger Relative

Tukey HSD Post Hoc Test Results Significant multiple comparisons

Note: The symbol ‘>’ in the last column denotes a significantly greater use of Malay and ‘<’ denotes significantly less Malay.

49.95

Mean

Older Relative

>45

G r o up

Interlocutor Same Age Relative Mean SD

Table 9. Group statistics, relatives – total language use by age group (p < 0.05 = statistically significant; ns = not significant)

Language use and shift among Malays in Singapore

147

148

Francesco Cavallaro and Stefan Karl Serwe

Figure 4. Language use with relatives by age group

compulsory Malay language classes in school and might therefore use their ‘mother tongue’ more actively. The use of English increases when these people start higher schooling or find employment. The lower Malay scores for the 18 to 24–year olds support such an interpretation. The high use of Malay by the older participants and to the older members of the family is not surprising. Kloss (1966) and Clyne (1982, 1985 and see 2003: 29) placed grandparents as crucial factors in the promotion of LM. Dorian (1977, 1980) reported high maintenance rates of Gaelic in families with living grandparents. Numerous studies around the world have highlighted grandparents and older relatives as important factors in language maintenance (Tosi 1986; Luo and Wiseman 2000; Ishizawa 2004). In an investigation on the Chinese community in England, Li (1994) found that the Chinese dominant are the older speakers and the bilingual or English dominant are the younger speakers. Cavallaro (forthcoming) found the same in his investigation of the Sicilians in Australia. That is, the members of the community who speak the most Sicilian

Language use and shift among Malays in Singapore

149

Figure 5. Language use with relatives (all age groups)

and Italian are the older members of the community and the younger generations are either bilingual or almost totally monolingual in English. A similar process seems to be underway for the Malays in Singapore. 9.2.

Income

Figure 6 relates monthly family household income to the respondents’ language use with immediate family members. According to the latest detailed government sources, 78.6% of Singaporean households have a median monthly income of between S$2700 and S$5400 (Department of Statistics 2009b). This is why our indices assume a low-income household to be below $2000, while a highincome household will earn above $7000. Our data indicate a number of interesting aspects. Similar to findings above, interactions with grandparents are almost an exclusive Malay affair regardless of the speaker’s household income, the domain or topic. For all other interlocutors

150

Francesco Cavallaro and Stefan Karl Serwe

Figure 6. Language use with family by income (all age groups)

some significant stratification across income levels is visible (Table 10) and, as expected, the amount of Malay decreases with decreasing age of the interlocutor, while no difference between older and younger siblings appears. The statistical data presented in Table 10 prove that speakers from families with a household income of $1999 and below use significantly more Malay with all core family members compared to speakers from other income levels. Only in conversation with parents do middle and low income groups not show any differentiation, but together they use more Malay than the high income group. This echoes our analysis of data concerning Malay households from the 2001 census (cf. above) and Bokhorst-Heng’s (1998) findings for earlier census data that English proficiency and household income are directly correlated. However, our results paint a more nuanced picture, because, regardless of their income bracket, speakers report adjusting their speech style to the interlocutor. Among our participants, it is the speakers from high-income families who exhibit the greatest range of variation. Low income families show less variation, but there is a notable amount of English used within the family contrary to what results from the official population censuses suggest. Domain bears no influence on language choice, while the topic is significant only for the middle-income group ($2000–$6999). This group uses significantly more Malay than the other income groups only when the topic is Malay Media in public, rather than English Media, and the interlocutors for this topic are

Language use and shift among Malays in Singapore

151

Table 10. Group statistics, family and relatives – total language use by income (p < 0.05 = statistically significant; ns = not significant) Household Income

Grandparents Parents Older Siblings Younger Siblings Older Relatives Same Age Relatives Younger Relatives Malay Friends

$2000–$6999

Oneway ANOVA Result

below $1000–$1999 Mean SD 51.54 8.58

$7000 and over

Mean 52.92

SD 5.68

Mean 52.24

SD 6.67

46.33

11.66

45.24

10.67

39.59

15.14

42.64

11.83

36.86

12.75

32.12

14.39

40.93

12.23

35.09

13.25

31.76

14.67

43.10

11.18

41.27

11.75

40.11

12.23

41.98

11.90

34.59

12.83

33.75

15.00

37.77

11.99

32.18

12.16

30.50

15.48

39.3

12.24

35.01

12.79

31.29

13.32

F(2,184) = 0.65, ns

Tukey HSD Post Hoc Significant multiple comparisons None

F(2,213) = 3.32, p < .05 1) A > C 2) B > C F(2,199) = 5.53, p < .05 1) A > B 2) A > C F(2,202) = 4.37, p < .05 1) A > B 2) A > C F(2,216) = 0.60, ns None F(2,217) = 5.78, p < .05 1) A > B 2) A > C F(2,220) = 3.96, p < .05 1) A > B 2) A > C F(2,219) = 3.54, p < .05 1) A > C

Key: A = below $1000–$1999; B = $2000–$6999; C = $7000 and over

parents (F(5, 884) = 3.19, p < .05), older siblings (F(5, 837) = 3.86, p < .05) and younger siblings (F(5, 834) = 3.04, p < .05). Language use with relatives over income (Figure 7) resembles the findings on the family. There are no differences between income groups in interactions with the oldest group of relatives, but with same-age and younger relatives members of the lowest income group speak more Malay than the middle and high income households (Table 10). On the other hand, the amount of Malay used is only marginally affected by the age of the relative. The most senior relatives appear to encourage the use of Malay, while same-age and younger relatives are addressed equally in English and Malay. Interestingly, speakers from the lowest income bracket have almost identically high scores for older relatives and same-age relatives. Overall, they exhibit the least variation in comparison to the other groups in this category. Generally, the graphs suggest a move towards a more balanced use of English and Malay for all income groups. Topic is only a significant factor (p < .05) in the language choice for members of the $2000–$6999 group again. Similar to their results with immediate family members above, they demonstrate a significantly higher use of Malay than the other income groups, when the topic is Malay Media

152

Francesco Cavallaro and Stefan Karl Serwe

Figure 7. Language use with relatives by income (all age groups)

in public as compared to English Media, with their older relatives (F(5, 837) = 3.86, p < .05) and younger relatives (F(5, 834) = 3.04, p < .05). 9.3.

Educational level

To measure the influence of speakers’ education onto their language use, four indices where constructed from the eight questionnaire items. Since respondents were asked to indicate their education level currently attained, the first category “Primary School” includes those with a minimum of six years of education and a proportion of the youngest age group (12–17 year-olds), who have not completed secondary school yet. The second category “Secondary School” is made up of those respondents who have obtained a Singaporean secondary school qualification, as well as those with vocational qualifications. Graduates from institutions that offer post-secondary degrees are grouped under “Post-Secondary”. Finally, university graduates of any level are considered in the category “Tertiary”. Respondents without any educational qualification and those with degrees from religious schools are not considered here, because the numbers were not substantial in our sample. Figures 8 and 9 show the graphs for language used with different family members and relatives by the educational level currently attained. In line with findings above, grandparents receive almost purely Malay input from all respondents, independent of their education level. However, when

Language use and shift among Malays in Singapore

Figure 8. Language use with family by educational level (all age groups)

Figure 9. Language use with relatives by educational level (all age groups)

153

154

Francesco Cavallaro and Stefan Karl Serwe

the interlocutor is a sibling or a relative there are significant differences between the languages used according to educational level achieved (Table 11). Figures 8 and 9 reveal similar trends for education to the ones on income mentioned above. The group with the lowest education level uses more Malay with all siblings and relatives than any other group. Those with secondary, postsecondary and tertiary education exhibit the same patterns of language use. Yet, secondary school graduates tend to speak more Malay with their siblings as compared to university degree holders. Most probably due to their close proximity in age there is no significant difference between speakers who have completed secondary school and those who have completed post-secondary education. Considering the degree of language variation between the different groups, it can be detected that, when interacting with family members and relatives, those participants with secondary school qualifications and higher show more adaptability than respondents with only primary school qualifications. The latter group scores consistently high on the scale, which means that they use mostly Malay. The participants who have attained the next three highest levels of education seem to adjust their choice of language according to the age of their interlocutors. The greatest variation in language choice is found among respondents who have completed tertiary education, though a clear shift towards more English is noticeable here. While domain is not significant, topic does seem to influence the language use of only the post-secondary degree holders when speaking to siblings. Members of this group use significantly more Malay when the topic is Malay Media versus Table 11. Group statistics, family and relatives – education (p < 0.05 = statistically significant) Interlocutor

Oneway ANOVA Results

Older Sibling Younger Sibling Older Relatives Same Age Relatives Younger Relatives

F(3,190) = 8.34, p < .05 F(3,194) = 11.41, p < .05

Tukey HSD Post Hoc Test Results Groups with Significant multiple comparisons 1) A>B; 2) A>C; 3) A>D; 4) B>D 1) A>B; 2) A>C; 3) A>D; 4) B>D

F(3,207) = 4.10, p < .05 F(3,206) = 9.28, p < .05

1) A>B; 2) A>C; 3) A>D 1) A>B; 2) A>C; 3) A>D

F(3,2068) = 10.59, p < .05

1) A>B; 2) A>C; 3) A>D

Key: A = Completed Primary School; B = Completed Secondary School; C = Completed PostSecondary School; D = Completed Tertiary Studies

Language use and shift among Malays in Singapore

155

English Media at home with their older sibling (F(5, 377) = 3.90, p < .05) and Malay Media versus English Media in public with their younger sibling (F(5, 408) = 3.50, p < .05). 9.4.

Gender

Since Gal’s (1979) classic study, gender as a factor in LM or LS has been investigated in a number of studies worldwide (Pauwels 1995). Previous studies report that in more established communities older males shift to the majority language faster than older women (Clyne 2003: 34–35). This has been attributed to the high rates of exogamy or the more traditional practices that see the males going out to work and women staying at home to raise the family. Mukherjee (1996), for example, studied the situation among the Panjabi and Bengalis in New Delhi, where the men (of all ages) shifted to Hindi faster than the women. He attributed this to the fact that most of the women in these groups did not work and, therefore, were not under any pressure to shift to the majority language. He posits that in some cases employment and the language associated with it prevail over other factors. An analysis of our results presents a complex and dynamic situation. The analysis of our data points to a slightly different situation for the Malays in Singapore than those reported in the studies cited above. For example, there is hardly any variation due to gender for the oldest group in our sample (> 45) and the youngest (12–17). The 25–45 group also shows little variation according to gender when the interlocutor is an older member of the family. However, Figure 10 illustrates that in the cases when the interlocutor is a younger relative (t(62) = 3.31, p < .05) or a relative of the same age (t(62) = 3.50, p < .05), women seem to speak much more English than males do. This pattern in the use of English is also apparent in the next younger group (Figure 11). The 18–24 year old female participants use more English than their male peers when interacting with all members of their families and relatives except for grandparents (Figure 11). Although the higher use of English by the women is only significant when the participants are speaking with their parents (t(62) = 3.05, p < .05). Figure 12 shows significant differences between the amount of Malay and English used by males and females when the participants are in the middleincome group and they speak to their older siblings (t(136 = 2.76, p < .05) and younger siblings (t(138 = 2.76, p < .05), with the women speaking more English than the men. It is interesting that in our study women from the middle-income group show a significantly higher use of English than men. Among our participants in the

156

Francesco Cavallaro and Stefan Karl Serwe

Figure 10. Language use by gender (25–45)

lower- and higher-income groups the men seem to use more Malay than the women, but statistically the difference was not significant. Other studies have also reported differences in language behaviour between women of different ages. For example, in a study carried out in Malaysia, Mukherjee (2003) reports that her older Bengali participants who did not work had a higher rate of maintenance than the younger women who were pursuing a professional career. She also found that the older Bengali women in Malaysia identify themselves more as Bengalis than Malaysians and saw themselves as the bastions of Bengali culture and traditions. Hence they had a sense of pride in maintaining their linguistic heritage. The younger women instead adopted English quite openly and consciously as a means to access its economic power. In migrant groups or communities with a younger profile and among the younger

Language use and shift among Malays in Singapore

157

Figure 11. Language use by gender (18–24)

people in more established communities, usually it is the women who speak the majority language more and shift to it faster than males (Pauwels 1995). This seems to be what is reflected in our data. All the women in our study use Malay with the older members of the community because it is what the older members expect. At the same time, the younger women who have higher aspirations as far as educational and career prospects are concerned use more English with people of the same age or younger. Bourdieu (1991) made the claim that the language we use can improve our social standings and increase our material gains. There is no doubt that in Singapore the English language enjoys a much higher status and prestige than Malay, and English is associated with higher economic values and education. Studies have linked women’s use of more standard speech norms with women’s

158

Francesco Cavallaro and Stefan Karl Serwe

Figure 12. Language use by gender and income (all groups)

sensitivity to prestige varieties and their more apparent drive for upward social mobility. Nichols (1978) found that the Gullah-speaking African American women in her classic study used far more prestige or standard forms than the men. She attributed this linguistic behaviour to the women’s effort to move up the social ladder. Gal (1979) found that younger women rejected the ‘peasant’ status of Hungarian in favour of German, and more importantly of Germanspeaking men. Though these views which portrayed women as ‘social climbers’ have been critiqued (see Cameron and Kulick 2003), it is an incontrovertible fact that there is robust evidence that women are more sensitive to prestige varieties and more likely to converge to socially preferred norms even though there are many socially motivated reasons for these observations. Smith-Hefner (2009) asserts that young women in Java “cultivate forms of speech which afford

Language use and shift among Malays in Singapore

159

them greater opportunities for social and economic advancement” (2009: 71). Women’s receptiveness to these social issues does not always mean that they shift to the majority language faster than the men. Herbert (1992) reported how in KwaZulu-Natal in South Africa the women among the Thonga immigrants from Mozambique shifted to Zulu much later and more slowly than the men, and when they did speak Zulu, they spoke it with a very noticeable Thonga influence. According to Herbert, this was due to the fact that the women were accorded a much higher status within the Thonga groups than did women in Zulu groups. 9.5.

Other domains

One of the aims of this study was to identify whether a domain exists that is naturally more Malay. Figure 13 plots the language used with other Malays in the wider community. Participants were asked to state how much Malay or English they would use when interacting with Malay staff at work or school; with Malay teachers (but not Malay language teachers); Malay professionals (doctors, lawyers, dentists), and Malay and non-Malay religious scholars.

Figure 13. More general language use (all age groups)

Again we observe a similar stratification among all age groups as reported with the family and relatives. The participants in this study have identified two interlocutors with whom they clearly speak more Malay. One group of people is the Malay shopkeepers and other service providers, such as hawker stall workers,

160

Francesco Cavallaro and Stefan Karl Serwe

Figure 14. Personal language use (all age groups)

barbers and cleaners. The other group of people is the Malay Ustaz or cleric. As observed in the analysis above, the highest users of Malay are the older participants, the participants from the lower socio-economic group and those with the lowest educational qualifications. The highest users of English instead are those from the highest socio-economic group and those with the higher educational qualifications. As far as age is concerned, the 18–24 and the 25– 45 groups seem to show a similar pattern of language use, while the 12–17 show a higher use of Malay in comparison to the 18 to 24-year olds (F(3, 221) = 8.76, p < .05). Some introspective data from the participants of all age groups corroborates the trends discussed above (Figure 14). Again we observe similar stratification among all groups according to age, income and education, as reported earlier in this article. The > 45 group maintains a fairly high level of Malay use except when counting. The other groups only show a significant increase in Malay use when they pray. That is, when they do not pray in Arabic.

10. General conclusions The census data report that 13.2% of the Malays in Singapore use Malay as their home language, and the many studies based on this data cloud the extent to which code variation is taking place within the Malay home, and somewhat obscures the real extent of the LS that is going on in the community. This is

Language use and shift among Malays in Singapore

161

due to the specific wording of the census question, which only asks for the home language. The results from this study clearly show that the home or family settings are not ‘natural triggers’ for the use of Malay. Indeed there seems to be a substantial amount of English spoken within the immediate family and close relatives. The age of interlocutor seems to be the most important factor with regard to choosing the language of interaction. Grandparents serve as the catalyst for almost total use of Malay independent of any topic, place, or any socio-economic variable. This clearly underscores the importance of grandparents in LM discussions in the Malay community. Yet, our results also show that for certain groups media input in Malay is significant and encourages the members of the speech community to use their ethnic language rather than English. The results of this study show an increase in the use of English to and by the younger people, with the lowest Malay use reported by the 18–24 year olds. The youngest group (12–17) consistently reported a higher use of Malay than the next older group and at times of the 25–45 year olds as well. This could be due to this group using Malay as an in-group language, or simply because they are currently actively studying Malay at school. However, as they grow and join the workforce or pursue higher education, their social network goes through further transformation and they then tend to use more English. The results also demonstrate some gender differences. Our data fall in with those from studies on gender issues. That is, older Malay women in Singapore speak more Malay than younger ones, but whether this is significantly more than the men will need further investigation and analysis. However, our data do seem to show that younger women from the middle-income group are leading the shift to English. The women in this study from the age groups that contain the most working people, the 18–24 and 25–45 age groups, use much more English with their siblings and relatives of the same age or younger than do the men. Speakers from the highest socio-economic level and highest education level show the highest rates of English use in the domains explored in this study. The results of our study also support Bokhorst-Heng’s (1998) observations that English use is highly correlated with academic success. Our results clearly indicate that those with higher educational levels use much more English than those with lower educational levels. This group also shows the greatest range in use across both languages. They seem to be at ease with speaking almost totally in Malay with their parents and grandparents and very high levels of English to the younger members of the family. This behavior mirrors Platt and Weber’s (1980) as well as Pakir’s (1991) models for English language use in Singapore that claim that education and socio-economic status is crucial for acquiring the whole spectrum of varieties of Singapore English. While Platt and Weber (1980)

162

Francesco Cavallaro and Stefan Karl Serwe

and Pakir (1991) thought of the speaker’s assessment of formality as the deciding factor in their choice of English, our data identifies age of interlocutor as the overarching determinant for Malay Singaporeans in their choice of language. It is also apparent that, within the context of the immediate family, speakers with higher educational levels exhibits a sensitivity and flexibility to adjust their language(s) of interaction to the different interlocutors and, occasionally, the topics. Therefore, one may conclude that the home language has been and still is important when it comes to high levels of education achievement. The English language will guarantee these achievements to Singaporeans much more than Malay. A relationship between language use and income proposed by BokhorstHeng (1998) is partly supported by our data. Respondents from low income households tend to use more Malay with their family members and relatives, while speakers from high income families report using significantly less Malay. However, we cannot report a shift away from using Malay for the latter. Rather, these speakers demonstrate a greater degree of adjustment of their language repertoire according to the interlocutor and the topic of the discussion than the participants from the lower income group; much in the same way as those participants with higher educational qualifications. All participants report high use of Malay with older interlocutors. However, it is the younger participants and those with a higher education and household income that demonstrate greater proficiency in English. From an LM point of view the question here is what will happen when these older family members and relatives pass on? Our data shows that once the obligatory domain of the grandparents is gone, the only people left that elicit high rates of Malay are the shopkeepers, such as hawker stall owners, and Malay religious clerics. Effective language maintenance is a very difficult issue, as we can see from the poor LM results in the U.S. (Veltman 1984; Fishman, et al. 1985; Hakuta 1986) and in Australia (Clyne 2005; and see summary in Cavallaro [forthcoming]). However, the more positive results from Canada (Swain and Lapkin 1986) and Europe (Baetens Beardsmore 1993) highlight the fact that with careful planning LM is possible. Singapore’s linguistic landscape today is a product of its language policies. These policies were set into practice in the tense 1950s and 60s when the country was just finding its feet. The English plus ‘mother tongue’ policy was implemented to reduce the interethnic divisions, to promote a common Singaporean identity and promote economic growth (Gopinathan 1988, 1998, Gopinathan et al. 2004). Rapid growth and economic success have come hand in hand with and are vindications of the country’s emphasis on English and, later, Mandarin. English is today the de facto national language in Singapore and is perceived as the

Language use and shift among Malays in Singapore

163

key to access world markets. The status of English against that of all the other languages in Singapore has meant that there is no domain in Singapore where English is not spoken. The Malay community relies on extensive codeswitching for all its functions and once the older generation dies out, there will be much less need to speak exclusively in Malay. Many linguists around the world have put forward compelling arguments for the maintenance of minority languages (see Cavallaro [2005] for a summary). However, it is undeniable that official sanctions, together with economic forces, have created a sweeping momentum in favour of English and at the expense of the other languages of Singapore. Our study confirms that young Malays in Singapore speak a notable amount of Malay, but that they then shift to using more English when they join the work force or attend a tertiary institution. Projecting this trend further down the time line, one can only see continuing erosion of the domains where Malay used to be a stronghold and decreasing use of Malay across all age groups. Can something be done to address some of these concerns? Information should be made more readily available to these young people and their parents on the benefits of bilingualism, stressing that acquiring a language other than English does not jeopardize the children’s acquisition of English. Research studies have shown that transgenerational language shift among minority communities is almost inevitable (Veltman 1984; Fishman et al. 1985; Hakuta 1986; O’Bryan et al. 1976; and Wardhaugh 1983). However, as pointed out by Cavallaro (2005) language shift is no more natural than language maintenance. In other words, we should resist the attitude to view LS as a natural consequence of progress and globalization. Though the present generation may not experience LS as a loss, when they finally realize it, it may be too late. We believe that scholars and academics have the duty to provide the community with the necessary information to enable them to make informed decisions regarding their linguistic heritage. This way of thinking has, in recent times, been derided as instigating linguistic ‘moral panic’ (Heller and Duchˆene 2007). However, it remains that LS in such situations is never totally voluntary and though the speakers are active agents in the process, they may choose another course of action, if presented with the relevant information and facts. If we were to be fully committed to the idea of bilingualism, the next phase of educational policies in Singapore should be directed at practical, effective and achievable strategies to reverse the language shift among speakers of all non-English languages spoken in the community. As far as Malay is concerned, there are still enough speakers in the community that if the right steps are taken, the shift to English can be arrested and a more stable form of bilingualism be achieved. In Singapore there is some urgency to promote additive bilingualism and to ensure that minority languages continue to be relevant in the contexts of

164

Francesco Cavallaro and Stefan Karl Serwe

the speakers’ environment. While speakers will always be motivated to make pragmatic choices, enlightened agencies can help by making the environment more conducive to choices that enhance linguistic diversity. Nanyang Technological University, Singapore Notes * A very big thank you to Marliana Binte Mohamed Aron for her help in the data collection and then with the statistical analysis. 1. Note that the term ‘mother tongue’ in the Singapore contexts refers to Mandarin Chinese, Malay and Tamil even if one’s mother tongue is English. Henceforth, the Singaporean usage of the term ‘mother tongue’ will be flagged by single quotation marks.

References Abdullah, Kamsiah & Bibi Jan Ayyub. 1998. Malay language issues and trends. In S. Gopinathan, Anne Pakir, Ho Wah Kam & Vanithamani Saravanan (eds.), Language, society and education in Singapore, 179–190. Singapore: Federal Publications. Alsagoff, Lubna. 2008. The commodification of Malay: Trading in futures. In Peter K. W. Tan & Rani Rubdy (eds.), Language as commodity: Global structures, local marketplaces, 44–56. London: Continuum. Aye, Daw Khin Khin. 2005. Bazaar Malay: History, grammar and contact. Singapore: National University of Singapore doctoral dissertation. Baetens Beardsmore, Hugo. 1993. An overview of European models of bilingual education. Language, Culture and Curriculum 6(3). 197–208. Bokhorst-Heng, Wendy. 1998. Language planning and management in Singapore. In Joseph A. Foley, Thiru Kandiah, Lionel Wee, Bao Zhiming & Anthea Fraster-Gupta (eds.), English in new cultural contexts: Reflections from Singapore, 287–309. Oxford: Oxford University Press. Bourdieu, Pierre. 1991. Language and symbolic power. Cambridge, MA: Harvard University Press. Cameron, Deborah & Don Kulick. 2003. Language and sexuality. Cambridge: Cambridge University Press Cavallaro, Francesco. 2005. Language maintenance revisited: An Australian perspective. Bilingual Research Journal 29(3), 509–530. Cavallaro, Francesco. Forthcoming. Transgenerational language shift: From Sicilian and Italian to English in Australia. Melbourne: The Italian Australian Institute, La Trobe University.

Language use and shift among Malays in Singapore

165

Chew, Phyllis Ghim-Lian. 2006. Language use and religious practice: The case of Singapore. In Tope Omoniyi & Joshua Fishman (eds.), Explorations in the sociology of language and religion, 213–236. Amsterdam & Philadelphia: John Benjamins. Clyne, Michael. 1982. Multilingual Australia. Melbourne: River Seine. Clyne, Michael. 1985. Language maintenance and language shift: Some data from Australia. In N. Wolfson & J. Manes (eds.), Language of inequality, 195–206. Berlin & New York: Mouton. Clyne, Michael. 1991. Community languages: The Australian experience. Cambridge: Cambridge University Press. Clyne, Michael. 2003. Dynamics of language contact. Cambridge: Cambridge University Press. Clyne, Michael. 2005. Australia’s language potential. Sydney: University of New South Wales Press. David, Maya Khemlani, Francesco Cavallaro & Paolo Coluzzi. 2009. Language policies – impact on language maintenance and teaching: Focus on Malaysia, Singapore, Brunei and the Philippines. The Linguistics Journal September 2009 [Special issue: “Language, culture and identity in Asia”]. 155–191. Department of Statistics. 2001a. Census of population 2000 statistical release 2: Education, language and religion. Singapore: Department of Statistics, Ministry of Trade and Industry. Department of Statistics. 2001b. Census of population 2000 statistical release 5: Households and housing. Singapore: Department of Statistics, Ministry of Trade and Industry. Department of Statistics. 2006. General household survey statistical release 1: Sociodemographic and economic characteristics. Singapore: Department of Statistics, Ministry of Trade and Industry. Department of Statistics. 2009a. Monthly digest of statistics Singapore: May 2009. Singapore: Department of Statistics, Ministry of Trade and Industry. Department of Statistics. 2009b. Occasional paper on income statistics: Key household income trends, 2008. Singapore: Department of Statistics. http://www.singstat.gov.sg/ pubn/papers/people/op-s15.pdf (accessed 23 October 2009). Dorian, Nancy C. 1977. The problem of the semi-speaker in language death. International Journal of the Sociology of Language 12. 23–32. Dorian, Nancy C. 1980. Language shift in community and individual: The phenomenon of the laggard semi-speaker. International Journal of the Sociology of Language 25. 85–94. Fishman, Joshua. 1991. Reversing language shift: Theoretical and empirical foundations of assistance to threatened languages. Clevedon et al.: Multilingual Matters. Fishman, Joshua, M. Gertner, E. Lowy & W. G. Milan (eds.). 1985. The rise and fall of the ethnic revival: Perspectives on language and ethnicity. Berlin & New York: Mouton de Gruyter. Gal, Susan. 1979. Language shift: Social determinants of linguistic change in Austria. New York: Academic Press.

166

Francesco Cavallaro and Stefan Karl Serwe

Giles, H., R. Y. Bourhis & D. M. Taylor. 1977. Towards a theory of language in ethnic group relations. In H. Giles (ed.), Language, ethnicity and intergroup relations, 307– 348. London: Academic Press. Gopinathan, Saravanan. 1988. Bilingualism and bilingual education in Singapore. In Christina Bratt Paulston (ed.), International handbook of bilingualism and bilingual education, 391–404. New York: Greenwood Press. Gopinathan, Saravanan. 1998. Language policy changes 1979–1997: politics and pedagogy. In S. Gopinathan, Anne Pakir, Ho Wah Kam & Vanithamani Saravanan (eds.), Language, society, and education in Singapore: Issues and trends, 2nd edn., 19–44. Singapore: Times Academic Press. Gopinathan, Saravanan, Ho Wah Kam & Vanithamani Saravanan. 2004. Ethnicity management and language:Towards a modified model of language education in Singapore schools. In Lai Ah Eng (ed.), Beyond rituals and riots: Ethnic pluralism and social cohesion in Singapore, 228–257. Singapore: Eastern University Press. Gordon Jr., Raymond G. (ed.). 2005. Ethnologue: Languages of the world, 15th edn. Dallas, Tex.: SIL International. http://www.ethnologue.com/ (accessed 2 July 2009). Gupta, Anthea Fraser. 1994. The step-tongue: Children’s English in Singapore. Clevedon: Multilingual Matters. Gupta, Anthea Fraser. 1998. The situation of English in Singapore. In Joseph A. Foley, Thiru Kandiah, Lionel Wee, Bao Zhiming & Anthea Fraser Gupta (eds.), English in new cultural contexts: Reflections from Singapore, 106–126. Oxford: Oxford University Press. Gupta, Anthea Fraser & Siew Pui Yeok. 1995. Language shift in a Singapore family. Journal of Multilingual & Multicultural Development 16(4). 301–314 Hakuta, K. 1986. The mirror of language: The debate on bilingualism. New York: Basic Books. Heller, Monica & Alexandre Duchˆene. 2007. Discourses of endangerment: Sociolinguistics, globalization and social order. In Alexandre Duchˆene & Monica Heller (eds.), Discourses of endangerment, ideology and interest in the defence of languages, 1–13. London: Continuum. Herbert, R. K. 1992. Language in a divided society. In R. K. Herbert (ed.), Language and society in Africa: The theory and practice of sociolinguistics, 1–19. Johannesburg: Witwatersrand University Press. Ishizawa, Hiromi. 2004. Minority language use among grandchildren in multigenerational households. Sociological Perspectives 47(4), 465–48. Kamwangamalu, Nkoko M. 2006. Religion, social history and language maintenance: African languages in post-apartheid SouthAfrica. In Tope Omoniyi & Joshua Fishman (eds.), Explorations in the sociology of language and religion, 86–96. Amsterdam & Philadelphia: John Benjamins. Kassim, Aishah Mohamad. 2008. Malay language as a foreign language and the Singapore’s education system. GEMA Online Journal of Language Studies 8(1). 47–56. Kloss, Heinz. 1966. German-American language maintenance efforts. In Joshua Fishman (ed.), Language loyalty in the United States, 206–252. The Hague: Mouton.

Language use and shift among Malays in Singapore

167

Kong, Chong Hoi. 1977. A study of language maintenance and shift in Singapore as a multilingual society. RELC Journal 8. 43–62. Kuo, Eddie C. Y. 1980a. Population ratio, intermarriage and mother tongue retention. In Evangelos A. Afendras & Eddie C.Y. Kuo (eds.), Language and society in Singapore, 254–264. Singapore: Singapore University Press. Kuo, Eddie C. Y. 1980b. The sociolinguistic situation in Singapore: Unity in diversity. In Evangelos A. Afendras & Eddie C.Y. Kuo (eds.). Language and society in Singapore, 39–62). Singapore: Singapore University Press. Kuo, Eddie C. Y. & Bjorn H. Jernudd. 2003. Balancing macro and micro sociolinguistic perspectives in language management: The case of Singapore. In J. Lindsay & Tan Ying Ying (eds.), Babel & behemoth: Language trends in Asia, 103–123. Singapore: Asia Research Institute, National University of Singapore. Kwan-Terry, Anna. 1989. Education and the pattern of language use among ethnic Chinese school children in Singapore. International Journal of the Sociology of Language 80. 5–31. Kwan-Terry, Anna. 2000. Language shift, mother tongue and identity in Singapore. International Journal of the Sociology of Language Vol. 143. 85–106. Lee, Huiying Nala, Ai Ping Ling & Hiroki Nomoto. 2009. Colloquial Singapore English got: Functions and substratal influences. World Englishes 28(3). 293–318. Li Wei. 1994. Three generations, two languages, one family. Language choice and language shift in a Chinese community in Britain. Clevedon: Multilingual Matters. Li, Wei, Vanithamani Saravanan & Julia Ng Lee Hoon. 1997. Language shift in the Teochew community in Singapore:A family domain analysis. Journal of Multilingual and Multicultural Development 18(5). 364–384. Llamzon, Teodoro A.. 1978. English and the national languages in Malaysia, Singapore and the Philippines: A sociolinguistic comparison. Cross Currents 5(1). 87–104. Lowenberg, Peter. 1988. Malay in Indonesia, Malaysia, and Singapore: Three faces of a national language. In Florian Coulmas (ed.), With forked tongues: What are national languages good for?, 146–178. Ann Arbor: Karoma Publishers. Luo, Shiow-Huey & Richard L. Wiseman. 2000. Ethnic language maintenance among Chinese immigrant children in the United States. International Journal of Intercultural Relations 24(3). 307–324. Mukherjee, Aditi. 1996. Language maintenance and language shift: Panjabis and Bengalis in Delhi. New Delhi, India: Bahri Publications. Mukherjee, Dipika. 2003. Role of women in language maintenance and language shift: Focus on the Bengali community in Malaysia. International Journal of the Sociology of Language 161. 103–120. Nichols, Patricia C. 1978. Black women in the rural South: Conservative and innovative. International Journal of the Sociology of Language 17. 45–54. O’Bryan, K. G., J. G. Reitz & O. M. Kuplowska. 1976. Non-official languages: A study in Canadian multiculturalism. Ottawa: Supply and Services Canada. Pakir, Anne. 1991. The range and depth of English-knowing bilinguals in Singapore. World Englishes 10(2). 167–179.

168

Francesco Cavallaro and Stefan Karl Serwe

Pakir, Anne. 1998. Language and society. In Lubna Alsagoff, Zhiming Bao, Anne Pakir & Lionel Wee (eds.), Society, style and structure in language, 3–107. Singapore: Prentice-Hall. Pakir, Anne. 2004. Medium-of-instruction policy in Singapore. In J. W. Tollefson & A. B. M. Tsui (eds.), Medium of instruction policies: Which agenda? Whose agenda?, 117–133. Mahwah, NJ: Lawrence Erlbaum Associates. Pauwels, Anne. 1995. Linguistic practices and language maintenance among bilingual women and men in Australia. Nordlyd 11. 21–50. Platt, John & Heidi Weber. 1980. English in Singapore and Malaysia. Kuala Lumpur: Oxford University Press Reagan, T. 2001. The promotion of linguistic diversity in multilingual settings: Policy and reality in post-apartheid South Africa. Language Problems & Language Planning 25(1). 51–72. Riney, Timothy. 1998. Toward more homogeneous bilingualisms: Shift phenomena in Singapore. Multilingua 17(1). 1–23. Saravanan, V. 1995. Linguistic and cultural maintenance through education for minority groups in Singapore. In Makhan L. Tickoo (ed.), Language and culture in multilingual societies: Viewpoints and visions, Pathology series 56, 139–151. Singapore: SEAMEO Regional Language Centre. Saravanan, V. 1999. Bilingual Chinese, Malay and Tamil children’s language choices in a multilingual society. Early Child Development and Care 152. 43–54. Schiffman, Harold F. 1998. Standardization or restandardization: The case for ‘Standard’ spoken Tamil. Language in Society 27. 359–385. Schiffman, Harold F. 2002. Tongue tied in Singapore: A language policy of Tamil? Journal of Language, Identity and Education 2(2). 105–125. Schiffman, Harold F. 2007. Tamil language policy in Singapore: The role of implementation. In Viniti Vaish, S. Gopinathan & Liu Yongbing (eds.), Language, capital, culture: Critical studies of language in education in Singapore, 209–226. Rotterdam: Sense Publishers. Smith-Hefner, Nancy J. 2009. Language shift, gender, and ideologies of modernity in Central Java, Indonesia. Journal of Linguistic Anthropology 19(1). 57–77. Stroud, Christopher. 2007. Multilingualism in ex-colonial countries. In Peter Auer & Li Wei (eds.), Handbook of multilingualism and multilingual communication, 509–538). Berlin & New York: Mouton de Gruyter. Swain, Merril & Sharon Lapkin. 1986. Secondary French immersion: The goods and the bads. Contact 5(3). 2–9. Tan, Jason. 2007. Schooling in Singapore. In Gerard A. Postiglione & Jason Tan (eds.), Going to school in East Asiaý, 301–319. Westport: Greenwood Publishing Group. Tollefson, J. W. 1991. Planning language, planning inequality. London: Longman. Tosi, Arturo. 1986. Home and community language teaching for bilingual learners: Issues in planning and instruction. Language Teaching 19. 2–23. Vaish, Viniti. 2007. Bilingualism without diglossia: The Indian community in Singapore. The International Journal of Bilingual Education and Bilingualism 10(2). 171–187.

Language use and shift among Malays in Singapore

169

Vaish, Viniti. 2008. Mother tongues, English and religion in Singapore. World Englishes 27(3/4). 450–464. Veltman, C.. 1984. Language shift in the United States. Berlin, Germany: Mouton. Wardhaugh, R.. 1983. Language and nationhood: The Canadian experience. Vancouver, Canada: New Star Books. Wee, Lionel. 2003. Linguistic intrumentalism in Singapore. Journal of Multilingual and Multicultural Development 24(3). 211–224. Xu, Daming, Cheng Hai Chew & Songcen Chen. 1998. Language use and language attitudes in the Singapore Chinese community. In Saravanan Gopinathan, Anne Pakir, Ho Wah Kam & Vanithamani Saravanan (eds.), Language, society and education in Singapore, 133–155. Singapore: Times Academic Press. Francesco Cavallaro teaches sociolinguistics in the new Division of Linguistics and Multilingual Studies. He has published on language maintenance and shift, and language policy and attitudes in Singapore. He is currently researching the language use in the Malay and Hindi communities and the social aspects of bilingualism in Singapore’s multilingual context. His key publications include Transgenerational language shift in an immigrant setting: From Sicilian and Italian to English in Australia (The Italian Australian Institute, La Trobe University, forthcoming). Stefan Karl Serwe <[email protected]> is a lecturer in English communication skills and German language in the Language and Communication Centre at Nanyang Technological University. His research interests are in the areas of (interactional) sociolinguistics, pragmatics, conversation analysis and language contact.

Family language policy: Core issues of an emerging field* MILA SCHWARTZ

Abstract Research in the field of minority language maintenance and loss regards the family as the central driving force in children’s language socialization within the context of both minority and majority languages. The present review of current literature focuses on the emerging field of family language policy and addresses the following three components: ideology, management and practice. The integrated survey reviews research on the intra-family factors that can drive family language policy, the manner in which parents (mostly immigrants) and children cope with the minority-majority language reality, and the ways in which parents conduct their language policy at home. Keywords: language maintenance; family language policy; research methodology.

1. The scope of the present review From its inception, research in the field of language maintenance and shift has underscored the critical role of the family in the preservation of immigrant and ethnic minority languages (also referred to here as L1; home language, or heritage language). The review focuses on the newly emerging field of family language policy, which according to King et al. (2008) “provides an integrated overview of research on how languages are managed, learned and negotiated within families” (2008: 907). In particular, the present paper focuses on how parents (mostly immigrants) face the challenges of the minoritymajority language reality, in which children generally grow up with the minority language spoken at home and the majority language spoken in the community.

172

Mila Schwartz

Research on family language policy (FLP) incorporates analysis of language ideology, practice and management, which were classified by Spolsky (2004) as components of the language policy model with respect to the speech community. In distinguishing these three components, Spolsky (2004: 5) notes “. . . language practices – the habitual pattern of selecting among the varieties that make up its linguistic repertoire; its language beliefs or ideology – the beliefs about language and language use; and any specific efforts to modify or influence that practice by any kind of language intervention, planning or management.” Using this model at the family level enables us to integrate the separate components within a structural, flexible, and expandable framework. Although the concept of FLP has been defined only recently, existing research surveys reveal that a focus on family language ideology, practice, and management arouses keen interest worldwide. The present review of a narrow period (approximately 1998–2008) includes research that takes a close look, from a socio-linguistic and socio-psychological perspective, at the full complexity that parents face in their efforts to transmit the home language to the second-generation immigrants and ethnic minority children. Therefore, the review addresses the studies that investigated: (a) the role of the family in home language maintenance; (b) intra-family factors related to FLP; (c) family language ideology and practice; (d) family language practice and management; (e) the challenges of FLP; (f) methodologies of FLP research; and (g) further research directions. Note that the questions of language contact, social networks, and friendship networks are not within the scope of the present review.

2. The role of the family in home language maintenance The family is considered to be an extremely important domain for studying language policy because of its critical role in forming the child’s linguistic environment. Fishman (1991), an early proponent of proactive language maintenance research, called for a reversal of language shift (RLS) through efforts to retain ethnic languages at the level of the family and the community. According to Fishman (1991), the family contains a natural boundary that serves as a bulwark against outside pressures. Association with intimacy and privacy makes the family particularly resistant to outside competition and substitution. Although the modern urban family has lost much of its socialization power, it is nevertheless “the most common and inescapable basis of mother tongue transmission, bonding, use and stabilization” (1991: 94). Fishman (2001) showed that the desire to maintain and transmit the home language is not anti-modern and represents a welcome alternative to complete globalization. More specifi-

Family language policy

173

cally, Fishman (2000) identified the most important point of intergenerational language transfer as the use of the ethnic language at home by women of childbearing age with their children, because the family and community are critical for the maintenance of the home language. Therefore, focusing on the nuclear traditional family with children we can explore more closely the children’s language socialization within the context of both minority and majority languages (Spolsky 2007), in other words, the way in which “younger children . . . through interactions with older and more experienced persons, acquire the knowledge and practices that are necessary for them to function as, and be regarded as, competent members of the communities” (Garrett and BaquedanoLopez 2002: 341).

3. Intra-family factors of FLP Several elements in the family’s background appear to be related to FLP. In exploring the factors closely linked to the parents’ decision on whether to abandon or reinforce the heritage language, some of the studies refer to elements in the social and community background, which are outside the family domain, as for example ethnolinguistic vitality aspects (Giles et al. 1977) and community language acculturation patterns (such as power relationships) (Lambert and Taylor 1996). The present review focuses on several demographic, socio-cultural, and psychological factors that were found to be directly related to family background and could drive FLP. 3.1.

Family structure

Family structure, in particular the presence of older children and sibling position, plays an importance role in intergenerational L1 transmission (Baker 2001; Fishman 1991; Harris 1995; Spolsky 2007; Wong Fillmore 1991). Older siblings seems to play an essential role in the language socialization of the younger ones (Kyratzis 2004; Spolsky 2007; Zhu and Li 2005), but the direction of their influence remains equivocal. Spolsky (2007) argues that the older children bring the majority language into the home and speak it occasionally with the parents and regularly with younger siblings. There is accumulating evidence about the role of older siblings as mediators in the younger siblings’ majority language literacy among immigrant families (Gregory 2004; La Piedra and Romo 2003). However, there is also the opposite example of interaction between older and siblings in the case of the multi-children family described by Kopeliovich (forthcoming). In this family, all the siblings followed their mother’s rules of strictly

174

Mila Schwartz

speaking the home language with infants and toddlers at least until they started their formal preschool education. We know that the older siblings affect FLP, but we lack detailed studies and have few clear indications regarding actual language interactions between siblings at home. 3.2.

Parental education

Findings concerning parental education are inconsistent. It has often been claimed that ethno-linguistic minorities need a strong educational experience in their own language and tradition in order to maintain their mother tongue and ethnic identity (Kloss 1966; Lambert and Taylor 1996; Allard and Landry 1992). King and Fogle (2006) found a high level of education relative to the total population among American families promoting heritage language retention and bilingual education. At the same time, Doucet (1991) and Harres (1989) found that the opposite was true, so that the higher the educational level of the informants was, the greater their shift away from L1 usage. Results concerning socio-economic status (SES) were also contradictory: some families of lower SES were favorably disposed toward language shift, whereas families from higher SES favored language maintenance (Williams 1987; Lambert and Taylor 1996). However, other studies point in the opposite direction, indicating that high SES is likely to result in language shift (Harres 1989). 3.3. Acculturation of the parents Another important factor affecting FLP is the parents’ acculturation to the host country’s culture. Acculturation is the process of adapting to a new culture. It has been found that the younger the age of the immigrant at arrival, the greater the language shift is (Doucet 1991). Age at arrival was also found to correlate with linguistic habits and behaviors, with older members of the community being the more traditional speakers and using L1 more frequently (Clyne 1982). Similarly, the length of residence of immigrant families in the host country was observed to be strongly associated with both L2 proficiency and L1 attrition among immigrant children (Baker 2001). The longer the time spent in the host country, the better the command of L2 was and the greater the language shift. Finally, cultural identification with the host country and the country of origin are significant factors in the formation of FLP among immigrants. For example, Pease-Alvarez (2003) conducted in-depth interviews with 63 parents who were first and second generation immigrants from Mexico in California. The findings point to a strong tendency of some participants to abandon the use of

Family language policy

175

Spanish with their children and adopt English monolingual norms and Anglo values in an effort to improve their social status and new cultural identity, and to enjoy the benefits associated with becoming Americans. In another example, research on the recent wave of Russian-Jewish immigrants to Israel shows both the powerful assertion of the immigrants’ original cultural identity (Russian) and their openness to possessing a strong Jewish identity (Ben-Rafael et al. 1997; Donitsa-Schmidt 1999). In sum, it is clear that the cultural identity of the immigrant parents is strongly linked to interfamily aspects (e.g., economic and social status of the language minority group, language and culture pride and awareness). 3.4.

Family cohesiveness and emotional relations

Research conducted by Wong Fillmore (2000), Okita (2002) and Tannenbaum (2005) addressed directly for the first time the emotional aspects of home language maintenance or loss. The preservation of L1 was found to be relevant not only to the survival of the minority language from a purely linguistic perspective, but also as a link between the generations and cultural values of the ethnolinguistic group (Wong Fillmore 2000; Tannenbaum and Howie 2002). Children are brought up to become members of their cultural group in part by the way in which their parents interact and use heritage language with them, especially in early childhood. Parents often view the children’s socialization into their culture through use of the home language as a positive symbol of cultural pride and a tool that strengthens family cohesion. By contrast, a language shift in the family leads by the children (Spolsky 2004, 2007), can be expressed in the conflicting intergenerational talks about social, cultural and linguistic practice (Caldas and Caron-Caldas 2002; Hua 2008) and has a negative effect on family relations if adults and children speak different languages (Wong Fillmore 2000). The parents’initial decision on language maintenance or shift may be strongly related to complex emotional processes. Tannenbaum (2005) analyzed the link between past and present family relations on one hand and home language maintenance on the other among immigrant families in Australia in light of psychological motives and emotional aspects. To the extent that home language maintenance can serve as a powerful tool for cohesion between generations of immigrants, its loss can contribute greatly to creating emotional distance between past and present. Tannenbaum (2005) brings several examples of immigrant narratives reflecting a tendency to build a barrier between their painful childhood experiences in the country of origin and their present “rehabilitation” in the host country through L1 loss and shifting to L2 in family communication. Lina, who immigrated to Australia from the former Soviet Union (FSU), ac-

176

Mila Schwartz

knowledged her conscious choice of English (L2) for family language practice as her way to build a “different kind of family” than hers was: I realized that I had hoped to have a completely different life, a different family. I didn’t want to raise my boys like my parents had raised me . . . I was never very close to my brother and sister, and never knew much about my parents’ own life. They did not tell me much about themselves, about their problems, about their own childhood. (Tannenbaum 2005: 237)

Tannenbaum (2005) noted that the barriers immigrants erect between their past and present as a way to establish a “different family” do not work: “Their failure to use their mother tongue with their children may be one aspect of the tendency to detach themselves from their own past and both [parents] feel that, to some extent, it is alienating them from their sons.” (Tannenbaum 2005: 245). Okita (2002) also described the phenomenon of native language avoidance among Japanese mothers in the UK married to English men. These mothers decided at the outset to use English with their children. The reasons for parents not to speak their native language with their children have to do with their attitudes and personal experiences with ethnicity. Okita identified two reasons underlying the choice of English, both linked to the mothers’ childhood experiences: dislike of the “Japanese way” of relating to children and avoiding the preservation of old ties with their families in Japan. These mothers reported negative experiences with their families in Japan. Well, it was always work, work, work. We hardly had dinner together, just on New Year’s Day . . . Family ties . . . I didn’t have it . . . My father did what he wanted to do and my mother suffered from it. (Fumi Findlay in Okita 2002: 92)

Moreover, the Japanese language is strongly associated with the behavior patterns of the “Japanese way”. When I speak in Japanese, I become like a Japanese mother and I don’t like it. (Mrs. Inwood in Okita 2002: 92)

In sum, in-depth investigation of the interaction between the parental decision about the language of childrearing is a critical step in the realization of FLP. Focusing on the inner world of immigrants by means of ethnographic interviews contributes a new perspective to the study of home language retention as a social and psychological phenomenon.

Family language policy

177

4. Family language ideology and practice Resent research on FLP exposes the full complexity and non-linearity of relationships between parental language ideology and actual language practice and management. On one hand, developmental psychologists suggest that parents’ ideas motivate their practices, which in turn are strong determinants of the children’s development (De Houwer 1999; Johnson and Martin 1985). In both monolingual and bilingual contexts, children’s linguistic environments are shaped to a large degree by the parents’ beliefs and attitudes, which constitute the primary environments of early childhood. Parental beliefs about how children acquire language or languages (L1 or L2) and about their own roles in this process appear to have a substantial effect on the parents’ linguistic behavior (language practice and management) toward their children (De Houwer 1999; Spolsky 2007). For example, Barkhuizen (2006) found that Afrikaans-speaking South African immigrant parents in New Zealand believed that their children’s transition to English would be smooth if they were socialized in English during the pre-immigration stage. On the other hand, the declared language ideology of one or both parents does not necessarily coincide with the strategies followed consciously or unconsciously in language practice with children (De Houwer 1999; Goodz 1994; King 2000; Kopeliovich forthcoming; Schwartz 2008; Spolsky 2004). For example, in a dramatic description of the parents’ feelings of dissatisfaction with L1 transmission, Kopeliovich (forthcoming) investigated FLP in one immigrant family from the FSU in Israel. Using in-depth interviews and observations, Kopeliovich uncovered the hidden sides of FLP and showed how the pro-activist mother gradually reconsidered her LP and renounced her strategy of overt interference with the children linguistic behavior. The children’s resistance to use of Russian (L1) at home after immigration to Israel, and sharp conflicts within the family forced the mother to look for new solutions and change her “Russian only” practice at home. Furthermore, her awareness of the inconsistency between her ideology and the actual language practice was painful and disappointing. This discrepancy has also been observed at the structural linguistic level. For instance, contact-linguistic analysis (Myers-Scotton 2002) of mother-child communication revealed many types of contact varieties combining Hebrew and Russian elements within the mother’s discourse. Along the same lines, Schwartz (2008) found discrepancies between immigrant parents’ declared commitment to L1 maintenance among second generation Russian-Hebrew speaking children and their reports of actual language practice with their children and of written language management at home. Although almost all parents reported having positive attitudes toward Russian

178

Mila Schwartz

language preservation, only 27% stated that they themselves conducted reading instruction. Moreover, in families that practice code-mixing, the second generation showed lesser knowledge of Russian (L1) vocabulary. At the same time, it was found that the children’s positive attitude toward the development of both languages at home and toward heritage literacy acquisition at school contributed significantly to their lexical knowledge of L1. Thus, the children’s reports about their attitudes toward L1 maintenance, in contrast to those of their parents, seem to reflect the real family language policy and appear to be an indicator of the effectiveness of their parents’ efforts to preserve the Russian language at home. To conclude, the data reveal that the links between the parents’ language ideology, practice, and management may be indirect and even conflicting.

5. Practice and management Family language practice refers to patterns of language choice and preference within the family and in different contexts. This practice could reflect sociocultural changes in intergenerational interactions within immigrant families. Hua’s (2008) qualitative study of Chinese immigrant families in the UK vividly displayed conflict talk between Chinese mothers and their second generation adolescent children. The mothers positioned their cultural norms and customs (i.e., the cultural principle of reciprocity) which are valued in Chinese ethnic communities while the children, in turn, tried to challenge the family’s sociocultural values. Both the mothers and the children used a wide repertoire of linguistic resources (e.g., codeswitching from English to Mandarin and vice versa, children’s choice of the English pronoun you instead of the Chinese polite form nin when speaking to their mothers) in order to stress their socio-cultural perspectives. Furthermore, recent studies by Caldas and Caron-Caldas (2000, 2002) pointed out that children’s preferences for either of their two languages is highly sensitive to environmental context. The researchers coined the term “children’s bilingual preference”, which they measured by constructing a bilingual preference ratio (BPR), aiming to understand the effect of external and internal influences on the language preferences of their own three English-French speaking bilingual children as they progressed through adolescence. They calculated the BPR weekly using audio-taped recordings of family mealtime conversation over a period of six years in which the family spent the school years in Englishspeaking Louisiana and summer vocations in French-speaking Qu´ebec. The parents, a native French-speaking Canadian mother and a native English-speaking father, decided to raise their children in a predominantly French-speaking home.

Family language policy

179

Consequently, both parents strongly supported a preference for French in family communication. During the academic year, the children were exposed to predominantly English speaking in Louisiana, but the twin girls were enrolled in a French immersion program during 5 years of the 6 years of the study, whereas their older brother spent only one semester in that school. In Qu´ebec, the children lived in an entirely French-speaking social environment. Upon entering adolescence, the older son’s language behavior underwent a striking shift in preference to the socially dominant English, as an expression of disengagement from family control. The boy also began to tease his sisters for speaking French. In time, the twins’ French-speaking also dropped dramatically despite the fact that they were enrolled in a French immersion program and had friends from French-speaking homes. Thus, despite the predominantly native French- speaking milieu, the adolescents preferred English as the vernacular for both peer-peer social function and communication with their parents. At the same time, during their relatively short involvement within the French-speaking Qu´ebec context, all three children manifested massive growth in the number of French words spoken around the family dinner table, in particular after returning from summer camp. In sum, this in-depth case study underscored the overwhelming influence of peer control on language practice when the children enter adolescence. The data support the Group Socialization Theory (Harris 1995). It stresses that children favor the behavioral system of the peer group outside the home over the one they acquire at home. Harris used the case of bilingual families to illustrate this claim because ethnic minority children tend to shift from the heritage language to the dominant one and acquire the pronunciation of their peers rather than that of their parents. Although the parents were shocked by the pervasive nature of the social force and felt helpless in their efforts to maintain French, they did not attempt to impose their language ideology by rigid rules. There is sufficient data on immigrant adults’ language preferences (DonitsaSchmidt 1999; Fishman 1991; Hakuta and D’Andrea 1992; La Piedra and Romo 2003). These studies provide clear evidence for domain separation, that is, a preference for the home language in the private domain (for example, in intimate communication with spouse and children) and the use of L2 largely in public domain (at work and within the framework of host country institutions, such as government agencies). There is less data about the children’s language preferences. To investigate this issue, Tannenbaum (2003) administered questionnaires to 307 children, aged 8 to 11 years, living in Sydney, Australia, and to one of their immigrant parents from various ethnic backgrounds. The findings suggest that parents and children differ in their language maintenance patterns: whereas parents differentiated between the public and intimate domains even when com-

180

Mila Schwartz

municating with their child, and were rather context sensitive, the children’s choices of language were not linked to specific domains of interaction with the parents, with a strong tendency to use the majority language across domains. These findings have been attributed to the possible lack of relevance of domain separation for children for whom intimate interactions do not result in higher use of the home language and may not be tied to closeness, as is the case for their parents. In analyzing this finding, Tannenbaum (2003) suggested that the lack of domain effect among the children may also be explained by methodological constraints, namely by the fact that the focus was only on the children’s language interactions with parents, which may restrict the generalizability of the data. Family language management refers to “efforts to control the language of family members, especially children” (Spolsky 2007: 430). It starts with the parents’ decision about the language choice to be used with the children. This initial decision is considered to be a crucial factor in L1 retention (Fishman 1991; Spolsky 2007). At the same time, the absence of an explicit decision concerning initial language choice in communication with the children may be interpreted, according to Spolsky and Shohamy (1999), as an absence of a conscious and knowledgeable FLP. Okita (2002) and Schwartz et al. (forthcoming) found that in reality bilingual family decisions regarding initial language use do not always involve clear processes and arise at times spontaneously, without discussion. The unplanned pattern of parental behavior may simply reflect the fact that most of the parents are not professional linguists (Okita 2002). This pattern can also be attributed to the shortage of educational professionals and of specialized literature on raising bilingual children in countries without a longstanding tradition of training professional staff for bilingual education (e.g., Israel, France). Finally, as was suggested by Schwartz (2008), the absence of a clearly defined decision at the family level may also reflect a common situation in the country or region of the family’s residence regarding language practice, as in the case of Israel, where there is no law defining state LP (Spolsky and Shohamy 1999). We distinguish two central tendencies in the data on family language management: first, seeking external control for FLP by searching for a supporting socio-linguistic environment; second, controlling the home language environment (establishing family cultural traditions and rituals strongly associated with L1 or a regime of penalties and rewards for using a particular language at home). Note that all these tendencies may co-exist within a single family and appear in considerably modified presentations under different conditions.

Family language policy

5.1.

181

External control for FLP

It has been found that parents, searching for external control for a supporting socio-linguistic environment, can plan several relevant strategies and implement them. For example, in his study of FLP among Afrikaans-speaking South African immigrants living in New Zealand Barkhuizen (2006) used in-depth narrative interviews and found a tendency among parents to plan for their children’s maximum exposure to L1, including the choice of suburbs with a high concentration of South African immigrants. Recall that in the case of the three bilingual English-French speaking adolescents (Caldas and Caron-Caldas 2002), it was the external situation (enrolling the children in an all-French speaking biking excursion and summer camp in Qu´ebec) rather than the parents’ practices and wishes that accounted for their preference for French over English at home. The choice of bilingual education serves as an important link in the practical realization of family language ideology. There is clear consensus about the critical role of early education in the maintenance and intergenerational transmission of the minority language (Baker 2001; Fishman 1991; Spolsky 2007; Wong Fillmore 1991). Baker (2001) distinguished two forms of early education for second generation immigrants: weak forms, in which children are educated only through the medium of the majority language, and strong forms, that permit substantial support for the children’s L1 maintenance in an environment of two reciprocally enriching languages and cultures. Note that ‘strong’ bilingual education creates an additive bilingual environment (Lambert 1975), in which there is substantial support for children to maintain their L1 as they acquire an additional language. By contrast, acquisition of the majority language (L2) at the expense of the children’s L1, as in a case of ‘weak’ bilingual schooling, results in a subtractive bilingual environment. What are the distinctive characteristics of parents who choose bilingual education for their children? King and Fogle (2006) aimed primarily at providing an in-depth description of how parents initially establish their family language policy and decide about the children’s education. Twenty-four middle-class, highly educated parents in the U.S. participated in the ethnographic interviews and reported on their motives for choosing a bilingual environment for their SpanishEnglish speaking children. The study found that parents often relate critically to any source of available advice or information concerning the education of minority language children. Moreover, even when the parents explained their choice of additive bilingualism by reference to the popular press and parenting advice literature, they were motivated primarily by their personal experience with language learning and practice in their childhood and in their extended

182

Mila Schwartz

family. Many parents believed that they had missed opportunities by not being educated bilingually, which shaped their decision to raise their children within an additive context. Furthermore, the parents’ personal sources of information were often linked to a critical evaluation of their relatives’ decisions not to expose their children to the heritage language. Thus, the parents’ decisions about bilingual education emerged as an effort to provide good parenting to their children against a background of unsuccessful examples from their own childhood or that of other family members. To explore the factors favoring the choice of early bilingual vs. monolingual education among second generation immigrants, Schwartz et al. (forthcoming) proposed an extended model of FLP by combining two components: the context of FLP and its consequences. The context refers to the structure of the family (marital status and the number and order of children), its socio-economic status and educational level; variables related to the parents’ acculturation in the host country (duration of residence in the host country, age at immigration, education in the host country, and cultural self-identification); the child’s history; the parents’ language competence, and their attitudes and expectations regarding the child. The consequences of FLP refer to the child’s language competence; changes in the child’s language competence and practice upon entering an educational institutions; and the parents’ satisfaction with these institutions. The research population consisted of young adult Russian-speaking immigrants to Israel from the former Soviet Union (FSU). The study has shown that variations in the parents’ choice of bilingual vs. monolingual education can be explained in part by three factors related to FLP and its background: the number of children in the family, the parents’ identification with Russian culture, and the children’s well-being as a motivating factor in the choice of educational setting. The significant link between the parents’ preference for bilingual education and the relatively small number of children in the family has been explained by two complementary factors. First, it is possible that couples with fewer children have more opportunities for offering early bilingual education to their children than have couples with more children because of diminished time pressure and fewer competing demands. Second, it is possible that in a single-child family language management is free of the sibling effect that promotes L2 shift, which may have a covert effect on the parents’decision on educational choice. Findings also suggest that the parents’ choice of bilingual education has been motivated not only by pure reasons of L1 maintenance but also by so-called ‘non-language’ related factors such as the quality of education, student-teacher relations, and the quality of the educational facilities. In establishing outside control of the socio-linguistic environment, parents can rely on ethnic religious institutions to provide second generation children

Family language policy

183

with a context for enhancing both their home language and their identity. Park and Sarkar (2007), who explored Korean immigrant parents’ attitudes toward heritage language maintenance in Canada, found that parents view the Korean church, which promotes Korean heritage schools and worship services, as a key instrument for “creating social, linguistic, and cultural centers” (Park and Sarkar 2007: 229) for immigrants and the second generation. 5.2.

Internal control for FLP

When parents enforce their authority, especially when the children are very young, they may insist on the children complying with the language policies they favor. The pro-activist mother in Kopeliovich’s (forthcoming) study reported on her vigorous interference with her children’s language choice, coercing them to use Russian (L1), upbraiding them for code-switching to L2, and correcting L1 mistakes during the first years in the host country. The mother acknowledged, however, these strategies were “futile and exhausting” (Kopeliovich forthcoming: 17), and resulted in confrontations with the children. At the same time, according to the children’s reports, the father’s non-deliberate tactics of language management, by establishing family cultural traditions and rituals strongly associated with L1, were rather fruitful. The father succeeded in influencing the children to use Russian by choosing attractive texts from Russian literature, becoming “an infinite source of bilingual humor based on Hebrew-Russian word puns, rhymes, deliberate misinterpretation of culturally unique idioms . . . ” (Kopeliovich forthcoming: 14). Kopeliovich concluded that the father’s approach to L1 retention was based on stressing the bilingual phenomenon as asset and not as a burden by intermingling the two languages in joyful play.

6. Challenges of FLP Some of the studies presented above suggest that the process of raising children bilingually requires intellectual and emotional investment and appears to be frustrating and burdensome. Okita (2002), Pease-Alvarez (2003) and Caldas and Carol-Caldas (2002) have examined the efforts related to heritage language transmission with reference to the notion of ‘invisible work’ or emotionally demanding work. As noted above, Okita (2002) analyzed the invisible work of minority language transmission in intermarried Japanese-British families. Her approach to Japanese language transmission within these families is greatly inspired by the views of DeVault (1987) and Delphy and Leonard (1992) on

184

Mila Schwartz

the nature of housework as invisible and emotionally demanding work. According to De Vault (1987), family work includes complicated invisible work such as planning and management (monitoring and controlling the children, coordinating schedules, and organizing events). This work is invisible “because it is largely mental”, distributed over time, and becoming visible only when it is not completed. Okita’s (2002) concept of invisible work includes the term “pro-activist mothers” who are highly motivated to transfer Japanese to their children. These full-time homemakers reported having difficulty in coping with conflicting language demands, especially when they felt personal responsibility for their children’s perceived limitation in English because of the mothers’ Japanese language project. Feeling that the school’s acceptance of bilingualism was contingent upon “doing very well in English”, part of them gradually abandoned their use of Japanese. At the same time, they experienced a conflict when it later became apparent that their children did not learn Japanese. Another outcome of active Japanese nurturing has been discouraging their English-speaking partners from active participation in childrearing. Okita stressed the emotional demands involved in raising bilingual children in intermarried families. The demands were manifest in at least four areas: dealing with internal conflicts such as the ones described above, balancing the various needs of the family, the need for continuous monitoring, and feeling responsible for the children mastering both languages. The pro-activist mothers studied by Okita’s were able to devote themselves full-time for child rearing. By contrast, in the families of Mexican descent investigated by Pease-Alvarez (2003), the immigrant mothers who continue to be the active language-socialization agents available to their children had to work outside the home, resulting in weaker Spanish (L1) retention at home. For example, Ms. Suarez described how the need to work outside the home hampered her efforts to maintain the heritage language: I think a lot of parents are working and I don’t think they have the time to get their kids . . . . It’s a lot of work. And I have to say that first hand that I wish I could sit and spend a couple of hours a day because I’m sure I could teach them as well as the school could and you know that’s so expensive. We don’t have the time. You know we are living at such a fast paced life. Everything is so expensive, two working parents, you are constantly going, so you basically just let it go, and they start to lose it. (Pease-Alvarez 2003: 18)

In addressing the emotional demands of raising children bilingually, Caldas and Caron-Caldas (2002) noted their astonishment and shock when despite their efforts to maintain the French language in the home (the mother’s L1) in a context of English dominance, the children’s command of French deteriorated

Family language policy

185

substantially when they entered adolescence, to the point of causing irritation to the parents. The data suggest the presence of a gap between the parents’ role as language teachers (see D¨opke 1988, 1992) who are expected to insist on minority language use and employ language teaching techniques possibly acquired through professional counseling, and the reality within authentic families.

7. FLP research methodologies There is great diversity of methodological tools used to investigate FLP, which may constrain the ability to compare the data and generalize the findings. At the same time, the variety of tools reflects the complexity of FLP research, which addresses a wide range of socio-linguistic contexts and demands an interdisciplinary approach. Perhaps the most frequent method used in FLP research is the qualitative approach manifest in in-depth, semi-structured interviews. The importance of interviews cannot be over-emphasized because they provide a sensitive method for understanding the processes taking place within the family. At the same time, there is a growing tendency for methodological triangulation in FLP research, with multiple methods required to explore the largely invisible processes and influences that arise in the course of intergenerational language transmission within families (Kopeliovich forthcoming; Okita 2002; Tannenbaum 2003). For example, Okita (2002) proposed a two-stage approach for data collection, first investigating the distinctive features of the target community (JapaneseBritish intermarried families in the UK) in a general sense through an exploratory survey, then, providing in-depth, qualitative insight into the family language policy and childrearing, using the life story method in separate, semistructured interviews of mothers and fathers. Combining quantitative and qualitative approaches, FLP researchers can identify the common characteristics of families belonging to a distinct community or sub-group. The rich source of descriptive data obtained from the survey forms the background for the deeper understanding of unique processes involved in FLP within one or several families of the target group. Discussing the mixing of methods, Brannen emphasized the importance of these strategies, generally understood as “more than one method of investigation and hence more than one type of data” (Brannen 1992: 11). Mackey and Gass defined methodological triangulation as a methodology that “entails the use of multiple, independent methods of obtaining data in a single investigation in order

186

Mila Schwartz

to arrive at the same research findings” (Mackey and Gass 2005: 181). Johnson saw the value of triangulation in that it “reduces observer or interviewer bias and enhances the validity and reliability of the information” (Johnson 1992: 146). In sum, the growing practice of using qualitative and quantitative data in FLP research demonstrates that the two research approaches should not be viewed as opposing poles in a dichotomy but rather as complementary tools for investigating complex phenomena. Another important methodological issue in FLP research is the incorporation of the children’s perspectives in the parental data. Until now, relatively few studies collected data on FLP from both parents and children. At the same time, using the children’s reports on FLP, observing their language socialization, and measuring their L1/L2 mastery can strengthen considerably the validity of data collected from parents. When considering the strengths and limitations of the data derived from children’s reports, we should be take into account that (a) even if we cannot assume that the children’s views are fully reflected in what they say, they are not likely to try to please the researcher by providing expected answers during the interviews (i.e., the halo effect), and (b) the children’s language ideology seems to be affected considerably both by parental language ideology and by the actual implementation of the language policy at home.

8. Further research directions The present review focused on the social, psychological, educational, and linguistic facets of intergenerational language transmission by applying Spolsky’s model of LP in a family context. The empirical data reflect the complexity and sometimes inconsistency of the parents’ invisible work invested in the children’s socialization in the home language. Although the notion of FLP is relatively new, the studies described above contributed significantly to our understanding of the FLP phenomenon and established a productive basis for the future research. At the same time, questions still outnumber the answers. First, there is a need for focused research on FLP to address the links between its components, taking into accounts its background and consequences by applying methodological triangulation. It is still unclear to what extent the parents’ ethnic roots are related to their FLP, what are the unique FLP features of various ethno-linguistic communities, and what are those applicable to other ethno-linguistic communities. Additional questions concern the longitudinal consequences of FLP and the manner in which it changes over time and possible directions in modifying the FLP as the children grow older. It is reasonable to assume that the parents’ language strategies with pre-school children differ from those with children

Family language policy

187

in elementary school and with adolescents. Given that the role of peers in the language socialization of bilingual children increases, as the children grow older, it would be important to observe, for example, parents’ tendency to seek better suited environments for L1 retention (Caldas and Caron-Caldas 2002; Spolsky 2007). Finally, as the main criterion of successful FLP is the children’s progress in L1, adding objective measures of linguistic behavior (e.g., conducting lexical and morpho-syntactic knowledge tests) could verify and reinforce the FLP data collected based on the parents’ and children’s perceptions. There is a need for insight into bilingual homes to examine language strategies within authentic environments, using ethnographic observations. These methodological approaches can contribute to the validity of research. Finally, future research must also take into account the interdisciplinary nature of FLP and bring together such disciplines as linguistics, sociology, and education to promote research in this area. Oranim College of Education and University of Haifa

Note *

This paper was supported by the grants awarded to Dr. Mila Schwartz by Kreitman Postdoctoral fellowship (Ben-Gurion University, Beer-Sheva, Israel) and by the Edmond J. SAFRA Foundation. The funders had no role in decision to publish, or preparation of the manuscript. I am grateful to Dr. Victor Moin for his useful comments.

References Allard, R. & R. Landry. 1992. Ethnolinguistic vitality beliefs and language maintenance and loss. In W. Fase, K. Jaspaert & S. Kroon (eds.), Maintenance and loss of minority languages, 173–195. Amsterdam: John Benjamins. Baker, C. 2001. Foundation of bilingual education and bilingualism. Clevedon: Multilingual Matters. Barkhuizen, G.. 2006. Immigrant parents’ perceptions of their children’s language practices: Afrikaans speakers living in New Zealand. Language Awareness 15. 63–77. Ben-Rafael, E., E. Olshtain & I. Geijst. 1997. Identity and language: The social insertion of Soviet Jews in Israel. In N. Levin-Epstein, Y. Ro’i & P. Ritterband (eds.), Russian Jews on three continents: Migration and Resettlement, 364–388. London: Frank Cass. Brannen, J. (ed.). 1992. Mixing methods: Qualitative and quantitative research. Aldershot: Avebury.

188

Mila Schwartz

Caldas, S. & S. Caron-Caldas. 2000. The influence of family, school, and community on bilingual preference: Results from a Louisiana/Quebec case study. Journal of Applied Psycholinguistics 21(3). 365–381. Caldas, S. & S. Caron-Caldas. 2002. A sociolinguistic analysis of the language preferences of adolescent bilinguals: Shifting allegiances and developing identities.Applied Linguistics 23(4). 490–514. Clyne, M. 1982. Multilingual Australia. Melbourne: River Seine. De Houwer, A. 1999. Environmental factors in early bilingual development: The role of parental beliefs and attitudes. In G. Extra & L. Verhoeven (eds.), Bilingualism and migration, 75–96. Berlin & New York: Mouton de Gruyter. Delphy, C. & D. Leonard. 1992. Familiar exploitation: A new analysis of marriage in contemporary Western societies. Cambridge: Polity Press. DeVault, M.. 1987. Doing housework: Feeding and family life. In N. Gerstel & H. Gross (eds.), Families and work. Philadelphia, 178–191. Philadelphia: Temple University Press. Donitsa-Schmidt, S. 1999. Language maintenance or shift – Determinants of language choice among Soviet immigrants in Israel. Toronto: University ofToronto unpublished doctoral thesis. D¨opke, S. 1988. The role of parental teaching techniques in bilingual German-English families. International Journal of the Sociology of Language 72. 101–112. D¨opke, S.. 1992. One parent one language: An interactional approach. Amsterdam: John Benjamins. Doucet, J. 1991. First generation of Serbo-Croatian speakers in Queensland: Language maintenance and language shift. In S. Romaine (ed.), Language in Australia, 270– 284. New York: Cambridge University Press. Fishman, J. A. 1991. Reversing language shift: Theoretical and empirical foundations of assistance to threatened languages. Clevedon: Multilingual Matters. Fishman, J. A. 2000. Reversing language shift: RSL theory and practice revisited. In G. Kindell & M. P. Lewis (eds.), Assessing ethnolinguistic vitality: Theory and practice; Selected papers from the Third International Language Assessment conference, 1–25. Publications in Sociolinguistics 3. Dallas: SIL International. Fishman, J. A. 2001. From theory to practice (and vice versa): Review, reconsideration and reiteration. In J. A. Fishman (ed.), Can threatened languages be saved?, 451–483. Berlin & New York: Mouton de Gruyter. Garrett, P. B. & P. Baquedano-Lopez. 2002. Language socialisation: Reproduction and continuity, transformation and change. Annual Review of Anthropology 31(1). 339– 361. Giles, H., R. Bourhis & D. Taylor. 1977. Towards a theory of language in ethnic group relations. In H. Giles (ed.), Language, ethnicity and intergroup relations, 307–348. London: Academic Press. Goodz, N. 1994. Interactions between parents and children in bilingual families. In F. Genesee (ed.), Educating second language children, 61–81. Cambridge: Cambridge University Press.

Family language policy

189

Gregory, E. 2004. “Invisible” teachers of literacy: Collusion between siblings and teachers in creating classroom cultures. Literacy 38(2) 97–105. ´ Andrea. 1992. Some properties of bilingual maintenance and loss in Hakuta, K. & D. Mexican background high school students. Applied Linguistics 13. 72–99. Harres, R.. 1989. The search for minority teachers. Los Angeles Times September 27. A1, A18. Harris, J. R. 1995. Where is the child’s environment? A group socialization theory of development. Psychological Review 103(3). 458–489. Hua, Z.. 2008. Duelling languages, duelling values: Codeswitching in bilingual intergenerational conflict talk in diasporic families. Journal of Pragmatics 40. 1799–1816. Johnson, R. C.. 1992. Offspring of cross-race and cross-ethnic marriages in Hawaii. In M. P. P. Root (ed.), Racially mixed people in America, 239–249. Thousand Oaks, CA: Sage. Johnson, J. & C. Martin. 1985. Parents’beliefs and home language environments: Effects on cognitive development. In I. Sigel (ed.), Parental belief systems. The psychological consequences for children, 25–50. Hillsdale, NJ: Lawrence Erlbaum Associates. King, K. 2000. Language ideologies and heritage language education. The International Journal of Bilingual Education and Bilingualism 3(3). 167–184. King, K. & L. Fogle. 2006. Bilingual parenting as good parenting: Parents’ perspectives on family language policy for additive bilingualism. The International Journal of Bilingual Education and Bilingualism 9(6). 695–712. King, K., L. Fogle & A. Logan-Terry. 2008. Family language policy. Language and Linguistics Compass 2(5). 907–922. Kloss, H. 1966. German-American language maintenance efforts. In J. A. Fishman (ed.), Language loyalty in the United States: The maintenance and perpetuation of nonEnglish mother tongues by American ethnic and religious groups, 206–252. The Hague: Mouton. Kopeliovich, S. Forthcoming. Family language policy: From a case study of a RussianHebrew bilingual family towards a theoretical framework. Diaspora, Indigenous and Minority Education 4(2). Kyratzis, A. 2004. Talk and interaction among children and the co-construction of peer groups and peer culture. Annual Review of Anthropology 33(1). 625–649. Lambert, W. E. 1975. Culture and language as factors in learning and education. In A. Worlfgang (ed.), Education of immigrant students. Toronto: Ontario Institute for Studies in Education. Lambert, W. E. & D. M. Taylor. 1996. Language in the lives of ethnic minorities: Cuban American families in Miami. Applied Linguistics 17(4). 477–500. La Piedra, M. & H. Romo. 2003. Collaborative literacy in a Mexican immigrant household: The role of sibling mediators in the socialization of pre-school learners. In R. Bayley & S. Schecter (eds.), Language socialization in bilingual and multilingual societies, 44–61. Clevendon: Multilingual Matters. Mackey, A. & S. M. Gass. 2005. Second language research: Methodology and Design. Mahwah, NJ: Lawrence Erlbaum Associates.

190

Mila Schwartz

Myers-Scotton, C. 2002. Contact linguistics: Bilingual encounters and grammatical outcomes. Oxford: Oxford University Press Okita, T. 2002. Invisible work: Bilingualism, language choice and childrearing in intermarried families. Amsterdam: John Benjamins. Park, S. M. & M. Sarkar. 2007. Parents’ attitudes toward heritage language maintenance for their children and their efforts to help their children maintain the heritage language: A case study of Korean-Canadian immigrants. Language, Culture and Curriculum 20(3). 223–235. Pease-Alvarez, L. 2003. Transforming perspectives on bilingual language socialization. In R. Bayley & S. Schecter (eds.), Language socialization in bilingual and multilingual societies, 9–24. Clevendon: Multilingual Matters. Schwartz, M. 2008. Exploring the relationship between family language policy and heritage language knowledge among second generation Russian-Jewish immigrants in Israel. Journal of Multilingual and Multicultural Development 29(5). 400–418. Schwartz, M., V. Moin, M. Leikin & A. Breitkopf. Forthcoming. Immigrants’ family language policy toward children’s preschool bilingual education: parents’perspective. International Journal of Multilingualism. Spolsky, B. 2004. Language policy. Cambridge: Cambridge University Press. Spolsky, B. 2007. Family language management: Some preliminaries. In A. Stavans & I. Kupferberg (eds.), Studies in language and language education: Essays in honor of Elite Olshtain, 429–449. Jerusalem: The Magnes press, Hebrew University. Spolsky, B. & E. Shohamy. 1999. The languages of Israel: Policy, ideology and practice. Clevedon: Multilingual Matters. Tannenbaum, M.. 2003. The multifaceted aspects of language maintenance: A new measure for its assessment in immigrant families. International Journal of Bilingual Education and Bilingualism 6(5). 374–388. Tannenbaum, M. 2005. Viewing family relations through a linguistic lens: Symbolic aspects of language maintenance in immigrant families. The Journal of Family Communication 5(3). 229–252. Tannenbaum, M. & P. Howie. 2002. The association between language maintenance and family relations: Chinese immigrant children in Australia. Journal of Multilingual and Multicultural Development 23(5). 408–424. Williams, G. 1987. Bilingualism, class dialect and social reproduction. International Journal of the Sociology of Language 66. 85–98. Wong Fillmore, L. 1991. When learning a second language means losing the first. Early Childhood Research Quarterly 6. 323–346. Wong Fillmore, L. 2000. Loss of family languages: Should educators be concerned? Theory into Practice 39(4). 203–210. Zhu, H. & W. Li. 2005. Bi- and multi-lingual acquisition. In M. Ball (ed.), Clinical Sociolinguistics, 165–179. Oxford: Blackwell. Dr. Mila Schwartz <[email protected]> completed her graduate her work at the University of Haifa in Learning Disabilities and Literacy Development among Bilingual and

Family language policy

191

Trilingual Children. She then continued her post-doctoral specialization at Ben-Gurion University and at the University of Haifa in the Edmond J. Safra Brain Research Center for the Study of Learning Disabilities. Today, her research interests include studying of bi-literacy; early bilingual education; language and cognitive development of early sequential bilinguals; family language policy; and immigrant teachers’ pedagogical development.

Is ergativity always a marker of agency? Toraja and Samoan grammar of action and the contribution of emancipatory pragmatics to social theory* AURORA DONZELLI

Abstract This article argues that looking at the grammatical encoding of agency in languages other than English helps us understand vernacular theories of action and enhances our critical awareness of the influence played by Western linguistics on our interpretations of cultural realities and on the shaping of social theoretical categories. Duranti’s (1990, 1994) ethnopragmatic analysis on how agency is grammaticized in Samoan political oratory showed that the usage of ergative constructions was strongly associated with the attribution of agency and responsibility to the referent of the noun phrase to which the ergative preposition /e/ was prefaced. Drawing on a corpus of political meetings video-taped between 2002 and 2003 in upland Sulawesi (Indonesia), this article describes the linguistic encoding of agency in Toraja, a language that like Samoan belongs to the Austronesian language family and presents ergative features. Unlike what was shown by Duranti for Samoan, my data reveal that ergative constructions in Toraja mitigate instead of foregrounding the referent’s agentivity and responsibility. While describing how agency is encoded in alternative grammatical patterns, the analysis shows how an understanding of agency informed by semantic notions of transitivity is not completely adequate for the Toraja ethnolinguistic context and invites a reflection on the relation between linguistic and anthropological theory. Keywords: agency; emancipatory pragmatics; ergativity; political speech; Austronesian.

194

Aurora Donzelli

1. Introduction During the last three decades, the notion of agency has become ubiquitous within empirical and theoretical discussions in the humanities and the social sciences (see Ahearn 2001a; Donzelli and Fasulo 2007; Duranti 2004; Emirbayer and Mische 1998; Ortner 2006 for useful reviews). Much of this literature focuses primarily on the role of human action in the reproduction of social structure and on the debate between social determinism and individual voluntarism, effectively summarized by Archer (2000: 4–5) as the clash between “Downward Conflation” (“we are nothing beyond what society makes us”) and “Upward Conflation” (where “the powers of the people are held to orchestrate those of the parts”). However, at a broader and more general level, the notion of agency refers to the human faculty to act and to the dialectics of control and affectedness, action and recognition, intentionality and responsibility, potentiality and actuality that underlies human action. Understood in this broader sense, the notion of agency captures the idea that humans cannot be studied in their own terms independently from the acts they perform and points to the fact that humans exist (and need to be understood) only in reference to other situated acting subjects. Indeed, it is hardly questionable that humans, with different degrees of awareness and willingness, are constantly engaged in performing actions, evaluating the potential results or regretting the actual outcomes of their own or other people’s deeds, assuming or disclaiming responsibility for the acts they actually perform or imagine to perform, debating whether to act or to refrain from action or whether they should act in a certain way or another. And even if it was the case that they could prevent themselves from been implicated in all this, they would still be in some way or another affected by other people’s acts. Different ascetic schools, in different historical periods, in different parts of the world have devised technologies for escaping this never ending dynamic of actions, actuations, and projections, but the fact remains that being in the world mainly means to be caught up in multiple (and probably endless) temporal and spatial chains of action. Several contemporary philosophical traditions have discussed human agency. Action theorists (Davidson 2001; Sewell 1992) highlighted the central role played by intentions in differentiating actions from events and “agency from routine practice” (Ortner 2006: 136), while German and French phenomenologists thematized the primacy of acts over objects and reformulated the question of being as a question about the experiential, existential, and perceptual accessibility of things. Martin Heidegger (1978) argued that humans as ontological entities (Dasein) only exist as embedded in relations with other entities (with

Is ergativity always a marker of agency?

195

other humans and with other objects). Humans’ involvement with the world is essentially manifested in what he termed care (Sorge) or concern (Besorgen), which appears under several specific and contingent forms such as “having to do with something, producing something, attending to something and looking after it, making use of something, giving something up and letting it go, undertaking, accomplishing, evincing, interrogating, considering, discussing, determining . . . ” (Heidegger 1978: 57). Heidegger’s point is that care and concern constitute fundamental ontological characteristics (existentialia) of the Being of human beings (Dasein). In this view, humans’ Being-in-the-world (In-der-welt-sein) is understood as a being-towards-the-world. In other words, human life is mostly a matter of intentional action.1 The idea that humans are not monadic entities, existing by themselves prior to their being projected towards the world through their actions, also shaped Merleau-Ponty’s reflections on the dialectic of intersubjective recogniton (1964b, 1964c) and the metaphysics of the flesh that he elaborated in the Visible and the Invisible (1964a). Indeed, agency constitutes a universal, intrinsic, and inevitable dimension of humanness (Archer 2000; Duranti 2004: 468; Ortner 2006: 136; Sewell 1992: 20). However, despite its cultural pervasiveness, social ubiquity, and existential inevitability, the notion of agency, apart from a few exceptions, has only rarely been thematized as a distinct object of ethnographic investigation,2 hence remaining mostly confined to a theoretical level. The emphasis on theory at the expense of ethnographic investigation often resulted in conflating agency with other notions such as resistance, autonomy, choice, intentionality, and creativity and contributed to making the term rather ambiguous (Emirbayer and Mische 1998), thus weakening its heuristic potential for the exploration of native practical philosophies. This paper wants to be an invitation to take a closer look at the intersection between grammar, pragmatics, and social theory.As Duranti (2004: 467) pointed out, “all languages have grammatical structures that seem designated to represent agency”. Hence, studying how speakers of different languages use these grammatical structures to express agency can shed light on the many diverse “folk theories of action” (Jackendoff 2007) existing in the world. As I will try to show in the next pages, this cross-linguistic analysis of the “encoding” and “performance” (Duranti 2004) of agency can also help “provincialize” (Chakrabarty 2000) Western linguistic theory and denaturalize “views of language derived from Euro-American languages” (Hanks et al. 2009), thus contributing to the critical agenda of the newly emerging field of emancipatory pragmatics. Moving from the idea that detailed ethnographic description of the pragmatics of political interaction will enhance our understanding of locally specific

196

Aurora Donzelli

cognitive and practical structures of agency, this article aims at describing the micro-processes through which social actors in a Toraja village of upland Sulawesi (in Indonesia) used specific linguistic resources to represent their faculty to act and to assume on themselves or assign to (present or absent) others intentions and responsibility for the actions or the state of affairs being discussed. Toraja,3 like most languages spoken in south Sulawesi, is characterized by an ergative-absolutive distribution of clitic particles, which index person, number and grammatical role on the verb, a pattern common to “Mayan languages of Mexico and Central America” (Comrie 1978: 339). This particular (ergative) morphological configuration of the system of verb-agreement entails the differential marking of the subject of the transitive clause, as opposed to the alignment of patients and intransitive actors, which receive the same clitic marker on the verb. A key question is therefore whether this specific mode of marking grammatical relations plays a role either in the cognitive (Goldin-Meadow 2003) or pragmatic (Duranti 1990, 1994) pattering of semantic roles. In his influential research on Samoan, Duranti (1990, 1994) showed that the usage of ergative markers by Samoan orators was strongly associated with the attribution of agency and responsibility. The present analysis shows that ergativity in Toraja oratory plays a very different role. Both Toraja and Samoan belong the Austronesian language family. However, the former is included in the Western Malayo-Polynesian branch (which is in turn a sub-ramification of the Malayo-Polynesian sub-family), while the latter is an Oceanic language within the Eastern Malayo-Polynesian branch (Blust 1977). Both languages present ergative features. However, in Toraja ergativity is expressed through the ergative-absolutive alignment of two different sets of pronominals that cliticize on the predicate and that have become “integrated” into a system of voice alternations, while in Samoan ergative-absolutive distinctions are expressed through a differential marking of the nouns (i.e. ergative preposition /e/ and absolutive zero marking). Although these genetic relations and typological commonalities are worth mentioning, the point here is not to draw conclusions about the typological or genetic position of the two languages, but to examine and reflect on the different ethnopragmatic value of ergative constructions in Samoan and Toraja political discourse. As we will see, unlike Duranti’s findings, ergative constructions in Toraja highlight the patient’s affectedness and at the same time downplay the causative role of the agent. Quite surprisingly, both the metalinguistic understanding and the semantico-pragmatic value of ergativity in Toraja oratory points toward a mitigation rather than a foregrounding of the role of agent. Indeed, the data drawn from political interactions reveal that ergative sentences constitute the unmarked and natural grammatical choice, while speakers rely on alternative

Is ergativity always a marker of agency?

197

resources to foreground the agentivity and responsibility of the referent of the noun phrase (and of the speaker). This reappraisal of Duranti’s work will not only provide pragmaticallygrounded evidence against the cross-linguistic equivalence of agency and ergativity, but will also question the association between the notion of agency and that of (semantic) transitivity. As will be shown, the grammatical construction, which encodes the highest degree of agency in Toraja does not correspond to an increase of the valency (or semantic transitivity) of the sentence. The mismatch between agency and transitivity in the pragmatics of Toraja political speech triggers a broader reflection on how semantic and linguistic notions tacitly shape the interpretations of social realities within both linguistic and socio-cultural anthropology. Linguists Hopper and Thompson (1980) proposed to understand transitivity on a semantic and pragmatic ground (rather than as a merely syntactic phenomenon) through a set of loosely co-occurring and co-varying parameters. In their view, higher degrees of transitivity correlate with an enhancement of the activeness and volitionality of the agent, as well as with higher levels of individuation and affectedness of the object. According to their famous analysis, agency is thus one of the ten basic components of the notion of transitivity. As it seems to me, social theorists seem to unwittingly rely on a notion of agency based on semantic ideas of volitionality, activeness, and effectiveness. In other words, the notion of agency used in the analysis of socio-cultural processes is patterned on an implicit idea of transitivity intended as a transfer of activity from an agent to a patient. However, the study of how grammars of non EuroAmerican languages deal with agency and transitivity can help question the naturalization of this equivalence. Problematizing the conflation between agency and transitivity can in turn contribute to elaborate a different philosophical understanding of the very notion of human agency. As we will see below, one of the most interesting properties of Toraja grammar is that it allows two strategies for expressing a two arguments (i.e. transitive) clause, that is, to describe the interrelation between an acting subject (Agent) and its object (Patient): The ergative construction (also called patient voice or PV) is endowed with a higher degree of transitivity and is used to foreground the ontological and pragmatic saliency of the object and its affectedness. The actor voice (or AV) instead is lower in transitivity and it enhances the assignment of agency and responsibility to the acting subject, back-grounding the saliency of the object and conveying the sense of an agent that is affected by its own actions. As I will try to demonstrate, the way in which the representation of the relation between agent and patient through the performance of an action is encoded in Toraja along a centripetal/centrifugal continuum through the paradigmatic

198

Aurora Donzelli

alternative between patient and actor voice constructions can tell something to social theorists and philosophers. Indeed, Toraja grammar of action highlights how every form of action does not only involve a centrifugal transfer of activity from an agent to a patient (as in PV), but also entails a centripetal feedback effect for the agent who aside from affecting the object/patient is also affected by the consequences of its own actions (as in AV). This, I believe, constitutes an important challenge the general understanding of agency and prompts a reflection on the relation between linguistic and anthropological/social theory. The following analysis is based on a corpus of linguistic data collected in a series of political meetings and electoral rallies in which I participated during several periods of ethnographic research aimed at documenting Toraja political language and interaction. A large part of my fieldwork (2002–2004)4 coincided with an important process of decentralization and administrative reform, which had started inToraja at the end of 2001 after three decades of authoritarian regime under President Soeharto’s rule. This particular historical conjecture gave me the chance to collect fifty hours of audio and thirty hours of video recording of electoral rallies (kampanye, ma’pakande), official (rapat) and more ‘traditional’ (kombongan) political meetings, as well as instances of formal and spontaneous interaction that occurred during ceremonies and rituals. Before proceeding with analyzing my Toraja material, let me first provide a basic definition of the notion of ergativity, as well as an overview of Duranti’s (1990, 1994) work on its encoding in Samoan.

2. Ergativity In linguistic typology, the category of ergative is employed to classify a variety of historical-natural languages across the world, which, despite remarkable differences in their morpho-syntactic or phonological structures, share a crucial similarity in their case-marking and/or verb agreement systems.5 Dixon (1979: 60) provided a general and concise definition of this phenomenon: “A language is said to show ergative characteristics if intransitive subject is treated in the same manner as transitive object, and differently from transitive subject”.6 Linguistic typologists are used to contrasting the ergative (or ergative-absolutive) pattern of case marking with the nominative-accusative one of which Latin constitutes an emblematic case. An example will make this clear:7 (1)

Puer venit Boy.NOM came ‘The boy came.’

Is ergativity always a marker of agency?

(2)

199

Puer puellam amat Boy.NOM girl.ACC loves ‘The boy loves the girl.’

In the Latin sentences above (adapted from Comrie [1978: 331]), we can see how the subjects of the intransitive (1) and transitive (2) sentences have the same morphological marker, while the direct object in (2) is marked differently. This pattern of case-marking, which, as we mentioned, is called nominativeaccusative is common to most European languages, even though they may not be as flexive as Latin in their nominal morphology. On the contrary, ergative-absolutive languages display a remarkably different morphology with respect to case-marking. This point becomes clear if we now contrast the two sentences (1) and (2) above with the Tongan examples below (adapted from Churchward [1953: 67–68]), which illustrate the differential marking of the grammatical role of subject in transitive (3) and intransitive (4) sentences typical of ergative morphologies: (3)

Na’e tamate’i ‘e Tevita ‘a Kolaiate PFV kill ERG David ABS Goliath ‘David killed Goliath.’

(4)

Na’e lea ‘a Tolu PFV speak ABS Tolu ‘Tolu spoke.’

As we can see, Tongan morphology is clearly ergative: the absolutive particle /‘a/ marks both the intransitive subject (4) and the direct object (3), while the transitive subject (3) is associated with the preposition /’e/. Surprisingly, the considerable attention given to the issue of ergativity within linguistics since the beginning of the 1970s, has not been paralleled by an equal interest on the part of linguistic anthropologists. The work undertaken in the late 1970s by Alessandro Duranti (1990, 1994), Elinor Ochs (1982, 1988), and Martha Platt in Western Samoa (Upolu) is without any doubt the most comprehensive and thorough ethnographic research on the pragmatic and social value of ergative constructions among a community of Samoan speakers.8 This research team was broadly concerned with exploring “the sociologiocal scope of ergative morpho-syntax within a language” (Ochs 1982: 651). While Ochs and Platt concentrated on documenting spontaneous interactions between adults and children, Duranti focused on language use among adults, paying particular attention to the formal and political speech used by chiefs and orators in the fono (politico-judiciary meeting). One of the most important findings of this

200

Aurora Donzelli

collective enterprise consisted of the discovery that the use of ergative markers among Samoan speakers was distributed according to differences in age, social status, and gender. These findings had important implications. Ochs’ research (1982, 1988) on child language showed for example that the late acquisition of ergative casemarking by Samoan children was mostly due to the fact that ergative casemarking rarely occurs in domestic language interaction among intimates, which drastically reduced the occasions in which children can acquire this structure. The analysis of the data clearly showed that expression of ergative casemarking among Samoan speakers was an index of social distance, formality, gender, social status, and thus was “sociolinguistically variable” (Ochs 1982: 646). But it was Duranti’s work on Samoan oratory to argue that this sociolinguistic distribution of ergative markers was deeply intertwined with the pragmatic assignment of agency and responsibility within the speech event. Duranti’s original contribution consisted of showing the existence of a clear relation between the morphological expression of ergative case-marking and the semantic and pragmatic encoding of agency in Samoan political speech (and praxis). His ethnopragmatic analysis of what kind of speech acts were accomplished through the use of ergative markers led him to conclude that besides being socially salient, ergative markers were also endowed with a particular semantic and pragmatic value. In this light, the scarcity of ergative constructions in Samoan adult speech appeared (at least partially) as a result of the semantico-pragmatic value of ergative markers. But let us take a closer look at the empirical and interpretative basis for this claim.

3. Ergativity as a marker of agency in Samoan Duranti’s analysis of Samoan speechmaking moved from an interest in studying the unfolding of the politics of linguistic representation within situated interactions. This enterprise focused on a notion of semantic roles that he derived from functional grammar (Fillmore 1968) and aimed at exploring how speakers use different morpho-syntactic and grammatical resources to negotiate different representations of extra-linguistic reality. In other words, he believed that the grammatical choices made by participants to spell out the reasons for the organization of a politico-judiciary meeting (fono), or the way in which – through linguistic means – they define the agenda of the encounter convey different configurations of semantic roles and hence different modes for the attribution of agency and responsibility to key participants.

Is ergativity always a marker of agency?

201

As already mentioned, Samoan has – like Tongan – an ergative-absolutive system of case-marking: the subject of the transitive clause (5) is marked by the ergative particle /e/; while the subject of the intransitive clause (6) and the object of the transitive clause (5) are morphologically identical in not being marked by any preposition (adapted from Duranti 1990: 651): (5)

‘ua fa’atau e le tama le suka TA buy ERG ART boy ART sugar ‘The boy has bought the sugar.’

(6)

‘ua alu le tama i le maket TA go ART boy to ART market ‘The boy has gone to the market.’

This differential marking for subjects of transitive clauses conveys a certain grammatical saliency to the encoding of the Agent role in Samoan, in that, it “offers . . . the possibility of explicitly and unequivocally assigning to a particular referent the role of ‘agent’, that is the willful initiator of an event that has consequences for either an object or an human patient” (Duranti 1990: 651). At the same time, the rarity of ergative case-marking in spontaneous language provides an additional social saliency of this type of construction. As the analysis of transcribed material drawn from political and domestic interaction revealed, ergative markers are rarely deployed and “transitive clauses with ergative agents are not very frequent” (Duranti 1990: 652). The data showed that Samoan speakers tend to encode the Agent role through other morpho-syntactical resources such as oblique objects and genitive modifiers. Quite interestingly, these alternative constructions seemed to be associated with semantic and pragmatic strategies to mitigate and downplay the assignment of agency and responsibility. For example, the use of oblique object case-marking represented the actor “as the initiator of an event . . . as a source of a transitive act rather than as an ergative agent” (Duranti 1990: 655), with the effect that the causative relationship between the human actor expressed through the oblique object and the resulting action described by the predicate was inferred but not explicitly affirmed. Another strategy used “to deemphasize someone’s contribution to a given task or achievement” (Duranti 1990: 656) consisted of the deployment of genitive modifiers, which provided the agent with the same morphological treatment of possessors, thus foregrounding the object (or patient) rather than the human agent (Duranti and Ochs 1990). Duranti (1990) noted that Samoan speakers made ample deployment of grammatical devices, which mitigated and downplayed the agency of the referents,

202

Aurora Donzelli

while transitive sentences with ergative markers mostly appeared in contexts where the actions of a group, an individual, or a deity were made accountable, or when the execution of a task was assessed. Retrospective analysis of transcribed material showed that speech acts such as negative or positive assessments, complaints and accusations displayed a higher concentration of ergative markers (Duranti 1990: 655–656). Hence, drawing on this empirical base, Duranti argued that ergative markers constituted important indexes for the assignment of responsibility and agency to the referent of the noun phrase to which the ergative preposition /e/ was prefaced. Moreover, tracking the distribution of the ergative markers in the speech of different chiefs and orators, he identified a sort of social division of labor among the speakers in the performance of speech acts in which agency and responsibility were assigned. The widest usage of ergative markers was associated not with the highest-ranking individuals but with orators of intermediate status. The higher ergative degree characterizing the orators’ speeches offered an important insight into how diarchy and social hierarchy are reproduced in Samoan society through micro-processes of communicative interaction. Duranti’s analysis displays a harmonious articulation of different linguistic and cultural levels of analysis in which linguistic formal features, their semantico-pragmatic value, and their social distribution perfectly overlap. However, one may wonder whether the perfect coincidence between grammar, semantics, pragmatics, and social reality is contingent on the ethnographic reality described by Duranti, whether it is the effect of the specific theoretical and methodological framework he employed, and whether it could hold in other ethnolinguistic contexts. Reflecting on the relation between agency and ergativity, Bernard Comrie (1978: 335), for example, argued that languages vary in the extent to which ergativity coincides with semantic agentivity. Comrie, however, grounded his claim against the equivalence between ergativity and agency on syntactic and semantic considerations, rather than on a pragmatic analysis of linguistic data drawn form spontaneous interaction. What is the pragmatic value of ergative constructions in Toraja political speech? Is ergativity in Toraja a marker of agency or does it have another pragmatic function?

4. Ergativity in Toraja The Toraja language (a.k.a. Toraja Tae’, South Toraja, or Toraja Sa’dan) is spoken by the people dwelling in the central highlands of the southern province of Sulawesi and belongs to the South Sulawesi subgroup, which constitutes one

Is ergativity always a marker of agency?

203

of the nine main language sub-groups of the island of Sulawesi (Noorydun 1991). Toraja is generally considered a morphologically ergative language in which two sets of pronominal clitics mark the person on the verb according to an ergative-absolutive pattern. (See Table 1.) Table 1. Ergative and absolutive clitics in Toraja

1s 1p INCL 1p EXCL 2s 2p 3 s /p

Set 1 Proclitics Ergative kutakimumi-(HON) miki-(HON) na-

Set 2 Enclitics Absolutive -na’ -ki’ -kan -kanni -ko -mi(HON) -kommi/komu -i (or zero)

As is apparent from examples below, the proclitic set plays the A function (i.e. Subject of a transitive verb) and marks the ergative case – as in (9) and (10) – while the enclitic set combines the S (i.e. Subject of the intransitive) – examples (7) and (8) – and P (i.e. Object of the transitive verb) functions – as in (9) and (10) – and marks the absolutive case.9 Intransitive sentences with pronominal clitics: (7)

lako pasa’ La=male=na’ market FUT=Go=1.s.ABS to ‘I am going/I go to the market.’

(8)

Ma’-jama=ko dio Rantepao INT-Work–2.s.ABS in Rantepao ‘You work in Rantepao.’

Transitive sentences with pronominal clitics: (9)

Na=kambei=ko 3.ERG=Beat=2.s.ABS ‘He beats you.’/ ‘They beat you’.

204 (10)

Aurora Donzelli

Na=sua=na’ 3.ERG=Order=1.s.ABS ‘He orders me.’/ ‘They order me’.

However, it should be noted that the label of ‘pronouns’ for these clitic sets can be partially misleading. Their status is between that of noun substitutes – as in the examples (7) to (10) above – and obligatory parts of the verb (see Valkama 1995: 49). In the latter case they occur with full NPs (noun phrases) and function as person markers that cross-reference an argument co-occurring in the same clause. In sentences with full nominal arguments – such as (11–13) – the proclitic set cross-references the NP that is inA function, while the absolutive set cross-references a NP that is in S and (optionally) P function. (11)

Puang Batu na=plie=i tau Puang Batu 3.ERG=choose=3.ABS people ‘The people chose/elected Puang Batu.’

(12)

Puang Matua tu masiang kumua melo Na=tiro=i that good 3.ERG=See=3.ABS Puang Matua REL day ‘God saw the light, that it was good.’ (Sura’ Madatu 1995: 1 [Kadadian Genesis])

(13)

Sule-m=i tu indo’=na Tulang Didi lako Return-PFV=3.ABS REL mother=DEF Tulang Didi to banua house ‘Tulang Didi’s mother returned home.’

As apparent from this outline, the Toraja ergative pattern is different from Samoan. Unlike Samoan where ergative-absolutive distinctions are expressed through a differential marking of the nouns, in Toraja free nominals have no markers and ergativity is expressed through verbal morphology.10 Another very important difference between the two languages is that Toraja ergative verbal morphology intersects with a system of voice alternations, which lacks in Samoan. Indeed, the examples of transitive clauses with pronominal clitics presented above illustrate only one (the patient voice or PV) of the two strategies available in Toraja for expressing a syntactically transitive clause (i.e. a two argument

Is ergativity always a marker of agency?

205

clause). Aside from the already discussed PV, a two argument clause can be expressed in Toraja by an alternative structure: the actor voice (AV). These two main voices types exhibit very different morpho-syntactic structures. We already saw how the patient voice (PV) is characterized by the presence of proclitic pronoun hosted by the verb and indicating person and number of the actor and sometimes11 an enclitic cross-referencing the patient. The actor voice (AV) is marked by the lack of ergative proclitics, the presence of the actor voice prefix uN - (where N is homorganic to the following first consonant of the verb root), and an optional enclitic cross-referencing P. The reminder of this article will offer examples of the pragmatic contexts where PV and AV are deployed in interaction, as well as of the metalinguistic treatment of the two clause types I observed during elicitation sessions with my Toraja consultants. In so doing my aim will be to analyze the pragmatics of Toraja ergative constructions. Looking at the actual use of ergative constructions by Toraja speakers reveals that the difference with Samoan is not only formal but also pragmatic.

5. Pragmatic value and metalinguistic treatment of ergative constructions in Toraja A first hint suggesting a lack of fit between the pragmatic function of ergativity in Toraja and in Samoan comes from the metalinguistic treatment of sentences with ergative subjects provided by my bilingual language assistants as we were engaged in translating the transcriptions of recorded speeches from Toraja into Indonesian. As most of my language assistants in the field were bilingual in Toraja and Indonesian, it came quite naturally to use Indonesian as a meta-language as we were engaged in transcribing and glossing Toraja linguistic material. It was in the course of these linguistic sessions that I came across the first piece of evidence against the equivalence between agency and ergativity. Far from confirming the relation between ergative person marking and the attribution of agency and responsibility to the subject/agent, the Indonesian glosses provided by my bilingual assistants revealed an opposite pattern. Ergative clauses inToraja were consistently turned into passives, patient voice12 , or impersonal constructions in Indonesian. The examples below illustrate this pattern of grammatical conversion, which appeared to be very consistent among all my language assistants. The first excerpt (14) comes from a traditional political meeting (kombongan) held at one ancestral houses of the village where I was residing, a participant is re-evoking the story of a mythical sword (the dosso) which he claimed was lost

206

Aurora Donzelli

by Puang Tumpa Patalangi’ when he went to gamble in the village Buakayu. The Toraja line (14): (14)

[December 23, 2002. Tape number 41]

94.

Eh. . . na=ala to Buakayu Eh 3.ERG=take people Buakayu ‘Eh. . . the Buakayanese took it.’

was turned by my assistant into the following Indonesian gloss (14a): v (14a)

Di-ambil orang Buakayu PAS-Take person Buakayu ‘It was taken by the Buakayu people.’

During another state-sponsored political meeting (rapat), Pak Batara, a very influential and authoritative noble man, deprecated the near-slavery conditions experienced by the Toraja who migrate in search for jobs to Malaysia and Singapore. Pak Batara’s intervention was aimed at highlighting the importance of building village schools and at the same time at emphasizing his merits in the promotion of the villagers’ education. In the excerpt below (15) he explains that his efforts to construct schools in the countryside are grounded on the idea that education is key to promoting local development and to preventing people from a future of migration and exploitation under Chinese employers in foreign countries. As he says: “I ‘work my butt out’ to build schools”: (15)

[February 24, 2003. Tape number 24]

2471.

supaya andi’ na-po-kaunan-komu baba’ So that NEG 3.ERG-CAUS-slave-2p.ABS Chinese ‘So that the Chinese will not turn you into (their) slaves’

2471a.

dio-lu tondok=na tau In-there village=DEF people ‘in their villages.’

2472.

Ia-mo-to ku=mati-mati-an-n=i That-PFV-DEM 1.s.ERG=die-RED-BEN-LK=3.ABS ‘Here (is the reason) I nearly kill myself (doing) it.’

2473.

saba’ ianna baba’ um-po-kaunan=komu Because if Chinese AF-CAUS-slave-2p. ABS ‘Because if the Chinese enslave you’

Is ergativity always a marker of agency?

2474.

tae’ ra na-sua-sua-manna=komu NEG LIM 3.ERG-give order-RED-LIM=2.p.ABS ‘they will not only give you orders’

2475.

tapi sae lako pessirrikan=na but until to toilet-3 ‘but even their toilets’

2475a.

um-base=i na=sua=komu 3.ERG-give orders-2.p. ABS AV13 -clean-3 ‘they will order you to clean!’

207

Here it is worth noticing how Batara conveyed a slightly victimized presentation of himself as an active agent of local development and education. While foregrounding his efforts, he also hints at his being affected by them (“I nearly kill myself”). However, what is highlighted here is not so much his role in the promotion of village education or that of the Chinese in the enslavement of the migrants. Rather, the emphasis is placed on the affectedness of the patients, of the people in the audience Batara is addressing to, who are represented as potential victims of Chinese exploitation. This is clearly conveyed by the Indonesian translation of the excerpt (15a), where the ergative constructions are consistently turned into what has been variously analyzed as a canonical passive or as a patient voice: (15a) 2471.

Supaya jangan kalian di-jadi-kan hamba oleh cina So that NEG 2.p PAS-become-CAUS slave by Chinese ‘So that you will not be turned into slaves by the Chinese’

2471a.

di kampung-nya orang In village-DEF people ‘in their villages.’

... 2474.

tidak hanya kamu di-suruh-suruh saja PAS-order-RED LIM NEG Only you ‘You will not just be given orders.’

2475.

tapi sampai pada tempat kencing But until to place piss ‘But even (their) toilets’

208 2475a.

Aurora Donzelli

kamu di-suruh men-cuci-nya you PAS-order meN-clean-DEF ‘you will be ordered to clean.’

Interestingly, the Indonesian gloss above (15a) presents a series of agentless passive constructions. This tendency of transforming Toraja ergative sentences into passive/patient voice constructions emerged also in my work with another language assistant who was fluent in English and preferred to use English as a metalanguage. Example (16) is taken from an electoral rally in which the speaker is arguing that the candidate he is supporting was democratically selected: (16)

[July 31, 2002. Tape number 48]

214.

Dadi karena na=usul to buda aspirasi to So because 3.ERG=propose person many aspiration person buda many ‘So since it was proposed by the many, (by) the aspiration of the many’

214a.

keluarga ki-noko-i 1.p.ERG=Sit-LOC family ‘we gathered the family in a meeting’

215.

kumua Ponja mo ta=pa-maju That Ponja PFV 1p.ERG=CAUS-push forward ‘(in which we decided) that it is Ponja that we support (we push forward)’

216.

saba’ nang ia na=ka-buda-i to buda 3.ERG-VBZ-many-3.ABS person many because really 3 ‘because he is really the one who is favored by the many.’

As apparent in the excerpt above, two out of four transitive active sentences with ergative proclitics are transformed into passives constructions in English (lines 214 and 216), in which Ponja (the candidate) is promoted from the position of direct object to that of subject, while the original subject (to buda, ‘the many the crowd’) expressed through ergative person markers in the Toraja original is transformed into an oblique argument and postponed agent in the English translation. Rather than drawing conclusions on patterns of morpho-syntactic equivalence between Toraja ergative constructions and their Indonesian or English counterparts on the grounds of the speakers’ metalinguistic analyses, what it

Is ergativity always a marker of agency?

209

is important to observe in the examples presented above is that Toraja ergative constructions clearly assign higher saliency to the patient noun phrase (NPP). The fact that ergative constructions should be understood as expressing the patient’s higher saliency is suggested also by word order. Although word order is not fixed in Toraja, NPA always occurs in post-verbal position and NPP is often fronted or topicalized through right dislocation. Moreover, it should be noted that in all the examples above, the ergative transitive subjects (NPA ) seem not to be marked for particular prominence (neither ontologically nor discursively) in the sentence as they are all indeterminate and generic: “the Buakayuese” in (14), “the Chinese” in (15), “the many/the crowd” in (16). What is fore-grounded is instead the semantic patient: the candidate chosen by the crowd in (16), the mythical sword in (14), the potential victims of the Chinese’s exploitation (15). Ergative constructions in Toraja are thus clearly associated with a process of object topicalization and/or focalization. In this respect they seem to share some functional properties of passive constructions. Indeed, as Shibatani14 (1985) showed on the basis of cross-linguistic evidence, passive constructions are associated with the pragmatic function of “agent defocusing”. While we should be cautious in drawing simplistic equations between ergative sentences in Toraja, di- passive constructions in Indonesian, and English passives, there is little doubt that, contrary to the Samoan case, the pragmatic value of ergative marking in Toraja is certainly not related to the enhancement of the agency of the NPA . As it was clear in the examples analyzed above, ergative constructions in Toraja seem to have a similar function to that of genitive modifiers in Samoan (Duranti 1990; Duranti and Ochs 1990), in that they convey an emphasis on the object (or patient) rather than the agent, thus resulting in a mitigation of the role of the agent as the cause of the event being described. The association between ergativity and object-focus constructions has already been argued for other neighboring languages with ergative morphology (see Friberg 1991; Himmelman 1996; Matti 1994: 72–73).15 Hence, the contrast with Samoan could not be more pronounced, while in Samoan ergative markers indexed the assignment of agency to the referent of the NP, in Toraja they correspond to strategies of agent-defocusing. While in Samoan, ergative case-marking was socially rare and pragmatically salient, in Toraja, rather than being exceptional, ergative proclitics on the verb (either crossreferencing a full NP or used in a pronominal function) are associated with the canonical and unmarked way of expressing a two arguments clause.

210

Aurora Donzelli

6. Encoding and performance of agency in Toraja political speech If ergative constructions described in the previous section represent the canonical unmarked choice in Toraja, how do Toraja speakers emphasize the agentivity of the NPA and assign responsibility its referent? As mentioned above, the encoding of agency, which in Samoan was produced through ergative casemarking, in Toraja is realized by another construction (called the actor voice or AV) associated with the verbal prefix /uN-/ and characterized by the lack of ergative proclitics and the optional presence of the enclitic suffix /–i/ which cross-references or indicates the object. AV appears much less frequently and constitutes the marked alternative to the canonical ergative transitive clause. A confirmation of the fact that /uN -/ type of clause expresses an emphatization of the Agent role is conveyed by the fact that my bilingual assistants consistently introduced a cleft sentence in the Indonesian translation of most Toraja /uN-/ constructions. The Toraja clause below (17) was for example turned into the Indonesian sentence (17a): (17)

Pak lurah un-jama=i te sura’ Mr Mayor AV-Work=3 DEF document ‘The mayor compile(s/d) the document.’

(17a)

ini Pak lurah yang meng-erja-kan surat Mr Mayor REL meN-Work-Kan document this ‘It is the mayor who compiles the document.’

AV constructions rarely appear in Toraja political speech. They are undoubtedly the marked choice and are clearly associated with the assignment of agency and responsibility to oneself or to a third party. Let’s consider a few examples drawn from naturalistic instances of political discourse. In excerpt (18) we can see an interesting instance of self-attribution of agency through the use of an AV construction in a speech delivered by Pak Batara during a rapat (state-sponsored political meeting). The meeting was the first of a series of encounters aimed at merging the two villages (desa) of Lemo and Marinding into a bigger administrative unit. However, the proceeding of the meeting got soon blocked by the obstructionist intervention of the Lemo delegation, who complained of having been excluded and marginalized by the Marinding people and threatened to abandon the meeting if the proposed name for the new administrative unit was not changed in order to incorporate the name of their village too. Seeking to re-establish control on the situation, Pak Batara argues that when he was the chief of an even bigger administrative unit (comprising the three villages of Lemo, Marinding, and Kandora) he had never

Is ergativity always a marker of agency?

211

favored either of the villages. He strongly emphasizes his impartiality as the ruler of the confederation of the three villages through the choice of a very strong construction with the /uN -/ prefix and the free personal pronoun, which strengthens the sense of his own personal agency (line 2237): (18)

[February 24, 2003. Tape number 24]

2237.

tonna kebetulan aku un-parenta=i When really 1.s AV-rule=3 ‘At that time I ruled.’

Although not in the Toraja original, the sense of heightened personal agency conveyed by this sentence could be well rendered by a cleft-sentence: “at that time I was the one who ruled!” Aiming at re-establishing his control over the Lemonese secessionist attempt, Batara proceeds with reminding them that it was his grandfather who had actually bought the land where the village of Lemo is now located. Therefore instead of advancing autonomist claims, they should remember that their village is a creation of his noble and powerful family. Hence, as he reminded the Lemo delegation that they are basically his own vassals, he describes the actions of his ancestors in a very agentive way using AV constructions. At line 2253, for example, he declares that it was his father who increased the number of the tongkonan (ancestral houses) in Lemo from two to five: 2253.

iatu um-pa-lima=i tongkonan lo’ ambe’-ku. That.TOP AV-CAUS-five=3 ancestral house there father-1.s ‘The one who turned the tongkonans into five down there (who increased the number of tongkonans to five units) was my father!’

Another pragmatic context for the deployment of AV constructions is constituted by threats and accusations. In excerpt (19) for example, an absent party is accused through the use of an AV construction of having demonstrated blasphemous behavior: (19)

[December 23, 2002. Tape number 41]

705.

Dia mangka-mo un-lutu aluk AV-destroy religion 3.s PFV-PFV ‘He has already attacked religion’

Another interesting locus for the analysis of the encoding and performance of agency in Toraja political discourse is constituted by electoral speeches. As a general rule, Toraja political rallies are marked by a high degree of indirection

212

Aurora Donzelli

and understatement. Speakers tend to deploy grammatical devices aimed at bracketing their own agency and the audience’s responsibility. However, against this general tendency to restrain the use of the explicit linguistic foregrounding of agency, it is worth noticing that the rare deployment of AV constructions in electoral rallies was generally related to an appeal to the emerging rhetoric of commitment and bottom-up democracy which characterized the process of administrative decentralization in the aftermath of the collapse of President Soeharto’s long-term authoritarian regime in May 1998. The excerpt below (20) was recorded at a village political rally in which several noble and authoritative supporters of Pong Jaka spoke in favor of his candidacy. Here we see how Massudi, the chief of a neighboring lembang and the son of a very powerful and noble man from Sangalla’, conveys his family’s support to the candidate (Pong Jaka) by arguing that he was selected through a democratic procedure. Making an explicit reference to the Indonesian term tanggung jawab (‘responsibility’), Massudi emphasizes the audience’s responsibility towards the candidate (line 82a) as a consequence of its involvement in the decisions that led to the candidate’s selection. He thus exhorts the audience not to withdraw its support of the candidate, by reminding them – through the use of an AV construction (line 87) – that they played an active role in choosing the candidate: (20)

[July 31, 2002. Tape number 48]

82.

tae’ mi-la ul-lamba’=i NEG 2.pl-FUT AV-neglect=3 ‘You will not neglect’

82a.

kumua iato tanggung jawab=na kale=na ia Pong Jaka That that.TOP responsibility=DEF body=DEF 3 Pong Jaka ‘that it is really a matter of responsibility towards Pong Jaka (himself)’

... 85.

Dadi yanna dako’to ke-den-sia upa’ So if later DEM if-Exist-LIM hope ‘So later if there is luck’

86.

tu Pong Jaka Tae’ mi=la un-tiro=i-tiro=i NEG 2.pl=FUT AV-look=3-look=3 REL Pong Jaka ‘you will not be reckless with Pong Jaka’

87.

Saba’ kita un-n-angka’=i Because 2.p.HON AV-LK-pick=3

Is ergativity always a marker of agency?

213

‘because you picked him (i.e. it is you who have chosen him)’

7. Conclusions The analysis developed in the previous pages has offered a description of the ways in which, in the course of political interaction, Toraja speakers use two different grammatical patterns (with ergative proclitics or with /uN -/ prefixes) for the representation of their actions. This dual strategy for expressing the relationship between an acting subject and its object allows a differentiation between a “technical” and a “deontological” agent.16 The “technical” agent is expressed through ergative constructions. These are the unmarked way of realizing a two arguments clause. Unlike what Duranti described for Samoan ergativity, they are not endowed with a moral/pragmatic force and they do not entail the assignment of agency and responsibility to the speaker or to another party. The “deontological” agent is instead realized through AV constructions, which, like ergative case-marking in Samoan, play an important function within the local political debate, in that, they are associated with the implications of being a moral and accountable agent. As was revealed by examples drawn from the recording of political interaction, speakers use AV constructions (marked by the prefix /uN -/) in a variety of contexts to perform accusations – excerpt (19) – or to assign praise and responsibility to themselves or to their interlocutors – excerpts (18) and (20) – that is, to express deontological agency. While presenting evidence against the cross-linguistic equivalence of semantic agentivity and morphological ergativity, the analysis corroborated the idea that the assignment of agency both to oneself and to a third party is a pragmatically salient act performed through specific morpho-syntactic resources. The relative rarity of this construction within spontaneous interaction may suggest that the assignment of agency constitutes a delicate matter, both in a crosscultural and in a cross-linguistic perspective. Hinting at the fact that speech acts that entail the assessment of the participants’ implication with certain state of affairs are universally bound to evoke careful political reasoning and linguistic behavior. A final observation to be made with respect to this inquiry in the Toraja pragmatics of action has to do with the fact that although AV constructions are undoubtedly associated with an emphasis on the (deontological) agency of the referent, they do not encode a high degree of transitivity. While Hopper and Thompson (1980) explicitly defined agency as one of the ten components of transitivity, an implicit notion of transitivity seems to shape the common under-

214

Aurora Donzelli

standing of the notion of agency both in social theory and in anthropological thinking. In an article that effectively summarizes how agency is defined in social theory, Sewell (1992: 19) points out that “to be an agent means to be capable of exerting some degree of control over the social relations in which one is enmeshed, which in turn implies the ability to transform those social relations to some degree”. Duranti’s (2004: 453) working definition of agency also highlighted how agency entails “control over one’s behavior”, capacity to affect other’s entities, and possibility of being evaluated for one’s actions. These notions seem to be closely related to the “prototypical transitive event” which involves: a “volitional, controlling, actively-initiating agent who is responsible for the event, thus [functioning as its salient cause]” (Giv´on 1994: 7). Indeed, as it seems to me, the general association of the notion of agency with the ideas of active-ness, voluntarism, and creativity is probably due to its tacit assimilation with conceptions and notions derived from Western semantic and linguistic theory, such as that of transitivity as a property of a clause which describes an action “which is carried over or transferred from an agent to a patient . . . [and] which is typically effective in someway” (Hopper and Thompson 1980: 251). Toraja grammar offers an interesting vantage point from which to rethink this tacit equivalence. Indeed, it presents an apparent contradiction: the AV grammatical construction, which encodes the highest degree of agency does not correspond to an increase of the valence (or semantic transitivity) of the sentence. AV constructions instead tend to occur in clauses where transitivity has been mitigated. They foreground the role of the agent but at the same time they convey a backgrounding of the object, which is generally (but not always!) indefinite or only partially affected. Toraja AV constructions refer to a “centripetal” of action in which the actor appears to be more affected than the patient. In this sense, they resemble the semantics of middle voice, that is a form indicating that the subject is affected by the action of the verb (a phenomenon that has been described for deponent verbs in Georgian, Latin, and ancient Greek, see Klaiman [1991]). The mismatch between agency and transitivity poses a challenge to the implicit equivalence between agency and transitivity understood as a transfer of activity or as a causative relation between the agent and the patient. This opens a critical reflection on how the interpretation of social realities may be unconsciously informed by semantic notions derived from Western linguistics and suggests that grounding our understanding of agency in terms of transitivity may be a partially misleading move for social theory. The idea of ‘involvement’ seems better suited to capture the dialectics of control and affectedness underlying human action, hence suggesting that agency implies being affected as much as being active.

Is ergativity always a marker of agency?

215

Sarah Lawrence College

Notes *

1. 2.

3.

4.

5.

6.

7.

The research on which this paper is based was generously funded by two postdoctoral fellowships from the Portuguese Foundation for Science and Technology (grants SFRH/BPD/40397/2007 and SFRH/BPD/21059/2004). My fieldwork in Toraja was conducted under the sponsorship of the Lembaga Ilmu Pengetahuan Indonesia and Universitas Hasanuddin in Makassar. I am grateful to Li Wei for having invited me to write an article for this issue of ALR and to Alessandro Duranti, Elinor Ochs and Ulrike Mosel for the insightful feedback they provided on some of the material presented in these pages. A special acknowledgement goes to my linguistic assistants: Clemens Malliwa, Daud Tanjong and Salmon Sirappa. Note that /uN-/ prefixes in Toraja are also obligatorily used for complements corresponding to what in English would be infinitives. The monographs by Ahearn (2001b), Duranti (1994), and Gell (1998) constitute rare cases in which the study of social change and emotions (Ahearn), art (Gell), and political practices (Duranti) is conducted from the vantage point of agency. Most of the participants in these interactions have competence in both Toraja and Indonesian (the national language). In this article however, I chose only to focus on material in Toraja. The data were collected in the course of two major periods of ethnographic fieldwork conducted in Toraja (ten consecutive months – between 2002 and 2003 – and five months – between May and October 2004), as well as during shorter stays in 1998, 2000, 2001, 2005–2006. The notion of ergativity I am dealing with here is restricted to morphological ergativity. I thus leave aside the issue of syntactic ergativity that concerns a significantly narrower number of languages. In passing I would like to point out that many linguists have argued that the notion of Subject is hardly applicable to ergative constructions. Comrie (1978: 330–331) for example tries to avoid the use of the notion of Subject and proposes a tripartite distinction (A, S, P); where A is used to refer to what is generally called the subject of a transitive clause (or ‘agent’), S refers to what is generally called the subject of an intransitive clause, and P stands for the direct object or ‘patient’. The abbreviations used for grammatical terms are based on the Leipzig Glossing Rules (http://www.eva.mpg.de/lingua/files/morpheme.html) with a few modifications: 1: 1st person; 2: 2nd person; 3: 3rd person; ABS: absolutive; ACC: accusative; ANTIP: antipassive;APPL: applicative;ART: article;AV: actor voice; BEN: benefactive; ber-: in Indonesian middle voice marker; CAUS: causative; CLF: classifier; COMP: complementizer; DEF: definite; DEM demonstrative; ERG: ergative; excl: exclusive; FUT: future; HES: hesitation; HON: honorific; -In: end-point or applicative

216

8.

9.

10.

11.

12.

Aurora Donzelli

voice marker (Jakartan Indonesian); incl: inclusive; INTR: intransitive; IPF: imperfective; -Kan: in Indonesian has several functions, such as applicative, causative, benefactive, and transitivizer; -Lah: in Indonesian this particle expresses a variety of meanings (i.e. imperative, concessive, and contrastive focus); LIM: limitative; LK: linker; LOC: locative; meN-: in Indonesian marks AV; MUT: mutual; NEG: negator; NMZ: nominalizer; NOM: nominative; NP: noun phrase; NVOL: non-volitional; -Nya: in Indonesian marks possession, definiteness or a generalized relationship of association; PASS: passive; PFV: perfective; pl: plural; PV: patient voice construction; PROG: progressive; QM: question marker; RDP: reduplication; REL: relativizer/relative clause marker; s: singular; TOP: topicalizer; TA: tense aspect; TR: transitive; VET vetative; VBZ: verbalizer. - affix; = clitic Other contributions to an ethnographic and pragmatic oriented analysis of ergativity were offered by: Jack Du Bois (1987), Clyfton Pye (1990), Bambi Schiefflin (1985, 1990). Susan Goldin-Meadow’s (2003) research on how deaf children use spontaneous sign languages to encode semantic relations has addressed the cognitive and prelinguistic notion of ergativity. The ergative-absolutive pattern is far from being perfectly coherent. Sometimes, negation or question markers, temporal or location adverbials, and certain conjunctions trigger the use of pronominal proclitics also on intransitive predicates. Another deviation from the ergative-absolutive distribution of pronominal clitics is displayed by antipassive constructions in which absolutive enclitics encode the A function and occur in transitive sentences. Scholars who work on other South Sulawesi languages such as Bugis, Mamasa, Mamuju, and Konjo noted similar exceptions to the ergative pattern (Friberg 1991; Matti 1994; Valkama 1995). This split pattern seems to be mostly based on syntactic grounds, even though the role of semantic factors should not be completely ruled out (see Valkama 1995: 47). The Toraja ergative-absolutive pattern demonstrates a close similarity to ergative Mayan languages of Mexico, such as Tzeltal and Sacapultec (see Du Bois 1987). Both in Sulawesi and in Maya languages ergativity is patterned in the verb-agreement. The similarity with Mayan ergative systems has been noted also by other scholars working on other Sulawesi languages. Martens (1988: 270) highlighted that Uma and Tzeltal have the same structure: ERG-verb-ABS for transitive clauses; and verb-ABS for intransitive clauses. Matti (1994: 67) noted that both in Mayan and in Sulawesi languages the ergative prefixes have the same form as the possessive set. It should be noted that the third person enclitic /-i/ cross-referencing the object is often dropped. Valkama (1995: 53) observed a similar tendency to drop the object indexing enclitic in the neighboring Duri language and argued that this occurs when the object is previously mentioned in the discourse. Friberg (1991) noted a similar tendency in Konjo and correlated the appearance and disappearance of the objectindexing clitic to the specific versus generic status of the object and to its degree of affectedness. The notion of the passive in Indonesian is particularly controversial in Austronesian linguistics. Chung (1976: 43) has no doubt in calling passive the process that

Is ergativity always a marker of agency?

13. 14.

15.

16.

217

“turns the direct object into a subject, removes the underlying subject to a prepositional phrase (with oleh ‘by’), and adds a prefix di- to the verb. The prefix replaces the active transitive prefix meng-”. However, Indonesian di-constructions are often analyzed as patient voice (Arka and Ross 2005; Austin and Musgrave 2008; Gil 2002; Himmelmann 2005; Wouk and Ross 2002). Space constraints prevent me from providing a detailed discussion of the debate. Therefore I opted for referring to Indonesian ‘di-’ constructions as passive/patient voice constructions. Note that /uN-/ prefixes in Toraja are also obligatorily used for complements corresponding to what in English would be infinitives. Advocating a pragmatic understanding of passive constructions, Shibatani (1985) showed that the main function of passives (shared by related constructions such as reflexive, reciprocal, honorific, potential and spontaneous formations) is that of backgrounding and defocusing the agent. Shibatani (1985: 834) insists that his view of passivization “as an agent-centered phenomenon” should not be confused with the idea that the main function of passive constructions is that of foregrounding the object. This is also confirmed by Cornelius Salombe’(1982), a Toraja linguist who in his description of Toraja grammar and verbal morphology called ergative constructions “the non-canonic passive” and assimilated it to the “object-preposing passive” in Indonesian (Salombe’1982: 88–91). I thank Elinor Ochs for having suggested to me this terminology to identify the two constructions.

References Ahearn, Laura. 2001a. Language and agency. Annual Review of Anthropology 30. 109– 137. Ahearn, M. Laura. 2001b. Invitations to love: Literacy, love letters and social change in Nepal. Ann Arbor: University of Michigan Press. Archer, Margaret. 2000. Being human: The problem of agency. Cambridge: Cambridge University Press. Arka I. Wayan & Malcolm D. Ross (eds.). 2005. The many faces of Austronesian voice systems: Some new empirical studies. Canberra: Pacific Linguistics. Austin K. Peter & Simon Musgrave (eds.). 2008. Voice and grammatical functions in Austronesian languages. Stanford: CSLI. Blust, Robert. 1977. The Proto-Austronesian pronouns and Austronesian subgrouping: A preliminary report. Working Papers in Linguistics of the University of Hawai’i 9. 1–15. Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial thought and historical difference. Princeton: Princeton University Press. Chung, Sandra. 1976. An object-creating rule in Bahasa Indonesia. Linguistic Inquiry 7. 41–87.

218

Aurora Donzelli

Churchward, Maxwell C.. 1953. Tongan grammar. London: Oxford University Press. Comrie, Bernard. 1978. Ergativity. In Winfred P. Lehmann (ed.), Syntactic typology: Studies in the phenomenology of language, 329–394. Austin: University of Texas Press. Davidson, Donald. 2001. Agency. In Donald Davidson (ed.), Essays on actions and events, 43–63. New York: Oxford University Press. Dixon, Robert M. W. 1979. Ergativity. Language 55. 59–138. Donzelli, Aurora & Alessandra Fasulo (eds.). 2007. Introduzione [Introduction]. In Aurora Donzelli & Alessandra Fasulo (eds.). Agency e Linguaggio: Etnoteorie della Soggettivit`a e della Responsabilit`a nell’Azione Sociale [Agency and language: ethnotheories of subjectivity and responsibility in social action], 11–27.Rome: Meltemi. Du Bois, John W. 1987. The discourse basis of ergativity. Language 63. 805–855. Duranti, Alessandro. 1990. Politics and grammar: Agency in Samoan political discourse. American Ethnologist 17(4). 646–667. Duranti, Alessandro. 1994. From grammar to politics: Linguistic anthropology in a western Samoan village. Berkeley & Los Angeles: University of California Press. Duranti, Alessandro. 2004. Agency in language. In Alessandro Duranti (ed.), A companion to linguistic anthropology, 451–473. Malden, MA: Blackwell. Duranti, Alessandro & Elinor Ochs. 1990. Genitive constructions and agency in Samoan discourse. Studies in Language 14. 1–23. Emirbayer, Mustafa & Ann Mische. 1998. What is agency? American Journal of Sociology 103(4). 962–1023. Fillmore, Charles J. 1968. The case for case. In Emmon Bach & Robert T. Harms (eds.), Universals of linguistic theory, 1–88. New York: Holt. Friberg, Barbara. 1991. Ergativity, focus, and verb morphology in several South Sulawesi languages. In Ray Harlow (ed.), VICAL 2 Western Austronesian and Contact Languages (Papers from the Fifth International Conference on Austronesian Linguistics), 103–130. Auckland: Linguistic Society of New Zealand. Gell, Alfred. 1998. Art and agency: An anthropological theory. Oxford: Clarendon. Gil, David. 2002. The prefixes di- and N- in Malay / Indonesian dialects. In Fay Wouk & Malcom Ross (eds.), The history and typology of western Austronesian voice systems, 451–474. Canberra: Pacific Linguistics. Giv´on, Talmy. 1994. Introduction. The pragmatics of de-transitive voice: Functional and typological aspects of inversion. In Talmy Giv´on (ed.), Voice and inversion, 3–47. Amsterdam: John Benjamins. Goldin-Meadow, Susan. 2003. Thought before language: Do we think ergative? In Dedre Gentner & Susan Goldin-Meadow (eds.), Language in mind: Advances in the study of language and thought, 493–522. Cambridge, MA: MIT Press. Hanks, William, Sachiko Ide & Yasuhiro Katagiri. 2009. Towards an emancipatory pragmatics. Journal of Pragmatics 41(1). 1–9. Heidegger, Martin. 1978. Being and time. Oxford: Blackwell. Himmelmann, Nikolaus P. 1996. Person marking and grammatical relations in Sulawesi. In Hein Steinhauer (ed.), Papers in Austronesian linguistics, 115–136. Canberra: Pacific Linguistics.

Is ergativity always a marker of agency?

219

Himmelmann, P. Nikolaus. 2005. The Austronesian languages of Asia and Madagascar: Typological characteristics. In K. A. Adelaar & Nikolaus P. Himmelmann (eds.), The Austronesian languages of Asia and Madagascar, 110–181. London: Routledge. Hopper, Paul J. & Sandra A. Thompson. 1980. Transitivity in grammar and discourse. Language 56. 251–299. Jackendoff, Ray. 2007. Language, consciousness, culture: Essays on mental structure. Cambridge, MA: MIT Press. Klaiman, H. Miriam. 1991. Grammatical voice. Cambridge: Cambridge University Press. Martens, Michael. 1988. Focus or ergativity? Pronoun sets in Uma. Papers in Western Austonesian linguistics. Canberra: Pacific Linguistics A (4) 79. 263–277. Matti, David F. 1994. Mamasa pronoun sets. In Ren´e van den Berg (ed.), Studies in Sulawesi Linguistics III (NUSA: Linguistic Studies of Indonesian and Other Languages in Indonesia), 65–88. Jakarta: Universitas Katolik Indonesia Atma Jaya. Merleau-Ponty, Maurice. 1964a. Le visible et l’invisible [The visible and the invisible]. ´ Paris: Edition Gallimard. Merleau-Ponty, Maurice. 1964b. Indirect language and the voices of silence. In Maurice Merleau-Ponty (ed.), Signs, 39–84. Evanston: Northwestern University Press. Merleau-Ponty, Maurice. 1964c. On the phenomenology of language. In Maurice Merleau-Ponty (ed.), Signs, 84–98. Evanston: Northwestern University Press. Noorduyn, Jacobus. 1991. A critical survey of studies on the languages of Sulawesi. Leiden: KITLV Press. Ochs, Elinor. 1982. Ergativity and word order in Samoan child language. Language 58. 646–671. Ochs, Elinor. 1988. Culture and language development: Language acquisition and language socialization in a Samoan village. New York: Cambridge University Press. Ortner, Sherry. 2006. Power and projects: Reflections on agency. In Sherry Ortner (ed.), Anthropology and social theory: Culture, power and the acting subject, 129–153. Durham, NC: Duke University Press. Pye, Clifton. 1990. The acquisition of ergative languages. Linguistics 28. 1291–1330. Salombe’, Cornelius. 1982. Bahasa Toraja Saqdan: Proses morfemis kata kerja. [Toraja Saqdan language: Verbal morphology]. Jakarta: Penerbit Djambatan. Schieffelin, Bambi. 1985. The acquisition of Kaluli. In Dan Isaac Slobin (ed.), The crosslinguistic study of language acquisition, 525–593. Mahwah, NJ: Lawrence Erlbaum. Schieffelin, Bambi. 1990. The give and take of everyday life. Cambridge: Cambridge University Press. Sewell, William Jr. 1992. A theory of structure: Duality, agency, and transformation. American Journal of Sociology 98(1). 1–29. Shibatani, Masayoshi. 1985. Passives and related constructions: A prototype analysis. Language 61. 821–848. Sura’ Madatu [Bible]. 1995. Jakarta: Lembaga Alkitab Indonesia.

220

Aurora Donzelli

Valkama, Karl. 1995. Person marking in Duri. In Ren´e van den Berg (ed.), Studies in Sulawesi Linguistics IV (NUSA: Linguistic Studies of Indonesian and Other Languages in Indonesia), 47–93. Jakarta: Universitas Katolik Indonesia Atma Jaya. Wouk Fay & Malcolm D. Ross (eds.). 2002. The history and typology of western Austronesian voice systems. Canberra: Pacific Linguistics. Aurora Donzelli is a linguistic anthropologist who has conducted ethnographic fieldwork in Indonesia. She has published on political oratory and vernacular practical philosophies (in Agency e Linguaggio, Meltemi, 2007), language ideologies and the aesthetics of verbal performance (in Text and Talk 27–4, 2007), and evangelization and colonial discourse genres (in Learning Religion, Berghahn, 2007). Before becoming Assistant Professor of Anthropology at Sarah Lawrence in New York, she worked in London (ELAP-SOAS), Lisbon (ILTEC) and Milan (Bicocca).

Opportunities for learning during storybook reading at preschool1 AMELIA CHURCH

Abstract Research in teacher-child interaction in early childhood settings typically focuses on the teacher’s talk, with a particular interest in the types of questions teachers use. This paper is interested in how teachers respond where children initiate the interaction, to explore the opportunities children may, or may not, have to get to the center of child-centered learning. Examples are provided here of young children’s self-selected questions or comments during shared bookreading and the responses made by teachers. The data illustrate that even though children demonstrate communicative competence in taking timely and appropriate turns in the interaction, the relevance of this talk is determined wholly by the teacher. Implications of extending children’s contribution to the talk-in-interaction are considered in relation to opportunities for learning. Keywords: teacher-child interaction; applied conversation analysis; emergent literacy.

1. Introduction There has been a tradition of ethnomethodological research in education which has been preoccupied with how elements of classroom interaction are done (for review see Baker 1997; Mori and Zuengler 2008). An interest in the ethnography of classrooms is pursued in the literature in primary and secondary school education research, but is yet to be developed early childhood contexts. This paper seeks to contribute to this space by analyzing particular features of interactions between children and teachers during shared storybook reading. The primary aim of this paper is to detail the opportunities children have to contribute to

222

Amelia Church

storybook reading events – specifically where children initiate topics – and how these contributions are responded to by teachers. Child-centered learning is a refrain of early childhood education, but there is little evidence of how children might actually get to the center of this learning. Research on reading in early childhood classrooms has overwhelmingly focused on text selection, and strategies or types of questioning that teachers use when reading with children. These strategies typically overlook the fact that this talk between teachers and children is co-constructed and constitutes the social action of the classroom; action that incorporates sharing storybooks. The aim of this paper is to see how the interactional work of responding to child-initiated topics is done during storybook reading. This research identifies instances where children initiate a comment or query about the text, and the focus of the analysis rests with how teachers respond. The data presented in this paper come from a larger study of effective literacy practices in early childhood (further details below). Teachers were videotaped reading books with small groups of four-year-old children, and the broader aim of the research is to determine effective strategies used by teachers. In transcribing this data, particularly striking were a number of instances in which the children made a bid for the teacher’s attention, but these bids were typically not taken up by the teacher. Given the current philosophy in early childhood education of child-centered learning, it is important to see what actually happens when the child attempts to get to the center, how the teacher responds, and what the implications of these interactions might be for children’s learning.

2. Reading in early childhood classrooms Children’s literacy experiences in preschool learning environments are unarguably related to later literacy development in the primary years of schooling (e.g. Wray et al. 2000; Connor et al. 2006). While variability in children’s opportunity for early reading development is mostly accounted for in the home environment, kindergarten programs ideally provide access to quality books and interactions with qualified teachers (e.g. Justice et al. 2008). In other words, “print knowledge is strongly influenced by the environments in which children are raised and, by extension, is an aspect of development that can be readily modified with changes to the environment” (Justice et al. 2009: 68) While literacy-rich environments in early childhood are multidimensional and require appropriate resources, meaningful opportunities to experiment and quality interactions with more knowledgeable others, this brief discussion focuses on storybook reading given the dataset in the present study. There are

Opportunities for learning during storybook reading

223

discrepancies in the degree of positive effect of shared book reading in children’s literacy attainment which can be accounted for in individual children’s home environment and their existing literacy knowledge (e.g. Wasik and Bond 2001). Furthermore, the focus of the interaction while sharing books can vary enormously from simply labeling pictures, to code-related talk (e.g. pointing out letters), to decontextualized talk (e.g. hypothesizing) which draws on children’s broader world knowledge (Hindman et al. 2008). Nonetheless, sharing books in early learning environments inevitably provides opportunities for children to develop concepts about print, extend vocabulary and participate in a literate community (Dickinson and Tabors 2001). Book reading in particular has been shown to promote vocabulary development, which supports later literacy attainment (Gerde and Powell 2009). Evidence also suggests that teachers can improve children’s print knowledge while reading stories aloud by explicitly drawing children’s attention to the text by pointing at words (and letter shapes) and commenting and asking questions about the print (e.g. Justice et al. 2009). More generally, interaction between teachers and children while reading stories enhances children’s language and early reading skills (Lonigan and Whitehurst 1998). Sharing books, in particular dialogic reading – that is “the use of probing, practice, teaching, feedback and repetition” (Lonigan and Whitehurst 1998: 282) – encourages children’s expressive language. Indeed, the quality of teacher-child interactions are not only a feature of storybook reading, but are a significant factor in developmental outcomes for children (see Howes et al. 2008) We know that early childhood teachers have the opportunity to extend children’s literacy knowledge (Raban and Coates 2004), but we know little about what teachers do when children spontaneously create these opportunities. The focus has been on what happens when teachers ask questions, rather than what happens when children ask questions or initiate talk about text. As stated in the introduction, the aim of this paper is to identify how teachers respond to child-initiated topics in the interaction. Research in classroom interaction has typically focused on triadic dialogue instigated by the teacher. Mehan’s (1979) three part sequence of Initiation – Response – Evaluation (IRE) remains a dominant model in educational research, with later work focusing on the possibilities in the evaluative or feedback/followup third turn (Sinclair and Coulthard 1975). Most notably in early childhood education, the notion of ‘scaffolding’ (Bruner 1986) underscores the role of adult-child interaction in enabling children “to operate at higher levels of cognitive functioning than they could achieve on their own” (Wray et al. 2000). More recently, the structure of teacher-child talk emerged as a significant factor in the quality of early childhood classrooms in the Effective Provision of Pre-School

224

Amelia Church

Education (EPPE) study in the UK. This work demonstrated the importance of the type of questions used by teachers to extend children’s conversation (and conceptual knowledge) describing high-quality interactions as “sustained shared thinking” (Siraj-Blatchford and Manni 2008). Evidently, ‘high-quality’ interactions in classrooms have been investigated in multiple paradigms, but as a study in applied conversation analysis, this work is invested in how teachers and children co-construct talk-in-interaction and the implications of this talk for learning. There has been a sustained interest in institutional talk in ethnomethodology and conversation analysis, particularly in medical, legal and interview interactions (e.g. Drew and Heritage 1992), but also classroom contexts (e.g. McHoul 1978, 1990; Mehan 1979; Heap 1985; Lerner 1995; Macbeth 2000; Austin et al. 2001; see Baker 1997 and Mori and Zuengler 2008 for review). Notably, the talk is conceived of as institutional not simply because it occurs in the classroom as an institution but rather that the talk is governed by context-specific rules that deviate from ‘standard’ conversation. In other words, it is what we hear teachers and children doing in systematic ways (cf. Hester and Francis 2001) that renders the talk institutional. A productive approach to understand the talk-in-interaction of the classroom is to consider how the talk is structured as evidence of how reading is to be done in that context (Baker 1991). However, the orientation in extant research has remained with teacher’s talk (and children’s subsequent talk) in light of the power dynamic assigning speaker roles in the classroom. It remains to be seen, however, how children use other opportunities to participate in literacy events, how the teacher responds to these contributions, and what this might mean for children’s understanding of storybook reading and classroom culture more broadly. While applied linguists will be familiar with conversation analysis (CA) of classroom interaction – particularly in the second language acquisition literature (see below) – this paper presents a novel approach to understanding the interactions between teachers and children in early childhood settings. As such, this paper seeks to contribute to the emerging body of work in early childhood that uses CA as a productive tool to uncover the local social order constructed by children and teachers (e.g. Danby and Davidson 2007).

3. Method In order to explore opportunities for literacy learning created by children in interaction with teachers, we look to examples of children’s spontaneous talk during shared bookreading events in early childhood classrooms. The data presented throughout this paper are taken from a larger study of early literacy teaching

Opportunities for learning during storybook reading

225

practices (see http://www.edfac.unimelb.edu.au/younglearners/) undertaken as part of a longitudinal study funded by the Australian Scholarships Group and the Australian Research Council, at the University of Melbourne. As part of this study, teachers from ten kindergarten programs2 contributed several data sets, which included video observations of three literacy events with a small group of four-year-old children. One of these events was a shared storybook reading with a group of three or four children, and the videorecordings of these events (duration ranging 5–15 minutes) serve as the data here. These videorecordings captured 70 examples of child-initiated turns during the storybook reading session. Of the ten teachers participating in this study, examples are provided from responses made by nine teachers, each of whom is identified by pseudonym in the relevant extracts. There were no examples of child-initiated utterances in the storybook session with the tenth teacher. This does not serve as an analytic claim about this particular teacher’s practice, nor the frequency of children’s self-selected turns in talk-in-interaction in kindergarten settings; the reader is reminded that the observations were a series of short activities. The methodology used in this study is best characterized as applied conversation analysis (Heritage 2005), as it uses the findings of structures and rules of talk-in-interaction to investigate a particular kind of institutional talk with a view to informing more effective practices in classroom interactions (see ten Have 2001: 3). Conversation analysis (for introductions, see Schegloff 2007; ten Have 2007; Liddicoat 2007) proves fruitful in this study because we are primarily interested in the emic features of the talk, that is, the features of the interactions that are treated as meaningful by the participants themselves. The turn-taking organization of talk-in-interaction allows us to see how the speakers treat each prior utterance and how each utterance is taken up (or not) in the ongoing interaction. The rule-governed nature of conversation not only accounts for the fact that one turn occurs after another but that each turn builds on the prior talk. The fact that speakers make the transition between turns seemingly effortlessly points to the cues provided for turn transition relevant places (Sacks et al. 1974), namely an intonational contour (falling or rising at the end), a syntactically complete string, often accompanied by nonverbal cues – e.g. a gaze shift to another (possibly next) speaker. In this study, the analysis is concerned with how teachers and children organize these turns at talk. At any turn transition relevant place – where one speaker finishes and another speaker may begin – there are a number of interactional possibilities. The speaker may choose to continue, the speaker may choose the next speaker, or the next speaker may choose her/himself to take the next turn.

226

Amelia Church

The extracts below provide examples of children’s self-selected turns and how these turns are responded to by the teacher. Each example is numbered for reference in the discussion but retains the teacher’s pseudonym and data extract number (e.g. Example 1 is an extract of the teacher Cathy reading with the children which records the sixteenth time a child self-selects as next speaker during the storybook reading event). In the extracts, the teacher is identified within the transcript as “ECT” (Early Childhood Teacher), children are identified by the first three letters of a pseudonym, the teacher reading directly from the text of the storybook is marked in italics, and the talk of particular interest is marked with ©. The extracts are taken from full transcriptions of the videorecorded storybook reading events, employing transcription conventions set out by Sacks et al. (1974); see Appendix. These conventions allow representation of features of talk which may prove meaningful to the participants, such as overlap, pauses and stress on particular words. The reason we are particularly interested in the features of talk in these classroom interactions, is that they point to opportunities for learning. As Markee and Seo point out “demonstrating that cognition can be directly observed in interaction is a difficult, perhaps impossible task” (2009: 39). For this reason, I will use the term ‘opportunities for learning’ throughout this paper to underscore the role of interaction in facilitating understanding, without making claims about cognition – what the children demonstrably understand about the learning object. Children’s knowledge of how one goes about participating in the talk-in-interaction and the rules of interaction specific to small group storybook reading, however, are more evidently on display. For the most part I am putting aside the debate that CA is “actually more cognitive (more honestly concerned with the mental worlds of participants) than most rival theories of discourse analysis” (Levinson 2006: 86), and the tension between psycholinguistic, sociocognitive and ethnomethodological theories learning. While we may avoid claims of cognition in discourse in this paper (see te Molder and Potter [2004] for further discussion), we will set out the very occasions, the very opportunities where learning can be co-constructed by children and teachers. In other words, “it is not, What does this talk tell us about underlying conceptions? Instead we ask, What is the contextually situated action being done here?” (Edwards 1993: 218). As such, this paper aligns with a non-cognitive stance in conversation analysis (see the discussion in Markee and Seo [2009: 44]; also Markee and Kasper [2004]), in that it is not seeking to dismiss the role of intrapersonal cognitive function nor claim that CA is the answer to every question. Rather that CA allows an emic lens to the interactions that take place in early childhood classrooms, providing an empirically-based analysis of classroom talk in early liter-

Opportunities for learning during storybook reading

227

acy events, as has proved fruitful in conversation analysis-for-second language acquisition (e.g. Richards and Seedhouse 2005; Markee 2008). The analysis that follows seeks to orient to an ‘ethnomethodological CA’ rather than a ‘linguistic CA’ (see Seedhouse 2007) to see how it is that teachers and children construct turn-taking as context-sensitive (e.g. Lerner 1995, 2003). Essentially, this paper argues for investigating the context-relevant features of interaction in early childhood settings to determine how opportunities for learning might be enacted by the participants themselves (see Firth and Wagner 1997).

4. Findings The observations in ten kindergartens revealed a variety of possible responses made by teachers to children’s self-initiated questions or comments during shared storybook reading. Following the 70 instances where children self-selected as next speaker, there was a variety of teacher behaviors, ranging from ignoring the child’s turn through to extending the child’s turn. The continuum of possible responses can be expressed as: (1) ignore; (2) postpone; (3) minimal acknowledgement; (4) confirmation; (5) expand. This does not propose a classification of the teacher’s responses per se, but rather outlines a progression of degree to which the child’s turn is oriented to in the interaction. The discussion below sets out this continuum, beginning with turns which are not taken up by the teacher, through to turns where the extension builds on the child’s turn – not only as an extension of the topic but in some cases adopted into the pedagogic script and becoming a focus of learning (e.g. vocabulary development) for the whole group. 4.1.

Ignore

In the observations recorded in this study, teachers would ignore comments or questions made by the children in a number of ways and for a number of purposes. This analysis does not claim to infer the teacher’s motivation, but rather details what talk of the children gets attended to (or not) and how the teacher responds. For example, teachers can ignore a child’s self-selected turn by resuming or continuing to read the story. In Example 1, Miranda (line 4) moves to provide an opinion on the illustrations (the teacher and four children are reading Where the Wild Things Are by Maurice Sendak) but the teacher does not treat this turn as a possible topic for the group, and continues reading (line 6). Karen’s commentary on the illustration (“that looks pretty”), is similarly ignored in Example 2.

228

Amelia Church

(1) Cathy.CI.16 01 ECT: show me the- pretend you’re a wild thing. 02 (1.5) ((all children use hands as claws)) 03 ECT: [ that’s it.] 04 Mir: [(I think that] thing is something else.) 05 (0.4) 06 © ECT: but Max stepped into his private boat, (2) Melanie.CI.5 01 ECT: a tv antenna? 02 Kar: that looks pretty¿ 03 (0.3) 04 © ECT: ↑ and a shiny pan.

Teachers were also heard to ignore children where a commentary on the narrative of the story is produced. In Example 3 (line 4), Bill provides a synopsis of the sea’s existence (i.e. how the sea came to be in David’s bedroom), but the teacher turns the page (line 5) and continues reading (line 6). (3) Jenny.CI.4 01 ECT: it’s just the noise the shell makes when you put it 02 against your ea: r. 03 (1.0) 04 Bil: the sea: came out of the shell .hhh (they are) x x. 05 (1.1) 06 © ECT: but David knew where there had been a sea in his bedroom.

In the same shared story book reading event, a child’s self-selected turn provides a decontextualized account (Example 4, line 3) that is aligned with the protagonist’s experience – i.e. like David in the story, she also has some sand at home. The teacher most likely hears this (evident by shift in eye gaze) but continues with the story without acknowledging the child’s turn. (4) Jenny.CI.5 01 ECT: I’m going to put the sa:n d into a ja:r . 02 (0.2) 03 ( ): I’ve got some sand at home. 04 (1.0) 05 © ECT: "should we go to the beach tomorrow" asked his dad.

Not only children’s commentary is ignored in storybook reading events. The question posed by Eric below (Example 5, line 5), is repeated (line 6) after abandoning the overlap, yet does not receive an answer (line 7). Again, it was evident that the teacher heard the child’s turn (indicated by a brief gaze towards this child), but chose to return to the text rather than take-up the child’s question.

Opportunities for learning during storybook reading

229

(5) Louise.CI.4 01 Xav: he’s saying "c c c c ". 02 ECT: he’s go- storming o:ff, those are bubbles they’re not 03 c’s. (0.5) he’s [storming off] with a grumpy look. 05 Eri: [why there-] 06 Eri: why there are bubbles. 07 © ECT: would you could you in a ca: r (.) eat them eat them 08 here they are.

Ignoring can be used by teachers as a strategy to allocate turns to children within the group. Example 6 is taken from a sequence in which Tim is dominating with a series of questions and offerings, many of which the teacher acknowledges. At some points, however, the teacher makes contributions from the two girls possible by ignoring Tim’s second guess (line 7) and selecting Ella as the next speaker (line 8). The teacher’s laughter (line 14) underscores the purposeful pursuit of a response from the reluctant Ella, despite Tim continuing to provide a correct response. (6) Sophie.CI.6 01 ECT: think about your book at h[o:me and what] animals are 02 in it. 03 TIM: [<white>] (0.4) <dog>? 04 ECT: hmmmm white dog it could be? wha[t do you think]. 05 Luc: [white dog]?= 06 ECT: =you think the white dog [(was next)]. 07 Tim: [GOLD FISH]! 08 © ECT: $what do you think it might be$ Ella? 09 (0.7) 10 ECT: ºshall I give you a little peepº? (0.3) 11 TIM: gold fish12 ECT: ºooh?º 13 Tim: purple cat. 14 Ell: ºpurple c[atº.] 15 © ECT: [hehuh]p(h)urple c(h)a:t. so what are the words?

In some cases, the teacher does not take up the child’s turn but re-directs the group’s attention, typically re-orienting the focus to the narrative. In Example 7, this is done with a minimal token of consequentiality (see Beach 1995), as the teacher uses the marker “okay” to return the groups’ attention to the story. (7) Melanie.CI.3 01 (0.8) 02 Kar: that’s got three wo:rds (like THat).

230

Amelia Church 02 03

(0.5) ECT: oka:y let’s have the story.

This re-orientation to the narrative of the story is also heard in Example 8. The teacher topicalizes Bill’s own narrative (i.e. when Bill goes to the beach he does somersaults), by tying the action of the subsequent turn to the protagonist of the story: “when he was at the beach”. (8) Jenny.CI.6 01 ECT: 02 03 ( ): 04 Bil: 05 06 ECT: 07

4.2.

x x x he does doesn’t he. (0.2) he’s (with) [x x x x. ] [a:nd when I:] go to the beach I: do sommersaults and it goes (like that). well do you wanna know something, when he was at the beach¿ (0.3) in his bedroom¿

Postpone

In some instances the teacher recognizes that the child has taken up the role of speaker but explicitly defers the proposed topic. In Examples 9, 10 and 11 (heard in the same reading of Dr Suess’ Green Eggs and Ham), the child’s self-selected turn constitutes a second attempt to gain the teacher’s attention. (9) Louise.CI.6 01 ECT: i want [you] to use your sensible thinking 02 [about what’s coming up]= 03 Eri: [can i sa:y some:thing]. 04 ECT: =okay? (.) what’s rhyming. 05 (0.8) 06 Eri: [(.hhh-)], ((moving towards book)) 07 © ECT: >[(hold on)] you can tell me in a second<. i will not eat them in a hou:se? i- >uh sorry< i will not eat them here or there i will not eat the:m?= (10) Louise.CI.10 01 ECT: I like the way you remembered all the words. 02 Eri: excuse m[e:¿ ] [so let’s-] we’ve al:most finished the story. 03 © ECT: 04 (1.0) I do not like, (11) Louise.CI.11 01 ECT: you d o not li ke them so you sa:y?= 02 Xav: excuse me? 03 © ECT: no were talki- we’re reading the ↑story. (0.2) try 04 them try them,

Opportunities for learning during storybook reading

231

In Example 9, Louise pre-empts Eric’s second attempt to take a turn as speaker – marked by the audible inbreath in line 6 – and urges him to “hold on, you can tell me in a second”. This turn allocation is not actually enacted, however, as Eric is not invited to “say something” in subsequent turns. A similar move to postpone the same child’s contribution is made by the teacher in Example 10 (line 3) – “we’ve almost finished the story”. The intimation that children’s questions or comments will only be welcome once the story has finished is further stated in Example 11 (line 3) in that “we’re reading the story” rather than talking about the story, asking questions about the story, or pausing for any other side sequence so close to the end of the story. 4.3.

Minimal acknowledgement

Minimal acknowledgements were made during storybook reading, where the child’s self-selected turn is acknowledged briefly before the teacher either returns to reading the story or moves to explicitly instruct the children to return their attention to the story. Minimal acknowledgement is taken here to indicate that the child’s contribution is acknowledged by the teacher in some way, but not expanded upon in the interaction. For example, in Example 12, the teacher audibly responds with laughter (line 5) to Genevieve’s commentary on a picture in the book, but resumes reading within the same turn. (12) Melanie.CI.4 01 ECT: "I’m drawing a: haterpillar (.) um: I mean a 02 cat erpillar" answered Jenny. 03 Gen: hehhe (0.2) $it’s a haterpillar because it has a hat 04 on$. 05 © ECT: ºhehehumº $.hhhh "oh de ar" sighed her mother.

Minimal acknowledgements were also heard where teacher recycled the child’s prior turn to affirm the prior statement. In Example 13, the teacher repeats Lucy’s turn in line 3 – overlapping Tim’s ratification of Lucy’s (correct) naming of one of the animals in the story. (13) Sophie.CI.2 01 Tim: .hhh there’s a red bird red bird? .hhh and w- .hhh 02 and a [fox .] 03 © Luc: [a horse]? a blue horse? 04 Tim: yeah? [you’re righ:t!]= [ºa blue horseº¿] 05 © ECT: 06 Tim= .hhh and .hhh and a red bird? (.) and a green frog,

232

Amelia Church

In Example 14, a sequence of children’s self-selected turns were minimally acknowledged by the teacher, (“thanks Bill you’ve read the first page”, “there is” and “it does look real”) until she explicitly directs the children (line 12) that the storybook reading will resume. (“Listening ears” is a phrase used by the teachers in this classroom to re-direct children to their role as audience.) (14) Jenny.CI.2 01 ECT: it’s ca:ll:ed there’s a sea (0.2) <[in my bedroom.] 02 ( ): [in my bedroom.] 03 ( ): [in my bedroom.] 04 © Bil: there’s a >sea in my bedroom<. 05 ECT: thanks Bill you’ve read the first page. 06 (0.8) 07 ( ): there’s a sea in my bedroom. 08 © ECT: there is. 09 ( ): that looks real. 10 © ECT: it does look real. 11 ( ): and this is= 12 ECT: =can I ºcan I (switch you around)º okay. got our 13 thinking ears and listening ears switched on now¿

Minimal acknowledgment was also produced by teachers in the form of closed questions. In Example 15, the falling intonation of the teacher’s “do you think so” (line 4) indicates the function of this turn as a receipt token rather than interrogation of Xavier’s suggestion that the character in the story/illustration is going to fall out of the tree, particularly as the teacher re-focuses the group’s attention (“okay”) in line 6. 4.4.

Confirmation

A more common form of acknowledgement heard in these storybook reading events with young children was a confirmation of the child’s prior turn. Although similar to the examples above of minimal acknowledgement, the teacher would typically add some new content in the feedback. In Example 16, Andrew inferred the reason for the Hungry Caterpillar having a stomach ache (“from eating all of that”) and the teacher responded with a positive evaluation and built on the semantic content of the child’s prior turn (line 4). (16) Michaela.CI.3 01 ECT: that night (0.3) he had a sto mach ache. 02 ( ): [x x ] 03 And: [from ea]ting all of that. 04 © ECT: yeah (0.7) so: full.

Opportunities for learning during storybook reading

233

A striking feature of these exchanges in which teachers provide confirmation, is that even where the child’s self-selected, self-initiated turn actually identifies a phoneme, grapheme, phonological feature, or word concept in the text, this is not necessarily taken up by the teacher. The child’s turn might be acknowledged for truth content, but often not developed as evidence of the child’s literacy knowledge. In Example 17, Bec identifies the identical rime in “Maurice” and “Horice”. Although the teacher affirms this observation made by Bec, she recycles the child’s “it’s close to” rather than explicitly marking the child’s phonological knowledge. (17) Cathy.CI.3. 01 ECT: and a person called Maurice wrote this story, 02 (Tom): hmm? 03 Jak: [ Maurice! ] 04 ECT: [and they’ve-] Maurice yeah and 05 >he [was clever because he-<] 06 Bec: [ it’s close to ] Hori:ce. 07 © ECT: it is close to Horice (.) and he did the pictures too.

In Example 18, Xavier identifies the letter “X” in the text of the book (lines 2 and 6) – it is the first letter of his name – and the teacher confirms the accuracy of this statement (“an ‘X’ for you”), but then continues to read the story (line 7). (18) Louise.CI.9 01 ECT: i do not [like them :?] 02 [‘x’ for ↑m ]e::. 03 (0.3) 04 ( ): a: [nywhe:re.] 05 ECT: [anywhere.] 06 Xav: but (.) there’s a ’xyz’’X’ for ↑me::. 07 ECT: an ’X’ for you:? (0.7) you do not like green eggs and ↑ham,

In Example 19, Karen identifies features of the word “hat” (line 3), which is confirmed by the teacher, but the letter concepts within the word are not expanded upon. (19) Melanie.CI.2 01 ECT: and there are the ↑words to tell us (.) Jen ny’ hat. 02 (1.1) 03 Kar: hat is a little wo:rd. 04 © ECT: that’s true it is a little word.

234 4.5.

Amelia Church

Expansion

Where the teacher does expand the knowledge expressed in children’s selfselected turns, we start to see a range of possibilities or opportunities for learning. In Example 20, the teacher encourages Miranda to go ahead with her stated intention to spell “Max” (line 5). Moreover, in the subsequent turn, the teacher scaffolds this attempt by drawing specific attention to the onset of the word (“what does it start with”). This expansion allows another child to respond to the question and identify the letter “M” (line 9). (20) Cathy.CI.10 01 ECT: and he sailed off through the night and day. 02 (0.7) 03 Mir: I’m gonna (.) I’m gonna spe:ll Max. 04 (0.4) 05 © ECT: alright off you go. 06 Mir: it’s easy (.) I’m gonna (do it later). 07 © ECT: okay fair enough. (0.4) ↑what does it start with. 08 ( ): u[m:.] 09 Tom: [m:.]

Responding to children’s self-selected questions also provides opportunities for semantic development. In Example 21, the teacher provides the pragmatic context to clarify the meaning of “marmalade”. (21) Karen.CI.3 01 ECT: because he liked:? 02 (1.2) 03 Jak: marma[lade.] 04 ECT: [marma]lade is what he liked¿ 05 (0.4) 06 Jak: what’s [marma]lade actually fo:r. 07 ECT: [(and)]08 © ECT: it’s like a- (0.3) well you put it on toast just like 09 hone:y? (0.4) and vegemite? what else do you like on 10 your toast.

In Example 22, the teacher picks up on an alternate word provided by Tim (the text of the book reads “looking”) and immediately provides positive feedback for this synonym (“staring, that’s a good word, instead of ‘looking’ we could change it to ‘staring”’). Later in the same activity, the teacher encourages the children to think of a variety of synonyms for the word “looking” which is productive in generating a number of words, including “glancing”.

Opportunities for learning during storybook reading

235

(22) Sophie.CI.5 01 ECT: i: see a:: ? 02 Tim: blue: [horse]. 03 Ell: [horse][[]]. 04 ECT: [[]]]. 05 Tim: [sta:ring at m]e::=. 06 © ECT: =sta:ring that’s a good word instead of looking we 07 could change it to staring. 08 (0.7) 09 ECT: blue horse blue horse what do you see. (0.2) i: see a::?

In Example 23, one child’s challenge to the notion that time cannot be stopped (the story is about a child who goes in search of time because her parents do not seem to have enough of it) prompts an extended sequence which explores the concept of a time. The teacher encourages this side sequence with a series of questions following Max’s self-selected turn (line 3). All children participate actively in this discussion, and this sequence develops understanding of temporal arbitrariness and social conventions of time, not to mention the humor marked by Max (line 32) about clocking off early from work. (23) Jodie.CI.3 01 ECT: 02 03 Max: 04 05 06 07 ECT: 08 09 10 ECT: 11 12 13 Max: 14 15 Ali: 16 17 ECT: 18 19 20 Bel: 21 22 ECT: 23 24 25 Max:

"no one (.) can stop time you know." (2.2) people can stop the time. (0.3) they can just- (0.5) turn the clock and hold it where it is (.) and then they- (0.2) (yeah) (.) people allow them to have time? (0.5) (well). maybe we could ↑try that one day. (1.4) and see what happens. (0.6) what will happen to the rest of (0.6) all the other children if we put the clock back. (0.8) um (.) be too much time (.) no one could [be-] [ ºxxx º ] [and](.) [and we’ll] be muddled up and we’ll have dinner at- (1.6)<↑breakfast ti:me.>= =we could¿ what do you think Bel? (0.3) if we turn my clock back for one hour? (0.7) yeah? (0.9) and we keep thinking (0.6) we’re an hour either behind the ti:me?(0.5) what would happen to the rest of everything else(.) though. they will have breakfast.

236

Amelia Church 26 27 28 29 30 31 32 33 34

ECT: hmmm $ it could be a mixed up day$. Tim: i know [ and the clock would be m]ixed up. Max: [it would be night time again.] ECT: it might be mixed up, what will happen to the traffic (0.5) people (0.4) and businesses. (1.2) Max: ehuhuh ·hhh they (xx) will sta:rt going ho:me. ECT: $they might start going home when they should be at work maybe Max.$ .hhh?

5. Discussion The observations of shared storybook reading in ten kindergartens recorded in this study have yielded an illustrative dataset of the type of talk that goes on in these interactions. While there are any number of analytic paths in this qualitative data, the focus here remains on the instances of child-initiated turns in the interactions, and the implications for (literacy) learning in how these turns are produced and responded to. Having outlined the range of responses made by teachers, the discussion moves to address three distinct noticings about this talk. Firstly, the manner in which children self-select as next speaker provides evidence of four-year-old children’s communicative competence. Yet these (appropriate) turns are not responded to as typical first pair parts in conversation, marking these sequences as examples of institutional interaction. The fact that the teacher determines whose turns get taken up is the second main finding of this work. The third point made below considers the implications for teaching and learning in responding to, and – more importantly – extending child-initiated talk. 5.1.

Children know the rules

One of the striking features of these extracts is the competence with which these four-year-old children attend to the rules of turn-taking in conversation. Children take the opportunity to self-select as next speaker at transition relevant places (see Sacks et al. 1974). In other words, the children contribute to the talk-ininteraction when the other speaker (i.e. the teacher) has finished and vary rarely in overlap. Children are – for the most part – not interrupting. Children are using the clues of intonational contours (rising or falling at the end of an utterance), syntax (a complete utterance), pauses and nonverbal cues (e.g. eye gaze) to anticipate the completion of the turn.

Opportunities for learning during storybook reading

237

For example, in the extract below, Eric’s turn is produced in overlap, having reasonably anticipated a turn completion (given the complete syntax, falling intonational contour and pause in the middle of the teacher’s prior turn). Eric immediately repairs this overlap by abandoning his turn (line 4) and restarting the question (line 5). (5) Louise.CI.4 01 Xav: he’s saying "c c c c ". 02 ECT: he’s go- storming o:ff, those are bubbles they’re not 03 c’s.(0.5) he’s [storming off] with a grumpy look. [why there-] 04 © Eri: 05 © Eri: why there are bubbles. 06 ECT: would you could you in a ca: r (.) eat them eat them 07 here they are.

Where children self-selected as next speaker they were invariably attuned to turn transition relevant places, always seeking to produce a first pair part where sequentially appropriate in the talk. Furthermore, children’s turns were typically topic-tied, and explicitly related to the content of the narrative. (3) Jenny.CI.4 01 ECT: 02 03 04 © Bil:

it’s just the noise the shell makes when you put it against your ea: r. (1.0) the sea: came out of the shell .hhh (they are) x x.

(12) Melanie.CI.4 01 ECT: "I’m drawing a: haterpillar (.) um: I mean a 02 cat erpillar" answered Jenny. 03 © Gen: hehhe (0.2) $it’s a haterpillar because it has a hat 04 on$. (16) Michaela.CI.3 01 ECT: that night (0.3) he had a sto mach ache. 02 ( ): [x x ] 03 © And: [from ea]ting all of that.

Even where the children’s comments extended beyond the immediate relevance of the text, they maintained the topic by reporting experiences similar to those of the characters in the story: (4) Jenny.CI.5 01 ECT: I’m going to put the sa:n d into a ja:r . 02 (0.2) 03 © ( ): I’ve got some sand at home.

238

Amelia Church

(8) Jenny.CI.6 04 © Bil: 05

[a:nd when I:] go to the beach I: do somersaults and it goes (like that).

These personal narratives produced by the children were not treated as relevant to the ongoing interaction, because they were not taken up by the teacher. Even the more compelling question-answer sequence does not necessarily facilitate a reply. In Example 5 above, Eric’s self-selected turn is framed as a question to solicit a teacher response (the question doing the soliciting by virtue of the demands of this first pair part of an adjacency pair anticipating the second pair part in the form of an answer; see Schegloff 2007). But the question in this context is ignored. The point being made here is that despite the children’s adherence to the rules of turn-taking, the interaction is shown here to be something other than typical conversation, as the children do not hold typical speaker rights in classroom interactions. There has been some criticism of the characterization of institutional interaction (e.g. Hester and Francis 2001), but the talk here is not defined as institutional because it occurs in a classroom between a teacher and a child, but rather because the children do not have equal opportunity in the talk-in-interaction. Indeed, the analysis presented here “takes us beyond conversational fragments to local negotiations of identity” (Austin et al. 2001: 183, editorial comment). It has been argued elsewhere that teacher roles are typically – and not necessarily helpfully – characterized as static (e.g. Richards 2006). The findings here do not claim that the identities of teachers and learners are necessarily fixed, but that in these reading events, the teacher and the children hold markedly different conversational status. The competence of children’s turn-taking does not afford the opportunity for children to successfully introduce topics, because the possibilities for the focus or topic of the talk-in-interaction are wholly determined by the teacher. 5.2. Teacher determines relevance Although the children make timely and topic-tied contributions to the ongoing activity, they are not in a position to determine the trajectory of the sequence – the teacher determines what talk is taken up, including the text of the story. In other words, despite following the rules of talk-in-interaction and making contingent contributions, the relevance of children’s contributions to the ongoing talk is not negotiated but rather directed by the teacher.

Opportunities for learning during storybook reading

239

The opportunity for children to self-select as next speaker is influenced by the structure of the sequence as a whole. The frequency with which the teacher comments on the text or seeks to engage the children with question about the text or their experiences or reactions to the story is related to the frequency of childinitiated utterances. Where the teacher is ‘straight reading’, typically speaker rights are not a negotiated space, in that children produce fewer self-selected turns. Regardless, the manner in which the teacher responds to children’s turns clearly marks the purpose of the talk-in-interaction. In other words, when we are doing reading, we focus on the story, both in terms of narrative structure and attend to the print (or pictures in some cases). We should be wary of making universal claims as to what ‘type’ of turns get taken up or extended by the teacher, as extrinsic factors (e.g. time available for the activity) are likely to influence opportunities to extend children’s contributions to the talk. Similarly, the book selection itself allows for particular focus on code or meaning, influencing the topic of talk about the text. The children’s experience, however, should be considered in the local context of learning what proves successful in attention-getting or at least maintaining rights as speaker. Typically, however, these contributions need be invited by the teacher in order to be ratified Methodological constraints of the observations reported here may also influence the frequency with or extent to which the teacher might develop the topic inititated by the children. Examples of postponing the children’s talk (Examples 9–11 above) were heard during the reading of a lengthy Dr Suess book. In this instance, all of the other children in the class were outside, the children participating in the storybook activity were keen to be outside, and no doubt the teacher felt the influence of time constraints for this activity. Furthermore, the text of the story itself, and the manner in which it is read may allow greater space for children to initiate topics. The point is not to criticize where children’s attempts to introduce topics are overlooked. Indeed, it is not feasible to extend every utterance in a classroom context. Of greater interest is where these turns do get taken up, how it is that the teacher responds and what this might mean for children’s learning. Overwhelmingly, in these extracts the teacher is orienting to the task or pedagogic purpose of the interaction as a whole (e.g. focusing on rhyme in a reading of a Dr Suess book) rather than children’s individual contributions to the talkin-interaction. This claim is made on the case-by-case evidence provided above of the teacher’s second pair parts to first turns initiated by children providing minimal acknowledgement but not a take-up of topic. Examples of ‘shared sustained thinking’ in early learning environments were found to be similarly rare by Siraj-Blatchford and Manni (2008). Markee (2008: 408) notes that that “an-

240

Amelia Church

alysts must demonstrate how particular behaviors actually promote language learning”; that data here demonstrate particular behaviors that promote the opportunity for learning, namely where the teacher takes up and extends the topic initiated by a child. 5.3. Turns (not) taken up as pedagogic resource Importantly, the analysis presented here is not designed as a critique of teacher practice; that is, the ‘not taking up of a turn’ does not stand for ineffectual practice. In emphasising the possible extension of children’s literacy learning by taking up children’s self-selected turns, we are perhaps overstating the feasibility of this extension during storybook reading. Only one person can talk at once, and not every speaker gains the right to topic simply by producing a first pair part. The findings presented here are not a critique of practice, but rather a vehicle to understanding the role that teacher responses play in creating learning opportunities and how these might be looked for in interactions in preschool settings. What matters, is what is done with those turns that are extended, and being aware that these choices have implications for what children learn is valued in a classroom context. Baker (1991: 180) used the phrase “subtleties of enthusiasm” to refer to critical readings of literature, but the notion applies nicely here to the conditioning of responses to offerings made by the children. While utterances may be acknowledged, it is those turns initiated by children which orient to the teacher’s pedagogic script and goal (i.e. what the teacher is setting out to achieve in the activity) that are likely to be met with upgraded enthusiasm. These findings underscore the importance of the sequential organization of classroom talk. It is not simply a matter of creating opportunities for children to initiate talk (cf. Dickinson and Tabors 2001), it is what happens after the child has initiated talk that creates opportunities for learning. The premise here is that if we are seeking to pursue child-centered learning, we should look to opportunities to extend children’s conceptual knowledge in the very context where they are demonstrating existing knowledge. Rather than teacher-elicited turns (i.e. children answering questions), child-initiated turns demonstrate the child’s existing knowledge and present attention. As seen in the examples above, these (teachable) moments provide opportunities for vocabulary development, phonological awareness, concepts about print and world knowledge, explicitly maximizing the benefits of shared storybook reading (see Hindman et al. 2008). It has been claimed elsewhere that preschool children seldom make comments about print (Ezell and Justice 2000, cited in Justice et al. 2009). Yet where children do articulate this knowledge about print, this is not necessarily built on as evidence of the very understanding teachers are seeking to develop in

Opportunities for learning during storybook reading

241

storybook reading activities. In the data presented here we see that children are orienting to concepts about print, but not necessarily the ones pursued by the teacher as the pedagogic goal. Although the teachers do acknowledge children’s self-selected turns in storybook reading events, this is typically done as a simple adjacency pair. Rarely is the child’s contribution to the talk-in-interaction, taken up as an opportunity to extend the literacy knowledge on display. Even where the teacher is not explicitly directing attention to a feature of the text, narrative or extension to children’s lives, this work is similarly done when the teacher responds only to those contributions made by the children that are relevant to the pedagogic script, relevant to the goal the teacher is pursuing. The flexibility to extend this script allows children to participate in the active construction of their learning.

6. Conclusion Although we might assume that small group interactions with teachers provide opportunities for sustained interaction and greater participation (Wasik 2008), any learning context which involves the teacher is inevitably driven by the teacher’s pedagogic intent or goal. In some cases this might indeed be extending topics introduced by children. The point of interest in early childhood education is where the teacher and the child are not orienting to the same goal, and to consider the implications of this misalignment for children’s learning. Essentially, the findings presented here demonstrate that although preschoolaged children attend to turn-taking rules in conversation and make contributions to the ongoing talk in appropriate ways at appropriate times, they do not hold sufficient status to self-select as next speaker, as the teacher determines what counts as relevant to the ongoing talk. Although the analysis here has in effect privileged the teacher’s voice in the interaction, this in itself mirrors the differential status of the speakers in classroom interactions in early childhood. The child’s role in the co-construction of opportunities for learning remains of interest, and should be taken up in other contexts within early learning environments. In exploring the notion of child-centered learning in early childhood classrooms, the data point to the difficulty faced by children in getting to the center in storybook reading activities. However, the size of the dataset in the present study precludes commentary on preschool teacher practice in general. We must acknowledge the context of these short exchanges – it does not follow that children are not more successful at introducing topics in other activities or when the membership group changes – e.g. dyadic interactions between teacher and child where there is perhaps greater space (time) to follow the child’s lead. Ev-

242

Amelia Church

idently the small dataset preclude observations about how and where children’s self-initiated contributions to classroom discussion are made relevant in other contexts, but suggests that future research would do well to go looking for it. The data presented in this paper point to the overriding influence of teacher authority on what constitutes relevance in storybook reading in preschool settings. This work points to the need to further detail how this authority is enacted – beyond responses to child-initiated utterances – and what the implications for learning might be. Even though opportunities for shared book reading are most significant at home, early childhood education and care provides a context for all children to have access to high quality reading interactions. Continuing research into what this quality is and how this quality might be done, will continue to improve our practice in supporting children’s early literacy development and facilitating opportunities for learning. Appendix: Transcription conventions (see Sacks et al. 1974) . , ? ¿ ! [] = (0.7) (.) : CAPS º º ↑↓ >< <> .hhh $ () (( )) x text ©

Falling terminal contour Continuing contour (incomplete) Strongly rising terminal contour Rising terminal contour Emphatic/animated utterance terminator Abrupt halt Overlapping speech Latching (contiguous stretches of talk) Pause measured in tenths of a second Pause timed less than 0.2 seconds Stress on the word/syllable/sound Lengthening of previous sound Increase in volume Decrease in volume Significant rise or fall in intonation Faster than surrounding talk Slower than surrounding talk Audible inhalation Laughing while talking (smile talk) Uncertain words (best guess) Comments e.g. quality of speech or intended hearer Unintelligible speech Italics mark teacher reading text Feature of interest

Opportunities for learning during storybook reading

243

Notes 1. This study is part of a larger project (The Young Learners’ Project) funded by the Australian Scholarships Group and an Australian Research Council Linkage Grant (Project number LP0883437). Acknowledgements are due to the Chief Investigators, Field Rickards, Bridie Raban, Margaret Brown and Esther Care, colleagues in the teacher research team, Janet Scull, Andrea Nolan, Robert Brown and Jan Deans, and particularly to Louise Paatsch for her interest and commentary on this work. Thanks are also due to the reviewers for pointing out flaws in the first version of this paper; those remaining are my own. 2. Kindergarten programs are publicly funded for four-year-old children in Victoria, Australia. A kindergarten program is attended by children in the year prior to beginning compulsory primary schooling.

References Austin, H., P. Freebody & B. Dwyer. 2001. Methodological issues in analysing talk and text: The case of childhood in and for school. In A. McHoul & M. Rapley (eds.), How to analyse talk in institutional settings, 183–189. New York: Continuum. Baker, C. D. 1991. Literacy practices and social relations in classroom reading events. In C. D. Baker & A. Luke (eds.), Towards a critical sociology of reading pedagogy, 161–188. Amsterdam & Philadelphia: John Benjamins. Baker, C. D. 1997. Ethnomethodological studies of talk in educational settings. In B. Davies & D. Corson (eds.), Oral discourse and education. Volume 3: Encyclopedia of language and education, 43–52. Dordrecht: Kluwer Academic. Beach, W. A. 1995. Conversation analysis: “Okay” as a clue for understanding consequentiality. In S. J. Sigman (ed.), The consequentiality of communication, 121–161. Hillsdale, NJ: Lawrence Erlbaum. Brouwer, C. E. 2003. Word searches in NNS-NS interaction: Opportunities for language learning? The Modern Language Journal 87. 534–545. Bruner, J. 1986. Actual minds, possible worlds. Cambridge, MA: Harvard University Press. Connor, C. M., F. J. Morrison & L. Slominski. 2006. Preschool instruction and children’s emergent literacy growth. Journal of Educational Psychlogy 98. 665–689. Danby, S. J. & C. Davidson. 2007. Young children using language to negotiate their social worlds. In L. Makin, C. Jones Diaz & C. McLachlan (eds.), Literacies in childhood: Changing views, challenging practice, 118–132. Marrickville: Elsevier Australia. Dickinson, D. K. & P. O. Tabors (eds.). 2001. Beginning literacy with language: Young children learning at home and school. Baltimore, MD: Paul Chapman. Drew, P. & J. Heritage (eds.). 1992. Talk at work: Interaction in institutional settings. Cambridge: Cambridge University Press.

244

Amelia Church

Edwards, D.. 1993. But what do children really think? Discourse analysis and conceptual content in children’s talk. Cognition and Instruction 11. 207–225. Ezell, H. K. & L. M. Justice. 2000. Enhancing children’s print and word awareness through home-based parent intervention. American Journal of Speech-Language Pathology 9. 257–269. Firth, A. & J. Wagner. 1997. On discourse, communication, and (some) fundamental concepts in SLA research. The Modern Language Journal 81. 285–300. Gerde, H. K. & D. R. Powell. 2009. Teacher education, book-reading practices, and children’s language growth across one year of Head Start. Early Education and Development 20. 211–237. Heap, J. L. 1985. Discourse in the production of classroom knowledge: Reading lessons. Curriculum Inquiry 15. 245–279. Heritage, J. 2005. Conversation analysis and institutional talk. In K. L. Fitch & R. E. Sanders (eds.), Handbook of language and social interaction, 103–147. Mahwah, NJ: Erlbaum. Hester, S. & D. Francis. 2001. Is institutional talk a phenomenon? Reflections on ethnomethodology and applied conversation analysis. In A. McHoul & M. Rapley (eds.), How to analyse talk in institutional settings, 206–218. New York: Continuum. Hindman, A. H., C. M. Connor, A. M. Jewkes & F. J. Morrison. 2008. Untangling the effects of shared book reading. Early Childhood Research Quarterly 23. 330–350. Howes, C., M. Burchinal, R. C. Pianta, D. Bryant, D. Early & R. Clifford. 2008. Ready to learn? Children’s pre-academic achievement in pre-kindergarten programs. Early Childhood Research Quarterly 23. 27–50. Justice, L. M., J. Kaderavek, X. Fan, A. Sofka &A. Hunt. 2009. Accelerating preschoolers’ early literacy development through classroom-based teacher-child storybook reading and explicit print referencing. Language, Speech and Hearing Services in Schools 40. 67–85. Justice, L. M., A. J. Mashburn, B. K. Hamre & R. C. Pianta. 2008. Quality of language and literacy instruction in preschool classrooms serving at-risk pupils. Early Childhood Research Quarterly 23. 51–68. Lerner, G. 1995. Turn design and the organization of participation in instructional activities. Discourse Processes 19. 111–131. Lerner, G. 2003. Selecting the next speaker: The context-sensitive operation of a context-free organization. Language in Society 32. 117–201. Levinson, S. 2006. Cognition at the heart of human interaction. Discourse Studies 8. 85–93. Liddicoat, A. J. 2007. An introduction to conversation analysis. London: Continuum. Lonigan, C. J. & G. J. Whitehurst. 1998. Relative efficacy of parent and teacher involvement in a shared-reading intervention for preschool children from low-income backgrounds. Early Childhood Research Quarterly 13. 263–290. Macbeth, D. 2000. Classrooms as installations. In S. Hester & D. Francis (eds.), Local educational order, 21–72. Amsterdam: John Benjamins.

Opportunities for learning during storybook reading

245

Markee, N. 2008. Toward a learning behavior tracking methodology for CA-for-SLA. Applied Linguistics 29. 404–427. Markee, N. & G. Kasper. 2004. Classroom talks: An introduction. The Modern Language Journal 88. 491–500. Markee, N. & Mi-Suk Seo. 2009. Learning talk analysis. International Review of Applied Linguistics 47. 37–63. McHoul, A. 1978. The organization of turns at formal talk in the classroom. Language in Society 7. 183–213. McHoul, A. 1990. The organization of repair in classroom talk. Language in Society 19. 349–377. Mehan, H. 1979. Learning lessons: Social organization in the classroom. Cambridge, MA: Harvard University Press. Mori, J., & J. Zuengler. 2008. Conversation analysis and talk-in-interaction in classrooms. In M. Martin-Jones, A. M. de Mejia & N. H. Hornberger (eds.), Encyclopedia of language and education, 2nd edn., Volume 3: Discourse and education, 15–26. New York: Springer. Raban, B. & H. Coates. 2004. Literacy in the early years: A follow-up study. Journal of Research in Reading 27. 15–29. Richards, K. & P. Seedhouse (eds.). 2005. Applying Conversation Analysis. Basingstoke: Palgrave Macmillan. Richards, K. 2006. ‘Being the teacher’: Identity and classroom conversation. Applied Linguistics 27. 51–77. Sacks, H., E. Schegloff & G. Jefferson. 1974. A simplest systematics for the organization of turn-taking for conversation. Language 50. 696–735. Schegloff, E. A. 2007. Sequence organization in interaction: A primer in conversation analysis I. Cambridge: Cambridge University Press. Seedhouse, P. 2007. On ethnomethodological CA and “Linguistic CA”: A reply to Hall. The Modern Language Journal 91. 526–532. Sinclair, J. & M. Coulthard. 1975. Towards an analysis of discourse. Oxford: Oxford University Press. Siraj-Blatchford, I. & L. Manni. 2008. “Would you like to tidy up now?” An analysis of adult questioning in the English Foundation Stage. Early Years 28(1). 5–22. te Molder, H. & J. Potter (eds.). 2004. Conversation and cognition. New York: Cambridge University Press. ten Have, P. 2001. Applied conversation analysis. In A. McHoul & M. Rapley (eds.), How to analyse talk in institutional settings, 3–11. New York: Continuum. ten Have, Paul. 2007. Doing conversation analysis: A practical guide, 2nd edn. London: Sage. Wasik, B. A. 2008. When fewer is more: Small groups in early childhood classrooms. Early Childhood Education Journal 35. 515–521. Wasik, B. A. & M. A. Bond. 2001. Beyond the pages of a book: Interactive book reading and language development in preschool classrooms. Journal of Education Psychology 93. 243–250.

246

Amelia Church

Wray, D., J. Medwell, R. Fox & L. Poulson. 2000. The teaching practices of effective teachers of literacy. Educational Review 52. 75–84. Amelia Church is a Lecturer in Early Childhood in the Melbourne Graduate School of Education at the University of Melbourne, and author of Preference Organisation and Peer Disputes: How Young Children Resolve Conflict in the series Directions in Ethnomethodology and Conversation Analysis (Ashgate, 2009). Her current research interests include developmental pragmatics, ethnomethodology in early childhood, and locally constructed learning through talk-in-interaction.

Negotiating language, literacy and identity: A sociocultural perspective on children’s learning strategies in a multilingual ESL classroom in Singapore* LAWRENCE JUN ZHANG

Abstract This article reports on part of a larger study investigating children’s language learning strategies (LLSs) for bringing to bear on the range and patterns of such strategies for enhancing children’s self-regulated engagement in biliteracy learning. Taking a case study approach, the study describes the process of the language learning activities in order to answer two research questions: (1) Do elementary schoolchildren use LLSs in meaningful interaction? (2) If they do, how do they negotiate language, literacy and identity in this process? A cloze passage task was set for 36 children to complete in groups of 5 each. Two methods were used in tandem in data analysis – data reduction and interpretation and connecting the inter-relationships and offering explanations. Results show that children use LLSs, as do adults, although the level of cognitive engagement differs; and that ‘successful’ and ‘less successful’ learners contribute to the meaningful interaction in different ways. Counting LLSs is not as significant as scrutinising learners’ effort for negotiating language, literacy and identity. I conclude that there is a need to nestle and reframe a cognitive view of language acquisition within a socially-imbedded system so that these commonly used constructs are not treated in isolation but in osmosis so that they are understood as “interactionally open and ecologically situated” (Canangarajah 2007: 936). Pedagogical implications of context-sensitive and culturally suitable strategiesbased instruction are also discussed. Keywords: bilingualism; bilingual learner strategies in classroom interaction; literacy; identity; sociocultural theory; Singapore.

248

Lawrence Jun Zhang

1. Introduction Language learning/learner strategy (LLS) research, since its initiation in the 1970s (Stern 1975; Wong Fillmore 1976; Naiman et al. 1978), has a tradition of examining second language (L2) learners’ patterns of strategy use in relation to language learning outcomes. The intention is to transfer the strategies that ‘successful’ language learners use to those who are ‘less successful’, in the hope that the latter will be assisted in their efforts for more effective language learning. This research has been revitalized these years, especially after the publication of Cohen and Macaro (2007a), where the authors propose a task-based approach to studying LLSs. Nevertheless, hitherto, the majority of the studies reported in the literature are about adult or young adult learners. Research on elementary schoolchildren or young learners is insufficient, if not scarce; how these learners process the linguistic input for acquiring L2 language skills through deployment of LLSs, if any, in interaction, remains unclear (cf. Chamot and El-Dinary 1999; Oliver 2002; Willet 1995; Wong Fillmore 1976). More importantly, how the process of language learning is related to their negotiation of literacy and identity is little explored (cf. Block 2007a; Norton 1995; Norton and Toohey 2001). It is especially so as regards elementary schoolchildren who are required to learn English as the main language subject in the school curriculum alongside their mother tongue within a broad context of governmentdesignated bilingual language policy, as is the case of Singapore (Alsagoff 2007; Pakir 2004). Worldwide, a few studies report on the strategies used by young learners, and these reports are mainly based on data from questionnaires, interviews, and think-aloud protocol analysis. This body of literature on young learners, although limited by the methodology, has offered interesting insights into how elementary schoolchildren use LLSs for acquiring an L2 (Chesterfield and Chesterfield 1985; Oliver 2002; O’Malley and Chamot 1990; Purdie and Oliver 1999; Zhang et al. 2008a). Given that social interaction is one of the many and perhaps most important ways in which children develop their language, literacy and identity (Norton 1997; Norton and Toohey 2001; Stroud and Wee 2007), and with a “social turn” (Block 2003) in second language research (e.g., the work of Norton 1995), this paper focuses on a group of Singaporean bilingual schoolchildren who were engaged in a cloze task that was supposed to trigger interaction. Necessarily, language, literacy and identity would be expected to be revealed in the process of completing the task that had been set for investigating the issues under discussion. Specifically, this study intends to bring to fruition what Cohen and Macaro (2007b) have recommended in LLS research, because a question that remains unanswered is what highly bilingual English learners do when they are engaged

Negotiating language, literacy and identity

249

in language learning activities that possibly invoke learning strategies. More importantly, in solving language learning tasks, what the participants do, as a way of their contribution to the completion of the learning task, deserves investigation. It is especially so when learning is regarded more as an interface of learners’ cognitive aspects and the sociocultural environment in which learning takes place (Lantolf and Thorne 2006; Vygotsky 1978). In particular, the study was motivated by the following two research questions: 1) Do elementary schoolchildren use learning strategies for meaningful interaction in language learning? 2) If they do, how do they negotiate language, literacy and identity in this process in relation to their English proficiency?

2. Defining language learning/learner strategies LLSs are usually defined as behaviors, actions, techniques, and thoughts learners engage themselves in for the purpose of understanding or using the target language (O’Malley and Chamot 1990; Oxford 1990; Rubin 1975). Although different scholars have proposed different taxonomies of LLSs in the seminal work in the field, in essence they are closely related. Wong Fillmore (1976) used two terms, social and cognitive, to be a point of departure for distinguishing the various LLSs that children used. Rubin (1975) did not present a clear classification system in researching mature learners, but Oxford (1990) used four terms to categorize all the possible LLSs: cognitive, memory, social, and metacognitive, with the former two directly contributing to language learning and the latter two indirectly playing their roles in helping the learner gain competence. O’Malley and Chamot (1990) used metacognitive, cognitive, and socialaffective as three main categories in their classification system. Not contradicting each other, the Oxford classification system and that of O’Malley and Chamot are the most widely adopted systems in the research community. Despite the highly over-generalized nature of these taxonomies, there is now general consensus that strategies are context-dependent and task-specific. Macaro (2006: 333) proposed that researchers adopt “a small strategy within a larger framework” conceptualization. So “strategies are conceived as occurring in working memory, are described through actions, goals, and situations at the lowest practical level [italic original], but then related to a broad framework of strategic plans, second language processes, and second language skills” (Cohen and Macaro 2007b: 279).

250

Lawrence Jun Zhang

In the mainstream SLA literature, interaction and negotiation are key terms that have gained prominence (Ellis 2008; Gass 2003; Long 1981). Research findings indicate that elementary L2 schoolchildren do and are able to negotiate for meaning in interaction (Oliver 2002). Cooperative interaction is beneficial because it provides learners with the opportunity to obtain comprehensible input that is uniquely modified for learners’ individual circumstances (Pica and Doughty 1985; Oliver 2002; Varonis and Gass 1985). Given the apparent importance of negotiation for meaning, a number of studies examined the conditions promoting such interaction (e.g., Pica et al. 1993), although they did not show any interest in literacy and identity in this negotiation process because of their declared disengagement with learners’ sociocultural aspects in acquiring a second language, as aptly pointed out by scholars such as Block (2003), Firth and Wagner (1997), and Zuengler and Miller (2006) in the field. Oxford (2007) argued that the possibility exists in SLA research, but, surprisingly, no research effort has been expended to explore how the gap between psychological and sociocultural perspectives can be bridged. Apparently, she has successfully shown that in LLS research much can be done to offer significantly meaningful insight into language learner behaviors and effective language teaching.

3. Children’s language learning/learner strategies For clarity and immediate relevance, my literature review focuses on elementary schoolchildren’s use of LLSs in ESL/EFL learning. As early as 1976, Wong Fillmore (1976) found that in learning a second language, social strategies were more important for children than cognitive strategies. O’Malley and Chamot (1990) found, however, that metacognitive strategies, comprising planning, monitoring and selective attention, among others, were the most important of all the three categories of strategies they had identified. Of the vast body of the LLS research literature, Chesterfield and Chesterfield .(1985), Sugeng (1997), Chamot and El-Dinary (1999), Purdie and Oliver (1999), and Lan and Oxford (2003) included elementary schoolchildren or younger learners. Chesterfield and Chesterfield (1985) managed to identify the LLSs of bilingual elementary schoolchildren and found differences between the successful and less successful learners. More recent studies all corroborate with the findings reported in the literature. In relation to the Singapore context, Zhang and his colleagues (Zhang et al. 2008a) presented the first of its kind that focused on identifying the LLSs used by Singaporean elementary schoolchildren based on analyzing the participants’ think-aloud protocols. Although their findings corroborate what has been reported in the literature, they found that expert learners were more flex-

Negotiating language, literacy and identity

251

ible in strategy choice and orchestration than their less successful peers when they were presented with reading texts, and their strategic repertoire was also richer. Chamot and El-Dinary (1999) used a combination of data collection techniques, including observation, interview and think-aloud; Zhang et al. (2008a) made use of the think-aloud technique; the others used questionnaires (Lan and Oxford 2003; Purdie and Oliver 1999), classroom observations (Sugeng 1997), or a combination of several data collection techniques. Findings from these studies all show that young language learners use learning strategies, but the frequency and variety of strategy use differ among learners of different proficiency levels. It is now agreed that the use of LLSs is positively correlated with second language learning success. These studies also point to the fact that good or expert language learners, who are most often high achievers in language learning, orchestrate and combine strategies for efficient language learning and it is not purely the quantity of the strategies they use that makes a difference; the qualitative differences in strategy choice and orchestration distinguish significantly the expert learners from novice learners (Macaro 2006; O’Malley and Chamot 1990; Oxford 1990). In reviewing 30 years of LLS research and practice, Cohen and Macaro (2007b: 283) recommend that in language learner strategy research “the taskbased approach should be encouraged as it appears to be a more fruitful trend than a more general-learning approach” in examining learners’ strategy use. They posit that: The goal of any language instruction program is not only to teach the L2 for the moment, but to instil within the learner a sense of what it is like to be a lifelong learner. We are increasingly living in a global world where communicative skills, in more than one language, can be not just an added bonus but rather a true necessity. Hence, having our language learners more skilled at using their strategy repertoire – whether at the elementary, secondary, tertiary, or graduate level – can ideally have both a short-term and a long-term impact. While the potential role that enhanced LLSs can have in this endeavor is great, it remains to be seen how effective the field is in harnessing that potential. (Cohen and Macaro 2007b: 284)

As Cohen and Macaro (2007a) rightly point out, the trajectory of LLS research has been mainly along a cognitive tradition, especially when different classification systems were developed (e.g., O’Malley and Chamot 1990; Oxford 1990), despite the fact that socio-affective strategies are also included in these systems (e.g., Oxford 2007). Nonetheless, studies on elementary schoolchildren’s use of strategies for negotiation of meaning, literacy and identity in interaction as a sociolinguistic phenomenon are still rather limited.

252

Lawrence Jun Zhang

4. Interaction, language, literacy and identity Li (2008) highlights that “sociolinguists see bilingualism and multilingualism as a socially constructed phenomenon and the bilingual and multilingual person is a social actor . . . language choice is not only an effective means of communication but also an act of identity” (2008: 13). Interaction is essential for language acquisition, and learners need opportunities for interaction not only for language development but also for socialization. Socialization takes different forms in terms of language choice. Scholars delineated the necessity of socialization in language acquisition and emphasized the intricate relationship between socialization and language acquisition (Block 2007b; Canagarajah 1995; Willet 1995). Canagarajah (1995) analyzed code-switching patterns in ESL in Sri Lanka. Results showed that code-switching served useful functions not only for classroom management but also for imparting the lesson content. He concluded that code-switching offered opportunities for students to negotiate values, identities and roles in the classroom, which in turn, would better prepare them for their sociolinguistic life outside the classroom. Young (2009) proposes that Practice Theory be better able to account for the practice of language learning and teaching because it views practice as taking language as essential and does not exclude factors such as the complexity of context. Practice Theory contrasts with traditional linguistic approaches to language structure and meaning which often treat language as context-free, occurring “nowhere, nowhen, and produced by nobody” (2009: 12). This is exactly because, as Young argues, social expectations are possible reflections of choice of language, and thus self-identity has to be explored. He posits that “all talk happens somewhere, at some time, and is produced somehow by somebody for some purpose, and the approach that practice theorists have taken is that talk and its context are inseparable” (2009: 49). Stroud and Wee (2007) highlight the differences in the onstage and offstage classroom interaction behaviors, thinkings, and positionings of Singaporean secondary school students when these students’ identities were revealed through the use of the variety of English with which they were familiar and comfortable and in which they were competent – Singapore colloquial English (SCE), usually expected to be used in daily informal communication in Singapore. SCE as a social and regional dialect of English is a basilectal form of the language, which deviates starkly from the standard norms of English used in the US or UK. Of course, the notion of ‘standard’ English can be equally debated in terms of what it really entails and what geopolitical connotations it carries. Stroud and Wee (2007) argue that instead of downplaying or even ignoring “the potential and very real negative ramifications of teaching in and through nonmainstream

Negotiating language, literacy and identity

253

languages” (2007: 50), the benefits of using nonstandard variety may not be fully understood. They maintain that: Attention to identity should encourage a considerable rethinking of multilingual classroom practices in the direction of seriously incorporating sociocultural and political awareness into them. But this rethinking, of course, raises problems of how to conceptually link social structuration, identity and language in ways that can be practically and feasibly implemented in the classroom . . . there is no doubt that English language teaching needs to pay much more attention to developing pedagogical strategies that take into account students’ identity work in learning . . . . (Stroud and Wee 2007: 50–51).

Block (2007a) defines identity clearly with reference to various aspects involving the individual’s exercising of agency. It is apparent that addressing identity in relation to the participant’s use of language alone tends to lead to the reseracher’s holding an essentilist understanding of identity. He summarizes that “performativity and presentation of self”, “positioning”, “ambivalence”, “hybridity”, “communities of practice”, and “power and recognition” (2007a: 11–45) need to be taken into consideration. Canagarajah’s (2006) proposition that identity in this post-modernism era is imbued with fluidity (i.e., identity changes over time) is an equally useful conceptualization of identity, because multidimensionality of one’s identity is common in communities of practice. The work that Norton and Toohey (2001) have done in Canada is directly relevant to the work reported here. Although good language learner (GLL) research has been conducted since the 1970s and cognitively-oriented SLA researchers have taken it for granted, Norton (1995), and Norton and Toohey (2001) challenge the notion of GLLs by taking a sociocultural approach to examining the issue (Vygotsky 1978; see also Lantolf and Thorne 2006). They argue forcefully that GLLs do not work in isolation. Instead, the proficiencies of the GLLs in their studies were not only tied up with their behavior of learning as individuals but also what their communities would offer them. They have therefore concluded that “understanding good language learning requires attention to social practices in contexts in which individuals learn L2s” (Lantolf and Thorne 2006: 318). The importance of learners’ exercising of agency needs to be considered because the very process of exercising agency comes with their forming and reforming of identities.

254

Lawrence Jun Zhang

5. Multilingual/multicultural Singapore Located on the southern tip of the Malaya Peninsula, Singapore is a nation state of 4.8 million population comprising mainly three ethnic groups – Chinese, Malay and Indian, according to a recent population census. Since its independence in 1958, Singapore has continued to use English as its major language of communication in the media, law, education, and government. In essence, English functions as a lingua franca for all the people despite the fact that English, Chinese, Malay and Tamil are designated as the four official languages. Due to its multilingual and multicultural features, the Government of Singapore had stipulated a compulsory bilingual education policy for all schoolchildren from elementary school onwards until junior college (senior high school). What this means is that all children starting from elementary schools are required to learn two languages as required curriculum subjects – English and a Mother Tongue. All children have to study all the subjects in English except the Mother Tongue subject and its related Good Citizen (moral education) course (Pakir 2004). Depending on whether the child comes from a Chinese family, Indian family, Malay family, the Mother Tongue subject varies according to the ethnic group to which the learner belongs. All the elementary schools follow the same national curriculum although some schools offer more core-curriculum activities (CCA) such as modern dance, sports, among others, than others and are traditionally known as better performers in academic subjects as examined through the Primary School Leaving Examination (PSLE) at the end of the 6year program. Used in a New English context, the English that serves a wider array of purposes in society at large for daily communication deviates from that of the US, UK, or any other “Inner Circle” countries (Kachru 1992). 6. Method 6.1.

Participants

Thirty-six schoolchildren from an ordinary neighborhood or community elementary school in Singapore were invited to participate in this study. All the pupils were in grade four, with ages ranging from 8 to 9 years, who had undergone kindergarten education where the two languages, English and a Mother Tongue, were taught informally. Upon their entry into elementary schools, English and the Mother Tongue were officially offered as compulsory curriculum subjects that were examined according to specific national syllabi. So the participants’ English proficiency varied in speaking, listening, reading, writing and vocabulary. Information on participant distribution is given in Table 1.

Negotiating language, literacy and identity

255

Table 1. Participant information in the study (n = 36) Participants Malay Chinese Indian Sub-total

Proficiency High achievers Low achievers 2 3 17 11 1 2 20 16

Total 5 28 3 36

The class from which these pupils were selected was chosen by the Head of the English Department of the school in consultation with their English teacher, who was also the form teacher. This particular class was chosen because its overall English scores were mediocre. The department head was asked to select one pupil each from the top and the bottom groups according to their latest English examination results based on the various assessment components. When there were no exam results, teachers’continuous assessment of pupil performance was taken as the criterion. The department head was also asked to select participants according to each pupil’s representativeness. 6.2. Task material A cloze passage of 191 words (see Appendix) was taken from an English textbook, Celebrate English (Unit 4, Book 4A), a textbook series that has been widely used in elementary schools in Singapore. The textbook series was written and produced by writers and educators based in Singapore in accordance to the English Language Syllabus (MOE 2001), where all the texts are presented according to the text types to which the texts belong. Unit 4 typically presents information texts and narratives on a common theme “Rivers to the Sea”. The cloze passage used in this study asked all the participants to fill in the blanks in groups of 6 children each. For the purpose of this study, the material was adapted slightly to remove all the potentially confusing blanks to make sure that the task would be clear to the participants. The underlying principle for using this cloze passage was to see if these children were to use these connectives such as “if”, “because”, “as”, “since”, “as a result of”, “but” , “however” and “either . . . or” typically found in these two texts, especially if they would ever engage in interactions for meaning-oriented negotiations. These connectives are usually regarded as being able to indicate the students’ academic literacy performativitity as and when the formal linguistic genre is required.

256 6.3.

Lawrence Jun Zhang

Data collection

Following Cohen and Macaro’s (2007b) recommendation for taking a task-based approach to LLS research, I collected data as part of a larger study intended to investigate young learners’ LLSs for bringing to bear on the range and patterns of such strategies that children initiated in an elementary school classroom, with an ultimate aim of developing these children into self-regulated learners (Zhang et al. 2008b). I solicited help from the classroom teacher. I gave a list of instructional procedures for her to assign group work after I talked with her about how to naturally observe her pupils working together to solve problems in language learning. All the children were assigned to different groups with a good mixture of learners of different ethnic backgrounds to make sure that they used English in negotiation or discussion. A typical group of six pupils consisted of three ‘successful’or high achievers and three ‘unsuccessful’learners or low achievers. Although there was no specific requirement for any seating arrangement, there was a specification for the number of children to be placed in each group. The participants were audio and video recorded. The audio-recordings of the task that was completed in one 30-minute session in a normal classroom environment were transcribed and verified against the video-recordings. The transcripts comprised mainly the conversations among the group members on the task at hand. Many times the transcripts were unedited representations of the local nativized and institutionalized variety of English in spoken exchanges recorded in conventional English orthography, and included all speech signals (e.g., hesitations, repetitions and insertions of utterance-final particles such as lah, lor, meh, hor and ah as used daily in verbal communication). Discrepancies were checked and confirmed by a colleague by viewing the videorecordings. The inter-rater reliability was 93%, a level that was considered highly acceptable. In order to follow the principle of anonymity, all the participants that are in the focus groups were assigned a code such as S1, S2, S3, and so on.

6.4.

Data analysis

Adopting a case study approach to providing “holistic” and “context-sensitive” descriptions of the learning activities (Patton 2001; Zhu and David 2008), I took field notes by keeping a distance from the children in order to describe “objectively” the process of the language learning activities which could invoke the use of learning strategies among 36 young learners and the process of how their Singaporean identity was performed in completing the learning task. Van Lier (2005: 205) explains that “cases are specific persons, places, or events that are interesting and worthy intensive study. The case is a real-life entry that

Negotiating language, literacy and identity

257

operates in a specific time and place.” So, I also resorted to video recordings as a way of illustrating typical patterns of schoolchildren’s strategies as well as their differences in strategy use in that particular location during that time. Two methods were used in tandem in data analysis: data reduction/interpretation and connecting the inter-relationships/offering explanations. The conversations among the members in different groups focusing on the same learning task were analyzed to look for typical patterns of strategy use in negotiation for meaning. The O’Malley and Chamot frame, which classifies LLSs into three broad categories (metacognitive, cognitive, and socio-affective) was used as the main framework in data deduction, and the negotiation strategies as described in Long (1981) and Pica and Doughty (1985), as adopted in Oliver (2002), were used as additional resources for categorizing the negotiation strategies when they surfaced from the data. I was aware that the strategies listed in Pica and Doughty (1985) and Oliver (2002) overlap with the strategies in the O’Malley and Chamot’s tripartite classification system, so they were re-examined according to the O’Malley and Chamot (1990) framework.

7. Findings and discussion To better illustrate the typical features for answering the two research questions, I focus on individual schoolchildren to discuss the differences between the ‘successful’ and the ‘less successful’ learners. I also intend to show how the ‘successful’ and the ‘less successful’ learners contributed to the progression of the language learning task for the purpose of improving their linguistic proficiency and competence in negotiating meaning in social interaction. I was clearly aware that the use of words such as ‘successful’ and ‘less successful’ could be loaded with other connotations, but for convenience, they were used to refer to two categories of learners, ‘high achievers’ and ‘low achievers’ in schools. 7.1.

Strategies for negotiating language, literacy and identity

The negotiation strategies as observed being used by the elementary schoolchildren were a mixture of metacognitive, cognitive and socio-affective strategies. Despite a difficulty in clearly distinguishing metacognitive from cognitive strategies, confirmation checks and comprehension checks are two clear metacognitive strategies. Clarification requests are made by the interlocutor to clarify what is said and include statements such as “what do you mean?”, “what’s that?”, “I don’t understand”, “is this what you said?” The use of such strategies also helps

258

Lawrence Jun Zhang

the participants to maintain group dynamics in the process of their getting feedback or information. Confirmation checks are statements by the interlocutor for making sure that the communication is correct. They may include repetitions accompanied by rising intonation, e.g., “it should be used, right?” Comprehension checks are made by the interlocutor to check that the preceding utterance has been correctly understood by the listener. So they are basically for monitoring comprehension. They usually comprise questions, either tag questions, repetition with rising intonation, or questions such as “[Do you] understand or not?”, or “can [you] read or not?” Cognitive strategies such as self-repetitions, re-reading, contextualization, inferencing and grouping help the learners in utilizing the cognitive resources available, so in many cases both high achievers and low achievers use them despite differences in quantity and quality. This is because, for example, selfrepetitions are the speaker’s partial, exact, and expanded repetitions of lexical items from his or her own preceding utterances within the speaking turns (see Oliver 2002 for more information). Socio-affective strategies include cooperating with one or more peers to obtain feedback, pool information, or model a language activity and questioning for clarification by asking a teacher to repeat, paraphrase, explain or offer examples. Requesting for clarification (clarification requests) often operates on the participants’ understanding of why there is such a necessity. So it is not only socio-affective but also metacognitive. We turn to S6, Tom, and then to S2, Jimmy, for a quick snapshot analysis of the strategies they used for completing the language learning task so that a comparison can be made between the ‘expert’ or ‘successful’ and the ‘novice’ or ‘less successful’ schoolchildren in terms of how they negotiated language and literacy, and how their identities were naturally brought out.

7.2. 7.2.1.

Differences between high and low achievers High achievers

Due to space, I present only one excerpt below and interpret the data within the frameworks explained earlier. Excerpt 1 typically presents a series of negotiation moves in the class activity with the cloze task as the organizing trigger. S6, as a high achiever, was proud of being asked to participate in this group activity. In fact, due to his high level of English proficiency that is usually associated with ‘successful’ learners, he was able to talk about things that interested him. The cloze task intended to solicit learners’ use of strategies, particularly global strategies for maintaining discourse continuity. Those who were categorized

Negotiating language, literacy and identity

259

into ‘high achievers’, namely, S4, S5, and S6, were able not only to initiate the conversation in negotiating meaning but they were right in most of the cases. They used a wide range of strategies for negotiating language for getting the meaning across. They predicted the content of the text, as shown in dyad 1 by S6 “the text is about clean rivers in Singapore”. “Re-reading” (dyad 6) and “contextualization” (dyad 8), “inferencing” (dyad 9), and “confirmation checks” (dyad 9) are other strategies that are easily identifiable. The orchestration of these strategies by these learners apparently suggests that they were more proficient in approaching a written text through the deployment of strategies although the language they used is SCE. It is noticeable that the high achievers were treated as such due to their higher levels of English proficiency as measured by the standard tests for assessing the candidates’ academic proficiency. This might be the reason why they stood out markedly as high achievers. Nonetheless, due to the multilingual and multicultural environment in which they lived and studied, even the high achievers frequently used features of SCE. Even English teachers spoke with such features in the classroom to enhance the ‘chemistry’ between them and their students sometimes despite the Ministry of Education’s (MOE 2001: 3) mandate that the English taught in Singapore should be “internationally acceptable English that is grammatical, fluent and appropriate for purpose, audience, context and culture”, and by definition, “internationally acceptable English refers to the formal register of English used in different parts of the world, that is, standard English” (ibid.). Excerpt 1 Children S6

S2 S5 S2 S3 S6 S1 S5 S4

This text is about clean rivers in Singapore. Hmm. . . Is it about Singapore River? I cannot see. I need to read. Eh, how about you (S2)? I think so lah. The text look like an information report. It must be. It mentioned rivers in Singapore. Let me read it again. Agree. We must try to fill in words first. Ok. I don’t know leh. Which word did you put in? I want to put in “if” for this blank, cos “fish and other living things will die”. “If” should be used, right?” Agree or not? Yah hor, read well with “if”. So what about next one? Can we fill in “but”? The earlier sentence says something different . . . Yes, but what about the coma there. I don’t see “but” used like this ah. Serious. It just doesn’t read right leh. Do you understand what I am saying or not? Hmm. . .

Dyad 1

2 3 4 5 6 7 8 9

260

Lawrence Jun Zhang

Excerpt 1 Children S5 All S6 S5 S3 S2

Can we put in “however”? Let us try it, ok? Can. Yes, this is it. Do you agree? Me too. .. Yes, agree. Yah loh. You meh (“you”, a Chinese word, meaning “have” or “there is a . . . ” and meh is a particle forming an interrogative sentence, so the meaning of this utterance is “is there such an answer?”)? Oh yah hor. I didn’t get it first, but now it’s clear... “however” is good, is it? Ok lah. . .

Dyad 10 11 12 13 14 15

As the English Language Syllabus (MOE 2001: 4) focuses on language use, learning outcomes, text types and grammar, students are expected to understand “how to communicate fluently, appropriately and effectively in internationally acceptable English. They need to understand how the language system works and how language conventions can vary according to purpose, audience, context and culture, and apply this knowledge in speech and writing in both formal and informal situations.” The way S6 and his other peers used SCE features in their discussions could be understood to indicate their intuitive knowledge about the difference between the formal and the informal registers. This explicit and implicit display of their literacy skills specifically represented by the use of particular literacy genres or text types does not avail them of any chance of giving up the use of SCE features. This alignment with a Singaporean identity says it all. 7.2.2.

Low achievers

As a low achiever in the group, S2, Jimmy, appeared lacking confidence at the onset. However, as is evident, this psychological constraint did not restrict his willingness to share. In fact, the very move he made to approach the text by asking his group members to look at the text type of it indicates that he was actually quite on task. As he said it, “the text looks like an information report” (dyad 2). Although it is now commonplace that all elementary schoolchildren had been well-exposed to a variety of text types, including narratives, recounts, information reports, explanations and expositions (arguments, discussions), by virtue of the English Language Syllabus (MOE 2001) explicating these terms, the fact that a low achiever like S2 mentioned it in approaching a written text deserves some attention. Normally, teachers’ expectation is that only good learners are able to regurgitate and apply the key concepts delivered in the language class-

Negotiating language, literacy and identity

261

room as per curriculum cycle, but ‘less successful’ learners such as S2 were also able to mention and apply the text type knowledge. Evidently, teachers’ pedagogical approach and the textbooks used with this group of pupils had a clear bearing, and they should. The question to be asked naturally is “why didn’t the ‘successful’ learner do this?” In fact, the answer to it is a provocative one. The assessment that was used to gauge the participants’ overall proficiency was based on their performativity of academic literacy – namely accuracy in using standard English, though the term standard is a highly contentious and debatable concept in this post-colonial and post-modernism era. This explains why they were categorized into low achievers despite the fact that they were able to contribute to their learning community to co-construct meaning. Different from the three high achievers, the ‘low achievers’, on the whole, were not as capable as their stronger peers in providing the right answers. Nonetheless, they were not deprived of any opportunity in the negotiation process, as it were. In most cases, their ability to negotiate for meaning was well brought out, but their relatively low linguistic ability disadvantaged them. They seldom overshadowed their stronger peers by offering the right choices for the blanks in the first place. Instead, many of the contributions were in the form of giving consent or expressing consensus through formulaic expressions such as “I agree”, “I think so”, or in some cases, simply a declaration of their lack of such linguistic knowledge. Their inability in linguistic expression and their prominent use of SCE as represented by the use of particles like “lah”, “loh”, and “meh” did not bar them from participation in the group activity. On the contrary, the negotiation process went on smoothly and as a matter of fact, none of the members noticed these features. Being Singaporeans, they might feel more comfortable with these features, as, typically, in addition to English being used in ways peculiar to the speech community in Singapore that can identify its speakers as inner group members, accent is necessarily a strong defining characteristic of their identities. Also, the informal task-based learning environment might have reinforced their understanding of the nature of the learning task. The ‘situated’ nature of the learning community should have its own practice, and it is commonly shared knowledge that served as the ground for the very productive completion of the task within the community of practice. Therefore, they resorted to the basilectal form for meaningful interaction for the construction of meaning, possibly in the spirit of building a learning community, where everyone was expected to contribute. The “social turn” (Block 2003) has made it imperative that learners’ interactive moves in language classrooms be examined for a better understanding of the more socially informed decisions they make. This is because socialization is crucial in second language acquisition (Block 2007a; Gass 2003; Pica and

262

Lawrence Jun Zhang

Doughty 1985). In a similar vein, Canagarajah (2007) posits that it is necessary to nestle and reframe a cognitive view of language acquisition within a sociallyembedded system so that these commonly used constructs are not treated in isolation. Although Firth and Wagner (1997) argued for rectifying the imbalance between the dichotomies that characterize SLA, we are now moving toward more radical options of reframing the constructs. The previously ignored or suppressed constructs are now becoming the basis for a new integration or synthesis. Language acquisition is based on performance strategies, purposive uses of the language, and interpersonal negotiations in fluid communicative contexts. The previously dominant constructs such as form, cognition, and the individual are not eradicated; they get redefined to adopt hybrid, variable, situational, and processual characteristics they did not have before. (Canagarajah 2007: 936)

This sociocultural view of language acquisition offers a new perspective on the language learning process in relation to this group of Singaporean schoolchildren being schooled in English. We can see clearly that there is a need for teachers and researchers to treat the cognitive “in a more socially embedded, interactionally open, and ecological situated manner” (Canangarajah 2007: 936) to expedite the acquisition process through language use in designing and implementing curricular materials. It is clear that these ‘less successful’ schoolchildren’s use of English is inaccurate, and their speech is frequently permeated with local dialectal features typical of any new variety of English. Canagarajah (2006) forcefully argues that “the changing pedagogical priorities suggest that we have to move away from a reliance on discrete-item tests on formal grammatical competence . . . and assessment would focus on strategies of negotiation, situated performance, communicative repertoire, and language awareness” (2006: 229). In this sense, it is evident that, more often than not, language learning did take place, or at least the negotiation process for learning did really proceed in interaction which was socially co-constructed by the learning community. The commonly accepted strategies for negotiation of meaning were well accepted in this community of practice, where these children’s willingness to share and contribute whatever they knew to the learning community can be taken as a positive attribute; their pragmatic use of the language in the negotiation process, despite its inaccuracy if framed with a grammaticality judgement test, suggests that successful ESL schoolchildren understood how to use the language for maintaining the conversation or discussion, and it is especially the case for those “more capable peers”, who were ready to pitch in without realizing that they were doing a service to the other peers. Even the less capable learners made the grade in jellying with their peers through typical utterances

Negotiating language, literacy and identity

263

used in the daily social life of Singaporeans. In this negotiation process, these schoolchildren used repair/self-repetition, clarification requests, and confirmation checks alongside “other consensus-oriented and mutually supportive practices” (Canagarajah 2006: 238). It is not only the number of strategies or the variety of strategies used by these children that distinguished them from one another despite the importance of strategy deployment and orchestration in language processing for linguistic gains. Rather, it is the ability to communicate through the deployment of strategies for social interaction for the purpose of learning the language material that did really make a difference, especially for the learners under discussion. Clarification requests were found to be the most frequent strategy used by children in their negotiation of meaning. It was particularly the case for the ‘successful’ learners. Nonetheless, the purpose of the kind of negotiation is usually framed according to the defining parameters of SLA: form, cognition and the individual (e.g., Oliver 2002). The limited nature of the learning material used in the present study could be a platform where ‘form’, ‘cognition’ and the ‘individual’ dominated. The patterns of the children’s negotiation suggest that instead of only focusing on the form of the language, meaningful discussion was “based on performance strategies, purposive uses of the language, and interpersonal negotiations in fluid communicative contexts” (Canagarajah 2006: 238). Sociocultural theory regards learning as essentially taking place in the zone of proximal development (ZPD). This is because, as Vygotsky (1978) explains, “the distance between the actual developmental level as determined by independent problem-solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers” (1978: 86). In this sense, learning is not an individual activity but one that involves others in the learning community and is to a great extent interpersonal, and collaboration is important. Language is a mediating tool between the individual and the learning environment (Vygotsky 1978). This is the reason why quite a number of participants started with “private speech” rehearsals before really engaging themselves in the interaction as a way of regulating any complex task that they would have to handle (see Lantolf and Thorne 2006).

8. Conclusion The study was set up to answer the Singapore Ministry of Education’s call of a “Teach Less, Learn More” initiative. As a start, it was interested in finding out how Singaporean elementary schoolchildren learn English (Zhang et al. 2008b).

264

Lawrence Jun Zhang

The preliminary findings suggest that strategies are important for negotiation of meaning, and that schoolchildren did really attempt at using strategies in different ways. The differences between the high achievers and low achievers as benchmarked against standardized tests administered by the school according to the academic norms were obvious. These differences lie in the larger number of strategies used by the high achievers or ‘successful’ learners. However, the low achievers or the ‘less successful’ schoolchildren also displayed their strength in having certain strategy repertoires that the successful schoolchildren did not orchestrate. This gives rise to the question of the validity of strategy number count. In fact, it was found that knowing how to negotiate as an important metacognitive strategy at the level of ‘planning’ was as important as knowing how a particular word was used, or how the language system works in its own right (see e.g., Cohen and Macaro 2007a; O’Malley and Chamot 1990; Zhang forthcoming). However, the patterns of the schoolchildren’s negotiation in this study suggest that their performance strategies, purposive uses of the language, and interpersonal negotiations in context-specific communication through the use of SCE features are more significant in helping them chart their learning journeys. This is because talking and socializing are important facets of language acquisition for children. In fact, the way that the learners resorted to the various strategies for negotiation of meaning might also be indicative of their exercising agency that is closely connected to their identities. Norton (1995) and Block (2007b) rightly posit that language learners’ relationship to the target language is socially and historically constructed. In Norton’s understanding, language learners’ ‘investment’ in the target language reflects their complex history and multiple desires; their investment in the language is also an investment in their own identities, which change over time and space (Norton 1997). As mentioned earlier, these schoolchildren were bilingual/biliteracy learners, who had to face challenges of learning the two required language subjects. Instead of only relying on counting their use of strategies in interaction, we have to see how effectively their communication needs were met. Thus, these Singaporean schoolchildren who were speakers and users of the New English variety have to be understood in light of the bidialectal and bilingual backgrounds against which they were raised in a multilingual society. Any attempt to despise them for the idiosyncratic use of the language unique in its own way may not put them to any advantage. Assessment that was based on the standardized test as the one used in this study might not have been able to reveal many aspects of the asset they had as bilingual, and in some cases, multilingual, users (Canagarajah 2006). This could be the case for schoolchildren in other similar contexts. Just as Li (2008) recommends, “a comprehensive understanding of any complex social

Negotiating language, literacy and identity

265

phenomenon as bilingualism and multilingualism required contributions from a variety of disciplines. The multidisciplinary and interdisciplinary approach has definitely generated research outcomes that challenge the received wisdom about the human mind and society. But being multidisciplinary or interdisciplinary does not in itself entail innovation” (2008: 16). My study resorted to sociocultural theory as a lens through which bilingual schoolchildren’s use of strategies for negotiating meaning, literacy and identity was explored. It is obvious that further research needs to address issues such as why bilingual schoolchildren use these strategies and what their preoccupations are when they are engaged in the learning task. Methods such as longitudinal observations, cross-sectional investigation, or immediate retrospective interview can be employed to find out the bilingual schoolchildren’s rationale for the communication behavior (see Zhu and David 2008). Once findings emerge, teachers will have evidence that will point to a necessity for helping these learners enhance their understanding of how language works in society, or more specifically, in classroom interaction. The aim of teaching is to offer the essential scaffolding which the schoolchildren need so that they can learn to make and construct meaning in the “Zone of Proximal Development” (Vygotsky 1978) to ultimately achieve high levels of linguistic and sociolinguistic performance competencies independently, which will serve the speech community in culturally suitable ways. The Singapore Ministry of Education (MOE 2001) mandates that “English is one of the four official languages in Singapore. As the language of public administration, education, commerce, science and technology, and global communication, it has become the medium by which most Singaporeans gain access to information and knowledge from around the world. The ability to speak and write English effectively, therefore, has become an essential skill in the workplace, and a mastery of English is vital to Singapore’s pupils” (2001: 2). While this policy statement is valid, when it comes to classroom practice, schoolchildren had no choice but use the nonstandard variety, SCE, if they wished to maintain the flow of the discussion (Alsagoff 2007; Stroud and Wee 2007). This is because, as Singaporeans, who are mainly speakers of English in a New English context, their use of the language is closely tied to their mother tongues, including their cultural traditions, values, and identities. These schoolchildren’s language development needs also to be understood within a framework of linguistic ecology in addition to the developmental stages that they have to go through as do schoolchildren in Australia, US, or UK. The findings from this study might shed some light on pedagogical practice in contexts similar to that of Singapore. Although differences exist in the specific definitions of LLSs, as mentioned above, instructional practices implemented by teachers rely heavily on these

266

Lawrence Jun Zhang

important findings (see Cohen and Macaro 2007a). Thus, the findings from this study might be able to offer further insights into elementary school classrooms where it might be necessary for teachers to incorporate strategies-based instruction into the language curriculum. Nonetheless, teachers must understand that a strategies-based instruction program should take into consideration the sociocultural context and schoolchildren’s lived experiences such as their bidialectal/ multilingual backgrounds, their daily use of the language as a lingua franca on top of the curriculum requirement as an academic subject in schools, their affiliations to the New English variety they use, among others. By doing so, teachers not only render good service in equipping the schoolchildren with the strategies of how to learn English more effectively; they also give these schoolchildren opportunities to explore possibilities for wider and more effective communication within their own means of the linguistic resources they are endowed with, and without being coerced to follow the exnormative academic standards. After all, as Zhang argues (2004), communicativity, intelligibility, and socially meaningful engagement are the core elements in human interaction and learning. Nanyang Technological University and University of Oxford

Appendix: Cloze task material Clean rivers in Singapore Many rivers in the world are in danger because of pollution. waste from factories and housing areas is dumped into the river, fish and other living things in the river will die. , we have a choice. We can vhelp to clean up our rivers as part of a group work on our own. In the 1970s, housing, factory and farm waste polluted the river. , it was not safe to fish or even swim in the water. In 1977, the government of Singapore challenged the people to make the river safe for fishing within ten years. It was a lot of work, it was well worth the effort. By 1987, the pollution had been removed the river was clean enough for fish. , work on the river did not stop. Walkways, restaurants, shops and parks were built along the riverbanks the idea was for people to use the river again. you go to the riverside today, you will see many Singaporeans and tourists enjoying themselves the river is clean and pleasant.

Negotiating language, literacy and identity

267

Note *

My sincere thanks go to Wai Yin Pryke for giving me an opportunity to work with her schoolchildren, and to Steven Tan and Victoria Chan for their assistance in data collection. I am grateful to all the pupils for participating in this investigation. This study was jointly funded by a grant, OER25/08/LZ, from the Office of Education Research, National Institute of Education, and the Ministry of Education, Singapore. An earlier version of the paper was delivered at an invited panel presentation, “Kid-Speak”, TESOL International Convention 2009, USA. A revised version was presented as a plenary paper at the Bloomsbury Student Conference 2009, Birkbeck, University of London, UK. I thank the discussant, Suresh Canagarajah, for his incisive comments on an earlier version, and Ernesto Macaro for hosting me as a Post-Doctoral Fellow in the University of Oxford, which has made the completion of this paper possible. I am equally indebted to Li Wei and David Block for their trust in me and constructive feedback which helped improve the clarity of the paper. All faults remain my sole responsibility.

References Alsagoff, L. 2007. Singlish: Negotiating culture, capital and identity. In V. Vaish, V. Gopinathan & Y. Liu (eds.), Language, capital, culture: Critical studies of language and education in Singapore, 25–46. Rotterdam: Sense. Block, D.. 2003. The social turn in second language acquisition. Edinburgh: Edinburgh University Press. Block, D. 2007a. Second language identities. London: Continuum. Block, D. 2007b. Socializing second language acquisition. In H. Zhu, P. Seedhouse, W. Li & V. Cook (eds.), Language learning and teaching as social inter-action, 89–102. London: Palgrave Macmillan. Canagarajah, S. 1995. Functions of code-switching in ESL classroom: Socialising bilingualism in Jaffna. Journal of Multilingual and Multicultural Development 6(3). 173– 195. Canagarajah, S. 2006. Changing communicative needs, revised assessment objectives: Testing English as an international language. Language Assessment Quarterly 3. 229–242. Canagarajah, S. 2007. Lingua franca English, multilingual communities, and language acquisition. Modern Language Journal 91. 921–937. Chamot, A. U. & P. El-Dinary. 1999. Children’s learning strategies in language immersion classrooms. Modern Language Journal 83. 319–338. Chesterfield, R. & K. Chesterfield. 1985. Natural order in children’s use of second language learning strategies. Applied Linguistics 6. 45–59. Cohen, A. D. & E. Macaro. 2007a. Language learner strategies: Thirty years of research and practice. Oxford: Oxford University Press.

268

Lawrence Jun Zhang

Cohen, A. D. & E. Macaro. 2007b. Conclusions. In A. Cohen & E. Macaro (eds.), Language learner strategies: Thirty years of research and practice, 275–284. Oxford: Oxford University Press. Ellis, R. 2008. The study of second language acquisition, 2nd edn. Oxford: Oxford University Press. Firth, A. & J. Wagner. 1997. On discourse, communication, and (some) fundamental concepts in SLA research. Modern Language Journal 81. 285–300. Gass, S. M.. 2003. Input and interaction. In C. Doughty & M. Long (eds.), Handbook of second language acquisition, 225–255. Oxford: Blackwell. Kachru, B. B. 1992. The other tongue, 2nd edn. Urbana, IL: University of Illinois Press. Lan, R. & R. L. Oxford. 2003. Language learning strategy profiles of elementary school students in Taiwan. International Review of Applied Linguistics 41. 339–379. Lantolf, J. & S. Thorne. 2006. Sociocultural theory and the genesis of second language acquisition. Oxford: Oxford University Press. Li, W. 2008. Research perspectives on bilingualism and multilingualism. In W. Li & M. G. Moyer (eds.), The Blackwell guide to research methods in bilingualism and multilingualism, 3–31. Oxford: Wiley-Blackwell. Long, M. H. 1981. Input, interaction and second language acquisition. In H. Winitz (ed.), Native language and foreign language acquisition, 250–278. New York: Annals of the New York Academy of Sciences. Macaro, E. 2006. Strategies for language learning and language use: Revising the theoretical framework. Modern Language Journal 90. 320–337. MOE (Ministry of Education, Singapore). 2001. The English language syllabus for primary and secondary schools. Singapore: Author. Naiman, N., M. Fr¨ohlich, H. H. Stern & A. Todesco. 1978. The good language learner. Toronto, Canada: Ontario Institute of Studies in Education, University of Toronto. Norton, B. Peirce. 1995. Social identity, investment, and language learning. TESOL Quarterly 29. 9–31. Norton, B. 1997. Language, identity, and the ownership of English. TESOL Quarterly 31. 409–429. Norton, B. & K. Toohey. 2001. Changing perspectives on good language learners. TESOL Quarterly 35. 307–321. Oliver, R. 2002. The patterns of negotiation for meaning in child interactions. Modern Language Journal 86. 97–111. O’Malley, J. M. & A. U. Chamot. 1990. Learning strategies in second language acquisition. Cambridge: Cambridge University Press. Oxford, R. L. 1990. Language learning strategies. Rowley, MA: Newbury House. Oxford, R. L. 2007. Bridging the gap between psychological and sociocultural perspectives on L2 learner strategies. In A. Cohen & E. Macaro (eds.), Language learner strategies: 30 years of research and practice, 47–65. Oxford: Oxford University Press. Pakir, A. 2004. Medium-of-instruction policy in Singapore. In J. Tollefson & A. Tsui (eds.), Medium of instruction policies, 117–133. Mahwah, NJ: Erlbaum.

Negotiating language, literacy and identity

269

Patton, Q. M. 2001. Qualitative research and evaluation methods, 3rd edn. London: Sage. Pica, T. & C. Doughty. 1985. Input and interaction in the communicative language classroom: A comparison of teacher-fronted and group activities. In S. Gass & C. Madden (eds.), Input in second language acquisition, 115–132. Rowley, MA: Newbury House. Pica, T., R. Kanagy & J. Falodun. 1993. Choosing and using communication tasks for second language instruction and research. In G. Crookes & S. Gass (eds.), Tasks and language learning: Integrating theory and practice, 9–34. Clevedon: Multilingual Matters. Purdie, N. & R. Oliver. 1999. Language learning strategies used by bilingual school-aged children. System 27. 375–388. Rubin, J. 1975. What the ‘good language learner’ can teach us. TESOL Quarterly 9(1). 41–51. Stern, H. H. 1975. What can we learn from the good language learner? Canadian Modern Language Review 31(3). 304–318. Stroud, C. & L. Wee. 2007.A pedagogical application of liminalities in social positioning: Identity and literacy in Singapore. TESOL Quarterly 41. 33–54. Sugeng, B.. 1997. A learning strategy profile of Indonesian elementary school students. RELC Journal 28. 82–106. Van Lier, L. 2005. Case study. In E. Hinkel (ed.), Handbook of research in second language teaching and learning, 195–208. Mahwah, NJ: Erlbaum. Varonis, E. & S. Gass. 1985. Nonnative/nonnative conversations: A model for negotiation of meaning. Applied Linguistics 6. 71–90. Vygotsky, L. 1978. Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Willet, J. 1995. Becoming first graders in an L2: An ethnographic study of L2 socialization. TESOL Quarterly 29. 473–503. Wong Fillmore, L.. 1976. The second time round: Cognitive and social strategies in second language acquisition. Stanford University unpublished doctoral dissertation. Dissertation Abstracts International 37, 10A, p. 6443. (University Abstracts International No. DDJ77-07085.) Young, R. 2009. Discursive practice in language learning and teaching. Oxford: WileyBlackwell. Zhang, L. J. 2004. Awareness-raising in the TEFL phonology classroom: Student voices and sociocultural and psychological considerations. ITL International Journal of Applied Linguistics 145. 219–268. Zhang, L. J. Forthcoming. A dynamic metacognitive systems account of Chinese university students’ knowledge about EFL reading. TESOL Quarterly, 44. Zhang, L. J., P. Gu & G. Hu. 2008a. A cognitive perspective on Singaporean bilingual children’s use of reading strategies in learning to read in English. British Journal of Educational Psychology 78. 245–271. Zhang, L. J., T. Kiss, M. Y. Ang, D. Zhang, W. Gong, G. P. Aw, C. K. Chin & K. W. Choong. 2008b. Enhancing Singaporean students’motivation, efficacy, engagement,

270

Lawrence Jun Zhang

and self-regulation for effective bilingual/biliteracy learning. Singapore: Office of Education Research, NIE, Ministry of Education. Zhu, H., &A. David. 2008. Study design: Cross-sectional, longitudinal, case and group. In W. Li & M. G. Moyer (eds.), The Blackwell guide to research methods in bilingualism and multilingualism, 88–107. Oxford: Wiley-Blackwell. Zuengler, J. & E. R. Miller. 2006. Cognitive and sociocultural perspectives: Two parallel SLA worlds. TESOL Quarterly 40. 35–58. Lawrence Jun Zhang is an associate professor at NIE, Nanyang Technological University, Singapore. His main interests are language learner strategies, metacognition and bilingual/biliteracy education, and his research work is published in British Journal of Educational Psychology, English Today, Journal of Psycholinguistic Research, Instructional Science, Language and Education and Language Awareness. He is an editorial board member of The Linguistics Journal, Metacognition and Learning and TESOL Quarterly.