Dominant Language Constellations: A New Perspective on Multilingualism [1st ed.] 9783030523350, 9783030523367

This volume is dedicated to the concept and several applications of Dominant Language Constellations (DLC), by which it

277 44 3MB

English Pages XIX, 282 [286] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Front Matter ....Pages i-xix
Introduction: The Dominant Language Constellations: A New Perspective on Multilingualism (Joseph Lo Bianco, Larissa Aronin)....Pages 1-15
Front Matter ....Pages 17-17
Dominant Language Constellations as an Approach for Studying Multilingual Practices (Larissa Aronin)....Pages 19-33
A Meeting of Concepts and Praxis: Multilingualism, Language Policy and the Dominant Language Constellation (Joseph Lo Bianco)....Pages 35-56
Where DLC Meets Multilingual Syntactic Development (Éva Fernández-Berkes, Suzanne Flynn)....Pages 57-74
Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual Contexts: African Perspectives (Felix Banda)....Pages 75-93
Front Matter ....Pages 95-95
Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present State and Challenges for the Future (Siv Björklund, Mikaela Björklund, Kaj Sjöholm)....Pages 97-115
Language Background Profiling at Canadian Elementary Schools and Dominant Language Constellations (Nikolay Slavkov)....Pages 117-138
Dominant Language Constellations in the Language Repertoires of Multilingual South African Students (Susan Coetzee–Van Rooy)....Pages 139-165
Front Matter ....Pages 167-167
The Evolution of Personal Dominant Language Constellations Based on the Amount of Usage of the Languages (Sarasi Kannangara)....Pages 169-186
Dominant Language Constellations of Russian Speakers in Cyprus (Sviatlana Karpava)....Pages 187-209
Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant Language Constellation (Stela Letica Krevelj)....Pages 211-229
A Dominant Language Constellations Case Study on Language Use and the Affective Domain (Richard Nightingale)....Pages 231-259
Quo Vadis, DLC? (Joseph Lo Bianco)....Pages 261-275
Back Matter ....Pages 277-282
Recommend Papers

Dominant Language Constellations: A New Perspective on Multilingualism [1st ed.]
 9783030523350, 9783030523367

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Educational Linguistics

Joseph Lo Bianco Larissa Aronin  Editors

Dominant Language Constellations A New Perspective on Multilingualism

Educational Linguistics Volume 47

Series Editor Francis M. Hult, University of Maryland, Baltimore County, Baltimore, USA Editorial Board Marilda C. Cavalcanti, Universidade Estadual de Campinas, Campinas, Brazil Jasone Cenoz, University of the Basque Country, Leioa, Spain Angela Creese, University of Stirling, Stirling, UK Ingrid Gogolin, University of Hamburg, Hamburg, Germany Christine Hélot, Université de Strasbourg, Strasbourg, France Hilary Janks, University of the Witwatersrand, Johannesburg, South Africa Claire Kramsch, University of California, Berkeley, USA Constant Leung, King’s College London, London, UK Angel Lin, Simon Fraser University, Burnaby, Canada Alastair Pennycook, University of Technology, Sydney, Australia

Educational Linguistics is dedicated to innovative studies of language use and language learning. The series is based on the idea that there is a need for studies that break barriers. Accordingly, it provides a space for research that crosses traditional disciplinary, theoretical, and/or methodological boundaries in ways that advance knowledge about language (in) education. The series focuses on critical and contextualized work that offers alternatives to current approaches as well as practical, substantive ways forward. Contributions explore the dynamic and multi-­ layered nature of theory-practice relationships, creative applications of linguistic and symbolic resources, individual and societal considerations, and diverse social spaces related to language learning. The series publishes in-depth studies of educational innovation in contexts throughout the world: issues of linguistic equity and diversity; educational language policy; revalorization of indigenous languages; socially responsible (additional) language teaching; language assessment; first- and additional language literacy; language teacher education; language development and socialization in non-­ traditional settings; the integration of language across academic subjects; language and technology; and other relevant topics. The Educational Linguistics series invites authors to contact the general editor with suggestions and/or proposals for new monographs or edited volumes. For more information, please contact the Editor: Natalie Rieborn, Van Godewijckstraat 30, 3300 AA Dordrecht, The Netherlands. All proposals and manuscripts submitted to the Series will undergo at least two rounds of external peer review. More information about this series at http://www.springer.com/series/5894

Joseph Lo Bianco • Larissa Aronin Editors

Dominant Language Constellations A New Perspective on Multilingualism

Editors Joseph Lo Bianco University of Melbourne Melbourne, VIC, Australia

Larissa Aronin Oranim Academic College of Education Tivon, Israel

ISSN 1572-0292     ISSN 2215-1656 (electronic) Educational Linguistics ISBN 978-3-030-52335-0    ISBN 978-3-030-52336-7 (eBook) https://doi.org/10.1007/978-3-030-52336-7 © Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To our parents Joseph’s – Vincenzo Lo Bianco and Antonia Ferraro and Larissa’s – Valeria Vasiljevna and Isar Borisovitch

Foreword: Multilingualism and Dominant Language Constellation

The realization amongst the international academic community in the last two decades of the significance of societal multilingualism has come as something of a surprise to the very many communities across the globe where multilingualism has always been an integral part of people’s everyday life. The unprecedented growth in the study of societal multilingualism is evidently due to the presence of large numbers of ‘misplaced’ populations who speak languages other than the assumed indigenous ones. These speakers are usually labelled as ‘immigrants’, ‘migrants, ‘refugees’, or ‘ethnic minorities’. Their presence seems to have disturbed what was otherwise perceived as a good, stable social order with clearly defined dominant national language(s). Yet, as Michael Clyne (2003) reminded us, language contact, and multilingualism that resulted from it, has always been a key contributing factor in the evolution of human languages. Crucially, language contact is a dynamic and ongoing process. Multilingualism, therefore, is also a dynamic and ever-changing phenomenon. To study multilingualism as a dynamic phenomenon, one must pay attention to the geopolitical, ideological, and historical contexts, as the present volume seeks to do. The chapters in the present volume address the meanings and implications of multilingualism in diverse contexts from the theoretical and conceptual perspective of Dominant Language Constellation (DLC). DLC applies to both societal language contact and multilingualism and individual multilingual practices. It invites us to ask critical questions as to how some languages become dominant in certain contexts but not in others; how societies manage multiple languages through language policy and language management agencies and institutions; what are the consequences of language policy and language management for individual languages and their speakers; how individuals function in society through their strategic use of different languages both in different contexts and over time; and how dominant language constellations change across generations, communities, and nations. We must remind ourselves that the newly ‘displaced’ languages may become ‘community’ or ‘heritage’ languages over time, but they may remain as national languages in a different context. The designation of a language and its speakers with labels such as ‘immigrant’, ‘minority’, ‘indigenous’, or ‘national’ has serious vii

viii

Foreword: Multilingualism and Dominant Language Constellation

consequences on the status of the language and the community in society. Policies and practices that are designed to support the minoritized languages and social groups usually assume that their status is a real one rather than imposed by society. For instance, in Britain and the USA, where English is the dominant language, people who have roots in another country are often designated as English-as-an-­ additional-language (EAL) speakers. EAL children are expected to struggle in the mainstream educational system. They need help with English. And if they do achieve well in schools, they will be celebrated as examples of success. For many speakers of Arabic, Chinese, or Spanish in Britain and the USA, the idea that their native language is a ‘minority’ language reminds them of their social status in an English-speaking society rather than the status of their languages on a global scale. Multilingualism has been used as good evidence of human beings’ capacity to learn and develop multilingual cognitive and knowledge systems that they can put to use according to context. The learning and use of multiple languages presume access to them. But access may not be equal. To understand multilingualism, one must look at the specific historical, sociocultural, environmental, and material conditions on access to languages for the individuals and communities concerned. Only in doing so can we understand the status of different languages in society and the implications of the designated status of the language for its speakers in society. Migration, which has hugely intensified across the globe in recent decades, is another condition that impacts on access to languages. The recent and ongoing migratory processes have not been balanced in the sense that it is often people from under-developed parts of the world migrating to the more developed countries and regions, though this is relative and by no means universal. The status of language and its speakers as a result of migration is a politically sensitive topic for exploration and has serious implications for policy and practice. The educational impacts of multilingualism are a major concern for both the speaker and the system. Some people are able to maintain all the languages in their repertoire and in fact, in some cases, can even increase the number of languages they can learn and use. Others have given up the languages they acquired in childhood and replace them with other languages that they use in their everyday communication. Reasons for language maintenance and language shift or attrition are complex and multiple. They are a fascinating topic for research. How the educational system responds to multilingualism is another hotly debated issue. The idea and ideal of education for all, or inclusive and equitable education, are put to real tests in the presence of diverse multilingual learners. Whilst many have argued that multilingualism is a rich and important resource for learning, it is fair to say that most of the so-called bilingual education programmes that exist in the developed world especially are educational programmes for people who are already bilingual rather than aiming to support learners to become bilingual or maintain their bilingualism and multilingualism. Institutional and policy support for multilingual pedagogies is still lacking. As research curiosity in multilingualism both as an individual and as a societal phenomenon continues to grow in the age of globalization and superdiversity, scholars have started to develop various competing theoretical and analytical frameworks.

Foreword: Multilingualism and Dominant Language Constellation

ix

With its emphasis on the dynamic and ever-changing nature of multilingualism and the need to embed our analysis and understanding in historical and ideological contexts, DLC makes an original and productive contribution to scholarship and knowledge. UCL Institute of Education, University College London  Li Wei London, UK

Reference Michael Clyne (2003) Dynamics of Language Contact. Cambridge: Cambridge University Press

Acknowledgement

The second editor of this volume wishes to thank the Research Authority of the Oranim Academic College for their support.

xi

Contents

 Introduction: The Dominant Language Constellations: A New Perspective on Multilingualism����������������������������������������������������������    1 Joseph Lo Bianco and Larissa Aronin Part I Current Developments of DLC  Dominant Language Constellations as an Approach for Studying Multilingual Practices ��������������������������������������������������������������   19 Larissa Aronin  Meeting of Concepts and Praxis: Multilingualism, A Language Policy and the Dominant Language Constellation����������������������   35 Joseph Lo Bianco Where DLC Meets Multilingual Syntactic Development ����������������������������   57 Éva Fernández-Berkes and Suzanne Flynn  Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual Contexts: African Perspectives����������������������������   75 Felix Banda Part II Institutional Expressions of DLC  Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present State and Challenges for the Future����������   97 Siv Björklund, Mikaela Björklund, and Kaj Sjöholm  Language Background Profiling at Canadian Elementary Schools and Dominant Language Constellations��������������������������������������������������������  117 Nikolay Slavkov  Dominant Language Constellations in the Language Repertoires of Multilingual South African Students ��������������������������������������������������������  139 Susan Coetzee-Van Rooy xiii

xiv

Contents

Part III Personal and Group Experiences with DLC  The Evolution of Personal Dominant Language Constellations Based on the Amount of Usage of the Languages ����������������������������������������  169 Sarasi Kannangara  Dominant Language Constellations of Russian Speakers in Cyprus����������  187 Sviatlana Karpava  Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant Language Constellation������������������������������������������  211 Stela Letica Krevelj  Dominant Language Constellations Case Study A on Language Use and the Affective Domain��������������������������������������������������  231 Richard Nightingale Quo Vadis, DLC?���������������������������������������������������������������������������������������������  261 Joseph Lo Bianco Language Index������������������������������������������������������������������������������������������������  277 Subject Index����������������������������������������������������������������������������������������������������  279

About the Editors

Joseph Lo Bianco  Professor Joseph (Joe) Lo Bianco holds the Chair of Language and Literacy Education at the Melbourne Graduate School of Education, University of Melbourne. He is a specialist in minority language rights, literacy and social opportunity, and language policy and planning, with a specific interest in conflict mitigation in multi-ethnic societies in South East Asia, educational equity for immigrant and Indigenous populations, multilingual and anti-racist education, and language revitalisation. He has extensive experience in real-world policy writing, policy advising, and community support over several decades in 25 international locations. In recent years, this has included an 8-year project in Southeast Asia for UNICEF under the title of Language, Education and Social Cohesion, focusing on conflict resolution in multi-ethnic settings in Malaysia, Myanmar, and South Thailand, and as a consultant to a 4-year project on Multilingual Cities in 12 European municipalities. In 1987, he authored Australia’s first national language policy, and between 1990 and 2002, he was the Director of Language Australia/The National Languages and Literacy Institute of Australia. Larissa  Aronin  is Associate Professor at the Oranim Academic College of Education, Israel. She is a founding member of International Association of Multilingualism (IAM) and has served as its Secretary for two terms (2010–2014), and is an editorial board member of a number of other peer-reviewed journals. Prof. Aronin has published in a range of international journals on a wide array of topics connected with multilingualism such as The International Journal of the Sociology of Language and The International Journal of Multilingualism and Language Teaching. She is the co-author of Multilingualism (John Benjamins 2012) and has co-edited The Exploration of Multilingualism (John Benjamins, 2009), Current Multilingualism: The New Linguistic Dispensation (De Gruyter, 2013), and more recently Twelve Lectures on Multilingualism (Multilingual Matters, 2019).

xv

About the Contributors

Felix Banda  is a senior professor of the Linguistics Department, University of the Western Cape, where he teaches undergraduate and postgraduate courses in sociolinguistics, multilingualism in society and education, critical media studies, and technology-mediated business communication. His research interests include linguistic landscapes, language practices in society and education, Bantu linguistics, multimodality, media and migration studies, and the social semiotics of personal and corporate identity branding. Mikaela  Björklund  is lecturer in foreign language education at the Faculty of Education and Welfare Studies, Åbo Akademi University. Her research and developmental work interests focus on the premises for teaching and learning English as a foreign language in Swedish-medium primary schools in Finland, Content and Language Integrated Learning (CLIL) at different levels in education, linguistic, and cultural integration patterns in comprehensive education, including teacher beliefs, pupils’ patterns of multilingual language use, and linguistic schoolscapes. The most recent research interest is teacher education as an arena for developing language awareness and multilingual language use. Siv Björklund  is professor of Swedish immersion at the Faculty of Education and Welfare Studies, Åbo Akademi University. Her research encompasses Swedish language immersion, Swedish as a second language, bilingual and multilingual learning, content and language integrated learning, and minority studies. Recent research projects focus on the relation between multilingualism and identity in immersion programs, pedagogical practices for language-diverse classes, and development of participatory writing among students with Swedish as a first or second language in different settings. Björklund is also one of the founders of the Journal of Immersion and Content-Based Language Education (JICB, John Benjamins Publishing Company).

xvii

xviii

About the Contributors

Susan Coetzee-Van Rooy  is a research professor at the North-West University in South Africa in the research entity, UPSET.  Members of UPSET study the “Understanding and Processing of language in complex SETtings.” Susan studies the multilingual repertoires of South Africans, and she approaches this work from the perspective of the Dynamic Multilingual Model (generated by Herdina and Jessner, 2002) and the notion of dominant language constellations (created by Aronin and her associates). She is a World Englishes scholar who contributes understandings of Outer Circle English from the multilingual South African context. She uses large-scale language surveys, language history interviews, language portraits, and social networks in her mixed methods studies. Susan teaches in applied language studies, macro-sociolinguistics and the sociology of language, with a specific interest in multilingualism with English. Éva Fernández-Berkes  studied theoretical linguistics and language acquisition at the Ortega y Gasset Graduate Institute (Madrid, Spain) and received her PhD from the Complutense University of Madrid (Spain) in 2002. She is currently a lecturer at the University of Applied Sciences of Burgenland (Pinkafeld campus). Her research interests concern multilingualism, syntactic development, and representation focusing on Hungarian, English, and German. Suzanne Flynn  is a professor of Linguistics and Language Acquisition at MIT. Her research focuses on the acquisition of syntax by both children and adults in bilingual, second and third language acquisition contexts. More recently, her work has focused on the neural representation of the multilingual brain. She also conducts research on language impairment and early onset of Alzheimer’s disease. Sarasi Kannangara  is a PhD student of the Institute of Linguistics and Literary Studies at Technical University of Darmstadt, Germany, where she conducts German as a second language course (A1–C1) for foreign students at the language resource center. She is also employed at the Department of Modern Languages, University of Kelaniya, Sri Lanka, as a probationary lecturer for German studies. She has also taken part in developing the national school curriculum for German Language in Sri Lanka, and she is a coauthor of the regional textbook, Kulturbeutel: eine interkulturelle Begegnung, for teaching German language in the field of Tourism. Her research interests focus on multilingualism and emotional aspects in third language acquisition. Sviatlana  Karpava  (PhD) is a lecturer of applied linguistics/TESOL in the Department of English Studies, University of Cyprus, and coordinator of the Testing, Teaching and Translation Lab. She is a management committee member and WG5 co-leader of the “European Family Support Network Cost Action. A bottom-up, evidence-based and multidisciplinary approach” (2019–2023) as well as a working group member of the “European Network for Combining Language Learning with Crowdsourcing Techniques” Cost Project (2017–2021). Sviatlana initiated Erasmus plus project “Planting languages-seeds of success” (2019–2021).

About the Contributors

xix

She is the general secretary of the Cyprus Linguistic Society (CyLing) and active member and research collaborator of Cyprus Acquisition Team (CAT). Her area of research is applied linguistics, morpho-syntax, semantics and pragmatics, first and second language acquisition, bilingualism, multilingualism and dialect acquisition, sociolinguistics, teaching, and education. She is interested in heritage language use, maintenance and transmission, language loss, shift and attrition, family language policy, and intercultural communication. Stela  Letica  Krevelj  is assistant professor at the SLA and TEFL section at the Faculty of Humanities and Social Sciences, University of Zagreb, Croatia. Her main areas of interest are psycholinguistic aspects of second and third language acquisition, multilingual language development, and methodological and ethical issues in applied linguistics research. She has published articles related to multilingualism and second language acquisition, and worked on national and international research projects involving young language learners. Richard Nightingale  holds a PhD in applied linguistics and is an assistant professor in the Department of English Studies at Universitat Jaume I, where he teaches at undergraduate and postgraduate levels. He is also a member of the LAELA (Applied Linguistics to the Teaching of the English Language) Research Group at the same university. His research mainly focuses on the interconnectivity between language, society, and the self. Within this framework, his main areas of interest are multilingualism, pragmatics, extramural learning contexts, affective factors, and the impact of networked digital media. Kaj  Sjöholm  is professor emeritus of Foreign Language Pedagogy in the Department of Teacher Education at Åbo Akademi University. His research interests center on second-language classroom research, multilingual education, and the cultural dimension of foreign language learning/teaching in language minority settings. Nikolay Slavkov  is director of the Canadian Centre for Research and Studies in Bilingualism and Language Planning at the Official Languages and Bilingualism Institute (OLBI) of the University of Ottawa. Prof. Slavkov’s contributions include edited volumes, articles, and book chapters on bi-/multilingualism, family language policy, language teaching, technology, and Slavic linguistics. He has published in Canadian Modern Language Review, Immersion Journal, International Journal of Multilingualism, International Journal of Bilingual Education and Bilingualism, Second Language Research, Canadian Journal of Applied Linguistics, TESL Canada, Lingua, and Journal of Slavic Linguistics. He is currently the Editor of OLBI’s bilingual journal, Cahiers de l’ILOB.

Introduction: The Dominant Language Constellations: A New Perspective on Multilingualism Joseph Lo Bianco and Larissa Aronin

Keywords  Multilingualism · Current multilingualism · Diversity and superdiversity · Translanguaging · Globalization · Language repertoire · Dominant language constellation · Third and multiple language acquisition · Education · Minority languages

This volume is devoted to examination of contemporary sociolinguistic practices from the perspective of Dominant Language Constellations (DLC). It is the first collection of studies guided by the DLC approach featuring contributors with expertise in applied linguistics, sociolinguistics, language policy and education. The authors of the volume introduce a variety of applications and interpretations of the DLC model leading to important insights. In this sense the volume breaks new ground by subjecting the DLC model to scrutiny from a wide variety pf perspectives. However, there is another sense in which we hope the volume is pioneering as well. This refers to a shift in general multilingualism studies which have been proliferating in recent years, so much so that in 2014 Stephen May described it as ‘the topic du jour’ (May 2014, p. 1), to the more targeted and specific conception inherent in the DLC concept. We welcome the vibrant interest in multilingualism research, and the rich array of perspectives brought to bear on its analysis and description. Particularly powerful has been the concept and applications of translanguaging (García and Wei 2014), unsettling established ways to perceive and practice education in our globalized world, and for researchers of multilingualism and bilingualism, especially in education environments, a stimulus to reimagine who learners are, what they do when they learn, and how we should teach them as a result. In repudiating the normalized ideological stances of past decades this work has been of major significance. J. Lo Bianco Language and Literacy Education, University of Melbourne, Melbourne, VIC, Australia L. Aronin (*) Oranim Academic College of Education, Tivon, Israel e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_1

1

2

J. Lo Bianco and L. Aronin

We see DLC research as a component part of this wider field of investigations into the language diversity of humankind, undertaken with the specific lens of scholars focused on the clusters of vehicle languages that individuals, institutions and societies utilize. This volume continues the research direction developed in the themed panels and symposia in Vienna 2016, Limerick 2017 and Lisbon 2018 on the Dominant Language Constellations and contains papers presented in these conferences as well as invited articles. The entire question of how we account for the language diversity of humanity (Lo Bianco  2014) becomes more important and topical as the focus of attention among scholars increasingly includes the reality of superdiverse societies and their linguistic contexts. When we look at superdiverse linguistic contexts it is clear that the terminology we have inherited in the language sciences is broadly inadequate to address the new forms of communication reality facing most of the world’s citizens today. Hence, the proliferation of new terms to describe, account for, and analyse our contemporary lingual world has expanded rapidly to include: multivocality, plurilingualism, translanguaging, metrolingualism and polylanguaging. In different settings these terms have resonance for language teaching, for programming of responses from various institutions, for curriculum writing, and of course for how we come to interpret and understand communicative behaviours and choices made by individuals and groups. How is DLC positioned in this environment of discussion and innovation in multilingualism studies? The DLC approach has arisen to reflect a critical and sometimes overlooked dimension of the global transformations in the use and acquisition of languages that have led to the distinctive nature of today’s multilingualism: the clusters of vehicle languages that individuals adopt, which serve their immediate needs, reflect their social environments demo-linguistics and which can be deployed according to the exigencies of practical life in a variety of domains. Operating as an inherent and central constituent of contemporary life is a key feature of current multilingualism (Singleton et al. 2013). In today’s reality of multilingualism, mobility, diversity and the technological advances feature in new forms. Migration has changed mainly with regard to trajectories, distances and destinations (Czajka and Hein de Haas 2014). The ensuing “Diversification of diversity”(Vertovec 2014: n.p.) is observable in a range of sociolinguistic facets including super diversification of language users, local linguistic practices, or language variants. These and the transformations produced by communication technology have brought crucial changes to linguistic practices both globally and locally. Although multilingual language practices exist side by side with mono- and bilingual social arrangements, it is multilingualism that has taken over not only numerically and territorially, but, more importantly, in its significance for the entire world. Multilingualism is defined as the organized and unorganized language practices with three and more languages and the handling of more than two languages by some or all members of a society, as well as the implications of these practices and this handling for the society and its members. ‘Handling’ involves language policies, attitudes, language behaviour and the assumptions underlying such behaviour

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

3

Table 1  Interrelated perspectives and concepts of contemporary multilingualism Concept or perspective What is it about Translanguaging Complex language practices Dominant language constellation Language repertoire

Emphasis applies to The informal process. The activity of enacting one’s language assets. Selected languages and their acquisition and use

Set of selected languages and skills. A model of language practices Totality of languages and skills of Totality of languages or skills an individual or community

in a particular community, all in the context of three and more languages being dealt with (Aronin 2019a). Affordances and challenges of the active presence of numerous languages in communities are also the realities of multilingualism (see e.g., Kramsch and Jessner 2015). It would be incorrect to say that multilingual language practices mean simply using many languages. The essence of current multilingualism is in how the many languages of the world are mastered and used. A set of languages is a prerequisite for individual and communal existence in contemporary human society because the human language faculty is now often expressed via a number of languages rather than a single language (Aronin 2007, 2019a; Aronin and Singleton 2008, 2012). The combinations of language skills that are either at one’s disposal or employed de-­ facto are expressed through the corresponding concepts of language repertoire, translanguaging and Dominant Language Constellation (see Table 1). It is clear from this that concepts of language repertoire and Dominant Language Constellation are cognates, they complement and mutually constitute each other, they are cognate notions because they aim to provide a description of a complex reality in non-reductive ways, but they differ in how they limit the focus of their attention. While a language repertoire aims to account for and include the totality of linguistic skills in all the languages possessed by an individual or by a community, a Dominant Language Constellation embraces only several languages (typically but not always three) that are deemed to be of prime importance. In other words, DLC is the active part of one’s language repertoire. One may say that a language repertoire is about the linguistic assets available in their totality, their plenitude or completeness, with, theoretically, no pre-defined limit, whereas a DLC refers only to active usage of an empirically verifiable cluster of languages. A DLC operates as a coherent whole, and therefore is the account of what an individual or wider grouping’s specific repertoire of used languages is in a given time and setting. Within the unit of a DLC languages play different roles and various linguistic and cognitive skills in several languages serve to carry out the necessary functions of a human language. One of the most salient discussions related to multilingualism has resulted in harmonizing to a certain extent the views of discreteness of languages and codeswitching (MacSwan 2017) on the one hand and translanguaging (García 2013; García and Wei 2015; Otheguy et al. 2015) on the other. Translanguaging refers to

4

J. Lo Bianco and L. Aronin

often informal social and pedagogical everyday linguistic practices (Li Wei 2018: 2) which does not exclude the awareness of the speakers of the existence of the idealized boundaries between languages and between language varieties (Li Wei 2018: 11). Recognizing the linguistic realities of the twenty-first century, the concept of translanguaging refers to “the fluid and dynamic practices that transcend the boundaries between named languages, language varieties, and language and other semiotic systems” (Li Wei 2018: 9). Both translanguaging and the DLC approach, demonstrate that the pattern of using several languages more or less concomitantly, and this overrides the previously sufficient pattern of employing one language at a time. While translanguaging is focused on the process and activity of using multiple languages, DLC provides a structure, a pattern that demarcates the active and crucial language skills deployed for a certain period of time. By shifting the focus from the investigation of separate languages to the exploration of their constellations, both translanguaging and the DLC approach radically depart from the monolingual perspective that has normatively prevailed in the past. Commenting on the reality of multilingualism Slavkov (this volume) notes that a monolingual perspective is still very much alive in practical dealings with dynamic and fluid multilingualism. Here we can identify an additional aim and purpose in the present volume, the wish to inject a practical and policy influencing conceptual innovation into discussions of human communication. While it is deeply true that some branches of the language sciences have embraced language pluralism, acceptance of multilingualism as a kind of ‘base condition’ from which to understand and account for human behaviours, is far from typically accepted across all academic fields. Multilingualism is not the assumed condition within all the language sciences, and  it is certainly not typical of how communication is addressed in education, public policy or management of services made available in multicultural societies (Lo Bianco, this volume). It appears that while awareness and encouragement of multilingual practices are on the rise and that many scholars are playing a leading role in bringing this about, there remains a steadfast monolingual habitus in both the academy and society. Even a cursory look at public debates in many nations shows that educational and social discourses are often still carried out within two great simplifications of social direction, the dominant one of the monolingual paradigm (practically the totality of European countries or other nations with nation-defined national languages) or the bilingual paradigm, in societies with a majority minority to which some concessions in policy have been made historically (such as Belgium, Canada, New Zealand). Contesting these simplifications can be assisted by the heuristically powerful notion of a DLC, which represents a clear expansion of both the monolingual and bilingual paradigms beyond the nationally imposed and expected languages. “Beyond”, here, includes the clusters of languages that citizens in such societies know and use, and which as sub-national groups might be of immigrant or indigenous origin, or they might be languages acquired and actively used for professional or recreational interest for various purposes. A multilingual nation, therefore, is not just multilingual in an undifferentiated way, it is a multilingualism comprised of many overlapping DLCs (Banda, this volume).

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

5

A Dominant Language Constellation is a group of one’s most important languages, functioning as an entire unit and enabling an individual to meet all their needs in a multilingual environment (Aronin 2006). As noted earlier, the DLC is closely related to a language repertoire yet differs from it because the DLC is selective, it includes only the most expedient languages or language skills that relate to a person or a group in their communication functioning at a given time and in a given environment. Put another way, the concept of a linguistic repertoire aims to include the totality of an individual’s, or a community’s linguistic skills, those used by them and potentially available to them for use, those identified with but not active and those which are passive yet still present. By contrast, analytically and descriptively, the DLC concept is concerned only with the active languages which stand out as being of prime importance and the vehicle of the communicative practices of that person’s daily life, business, career and identity expression (Aronin 2016, 2019b; Aronin and Singleton 2012: 59–75). In addition to being an individual matter, a DLC may operate across a whole community. As a social phenomenon, a DLC constitutes a complex of languages shared on a day-to-day basis by a group, an entire community or a population. An individual DLC is often imposed by a community, state or region of residence and includes the languages dominant there. From our consideration of the concepts that have arisen to account for super-­ diversity in the realm of languages it is clear that the DLC perspective expressly fits the general multilingualism research emphasis of contemporary scholarship. A methodologically important difference between language repertoire and DLC concepts lies in the degree of their particular aptness to monolingual, bilingual or multilingual paradigms. As explained previously, a language repertoire is an inventory of all the languages and language skills of monolingual, bi- and multilingual users. While the entire inventory works well describing language assets of monolinguals and bilinguals, considering all languages and skills of a tri- and more-lingual is also possible, but becomes too cumbersome and too complex to be manageable either in terms of practical use of one’s language assets or in terms of research. Moreover, language repertoire as a concept first appeared to describe the skills of a monolingual speaker and is not ‘tailored’ to account for the extreme complexity and special nature of multilingualism. In this respect, the Dominant Language Constellation approach is more conceptually economical in its focus on current linguistic practices, since it considers only active, most important part of one’s functional components of a language repertoire without losing sight of complexity and emergent qualities. An important feature of contemporary multilingualism is the global diffusion of English, the emergences of World Englishes and English as a Lingua Franca, and the politics of embrace and repudiation of English that have developed in response to its spread. The expansion of English and the various Englishes emergent across the globe is also functionally differentiated, since their presence in higher education and commerce, for example, is more concentrated than in general society and significantly more so than in home domains. Alongside the spread of English in these diverse ways there is a also a concomitant global diversification of other languages in single societies, noted already decades ago (Fishman 1998), so that more and

6

J. Lo Bianco and L. Aronin

more societies today are diverse, and previously diverse societies have intensified their linguistic diversity and communications technologies and mobilities in general have generated the understanding of super-diversity as the most credible account of new pluralism. Within this generalized picture of pluralization the DLC approach appears to capture and can account for apparently contradictory features. By accepting the subjectivity of individuals themselves determining their personal communication profile, the DLC approach acknowledges that within the English-influenced multilingualisms of societies individuals inhabit communication universes that select from and instantiate the wider external reality of languages, society and identity. In other words, the DLC is always a set of overlapping and communicatively effective choices that individuals and groups have made from what is available to them and useful for them. Hence, DLCs always emerge in particular contexts and are responses to those contexts agentively made by individuals and groups, as responses actively in their interests, and therefore display a multitude of patterns that instantiate the distribution of languages in geographical and political and educational environments. This is why there can be no single DLC, and no formulaic pattern of a DLC; the variation inherent in the notion of the DLC arises from the interaction between the constraining force of the communication regime the individual or group inhabit, and what choices and determinations they make of that environment. This pattern of an interaction between agency and constraint is a key feature in the chapters in this volume, and is central to the idea of the DLC. Hence a wide diversity of individual unique DLCs is found all over the world. Each continuously creates and over time recreates itself, by transforming or and replacing its components, in response to external social change and to internal subjective motivations and changed circumstances and needs. Yet, despite all this dynamism, there is stability and patterning in how DLCs operate, evident in most of the chapters in this volume, in relatively stable patterns of organization that detect and broadly describe an archetypal DLC, serving as an analytic model seeking that enriches our understanding of contemporary multilingual practices. One of the stabilities of a DLC relates to its internal coherence. A DLC is a unit of communication that transcends individual languages, it is therefore a cluster or a whole, and so, without depleting its composite parts, meaning the languages that comprise it, the DLC operates as a unit of communication, an integrated entity, both cognitively and sociologically. Being a general pattern of language acquisition and use therefore, a DLC exhibits systematic relationships between its component parts. Interactions inside a DLC lead to its constant transformations in configuration and the dominance of its component languages. Examples of such DLC makeovers can be found throughout this volume. The contributors to this volume explore the diverse patterns of language use and acquisition in various parts of the world and touch upon the productive interaction between languages and cultures of language teaching and learning. DLC alterations also mean the contextualized, glocal negotiation of identity, meaning in effect multiple and shifting identities that contemporary society makes available to us and which we can see people regularly exhibit. For this reason we have assembled a wide variety of national perspectives looking at both particular and general DLC

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

7

models and from this we can discern an instrumental value that the DLC concept offers in improving and enhancing language learning by enabling systems planning the future communication abilities needed by their citizens in a more intensely globalizing world. Further instrumental value is in relation to public policy (see Lo Bianco, this volume), and also behaviourally concerning complexity inherent in multilingual in functioning (see, e.g., Krevelj this volume). DLCs afford a rich opportunity for visual representation to support comprehension of complexities in relationships, hence verbal descriptions in many of the contributions in this volume are supplemented by visual expressions of various kinds and also of modelling. The analytic explanatory model is therefore refined by visual and tangible forms of representation through drawn replicas sketched or produced in various materials, such as maps and stellar constellations (Aronin 2016) (Fig. 1). The figure above describes the DLC of a member of a Russian-speaking community in Israel. The crucial languages are Russian (L1), Hebrew, the official language of the new country of residence, and English, the language of higher learning. The internal coherence of relations is evident in the depiction, gravitationally held together in orbit, and hence these all work together to allow Russian speakers to manage their daily lives and their education in the multilingual environment they inhabit. As part of the legend to allow deeper interpretation it is important to note that the 5-point stars in the figure denote languages of the repertoire; ‘planets’ represent repertoire languages known less well or seldom used, while 4-point stars

Fig. 1  A DLC showing Russian/Hebrew/English DLC in Israel

8

J. Lo Bianco and L. Aronin

indicate languages the subject of this DLC is exposed to in the more immediate living environment and often understood by them, at different levels of proficiency, whether separate words and phrases, or substantially more. These can be either a heritage language in a family or a language often heard due to social proximity, whether, by wish, or by circumstance. Another way of visual analysis of Dominant Language Constellations is DLC modelling. This term refers to the process of creating and manipulating external representations of individual or group dominant language constellations in the form of visual and/or tangible replicas. Such external representations of one’s language assets and practices boost cognition by shortcutting analytic processes, saving internal memory, creating persistent referents and providing structures that can serve as a shareable object of thought (Danaher 2016; Kirsh 2010; Shaw 2014). The simplest 3D models of personal Dominant Language Constellations are produced effortlessly by learners and users themselves with the help of pieces of plasticine of different colour and sticks. The activity of modelling one’s language unit is not only enjoyable and social cooperation; a 3D plasticine model of a personal Dominant Language Constellation can represent a cognitive extension of a person’s own language skills and is a material tangible symbol of a person’s subjective sense of sociolinguistic existence and the language skills which raises self- and language awareness (see Picture 1). Spheres of a different colour represent different languages, degrees of proficiency are represented by the size or dimension of the sphere. The greater the linguistic distance between languages, the longer the lines connecting the spheres. As opposed to hand-made models, which are important for awareness and emotional involvement that participants experience in engaging in these exercises in self-reflection and understanding, computer-generated models (Picture 2) have additional value because they afford increased capacities, such as consolidating images, unifying diverse representations, and thus enhancing comparability, but also for storage, analytical depth and various kinds of modification.

Picture 1  Hand-made Dominant Language Constellations produced at the DLC session at the Oranim Academic College of Education, December 2019

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

9

Picture 2  DLC computer-generated model of a Russian/English/Hebrew DLC.  Produced by Laurent Moccozet

Through these processes unified data available for modifications and further use by other researchers can be accumulated. As a result of practical experience of this kind, such as tracing the dynamics of individual DLCs in time and space, dealing with smaller and larger cohorts of language users, comparing and categorizing DLCs, identifying similarities and differing patterns in connection with certain events and in various settings, the design of specific DLC-related software is being undertaken. It is anticipated that DLC computer-assisted modelling will enhance comprehension of larger data sets, or more complex and fluid portions of information on the dynamics of multilingualism in the lived experience of diverse groups of people. Unlike depictions which might show on one chart the component languages of a DLC, the language repertoire and further languages that are less known or recognized, the DLC models discussed above focus on the patterned and coherent character of a DLC and reveal the interrelations between the languages, the linguistic distance between them, and the degrees of proficiency of a speaker in each component language. These maps and models underscore our argument for seeing the DLC as a unitary whole rather than a disparate collection of unintegrated languages and skills.

1  About This Volume This volume is one of the two interconnected edited volumes on the topic of DLC under the general supervision of Joseph Lo Bianco. The second forthcoming volume titled “Dominant Language Constellations Approach in Education and Language Acquisition” edited by Larissa Aronin and Eva Vetter will focus more directly on the role and relevance of DLCs in education. The present volume is divided into three parts. Part I, is dedicated to providing an account of current developments in DLC research. It includes four chapters that taken together outline the currently perceivable scope of the DLC concept and

10

J. Lo Bianco and L. Aronin

how it can form an approach to the wider examination of multilingulism in education and society. In Chapter 2 “Dominant Language Constellation as an Approach for Studying Multilingual Practices”, Larissa Aronin offers a new theoretical account for employing DLC as a supplement and extension to complement established methods of research in multilingualism and applied linguistics. In particular, Aronin discusses theoretical complexity and the longue durée perspective on events, as put forward by historian Fernand Braudel. Both complexity theory and the longue durée are examined as forms of new footing to ground the DLC approach in historical time and dynamics of social phenomena. In Chapter 3 “Meeting of Concepts and Praxis: Multilingualism, Language Policy and the Dominant Language Constellation”, Joseph Lo Bianco discusses the theory and practice of language policy and its connections, actual and potential, with the DLC. He explores policy making as forms of argument and policy as political action and reveals two sets of DLCs in dynamic tension: the policy makers intended ‘ideal state’ and the policy makers’ perceived ‘problem state’. The difference between what socio-linguistically empirically exists in the demography, as perceived by different interests, is the problem state. The intended, resolved, linguistic dispensation is the ideal state. These vary in the perception, political ideology, and level of information according to different interests. He argues that the concept of DLC holds the promise of aligning academic understandings of DLC more closely with those more commonly found among public officials and the wider non-specialist community, and argues that DLCs offer a potential increase in the traction that multilingualism research could have in the realm of public policy. Lo Bianco proposes the extension of the DLC concept to incorporate script and orthography issues. Chapter 4, “Where DLC Meets Multilingual Syntactic Development”, by Eva Fernández-Berkeś and Suzanne Flynn, meticulously and creatively fuses the DLC approach, the generative view on language and the Cumulative Enhancement Model for Language Acquisition (Flynn et al. 2004). This chapter reports a rare study of cross-linguistic influence at a syntactic level and is the first DLC research undertaken on developmental syntax. Their study aims to delineate the syntactic knowledge that learners can draw on when starting their next languages and uses DLC as an evaluation tool to this end. The authors stress the importance of discovering how languages might connect in a multilingual subject’s mind, as an alternative to comparing the achievements of multilinguals to those of monolinguals. The authors suggest that rather than focusing on the temporal sequencing of previous linguistic input, that developmental studies of language acquisition take different DLCs as the linguistic background for learners to acquire a subsequent language. In Chapter 5, “Shifting and Multi-Layered Dominant Language Constellations in Dynamic Multilingual Contexts: African Perspective”, Felix Banda explores the multi-layered DLCs of Zambia as they operate simultaneously across individual, household, community, regional and national boundaries. Advancing the field theoretically, the author provides the rich depiction of the established fluid and intense multilingual contexts found in many African settings, and shows how DLCs apply

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

11

to such complex configurations and multiple strata of communication. The chapter also looks into the historical and political circumstances of Zambian multilingualism to uncover the roots of its sociolinguistic diversity. The current roles of seventy-­ two indigenous languages, in particular the seven designated as official in the country  – Bemba, Nyanja, Lozi, Luvale, Tonga, Kaonde and Lunda and also of English are described and analyzed with regard of how they work within the typical DLC units in social and educational domains. Part II, Institutional expressions of DLC, contains three chapters dealing with language background profiling at Canadian Elementary Schools, societal versus individual DLC patterns in the Finnish educational context and DLCs in the language repertoires of multilingual South African students. Part II of the volume opens with Chapter 6, “Societal Versus Individual Patterns of DLCs in a Finnish Educational Context: Present State and Challenges for the Future” written by Siv Björklund, Mikaela Björklund and Kaj Sjöholm. The authors, teacher education specialists believe in the potential of officially bilingual Finland –with Finnish and Swedish the two national languages – for the development of successful policies and practices of a more wideranging multilingual education. The increase of linguistic diversity brought by the influx of immigration and recent international and internal events are changing the national linguistic dispensation and inherited expectations for languages of the country. They argue these developments render the current situation an opportune moment to study the emerging societal and individual Dominant Language Constellations. The chapter contains thorough analyses of the two national curricula of 2004 and 2014 yielding valuable findings regarding the competence and dominance, domain-range and functions, of different languages at a society level. The societally programmed DLCs are compared to the real-life individual DLCs of student teachers in Swedish-­ medium teacher education. The discussion on the identification of the participants as a monolingual, bilingual or multilingual self, demonstrates the need for developing multilingual awareness among student teachers in Finland. Chapter 7, “Language Background Profiling at Canadian Elementary Schools and Dominant Language Constellations”, by Nikolay Slavkov discusses in a theoretically rich way the critical issue of profiling in educational systems of the world. The study specifically examines Canadian contexts but its conclusions and recommendations for how profiling can be made more meaningful and accurate, while capturing linguistic diversity in open and inclusive ways, are valid for other multilingual educational contexts of the world. The practical suggestions for improving the language background profiling at public schools by deploying a convincing multilingualism rationale, than doing relying on an unexamined monolingual perspective, contributes to language policy considerations, administrative and political dealings with multilingual populations in Canada and in other localities. The chapter is also informative as to the use and limitations of language surveys. Chapter 8, “Dominant Language Constellations in the Language Repertoires of Multilingual South African Students, by Susan Coetzee-Van Rooy, broadens the range of methodology in multilingualism studies. The author presents the Specific Language Repertoire Survey Instrument as a productive tool allowing the students

12

J. Lo Bianco and L. Aronin

of urban multilingualism to gather a baseline information pool on their language repertoire and Dominant Language Constellations. Based on the survey data, this study delivers rich and essential information concerning the multilinguals with Southern Sotho and Zulu home languages. Coetzee-Van Rooy investigates the potential of combining methodologies using the language repertoire survey and the DLC approach. She demonstrates how the quantitative statistical information derived from the language repertoire, and DLC as a component part of this, obtained with the help of the SLRS instrument can enhance our understanding of the nature of current urban multilingualism in South Africa and enable comparisons across diverse contexts. Among the findings reported in the chapter is the estimation of the size of a typical DLC in South Africa, as well as theoretical claims about translanguaging as a phenomenon related to multilingualism. Part III of the volume is devoted to personal and group experiences with DLC. It consists of four chapters, describing and analyzing the learning and real life experiences of multilinguals with their DLCs in national settings as different as Germany, Sri Lanka, Cyprus, Croatia, Morocco and Spain. Chapter 9, “The Evolution of Personal Dominant Language Constellations Based on the Amount of Usage of the Languages” by Sarasy Kannangara traces her personal and professional DLC over time and across geographic space, between Sri Lanka and Germany. The author offers an emic perspective as language user-­ researcher. The description of Kannangara’s own language usage is richly illustrated by DLC maps in each meaningful life span. Beyond the personal account conveyed in the chapter, in its particularity and specificity, we can identify the potential of DLCs as a model of self-depiction for any individual in the globalized world of personal mobility and professional life. The author offers insights on the functions allocation for the DLC languages, and how individual DLCs emerge from the complex workings of personal aspirations, histories and conditions on the one hand, and global influences on the other hand. Chapter 10 “Dominant Language Constellations of Russian Speakers in Cyprus”, by Svaitlana Karpava examines the language use and identity dynamics in three groups of female Russian speakers in Cyprus. The author distinguishes between the participants in monolingual and mixed-marriage families and students. The analysis of Dominant Language Constellations of Russian, English and Greek revealed in the study that Karpava conducted shows social and linguistic trends in language behaviour and attitudes, family language policies and the wider language transmission policies of female Russian speaking immigrants in Cyprus. Chapter 11, “Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant Language Constellation”, by Stela Letica Krevelj, analyses and compares the experiences of the two groups of multilingual speakers with similar dominant language constellations. The chapter presents the results of quantitative research on simultaneous interaction of languages in two non-identical configurations of the common (‘the same”) DLC, which include Croatian, Italian and English. Examining the composition of the DLCs of Croatian and Italian speaking subjects, through the variables of proficiency, language exposure and mode of learning, as well as differing domains of language usage, the author captures the

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

13

sociolinguistic reality of two functional communities in the Istria county of Croatia. A detailed description and analysis of a common DLC of multilingual participants leads to a number of theoretical insights regarding the role of proficiency in cross-­ linguistic influence, research on identity in language contact zones and DLC as a research tool. The specificity and value of this study is that the author suggests studying cross-linguistic influence through DLC in retrospect and shows a new way for quantitative studies, using holistic approach. Chapter 12, is entitled by “A Dominant Language Constellations Case Study on Language Use and the Affective Domain”, by Richard Nightingale. This is a case study presented and analyzed by a sociolinguist through the reflections of a respondent protagonist, who is, himself a critical participant analyst with regard to the reasons and emotions of his own languages use. The chapter contributes both to the DLC line of research and to growing field of language research on attitudes, emotions and identity studies. Nightingale uses quantitative data and DLC analysis to investigate attitudes, emotion, and identity as they come about through the workings of the entire DLC of the participant. In a series of semi-structured interviews with the participant who speaks Spanish, Darija, Arabic and also Catalan, Italian, French and English the discoveries progressively shed light on at least one way of how a DLC operates in time and space in changing sociolinguistic contexts. The author convincingly shows how the reconfiguring of an individual’s DLC due to the changed environment significantly depends on the particular individual’s personality, choices, preferences and agency. In the concluding Chapter 13, “Quo Vadis, DLC?” Joseph Lo Bianco reviews the concept of the DLC as a heuristic, discusses the overall volume and summarizes the current development of concepts, new findings and their practical applications as canvassed in the research reported in the book. The chapter suggests possible lines of development contained by the methods and topics explored by research projects reported in the various chapters, and what future lines thinking for the DLC concept could be in the frameworks of sociolinguistics and multilingualism. This collection of research and reflection on DLC will be of use to researchers and practitioners in education and multilingualism as well as to undergraduate and postgraduate students studying education, educational linguistics, language policy and planning, applied linguistics, sociolinguistics, and multilingualism. We anticipate the volume will be also useful to students from a wide variety of disciplines, including social work, teacher training and psychology. The chapters related to globalization processes leading to changes in education and society may interest those working in globalization studies and political science. We hope the volume attracts the interest of readers who wish to acquaint themselves with current research and thinking in relation to particular aspects of multilingualism, such as the learning experience both within and beyond the classroom and governmental and institutional policymaking.

14

J. Lo Bianco and L. Aronin

References Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2007). Current multilingualism as a new linguistic world order (CLCS Occasional Paper #67). Dublin: Trinity College. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence. Cambridge: Cambridge University Press. Aronin, L. (2019a). What is multilingualism? In D. Singleton & L. Aronin (Eds.), Twelve lectures in multilingualism (pp. 3–34). Bristol: Multilingual Matters. Aronin, L. (2019b). Dominant language constellation as a method of research. In E.  Vetter & U.  Jessner (Eds.), International research on multilingualism breaking with the monolingual perspective (pp. 13–26). Cham: Springer. Aronin, L., & Singleton, D. (2008). Multilingualism as a new linguistic dispensation. International Journal of Multilingualism, 5(1), 1–16. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Czajka, M., & Hein de Haas, H. (2014). The globalization of migration: Has the world become more migratory? International Migration Review, 28(2), 283–323. Danaher, J. (2016). How do we enhance cognition through external representations? Five Ways. A post in Philosophical Disquisitions posted Ort 13, 2016. https://ieet.org/index.php/IEET2/ more/Danaher20161013. Last accessed 24 Aug 2019. Fishman, J. (1998). The new linguistic order. Foreign Policy, 113(Winter), 26–40. Flynn, S., Foley, C., & Vinnitskaya, I. (2004). The cumulative-enhancement model for language acquisition: Comparing adults’ and children’s patterns of development in first, second and third language acquisition of relative clauses. The International Journal of Multilingualism, 1(1), 3–16. García, O. (2013). Informal bilingual acquisition: Dynamic spaces for language education. In D. Singleton, J. A. Fishman, L. Aronin, & M. Ó Laoire(Eds.), Current multilingualism: A new linguistic dispensation (pp. 99–118). Berlin: Mouton de Gruyter. García, O., & Li Wei. (2014). Translanguaging: Language, bilingualism and education. Basingstoke: Palgrave Macmillan. García, O., & Li Wei. (2015). Translanguaging, bilingualism and bilingual education. In W. Wright, S. Boun, & O. García (Eds.), Handbook of bilingual education (pp. 223–240). Malden: Wiley. Kirsh, D. (2010). Thinking with external representations. AI and Society, 25(4), 441–454. https:// philpapers.org/rec/KIRTWE. Kramsch, C., & Jessner, U. (Eds.). (2015). The multilingual challenge. Berlin: De Gruyter. Li Wei. (2018). Translanguaging as a practical theory of language. Applied Linguistics, 39(1), 9–30. https://doi.org/10.1093/applin/amx039. Lo Bianco, J. (2014). A cerebration of language diversity, language policy, and politics in education. Review of Research in Education, 38, 312–331. MacSwan, J. (2017). A multilingual perspective on translanguaging. American Educational Research Journal, 54(1), 167–201. May, S. (2014). The multilingual turn: Implications for SLA, TESOL and bilingual education. New York: Routledge. Otheguy, R., García, O., & Reid, W. (2015). Clarifying translanguaging and deconstructing named languages: A perspective from linguistics. Applied Linguistics Review, 6(3), 281–307. Shaw, J. (2014). Why “Big Data” is a big deal. Harvard Magazine, March–April 2014, pp. 30–35 and 74–75. https://harvardmagazine.com/2014/03/why-big-data-is-a-big-deal

Introduction: The Dominant Language Constellations: A New Perspective on Multi....

15

Singleton, D., Fishman, J., Aronin, L., & Ó Laoire, M. (Eds.). (2013). Current multilingualism: A new linguistic dispensation. Berlin: Mouton de Gruyter. Vertovec, S. (2014). Reading ‘super-diversity’. In B.  Anderson & M.  Keith (Eds.), Migration: A COMPAS anthology. Oxford: COMPAS.  See u­ploads/2014/02/Vertovec_ COMPASMigrationAnthology.pdf.

Part I

Current Developments of DLC

Dominant Language Constellations as an Approach for Studying Multilingual Practices Larissa Aronin

Abstract  The Dominant Language Constellation (DLC) is a group of one’s most important, vehicle languages, functioning as a whole, and enabling an individual to meet all needs in a multilingual environment. Unlike linguistic repertoire, which embraces all the skills and registers of all the acquired languages, the Dominant Language Constellation includes only the most expedient languages for a person. (Aronin L. Dominant language constellations: an approach to multilingualism studies. In: Ó Laoire M (ed) Multilingualism in educational settings. Schneider Publications, Hohengehren, pp  140–159, 2006: 145; Aronin L. Multicompetence and dominant language constellation. In: Cook V, Wei L (eds) The cambridge handbook of linguistic multicompetence. Cambridge University Press, Cambridge, pp 142–163, 2016). DLC as a model delimits, specifies, and systematises the data regarding how multilinguals and groups deal with multiple languages concurrently. Communal DLCs are manifestations of how societal multilingualism occurs through linguistic practices. Grounding on the complexity approach and Braudel’s longue durée perspective the chapter explores theoretical foundations for employing DLC as a unit of analysis of language practices and additional perspective to the time-honoured methods of research in multilingualism. Keywords  Multilingualism · Super-diversity · Complexity · Dominant Language Constellation (DLC) · longue durée · Fernand Braudel · Visualizations · New linguistic dispensation

L. Aronin (*) Oranim Academic College of Education, Tivon, Israel e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_2

19

20

L. Aronin

1  Introduction Dominant Language Constellations (DLCs) is an emerging approach in applied linguistics and multilingualism for studying the current global distribution and use of languages. It is premised on global transformations that have taken place since the mid-twentieth well into the twenty-first century. The processes of glocalisation (global +local) and ensuing variously perceived developments of the period (e.g., Vermeulen and van den Akker 2010; Turner 2011, 2015; Kirby 2006, 2009) rendered different not only what we do, but also what we think, how we feel (Bauman 1999; Cilliers 2005; Urry 2000), and even, perhaps, who we are (Hayles 1993; Tofts et al. 2003; Kirby 2006). Social practices and activities that we have been performing for centuries have undergone a notable makeover. Language practices are no exception. The remarkable change in the role of languages in the contemporary world has become apparent (e.g., Fishman 1998; Maurais 2003). The volume that happened to be one of the last publications (if not the last) of the prominent sociologist of language Joshua Fishman (1926–2015) is titled ‘Current Multilingualism: A New Linguistic Dispensation’ (Singleton et al. 2013). Its thesis can be summarised as follows: ‘current multilingualism, although overlapping in very many respects with the multilingual arrangements of the past, represents an emergent and special linguistico-­social global condition’ (Aronin et al. 2013: 3). The condition branded the New Linguistic Dispensation is the modern-day distribution, use and treatment of languages which display new common patterns and interaction regularities throughout the world (Aronin 2007, 2017: 176; Aronin and Singleton 2008). Under the New Linguistic Dispensation, both qualitative and quantitative shifts have occurred from earlier ways in which people used and referred to their languages. In the first place, multilingualism in its present form has become an inherent and crucial property of society. This is in contrast to previous multilingual arrangements in the world, which although important too were far from being as central as they are now (Singleton et al. 2013). The multilingualism of the past was largely circumstantial, unevenly spread between groups and individuals. Being multilingual at that time was important for reaching specifically limited (but not unimportant) aims; multilingualism did not determine the development of humanity (see more on this Aronin 2007; Aronin and Singleton 2008, 2012). Changes concern a number of aspects of the language-society interface, including which language varieties we now call ‘languages’, which of them are accepted as a legitimate official communication medium, and which less-often-used languages, pidgins, urban dialects or mixed varieties captivate researchers’ attention. Not only languages and dialects experience a linguistic change, which is a normal process through time, but the ways of dealing with our linguistic assets have changed significantly. Today, multilingual practices comply with the current state of the world and new patterns are deployed anywhere from small groups to international, transnational corporations. In addition to the essential task of human communication, multiple languages have become inherent to the core undertakings of humankind, such as industry and business, medicine and earth sustainability activities, politics and state

Dominant Language Constellations as an Approach for Studying Multilingual Practices

21

development, education and arts. The change is noticeable in a wide range of geographical and social contexts across the globe and in local scales. With that being said, it would be wrong to think that language practices are limited to only multilingual arrangements. In fact, monolingual, bilingual and multilingual arrangements of today’s world intermix. There exist many bilingual spaces and less numerous monolingual niches, but multilingualism has proliferated significantly more in both importance and territorial distribution (Aronin 2015, 2019a). Multilingualism is the leading arrangement in the New Linguistic Dispensation. People always use languages for a reason which lies in the particular activity they are involved in. Purely linguistic activities for the sake of saying something hardly exist. One may say that utterances of young children, for example, may have a linguistic only aim of speaking, but even in this case drills in linguistic aspects do have a purpose and are performed with a concrete aim of language skills acquisition in mind. ‘Linguistic practices’ always and necessarily imply social practices interfaced with linguistic activities. Therefore, the term ‘multilingual practices’ here includes common human practices, both individual and group, ordinary and not so ordinary life activities, which are performed daily, regularly or sometimes, and which normally go together and are completed by language use. In other words, the term ‘multilingual practices’ broadly means performing work, communicating, engaging in entertainment, business, cooperation, tourism, teaching and learning, translation and interpreting activities, thinking and perceiving, or doing sport. For all these activities, both primaeval and newly appeared in modern days, languages are involved for the sake of carrying out various activities. Furthermore, understood broadly, multilingualism practices are carried out by people from all walks of life and of various professions, ages, belonging to various organisations, groups and societies. In the first place, these practices have a reference to the people who use multiple languages (two, three and more) in their daily lives. At the same time, those who might not be multilingual themselves but live in multilingual contexts are necessarily implicated in multilingual practices: contemporary human condition leaves no chance for most people to stay monolingual and monocultural. Therefore, we can safely assume multilingual practices refer to that what we do, in all the places across the globe, and not only to multilinguals but rather to the majority of the world’s population. Given the global changes, the new linguistic dispensation and the worldwide spread of multilingual practices a host of questions arise. Among them, the following general ones: In what ways have global multilingual practices changed to match the current globalisation transformations of the world? Is it possible to find an answer at all, given the super-diversity of populations, language varieties and sociolinguistic contexts? And if yes, how to capture the complex reality of these multilingual practices for the purposes of their examination? And which theoretical grounds could help this investigation best?

22

L. Aronin

This chapter attempts to provide a first approximation to the issue. It is devoted to probing the relevance of the DLC approach to the current research tasks in language policy, sociolinguistics, multilingualism, education and translation studies. To this end, in the first section of this chapter, I will address two scientific approaches that tap into the DLC approach: complexity and the longue durée. The second section of the chapter will be devoted to further developing the concept and approach of DLCs.

2  A  pproaches in Social Sciences Illuminating Current Multilingualism While many of the earlier and more recent perspectives in social sciences and philosophy are of value to multilingual research, in the present chapter I will confine the overview to the two distinct but compatible approaches: Complexity/Dynamic Systems and Longue durée introduced by a French historian Fernand Braudel in 1958 (Braudel 1958a, b, 1995). The two important perspectives explicitly overlap with each other in main premises and some sociologists see the longue durée as one of the important precursors for the systems view of the world (Lee 2018; Wallerstein 2013). While the complexity and dynamic systems theory has already been employed in multilingualism, the longue durée has not yet been introduced to multilingualism studies. In this section, I will attempt to show how both theoretical perspectives underlie the DLC approach and tap into it.

2.1  Complexity The complexity/dynamic systems perspective is an appreciated methodology in many fields of knowledge. In multilingualism and second language acquisition, it is most explicitly presented in the works of Larsen-Freeman (2012, 2016) and Jessner (2008, 2012; see also Herdina and Jessner 2002). Aronin and Jessner (2015) underline how multilingualism has a more complex nature than that of bilingualism and examine the implications of seeing multilingualism as a complex phenomenon. There are a number of features of complexity that are especially relevant with regard to understanding and dealing with contemporary largely multilingual language practices that are largely performed through Dominant Language Constellations. • Being an individual autonomous system and also as a part of a larger whole Complexity thinking sees phenomena (individual systems) as having a double nature (see, e.g. Capra 2015; Capra and Luisi 2014: 65). In other words, each aspect of language practices, e.g. languages, classes, teaching sessions, function both as

Dominant Language Constellations as an Approach for Studying Multilingual Practices

23

being an individual autonomous system and also as a part of a larger whole. Accordingly, a dialect or a language, language use in a particular family, language policy at a particular school, has individual autonomy. At the same time, it is always a part of a larger whole, such as language policy of an educational system and language policy of a country. In the same vein, a multilingual individual can be seen as both an autonomous multilingual phenomenon and a part of the entire population of multilinguals at a university or in a country. Multilingual subsystems can be split into smaller autonomous subsystems and be endlessly studied. A particular language, as a larger whole, contains phonetics, morphology, syntax, pragmatics and other components. Each discipline studying a particular linguistic subsystem has its own tenets, terminology and research methods. For example, sociolinguistics deals with language variation, while phonologists investigate different patterns of sounds in different positions in words within each language or in various languages. As Fernández-Berkes & Flynn (this volume, p. 58) rightly noted ‘[t]he endeavor to study multilingualism must lead linguistic disciplines to define their own scope where each can apply their own tools tom disclose a part of the puzzle that reveals what multilingualism is about’. Accordingly, when contemplating research, we have to bear in mind that, at each level of complexity, the observed phenomena exhibit properties that do not exist at the lower level. For instance, morphology study will not comprehensively describe linguistic practices in a school. In fact, it is not supposed to at the level of solely linguistic oriented research and it has other purely linguistic goals. Similarly, investigation of language use in one language of a DLC will, of course, yield results, but it will not yield a comprehensive, authentic picture of one’s language use. And this is the reason that research of the entire Dominant Language Constellations, rather than the study of learning or use of separate languages is proposed today. In a multilingual world, complexity occurs hand in hand with diversity. Sociologist Steven Vertovek addressed the diversity in the contemporary world by introducing the concept of ‘super-diversity’ (Vertovek 2007). The ‘diversification of diversity’ according to him “has not just occurred in terms of movements of people reflecting more ethnicities, languages and countries of origin, but also with respect to a multiplication of significant variables that affect where, how and with whom people live” (Vertovek 2014, no page, online). This, in its turn, accounts for the existence of super diverse populations of language users. The people of today are speakers, users and learners of numerous language varieties, ranging from international and spoken by millions to minority, lesser-used languages, territorial and social dialects and pidgins. On top of the extreme diversity of language users and languages, there is a vast range of intricately varied contexts of language use. It is obvious that in current multilingual practices, diversity is common and is a norm rather than an exception. Dominant Language Constellations, which follow the common pattern on the one hand, on the other hand, are variedly different and unique for each person or organization and represent the super diversity of today’s world as a drop of water represents the ocean.

24

L. Aronin

• The emergence of new qualities (emergent qualities) According to the complexity theory, the essential properties of an organism or a living system are properties of the whole which none of the parts have (Capra and Luisi 2014: 65). How do the essential properties of the whole different from the sum of its parts emerge? They arise from the interactions between the parts. In this way, the current linguistic dispensation emerges from the extreme variety of languages, individuals, communities, cultures and lifestyles that interact on a multitude of levels. Each multilingual individual is, in fact, a newly emergent phenomenon. The continuous wide-ranging interactions produce ever new situations, new language varieties and new language users, fresh patterns of language use and exponentially generate DLCs. The implication of the primacy of the whole over the parts constructing it is that in order to understand the multilingual practices and multilingualism itself, we have to ponder the emergent systems as units in their entirety rather than isolated parts of which the unit consists. This is because the parts, however important their separate study might be, do not possess the properties of the whole. From this follows that a Dominant Language Constellation is a whole having its unique emergent property and cannot be reduced to a sum of languages that constitute this DLC. • Connectedness and relationships Another central feature of the complexity paradigm especially important for us with regard to understanding and dealing with contemporary language practices is thinking in terms of connectedness and relationships. In multilingualism, these are traditionally understood as interactions between languages, speakers and the settings and also contacts between cultures and lifestyles. All the aspects and dimensions — linguistic, emotional, cognitive, physiological and material – are condensed and tightly interconnected within each DLC. Recent decades brought to the forefront of scholarly attention other exciting connectivity forms. A notable advance has occurred in studying interconnections and multilateral relationships with regard to the human brain, cognition and culture. Our understanding of interconnectedness in the progressively complex world was propelled by the emerging multidisciplinary fields of cultural neuroscience, neurophilosophy and neuroarchaeology which grounded philosophical inferences on the synthesis of archaeological and neuroscience data (Wolfe 2014; Neidich 2014; Malafouris 2010a). The new concept of metaplasticity, ‘the emergent properties of the enactive constituitive intertwining between neural and cultural plasticity’ (Malafouris 2010b: 56; Malafouris 2009; Malafouris and Renfrew 2008) and the notion of the brain-artefact interface relating to the point of intersection between cognition and material culture (Malafouris 2010b) have reified the tight indivisible connection and natural fusion of cognition, body and physical world. Neuroscientists, archaeologists, sociologists and philosophers alike arrived at the realisation that the brain is related to the body. Archaeologist Carl Knappett, the author of ‘Thinking Through Material Culture’ (Knappett 2005) persuades us that artefacts are fundamental to human cognition and perception. Lambros Malafouris, working in the

Dominant Language Constellations as an Approach for Studying Multilingual Practices

25

interdisciplinary areas of anthropology and archaeology of mind, philosophy and semiotics of material culture, in his book How Things Shape the Mind (Malafouris 2013), underscores the inseparability of thought, action and material things. Neuroscientist and Professor of Biological Engineering, Alan Jasanoff, in his most recent book The Biological Mind: How Brain, Body, and Environment Collaborate to make us who we are (Jasanoff 2018), convinces us that our emotions, as much as our brains, are key in mediating how we perceive and interact with the world. The connection and role of emotions in decision-making, memory and communication is the theme in the interdisciplinary works of neurobiologist Antonio Damasio (2010). Thus, a unifying conception of mind, matter and life has been established in the interdisciplinary research. ‘Mind and matter no longer appear to belong to two separate categories, but can be seen as representing two complementary aspects of the phenomenon of life – process and structure’ (Capra 2015: 246). The interconnectedness of cognition, body, language and material world accounts for the expanding of the ‘capacity’ of a DLC from just a linguistic phenomenon to social, physiological, cultural and materially grounded one. This richness of aspects defines a DLC as a multimodal entity, that is, a unit characterised by several different modes of activity or occurrence. Take a language, for example. Each language, in addition to its morphology, phonetics, semantics and syntax also necessarily includes cognitive, psychological, cultural and physical aspects by which it operates in real life. Each modality– be it physical or neural or material  – is worth being examined in the DLC framework.

2.2  Longue Durée The approach put forward by eminent historian Fernand Braudel (1958a, b) is also instrumental for studying current language practices. Braudel’s works explicating longue durée (1958b, 1995) are addressed not only to historians but to researchers in the social sciences whom he called for the unified vision of to his mind, too sharply divided disciplines such as history, geography, anthropology and economy. Unlike complexity, longue durée has not yet made its way to the sociolinguistic discussions, although its main points  – looking at the past and around an event instead of singling it out as an isolated one – are definitely observed by many studies, perhaps, on purely common-sense grounds. Braudel, very much like complexity scholars, emphasises connection and interrelation of systems that exist in reality and argues that they should be studied as such. The major methodology principle of the longue durée (long-term) approach is to engage in broader synthesis to concentrate on slowly evolving structures rather than on ‘eventual history’ (histoire événementielle, a term by François Simiand). Longue durée emphasises the importance of the extended, comprehensive purview reaching across time and space. From this follows that each event in multilingualism is to be treated as a segment of a broader context, long-time period, as having its origins, rich circumstances and reasons.

26

L. Aronin

Braudel also noticed that qualitative studies, even those very carefully carried out, may be and often are very detailed and consider all possible factors. Such separate studies, Braudel warns, lose a lot of their reliability if the researcher does not place his subjects in the context of time and surroundings. Normally, we are not aware of small gradual changes in a DLC concerning languages, context or a multilingual user that eventually leads to greater modifications; instead, we notice only the drastic ones. Following the longue durée vision, it becomes clear that each DLC is just a snapshot in time and place and that it has emerged from the previous situation. From the longue durée and complexity angles of view, any present DLC will be in the longer or shorter course of time followed by its next modified version. With the high probability of every case being totally different, as is often the case with the situations circumscribed by DLCs, continuity and seeing events in a long and broad perspective, in time and space, with their past and future, is a helpful approach. Both the complexity and longue durée approaches offer ways to manage the challenging features of current language practices. It is often the case that whilst the diversity of linguistic practices are mostly celebrated in words, and despite the tangible efforts made to integrate diversity into education, on the practical level diversity very often presents a problem. Unruly variations pose challenges for teaching and communication in and outside the classroom. Given the notorious unpredictability of educational and social outcomes in multilingualism, caused by sensitivity to initial conditions and constant flux in all the spheres of multilingual existence, we may think that contemporary language use and acquisition is in the state of chaos. But it is not. In fact, the multilingualism of today is exceptionally complex and diverse but not chaotic. The situations defined as ‘close to the state of chaos’ have patterns that can be discernible (Bossomaier and Green 1998). Common structural features and patterns are considered more relevant than the specific details of each system (e.g. Langton 1991). Patterns provide certain uniformity and congruity; they are commensurable. A recognisable pattern which allows for deriving meaningful findings and general direction predictions is a good source of knowledge about current language practices. A pattern of Dominant Language Constellation can work as a common denominator for the super diverse phenomena of current multilingualism characterised by a complex mixture of human actions and abilities, languages and political, territorial and social environments. Doing research through this pattern, we might be able to respond to its challenges with more success. In the next section of this chapter, the Dominant Language Constellation concept which serves as such a common denominator is discussed.

3  Dominant Language Constellation (DLC) Dominant Language Constellation is a set of languages that together carry out all the functions of the human language, thus enabling individuals and groups to persist in a multilingual environment (cf. Aronin 2006, 2016). The distinctiveness of

Dominant Language Constellations as an Approach for Studying Multilingual Practices

27

Dominant Language Constellation as a set of languages is in that it includes only the most expedient languages, rather than all the languages known to an individual or used in a community, the latter of which characterise language repertoire. Dominant Language Constellation and Language repertoire are complementary concepts. DLC constitutes an active, working part of the pool of language repertoire and typically comprises three languages. A language repertoire, a kind of linguistic resource or storage, may include a much longer list of languages and skills. Since the main features of DLC were described in earlier publications (Aronin 2006, 2016, 2019a, b, c; Aronin and Singleton 2012: 59–75). I will focus on the issues and topics that have not yet been given due attention.

3.1  How to Define which Languages Constitute a DLC? A frequent query coming from students of DLC and practitioners, teachers and social workers, is how to define which languages are the languages of the DLC and which are those remaining in one’s linguistic repertoire. This is one of the questions to which no definite ‘correct’ answer exists, at least, as yet. No doubt, each of us can articulate the composition of our own DLC. But it is also true, that it is legitimate for an onlooker and external person to have their own thoughts about their family member’s, acquaintance’s, or a neighbour’s DLC on the grounds of observation and sensible judgement which can differ from a self-defined DLC. Both emic (inside, from the perspective of a subject) and etic (outside, from the perspective of the observer) methods of identifying the exact make up of a certain DLC have the right to exist, and perhaps the correct answer to the question of particular DLC constitution would come from considering both perspectives. While there are no strict instructions or an algorithm on how to decide which language resources are included in a DLC and which are not, particular conditions or factors that draw several languages together into a tight workable unit do exist. Analysing experience in individual or cohort studies, we should be able to detect at least some criteria of involvement of a language or a dialect into a DLC. The following is the preliminary considerations regarding the possible criteria for delineating the composition of a DLC. • Carrying out the complete set of functions characteristic of a human language; The foremost condition for uniting of several languages in a DLC is that together they carry out the complete set of functions characteristic of a human language. The division of these functions between languages may be and frequently is uneven. One language can perform most of the functions, and the two others take the rest upon themselves. • ‘Reasonable immediacy’ In order for a DLC to be ‘in a working condition’, the languages should be immediately available for communication. We may tentatively call it the criterion of

28

L. Aronin

reasonable immediacy. The reasonable time of reaction is the one during which communication does not break or lose its initial purpose. A language should be readily available for its user either in order to produce a remark or a written piece (an output) or receive and understand a response. To visualise this criterion, imagine that you are writing SMS or communicating via a WhatsApp with three friends. Imagine also that one friend answers your messages almost immediately, another reacts after some time, but you can be sure that within a particular time period, half of an hour, or 2 h your message will be read, if not answered by your WhatsApp interlocutor. The third friend will not look at your messages for three or more days, and then, usually the communication renders senseless. You will definitely think twice about sending him your messages on an impulse, knowing that the message will miss the point. As a result, the communication with this third friend via this particular communication channel (technology of WhatsApp) would appear to you not time-­ effective, and you will not select this channel for communication. It seems that a similar mechanism of checking time immediacy is at work when languages of DLC are selected. A language, in which a conversation or a written dialogue is too slow or hardly available is not a working language in a real- time and cannot be part of a DLC. The principle of reasonable immediacy of input and output does not require the selected languages of a DLC to be mastered in the highest possible proficiency. In fact, if an L2/L3 beginner with a limited vocabulary can use a language for their daily activities, then this language, especially if it is an official or a native language of a country or community, enters their DLC. Sarasi Kannangara uses the criterion of ‘authentic communication’ saying ‘None of these languages were used for authentic communication; therefore, they do not belong to my DLC’ (this volume, p. 179). This criterion is also only indicated and would benefit from further explication. The above three conditions: (a) carrying out the complete set of functions characteristic of a human language; (b) ‘Reasonable immediacy’; and (c) ‘authentic communication’ are preliminary suggestions regarding the criteria for the definition of the DLC composition. They need further development to be specified into practically usable markers to serve as criteria for ‘admission’ to a Dominant Language Constellation.

Dominant Language Constellations as an Approach for Studying Multilingual Practices

29

3.2  D  LC as a Unit of Analysis of the Current Multilingual Practices Dominant Language Constellations are real and tangible phenomena of human life. They involve real heard and seen languages produced by people in natural physical world situations and real-life milieu. Along with that, a DLC is an abstract construct which represents, embodies, reflects, personifies and reifies multilingual reality. In this capacity, DLC serves as a model for research and a unit of analysis for educational, cognitive and social investigations (Aronin 2019b, c). A DLC–based research study places special attention on the emergent quality of this unit, not equal to the sum of its parts. When we see a set of three languages as an entity, we do not simply register the existence of certain languages but focus on relationships and configurations between them. For the complexity inspired DLC approach, the emergent qualities of the whole are significant aims in research. This can be exemplified by the way strict ballet competition judges look at the ballerina and dancer’s performance beyond registering the correct elements such as clean lines, hands, toes and feet work, turns in pirouettes; but the scoring reflects the performance as a whole – its soul, the emotions it conveys and the lasting impression it leaves. Another example is when a physician receives a blood test analysis, it is not only that he is interested in the list of blood constituents such as platelets, neutrophils and other elements that may potentially indicate health problems. In order to assess how various organs (e.g., kidneys and liver) are functioning, the doctor evaluates disease markers as well as nutrients and waste products in the blood with regard to a particular patient and the overall picture of his emotional and physical health. The overall configuration of parts that comprise the whole and make this whole what it is, is equally critical for studying multilingualism. Thus, the traditional language versus language comparisons simply stating the presence of a number of languages in a multilingual event, as is still the frequent case, is useful, but not fully sufficient to describe multilingualism. Consider by way of example, the fast advancing area of cross-linguistic interactions. In the recent decade, the field experienced the move from the traditional investigation of the interaction between the native language and an additional one, to the focus on the interaction between the non-native languages (De Angelis 2005, 2018; De Angelis and Dewaele 2009). Today the implications of the DLC approach encourage the attention of scholars to the interfaces between the entire sets of languages. Stela Krevelj (this volume) suggests examining the entire DLC units in retrospect, instead of examining the effect of separate variables on the phenomena under study. She believes that “for example, comparative measures of exposure to the languages within the DLC, rather than the variable of exposure to each of the three languages in much more informative in terms of its overall effect on multilingual behaviour” (Krevelj, this volume, p. 255). A propensity for presenting DLC units in visual and tangible forms contributes to the value of DLCs as a unit of analysis. It is a common knowledge that patterns are best perceived visually because pondering visual images helps to think and

30

L. Aronin

compare. From the first studies in the DLC research line, DLC maps have been used (cf. Aronin 2016, introduction, this volume) to schematically describe the languages of DLC, repertoire and also point to some context languages. Later they proved to be instrumental in in-depth case study investigations (see chapters by Richard Nightingale and Sarasi Kannangara in this volume). The most recent kind of visualisation, or rather, tangibilisation of Dominant Language Constellations is playdough modelling of a DLC. These 3D models appear to be instrumental in pondering the particular situations as well as in bringing across the idea of socio-linguistic super diversity and value of each and every DLC. Notwithstanding the variation in DLCs, the frame of DLC neatly categorises the dynamic, diverse and fuzzy reality in local and global sociolinguistic contexts. It provides a structure to rely on and a point of departure for research of languages operating in societies at particular time and space by helping to find similarities and patterns seemingly unmanageable variety of language users, languages and contexts of the contemporary world. Thus, on the one hand, DLC model detects and broadly describes the archetypal. On the other hand, DLC allows for taking a comprehensive look to a desirable extent of each particular case. As far as particular and individual is concerned, DLC model personifies and embodies the contemporary language use patterns in particular real-life representations.

4  Conclusions Multilingual practices of today have spread to most of the communities across the globe and differ from the language practices of earlier times. The major objective of this chapter was to give an account of the Dominant Language Constellation approach that shifts the focus from the investigation of separate languages to the exploration of their constellations. The DLC approach draws on a number of perspectives, in particular, the complexity and the longue durée, which, although originating from different areas of human knowledge, coincide in the important points that they stress. Both perspectives put on the forefront holistic methods of enquiry, interactions, interconnections and relationships between the multiple actors. The second section of the chapter specifically devoted to Dominant Language Constellation discusses DLC as a unit of analysis of the multilingual situations in applied linguistics and sociolinguistics and how some specific features of DLC are conducive to the examination of particular multilingual practices. DLC methodology offers an additional avenue into current multilingual practices a wide range of human activities: in societal management, for carrying out a realistic language policy, in a school or organisation, for communities in the making. It might also turn important for translators and interpreters, teachers and education officers, writers and critics, and other experts in their professional activities.

Dominant Language Constellations as an Approach for Studying Multilingual Practices

31

References Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2007). Current multilingualism as a new linguistic world order (CLCS Occasional Paper #67). Dublin: Trinity College. Aronin, L. (2015). Current multilingualism and new developments in multilingual research. In P. Safont-Jordà & L. Portoles (Eds.), Multilingual development in the classroom: Current findings from research (pp. 1–27). Cambridge Scholars Publishing. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L. (2017). Conceptualizations of multilingualism: An affordances perspective. Critical Multilingualism Studies, 5(1), 7–42. http://cms.arizona.edu/index.php/multilingual/article/ view/109/163. Aronin, L. (2019a). What is multilingualism? In D. Singleton & L. Aronin (Eds.), Twelve lectures on multilingualism (pp. 3–34). Bristol: Multilingual Matters. Aronin, L. (2019b). Dominant language constellation as a method of research. In E.  Vetter & U. Jessner (Eds.), International research on multilingualism: Breaking with the monolingual perspective (pp. 13–26). Springer. Aronin, L. (2019c). Challenges of multilingual education: Streamlining affordances through dominant language constellations. Stellenbosch Papers in Linguistics Plus, 58, 235–256. Aronin, L., & Jessner, U. (2015). Understanding current multilingualism: What can the butterfly tell us? In C. Kramsch & U. Jessner (Eds.), The multilingual challenge (pp. 271–291). Berlin: De Gruyter. Aronin, L., & Singleton, D. (2008). Multilingualism as a new linguistic dispensation. International Journal of Multilingualism, 5(1), 1–16. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Aronin, L., Fishman, J., Singleton, D., & Ò Laoire, M. (2013). Introduction. Current multilingualism: A new linguistic dispensation. In D. Singleton, J. A. Fishman, L. Aronin, & M. Ó Laoire (Eds.), Current multilingualism: A new linguistic dispensation (pp. 3–23). Berlin: Mouton de Gruyter. Bauman, Z. (1999). Globalization: The human consequences. Cambridge: Polity Press. Bossomaier, T., & Green, D. (1998). Patterns in the sand: Computers, complexity, and everyday life. Reading: Perseus Books. Braudel, F. (1958a). Histoire et Sciences sociales: La longue durée. In Annales. Économies, Sociétés, Civilisations. 13e année, 4: 725–753. Braudel, F. (1958b). History and the social sciences the longue durée. Review, 32(2), 171–203. Braudel, F. (1995). The Mediterranean and the Mediterranean world in the age of Philip II (Vol. I, 2 Vols., pp. 23–272). Berkeley: University of California Press. Capra, F. (2015). The systems view of life. A unifying conception of mind, matter, and life. Cosmos and History: The Journal of Natural and Social Philosophy, 11(2), 242–249. Capra, F., & Luisi, P. L. (2014). The systems view of life: A unifying vision. Cambridge: Cambridge University Press. Cilliers, P. (2005). On the importance of a certain slowness: Stability, memory and hysteresis in complex systems. In Paper delivered at the Complexity, Science and Society conference. Liverpool. Damasio, A. (2010). Self comes to mind: Constructing the conscious brain. Pantheon Books. De Angelis, G. (2005). Multilingualism and non-native lexical transfer: An identification problem. International Journal of Multilingualism, 2(1), 1–25.

32

L. Aronin

De Angelis, G. (2018). Cross-linguistic influence and multiple language acquisition and use. In D. Singleton & L. Aronin (Eds.), Twelve lectures on multilingualism (pp. 163–177). Bristol: Multilingual Matters. De Angelis, G., & Dewaele, J.-M. (2009). The development of psycholinguistic research on cross-linguistic influence. In L.  Aronin & B.  Hufeisen (Eds.), The exploration of multilingualism: Development of research on L3, multilingualism and multiple language acquisition (pp. 63–77). Amsterdam: John Benjamins. Fishman, J. (1998). The new linguistic order. Foreign Policy, 113(Winter), 26–40. Hayles, K. (1993). Virtual bodies and flickering signifiers, October 66, Fall, October Magazine and Massachusetts Institute of Technology. Herdina, P., & Jessner, U. (2002). A dynamic model of multilingualism: Perspectives of change in psycholinguistics. Clevedon: Multilingual Matters. Jasanoff, A. (2018). The biological mind: How brain, body, and environment collaborate to make us who we are. New York: Basic Books. Jessner, U. (2008). A DST Model of multilingualism and the role of metalinguistic awareness. Modern Language Journal, 92, 270–283. Jessner, U. (2012). Complexity in multilingual systems. The Encyclopedia of Applied Linguistics. Wiley-Blackwell. Kirby, A. (2006). The death of postmodernism and beyond. Philosophy Now, 58. https://philosophynow.org/issues/58/The_Death_of_Postmodernism_And_Beyond. Kirby, A. (2009). Digimodernism: How new technologies dismantle the postmodern and reconfigure our culture. New York: Continuum. Knappett, C. (2005). Thinking through material culture: An interdisciplinary perspective. Philadelphia: University of Pennsylvania Press. Langton, C. G. (1991). Life at the edge of chaos. In Artificial life II. Addison-Wesley. Larsen-Freeman, D. (2012). Complex, dynamic systems: A new transdisciplinary theme for applied linguistics? Language Teaching, 45(2), 202–214. Larsen-Freeman, D. (2016). Classroom-oriented research from a complex systems perspective. Studies in Second Language Learning and Teaching, 6(3), 377–393. Lee, R. E. (2018). Lessons of the Longue Durée: The legacy of Fernand Braudel. Historia Crítica, 69, 69–77. https://doi.org/10.7440/histcrit69.2018.04. Malafouris, L. (2009). ‘Neuroarchaeology’: Exploring the links between neural and cultural plasticity. Progress in Brain Research, 178, 251–259. Malafouris, L. (2010a). Metaplasticity and the human becoming: Principles of neuroarchaeology. Journal of Anthropological Sciences, 88, 49–72. Malafouris, L. (2010b). The brain-artefact interface (BAI): A challenge for archaeology and cultural neuroscience. Social Cognitive and Affective Neuroscience, 5(2–3), 264–273. https://doi. org/10.1093/scan/nsp057. Malafouris, L. (2013). How things shape the mind: A theory of material engagement. Cambridge: MIT-Press. Malafouris, L., & Renfrew, C. (2008). Steps to a ‘neuroarchaeology’ of mind: An introduction. Cambridge Archaeological Journal, 18(3), 381–385. Maurais, J. (2003). Towards a new linguistic world order. In J.  Maurais & M.  Morris (Eds.), Languages in a globalizing world (pp. 13–36). Cambridge: Cambridge University Press. Neidich, W. (2014). The mind’s eye in the age of cognitive capitalism. In C. T. Wolfe (Ed.), Brain theory: Essays in critical neurophilosophy (pp. 264–286). Basingstoke: Palgrave Macmillan. Singleton, D., Fishman, J. A., Aronin, L., & Ó Laoire, M. (2013). Current multilingualism: A new linguistic dispensation. Berlin: Mouton de Gruyter. Tofts, D., Johnson, A., & Cavallaro, A. (Eds.). (2003). Prefiguring cyberculture: An intellectual history. Cambridge: MIT Press. Turner, L. (2011). Metamodernist manifesto. http://www.metamodernism.org/. Retrieved April 15, 2019.

Dominant Language Constellations as an Approach for Studying Multilingual Practices

33

Turner, L. (2015, January 10). Metamodernism: A brief introduction. Berfrois. Retrieved April 15, 2019. Urry, J. (2000). Sociology of time and space. In B. S. Turner (Ed.), The Blackwell companion to social theory (pp. 416–443). Oxford: Blackwell. Vermeulen, T., & van den Akker, R. (2010). Notes on metamodernism. Journal of Aesthetics & Culture, 2, 1. Vertovek, S. (2007). Super-diversity and its implications. Ethnic and Racial Studies, 30(6), 1024–1045. Vertovek, S. (2014). Reading ‘Super-diversity’. In B.  Anderson & M.  Keith (Eds.), Migration: A COMPAS anthology. Oxford: COMPAS. http://compasanthology.co.uk/wp-content/ uploads/2014/02/Vertovec_COMPASMigrationAnthology.pdf. Wallerstein, I. (2013). World-systems analysis. Sociopedia.isa. https://doi. org/10.1177/2056846013114. Wolfe, C. T. (Ed.). (2014). Brain theory: Essays in critical neurophilosophy. Basingstoke: Palgrave Macmillan.

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and the Dominant Language Constellation Joseph Lo Bianco

Abstract  This chapter combines discussion of multilingualism and language policy, with the productive power of the Dominant Language Constellation concept. Multilingualism research has flourished and deepened in recent years, producing many analytically revealing and empirically robust accounts of the super-diverse communicative environment of our world. This body of multilingualism studies makes available to educators, public officials and ordinary citizens, as well as scholars, the lived realities of multiple languages and their presence in the everyday lives of citizens. In public policy settings however, multilingualism studies have had far less traction. This chapter pursues a line of questioning about language policy and planning and the DLC in an exploration of how multilingualism can be linked to public policy formation of states (a prime example is Vietnam), and the personal language planning of individuals and institutions. Key to the discussion is a conception of knowledge linked to the vita activa and praxis of enlightened individuals and scholars seeking linguistic justice, but also the vita contemplativa of conceptual clarification. The DLC is a promising conceptual innovation because it fosters productive dialogue between academic accounts of language diversity and the complex realm of policy and decision making. The chapter concludes by discussing how these domains can be aligned through a shared body of concepts to become mutually comprehensible, and the likely outcomes if academics furnish accounts of demo-­ linguistics that are persuasive and politically tractable. The chapter also offers some new additions to the stock of ideas within the DLC concept, such as the idea of a coherent script cluster within a language grouping, and thereby expands the concept itself. Keywords  Language policy and planning · DLC · Orthography · Script reform · Vietnam · Multilingualism · Praxis · Linguistic repertoire

J. Lo Bianco (*) Language and Literacy Education, University of Melbourne, Melbourne, VIC, Australia e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_3

35

36

J. Lo Bianco

1  Introduction: Concepts and Praxis In her meditation on the “human condition” political philosopher Hannah Arendt (1958) argued that intellectuals should strive for a vita activa, neglected in Western philosophical traditions in favor of the more socially disengaged vita contemplativa. This was a claim for restitution of active life, even for activity itself, in support of a different vision of how knowledge is constructed and how the collective and public lives of humans should to be lived. To these ends Arendt proclaimed the importance of praxis. Requesting that intellectuals engage with real world problems and apply philosophical ideas to everyday experience, recovers concepts about culture, politics and ruling ideas pioneered by Antonio Gramsci in the 1920s. It is not often recognized that Gramsci’s politico-cultural ideas were shaped by reflection on language, including his formal training in linguistics. He links patterns of hegemony in political and economic relations to language, helping to center political discourse and language problems more deeply in public affairs, and reflexively also directs attention to language as an object of social organization through its intergenerational reproduction (Ives 2004). As a Marxist, his interest is principally with social class, political conditions and their revolutionary transformation. However, his conception of the language order of the world we inhabit has far wider resonance (Borg et al. 2002). This concerns how language regimes index a wider order of social conditions, containing hierarchies of standard languages and their relation to non-standard and minoritized dialects, and all of these linked to forms of discursive authority required to maintain this order. How human subjects, especially subaltern classes, acquire knowledge of regimes of language (Kroskrity 2000) and are brought to conceive of these as commonsense and natural, rather than historically contingent and constructed, is critical to how these extant language orders can be remade in more democratic and socially inclusive ways. If a regime of language imposing marginality is retained through the operation of cultural hegemony, it can be disrupted by active, discursive, argumentative praxis, or at least made vulnerable to political action by critically engaged social agents. For Gramsci, the category of intellectuals and philosophers ranges beyond those located in institutions and tied to academic disciplines, to include intellectuals positioned to reflect on practical activity, akin to Arendt’s praxis. With the common aim of breaking the hegemonic power of current arrangements, which rely on acquiescence to these arrangements even by those most disadvantaged by them, both philosophers propose forms of organic intellectual work. Intellectuals must abandon cultural aloofness, but also intellectual aloofness, if they imagine that knowledge of reality can only be gained through removed contemplation. Language policy is an organized form of normative political action. The vita activa advocated by Arendt revives the notion of praxis from antique philosophy. It invokes aware action as solidarity and knowledge among human groups to contest and repudiate alienating forms of representation and governance. When applied to language problems, this constitutes a kind of language policy activism, e.g. policy

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

37

as discourse, directed to change attitudes and disrupt hegemony; and policy as argument, directed to persuade public officials to allocate resources to projects of language (Lo Bianco and Wickert 2001). In this chapter I discuss some aspects of the interaction between the concept of DLC and the activity of language policy. This is exemplified through a specific, and not widely known, historical case of language policy and language activism: the diachronic experience of Vietnam’s process of script reform, which I will compare briefly to script reform in Turkey and elsewhere. Vietnam’s reform of its writing system is an extended remaking of the inherited national language regime, a reform conducted over more than a millennium; cursory reference is made to the much more well known instance of script reform undertaken comparatively rapidly in the formation of the Turkish republic. The Vietnamese example of language policy highlights the powerful presence of an institutional and national DLC, a group or constellation of languages, tractable and specifiable according to the specific languages involved, the social roles they perform and the orthographic conventions they deploy. From this discussion I propose a modification to the DLC concept to add the notion of a ‘script cluster’.

2  DLC: Concept and Characteristics The theoretical force of the Dominant Language Constellation (DLC) concept, as conceived by Larissa Aronin and her colleagues, derives from its particular take on the phenomenon of multilingualism, a near ubiquitous condition of the contemporary world. In this volume Aronin points out that a “…DLC is an autonomous part of a bigger whole system of multilingualism”. I will argue that the DLC represents a significant conceptual corrective to currently popular but policy-undifferentiated characterizations of general multilingualism. The existence of multiple languages in society is currently a popular research topic celebrating diversity and heteroglossia as its most characteristic features. In reality, however, multilingualism is often riven with inequality, injustice and even violence. Multiple language abilities are not uniformly distributed across social groups or geographic areas, nor do they afford equal access to resources, material or symbolic. In fact, one of the most striking features of multilingualism is the sharp hierarchy in value accorded to different named languages in education (Skutnabb-­ Kangas et  al. 2009), the prejudice, inequity and discrimination that accompanies some languages and forms of speech (Eades 2013; Piller 2016), and the chronic and sometimes violent conflict that is associated with states’ refusal or inability to accommodate linguistic minority rights (Lo Bianco 2016, 2019). A DLC does not simply account for the observable presence of a cluster of tightly connected languages in the lives of individuals and institutions. Rather, as an analytical unit within multilingual studies, the DLC concept empirically studies the behaviors of groups who construct clusters of the most important vehicle languages

38

J. Lo Bianco

for themselves and forge these into a coherent meaningful whole. This allows analysts to discuss the impact of multilingualism on social arrangements in ways that are politically realistic, and linked to how policy makers intuitively imagine such arrangements through their own forms of praxis as power-holding officials responsible for the management of public affairs. In this way the DLC concept suggests a possible convergence between academic accounts of how multilingualism is operationalized in real world communication and how many policy actors understand communication. The promise of the DLC is that language activists, policy makers and academic researchers could develop an enriched debate about what should constitute an appropriate language policy, which would ideally extend to the entire scope of the multilingualism present in a given society. If we can more fully take on the vita activa through scholarship and utilize new work in sociolinguistics, perhaps we can improve dialogue between various interests about the content and implementation of language policies that today are so critical to social opportunity and equality, cosmopolitan diversity and transnational futures. In Aronin’s conception, the DLC of an individual, but also a DLC that might prevail in an institution or even an entire society, functions as a categorical whole; dynamic yet integrated into the fabric of communication, and meeting the needs of that individual, institution or society. The languages combined in the DLC function as a coherent and mutually reinforcing whole precisely because they are vehicle languages, selected for their utility and functionality from the available communicative resources. The languages of the DLC all serve to meet particular communication needs of an individual, who distributes the load of his or her specific communicative efforts by drawing on whichever language is appropriate or persuasive, according to the expectations and demands of the socio-economic setting the individual inhabits in general, or requires in a specific instance of communication. This functionality of the use of DLC languages means that the relevant language abilities can range from fully proficient to partial skill targeted at a specific communicative need. An example might be reading ability in one language for a recreational hobby or interest, full-range ability but mostly confined to domestic and community usage for a second language, and compartmentalized academic technical ability in a third language, but in this language the speaker might lack, because they don’t require, command of colloquial expression and informality. This absence of the native speaker cultural register does not impair the individual’s communication effectiveness, it simply reflects the utility criterion of the DLC notion. The DLC is unitary rather than fragmentary, the entire set of functional and coordinated language abilities are integrated, rather than random and unsystematic. However, because the surrounding social milieu is diverse and changing the DLC is itself also dynamic. A DLC reflects social changes but also helps to generate modifications to prevailing communication patterns, informally through how continual use of a cluster of languages creates expectations among interlocutors to do the same, or more deliberately through formal processes of language policy and planning. In such ways a DLC is affected by the processes and activity of explicit language policy and implicit language planning through structuring social arrangements. In turn the

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

39

presence of active DLCs in a community, generated by past implicit language planning, helps to shape assumptions that policy makers, activists, community members, and academic experts, have towards language and multilingualism which might then make their presence felt in education as overt designs for future communication arrangements. On these premises the Dominant Language Constellation has particular utility as a concept relevant to the processes involved in language policy making. Unlike other ways to observe and analyze multilingual sociologies DLCs are normatively constructed because they reflect choices and strategies made by speakers from among what is potentially available to them. In the New Linguistic Dispensation (Aronin 2007, 2017: 176; Aronin and Singleton 2008, 2012; Singleton et al. 2013) Aronin’s further innovation is to focus attention not on accounts of plurality among languages, nor on the landscape which languages inhabit. Instead, the New Linguistic Dispensation focuses attention on the specific configurations of languages and varieties, the ‘constellations’, that prevail in specific settings. This shifts attention away from undifferentiated accounts of linguistic pluralism to precise accounts of domains, settings or contexts, that are characterised by identifiable clusters of languages. It is this that I believe makes the DLC a ‘tractable’ concept with immediate and practical implications for language policy and planning. In policy discussions a tractable problem is one which is relatively solvable, able to be grasped, or that can be conceived of in ways that make it available for the tools of policy analysis to handle according to its conventionalized requirements of measurability, observability, monetization and so forth, from which some form of public administration, costing and intervention is then devised.

3  Multilingualism: Spread and Description The last quarter of the twentieth century and the first quarter of the twenty-first have been characterised by large scale and deep rooted demo-linguistic changes. Societies of former emigration have become societies hosting large numbers of new immigrants from culturally and linguistically different backgrounds. Societies of immigration have continued to receive new immigrants. The large population transfers implied in these changes have made issues of language diversity a characteristic and a challenge for ever larger numbers of institutions. Whether teachers in schools, nurses in hospitals, magistrates at court, public officials collecting tax revenue or regulating small business licenses, or transnational corporations. As a result, almost all vertical social activity (the administration of life) and more and more horizontal social interaction (citizen to citizen life), is imbued with the need to negotiate across linguistic differences. Potentially constituting a radical reconfiguring of the presuppositions of human communication, of what is taken as normal, multilingualism is today coming to be positioned as the “default human condition in terms of current worldwide

40

J. Lo Bianco

demography…arguably our primal human state” (Evans 2017: 34). The scientific constitution of linguistically diverse conceptions of society and communication norms as ineluctably multilingual is also gaining affirmation in international declarations. A recent and significant example is the Salzburg Statement for a Multilingual World, issued by the Salzburg Global Seminar and released globally on International Mother Language Day, 21 February 2018 (Salzburg 2017). The key mediator between academic research and public declarations is the percolation of scholarly ideas into policy frameworks or directly through political agitation in democratic and semi-democratic societies, which typically occurs through social agent networks of scholars, activists and community representatives in interacting (Oldfield and Lo Bianco 2019; Heugh and Skutnabb-Kangas 2010). The new multilingualism research agenda of the past decade has not always been designed with practical application in mind, and some of it seems barely aware of the possibility, yet through percolation processes it can still pose challenges to institutional life. Mass education systems are constructed by states and typically function with ‘nationing’ purposes (Fishman 1971). Historically most nationing has worked to control and limit, and in extreme cases to eradicate, cultural diversity and minority identity in the interests of securing the institutional dominance of a selected code. As a result, national institutions typically function as monocultural and mono-­ literate which in turn they tend to normalise. To take only education as an example we can observe worldwide agitation for multicultural and multilingual forms of curriculum to reflect demo-linguistic realities (UNESCO 2009), but the pressure they contest is the appeal and power of standardization and simplification, rationalized as efficiency and cost-effectiveness, that represent the dominant logic of public administration. In the main, public agitation for more responsive and culturally pluralistic institutions originates in citizen pressure (Lo Bianco and Bal 2016), as activists engage in language policy in the form of discourse and argument. Often this means utilizing scholarly findings, political pressure and new conceptual tools to demand more responsive institutions. The most common of these is the demand, now practically worldwide, to lessen the gap between multilingual social reality in out-of-school life, and the mono-cultural mono-literacy typical of mainstream schools, tempered only by compartmentalized teaching of prestige foreign languages. Widespread moves for acknowledgement of indigenous rights, in recent decades suppressed and often obliterated, are increasingly being supported in legal and administrative affairs and in a multiplicity of ways are shaping new conceptions of the polity in which these communities reside (Coronel-Molina and McCarty 2016). Alongside recognition of Indigenous claims to a culturally distinctive presence in settler colonial societies, there are more conventional ways that diversity is accelerating: migration in its many forms, including the expansion of temporary flows of population movement (international study, tourism), integrating economic markets in goods and services, and the facilitation of extra-local instantaneous communication via hand-held devices. Newly invigorated approaches to diaspora studies (Tölölyan 2012), are increasingly cited by public policy analysts working with sociological and economic researchers, including academies and think tanks, to

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

41

construct policy tractable concepts for national advantage. A clear recent example makes this point powerfully, exemplifying a trend in Australian policy which has traditionally been much more supportive of the teaching of languages framed as being “foreign” than in supporting the maintenance of minority languages. Over some years the Australian Council for Learned Academies built a case for construing immigrant language maintenance as an economic ‘diaspora advantage’ (Rizvi et al. 2016). In this logic a diaspora is defined as a self-conscious group with a claim to a country of family origin, with individuals who identify with this claim and develop emotional attachment to it, and who sustain connections despite dispersal. In the past, these conditions rendered such communities vulnerable to rejection and marginalization and subjected to pressure and demands for full assimilation into mainstream culture, including loss of active use of their languages. However, this repudiation by host country ‘natives’ of diaspora community efforts to retain intergenerational use of diaspora connections, and especially of the idea of language and culture maintenance, is increasingly being re-thought as a project of national advantage in its ability to contribute needed commercial skills and contacts. In this strategic reimagining of past social outsiders as offering commercial advantage to social insiders, social marginality of minorities is instrumentally rebuilt as something advantageous, as a positive asset of language ability, contacts, local knowledge and intimacy (Rizvi et al. 2016). This is a process of language planning which occurs through discourse and influence-seeking, led by intellectuals and supported by communities. It is still far from being realized but is already being taken up by prestige mainstream institutions, yet it too serves to elevate only some language interests, those coming from large economies and for speakers of standard literate forms of those languages. The field of ‘new diaspora studies’ (Cohen and Fischer 2018) increasingly conceives ethnic and linguistic difference as a utilitarian resource available to host societies as well as an identity characteristic of minority communities. The policy connection is to the attitude of receiving society institutions. When receiving societies marshal minority communities’ language and cultural skills, and the connections and interaction they retain in their family and community networks across transnational space, their language abilities are reimagined in policy settings no longer as a problem but as a resource. However, this ‘good fortune’ is reserved only for well-­ positioned communities and this inequality is replete with problematic issues of other kinds, such as the potential for exploitative state uses of community language abilities, as is true of any issue when it is drawn into the instrumental rationality and state interests of public policy settings (Ruiz 2010). Another kind of spread mechanism for contemporary multilingualism research is the urban metropolis. The city in particular, and municipal authority in general, are being asked to tackle language questions (King and Carson 2016; Chik et al. 2019). The rich vein of interest revealed in studies communication in superdiverse environments, especially cities, has generated a proliferation of new terminology and concepts for the practices of social interaction, recognising the dynamism of multivocality, plurilingualism, translanguaging, metrolingualism (Otsuji and Pennycook 2015) and polylanguaging.

42

J. Lo Bianco

Societal multilingual practices such as these are found among Indigenous Australians, across many but perhaps not all of Africa, South and South East Asia, and along the old and new trade routes of Central Asia, but not in all parts of East Asia or in Polynesia, other than among people of the vast liquid spaces of Oceania (Lo Bianco 2015). Across the vast expanse of Oceania there is a wide array of polities which preside over immense and probably unequalled levels of linguistic density in the Melanesian region, while other sociolinguistic realities in Polynesia and Micronesia exhibit radically reduced language diversity and relatively settled linguistic communicative regimes. In these ways multilingualism and place are positioned in more complex ways than historical attachment to land or territory (Elden 2013). The horizontal scattering of populations no longer inevitably produces cultural and linguistic dissolution. This is largely because instantaneous multi-modal communication technologies facilitate aggregation and allow physically dispersed populations to retain unique forms of difference, often centered around languages. The celebration of linguistic diversity that these new potentialities offer is tempered by the tenacious hold of singular national language ideologies that are only marginally less powerful today than in the past (Gramling 2016). Within nation-state structures, especially in education and its assessment and pedagogical routines which privilege single languages of instruction in official state languages and standard prestige literacies, only the taught ‘foreign’ languages, imagined as bounded and separate linguistics codes attached to foreign polities are permitted. In many or even most societies multilingualism is still discussed as threat or challenge to social cohesion. In recent policy documentary work (Lo Bianco and Bal 2016) this is true in Denmark, Thailand, Turkey, Myanmar. Even in previously more accommodating settings, such as Australia and Canada, policy makers increasingly describe linguistic diversity as a kind of society-dissolving threat to established traditions and social cohesion (see Joshee, Peck, Thompson, Chareka and Sears; Özsoy and Bilgi; Wang, Hong and Schapper; Horst in Lo Bianco and Bal 2016), while in Brazil and South Africa where race and cultural affiliation, rather than language diversity, have long been the central feature of national discussions about pluralism, language questions are coming to the fore but not always positively (see Soudien and McKinney; Chamlian and Kolwalewski; in Lo Bianco and Bal 2016). The still dominant way to think about multilingualism, in practice if no longer in theorisation, is for political systems and institutions to “manage” it (Little et al. 2014). Contemporary multilingualism therefore can be contrasted to previous multilingual arrangements in three main ways: dispersion, density and practical challenge. Today, there are fewer non-multilingual societies than at any time since the spread of nation-state ideologies premised on the idea that national unity required linguistic uniformity. In addition to population movements, this represents the defeat of extreme monolingual nation-making practices and of aggressive assimilation practices. This greater dispersion of multilingual contexts is so extensive that we can speak of a near universality of the phenomenon of demo-multilingualism, or language diversity at the population level. In addition to spread, we can also observe

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

43

greater density of multilingualism, involving intensity and depth, so that multilingual configurations merge with a proliferating number of semiotic systems and forms of difference and identity in single polities. The third characteristic of multilingualism today resides in the depth and persistence of the practical challenge that multiple languages and multilingual citizenries pose to public administration, education and commercial life. The study of multilingualism is currently enjoying a rich and productive period, of conceptual innovation, terminological expansion and new fields and methods of research (Conteh and Meier 2014; May 2014). Much of the prolific literature on linguistic diversity is inspired by a heteroglossic inclination (Choi and Ollerhead 2017), or aims to overthrow inherited views that equate linguistic difference with deficit, imposed historically by dominant and supremacist language ideologies on racialized minorities and other subaltern communication groups (Rosa and Flores 2017). In the multiplicity of multilingualism studies the linguistic repertoire is posed as a challenge to the separate specification of languages, countable as bounded entities. The term repertoire is used to characterise all the communicative skills and registers of an individual. This is one point of difference between repertoire and DLC, another extends beyond individual functional use to describe an autonomous part of a bigger whole system of multilingualism. The desire to depict the entire scope of the communicative register (full bounded languages and partial language abilities alike, and other semiotic resources used by an individual) distinguish this from the approach taken by the concept of the Dominant Language Constellation. This will be explored by comparison with the approach of Busch (2012), who advocates a poststructuralist reordering of linguistic repertoire and introduces the practice of individuals depicting their personal language portrait graphically. In this procedure individuals are brought to examine and display how they imagine and represent their range of communicative resources and the meaning of these in their lives, diachronically and synchronically, but also relationally to groups of significance to them. Busch characterises these practices as heteroglossic repertoires, a suggestive notion traced to the concept of linguistic repertoire, likely first conceived in seminal work by JJ Gumperz in the early 1960s. As a pioneer of the field of the ethnography of communication Gumperz described the verbal repertoire as “the totality of linguistic forms regularly employed in the course of socially significant interaction” (1964, p. 137) which he imagined as providing “the weapons of everyday communication. Speakers choose among this arsenal in accordance with the meanings they wish to convey” (p.  138), for which he used data from the agricultural village of Khalapur, near Delhi, India, and the small town of Hemnes, in Norway. The Khalapur and Hemnes repertoires show the “fruitfulness of the verbal repertoire as a sociolinguistic concept” (p.  151) bridging the “gap between grammatical systems and human groups” (p. 151). Busch complements this standard trajectory of anthropologically inspired sociolinguistics with consideration of the much acclaimed contemporary condition of ‘super diversity’ (Vertovec 2007) and post-structural theorising, to incorporate analytical categories of personal desire, political and personal power.

44

J. Lo Bianco

As an anthropologist focused on cultural diversity expressed in and produced by intercultural communication, one of Gumperz’s primary theoretical concerns was to shed light on language processes in social interaction. In this work on various speech communities he directed the attention of scholars who followed him on how boundaries and relations between social groups and their language characteristics and behaviours operate and are perceived and enacted. In regular and dense interaction through shared and exchanged verbal signs, groups cohere and establish their inner sense of boundaries, and hence their linguistic repertoire. Busch melds these Gumperzian notions with the challenge posed by conditions of contemporary, prototypically urban, super-diversity and poststructuralist approaches, exemplified in work by Jacques Derrida and Judith Butler, to explore previously neglected factors such as the power of categories. Many characterisations of communication that focus attention on heteroglossia are inspired by a desire to contest standardizing, normalizing and hierarchical valuations of the communicative order of society. These wish to establish formulations of multilingualism according to principles not merely of post structural critique but of a commitment to a social justice inspired re-ordering of the communicative resources of society. In Southeast Asia a key example of this is the movement for Mother Tongue Based Multilingual Education (https://bangkok.unesco.org/tags/ mtb-mle) directed at solutions to education problems in multilingual post-colonial societies. In keeping with the spirit and principles involved in a justice oriented education approach to multilingualism in marginalized mother tongues some scholars draw a division between a primarily vertical conception of bounded separate languages, and an observed practice of widespread multilingualism that is dynamic in usage, creative in communicative practices but fully functional and effective, yet mostly ignored or stigmatised by education authorities. An example of this is Kathleen Heugh’s “functional multilingualism” (2018) in which multiple language use is distinguished from the hierarchical view of multilingualism that she argues reflects and reproduces nation-state monolingualisms, or with multilingualism as it is understood in Western European settings which is premised on the mastery and use of separate and bounded discrete languages, essentially constituting multiple yet parallel monolingualisms. This latter characterisation of multilingualism is associated with education systems and authorities, legal systems and literary works, and with formal systems of translation and interpreting, with their inclination to strict separation of languages. Heugh reports practices of daily fluid use of multilingualism unconstrained by any desire for keeping to bounded or separation between named languages, as might be inherited by cultural authority or official policy. This is in effect an affirmation of multilingualism in use, of in-between spaces, among languages that are inflected for purpose and function, but is perhaps better understood not as multilingualism so much as the application of semiotic resources through the agency of individuals and communities. However, in Heugh’s work some multilingual individuals were observed to practice language separation as well, either with uninterrupted use of single languages for some purposes, or through language marked code-switching. Some of these people also advocate for dual medium bilingual education of

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

45

functional multilingualism in clearly named and demarcated languages, and this helped to inform the discourse of multilingualism in the immediate post-Apartheid period in South Africa. These analytical approaches are consistent with ‘Southern Theory’ perspectives (Connell 2007; Stroud and Mpendukana 2009) which aim to problematise and overturn asymmetrical patterns of knowledge production which are geopolitically inequitable and privilege systems of characterisation and understanding of social phenomena, including communication, designed by and reflecting the experiences of developed societies and their historically specific, rather than universal, experiences. Rama Kant Agnihotri (2014) has also advocated a similarly dynamic or fluid notion of multilinguality or innate multilinguality proposing that language itself and not just the societal phenomenon of multiple co-existing languages, be seen as a socially enacted multilinguality, constitutive of being human and an educational peace-promoting practice in conflict zones. This horizontal and usage based view of multilingualism contests both popular and scholarly views of language as standardized, homogeneous or pure, with Agnihotri pointing out that some 96% of humanity lives within social systems in which practices of language are highly fluid and mixed rather than rigidly compartmentalized. In Europe official documentation has also moved towards a semblance of Agnihotri and Heugh’s sense of two kinds of multilingualism, but with a strong eye towards formal language policy. The Language Policy Division of the Council of Europe (CoE) has promoted a distinction between multilingualism and plurilingualism, to capture a semantic difference between the multilingual abilities of individuals (plurilingualism), compared to the presence of multiple languages in society, in which there will likely also be many monolinguals, (multilingualism). The deployment of these terms to allow separate specification of individual/collective abilities in several languages was popularised by the CoE in its Guide for the Development of Language Education Policies in Europe (CoE 2007). First proposed in the 2003 draft report, this distinction was adopted formally in the 2007 final report as “…distinction between plurilingualism as a speaker’s competence (being able to use more than one language) and multilingualism as the presence of languages in a given geographical area…a shift, therefore, from a perspective focusing on languages (a state may be referred to as monolingual or multilingual) to one that focuses on speakers” (p.  10), and also marking off the term polyglot for exceptional individuals, since “…plurilingualism cannot be considered the privilege of a “gifted” elite”. (p. 38). In effect identification of a realm of horizontal linguistic fluidity, as distinct from institutional and vertical separation of languages, with its adherence to standard forms and primacy of writing and formal correctness over speech, of accuracy over fluency, can be observed in diverse settings in the world. It is possible to readily articulate sociolinguistic implications of fluid multilingualism, or, to use Agnihotri’s term, multilinguality. At the sociolinguistic level it can represent recognition of a co-operative, inclusive, bridging practice of communication, with fluid movement between named different languages but also between all forms of code integration (mixing, code-switching, multi-modal technology mediation, hybridity,

46

J. Lo Bianco

translanguaging). But how can multilinguality as a depiction of actual observed communication, link to a policy and political connections? How can it inform pedagogy and improve education? Can the horizontal fluid dynamism of communicative reality represent a force to reduce social inequalities and advance recognition of identity formations significant in the lives of large numbers of people, and ultimately to promote functional multilingual learning and enhanced learning? Before such questions could be addressed a further one should be posed. As Rymes (2013) shows, demonstrating from instances of communication drawn from mass media, in everyday life the existence and vitality of communicative repertoires cannot be confined to any kind of language resource alone. The question then becomes, if we are to accept that communicative repertories are fluid, as empirically it is clear that they are, should accounts of communication not incorporate the entire social semiotic expanse of dress, gesture, posture, haptics, kinesics, narration and referencing intertextually, diachronically and synchronically to shared information? Not only a multi-modal and multiple set of resources arises from this, but the forms of differentiation expand greatly, and perhaps to infinity. The creative potential and differentiation are situated and therefore dynamic, achieving specification according to the particular sets of participants, circumstances and purposes for which a specific encounter is understood to arise among participants. It is clear that ethnographically verifiable realities of communication offer potentially limitless nuance, according to the means for identification individual researchers deploy. What is the relation of DLC to these expanded conceptions of multilingualism? The array of rich, productive and interdisciplinary investigations of multilingualism discussed above are complemented by DLC, they are extended and differentiated by DLC, but also constrained by DLC. Unlike ever more particularized ways to see and analyze multilingual data and settings, either undifferentiated as to groups of languages, or overly differentiated according to repertoire diversity and modality, DLC delimits the analytical focus in a productive way by casting light specifically on the vehicle groups or clusters of languages selected by individuals or groups from within the multilingual and semiotically plural range of meaning resources available to them. In this, DLC differs substantively from the accounts just given, shifting attention away from overall or generalized multilingualisms or micro-communication repertories, to clusters of languages, with both extended and limited proficiencies, that have saliency and functionality in the lives of their users. As Aronin has explained the DLC comprises only the most expedient languages a person requires and chooses to activate (Aronin 2006: 145, 2016) and so as a model DLC delimits, specifies, and systematizes data regarding multilingual communication and how groups deal with multiple languages concurrently. In this resides its particular tractability for language policy discussion and its particular implication for language pedagogies.

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

47

4  Language Policy and the DLC The discussion of policy in this chapter is not limited to public policy, meaning the declared and explicit processes of language policy formation by public authorities entrusted with such responsibilities, but also to informal influence and planning. The formation of a DLC reflects private choices made by individuals for their personal needs as well as public determinations imposed on societies top-down through law or resource allocations made by governments. What counts as language policy and planning is widely accepted today to range well beyond formal processes of governmental decision-making, hence the DLC concept should also be considered relevant to less overt forms of language planning that reside in the projection of an individual’s future needs, personal desires, self-image, economic or symbolic interests and identity. Institutions and the entire society are aggregations of individuals, organized and structured into interest groups and various kinds of authority. In this understanding, therefore, we can extend the relevance of DLC to embrace both personal and small-group language planning as well as public and overt language policy making. We can see from this that any DLC represents a kind of settled outcome of processes of language planning, and is always subject to further change and development. Any language planning process can be more or less conscious, more or less autonomously chosen or constrained by social circumstances, more or less overt and formal. A DLC emerges in interaction with societal language resources and their hierarchies and roles, the rewards or benefits individual languages afford or deny. It follows that any DLC is an outcome of what the analytical unit (a person, a family, an ethnic minority group, or an institution or a whole society), needs, chooses, prefers or prioritizes as the order of available and necessary resources for their various exigencies, as they perceive them and to the extent that they have agency and ability to effect their choices. Most of what is ordinarily understood to be language policy is explicit, conscious and deliberate in character, typically associated with government decision making processes and public authority, referred to once by Joshua Fishman as “the authoritative allocation of resources to language”. Importantly, this process of authoritative allocation of resources is “arrived at and implemented in accord with the community’s norms” (Fishman 2009, p. 311). Academic studies in language policy originated in research on how societies address practical language problems and the canonical formulation became (a) the selection, codification, and standardization of norms for new languages to serve administrative, security, and unification agendas of newly independent states and (b) research to support integration and incorporation of Indigenous and immigrant children into mainstream schooling; and (c) academic interest in communication problems in post-colonial nation building in Africa and Asia in the post–World War II era (Spolsky 2012). The DLC accounts for a speaker’s or communities’ extant communicative patterns, while the object of language planners’ attentions to multilingualism existing in a society has usually been to evaluate its fit to the perceived needs of the authority

48

J. Lo Bianco

for whom the planner is designing a political solution. This potentially and often involves a plan to intervene in education and other domains to change existing DLCs informed by future shortfalls of national language skills, or to arrest or reverse shift or loss away from valued languages, or in other ways to modify the stock of available language abilities. Because formal language policies are future-oriented statements a key feature of language policy texts is to conform to a political desire of the issuing authority. The political objectives can be overtly stated or implicit in the text which announces policy and are typically objectives advancing some kind of national consolidation, some specific intention for economic change, some strategic response to geo-­ political security concerns, or some other national purpose believed to be served by a modification to the extant language capabilities of the citizen population over which the authority has political jurisdiction. These elements will be absorbed now in a discussion of a specific language policy experience that combines both formal and informal, explicit and implicit, forms of both policy and planning, and which is extended over centuries of the cultural and political life of a single political jurisdiction.

5  Extending the DLC An examination of the history of the Vietnamese orthography reform and repudiation of colonial language imposition is now explored briefly to provide a diachronic account of a DLC as it was historically constructed, and to recommend the expansion of the concept of DLC to incorporate orthography. In his extended study of the process of Vietnam’s script politics De Francis (1977) points out that while romanization pressure due to European colonial expansion into Asia was experienced by all East Asian societies, with the same claim and judgment as to the inherent superiority of alphabetic writing over alternative forms, Vietnam was and remains the only East Asian society that undertook the radical and society-changing step of full and permanent romanisation. At different times, China (Gottlieb and Chen 2001), Japan and Korea also experienced considerable pressure for script modification deriving from technological and commercial power located in Western modernity and projected by its imperial powers into Asia. In Vietnam, however, the process can be traced to multiple extended anti-colonial struggle. It commenced not with Western but Chinese domination, from the southward incursions of Chinese imperial rule, the installation of a Confucian court at Hanoi from around 939 CE and the depth of literary culture that took root and flourished within the native literati and the Confucian scholars (Lo Bianco 2001). Classical Chinese became the courtly language written in Chu Han, over time contested by a native dissident group using a script based on Chinese characters, and then by romanised writing in Vietnamese. Over a millennium with the progressive emergence of each of the contesting scripts a possibly unique pattern of language, script, colonial affiliation and domestic politics evolved.

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

49

The three writing systems that were in play in the final centuries of this period were Chu nho or Chu Han; Chu Nom or Nom; and the Chu Quoc Ngu or Quoc Ngu, and the effective sequencing between them is as follows: • Chu Han was based on traditional Han Chinese characters, and originated in southern China. It was introduced into northern modern-day Vietnam through the expansion of Chinese rule, and then dominated for hundreds of years under the local dynastic monarchies that succeeded it; • The Chu Nom rival was developed as an indigenous form of the Chu Han, the two were used concurrently, often in opposition to each other, for a period of 800 years; • The Quoc Ngu romanized script was devised by Portuguese and French Catholic missionaries in the mid 1700s, originally devised as a tool to assist their learning of the Vietnamese language, especially noted for its extensive diacritics as tone pronunciation guides. Since 1945 Quoc Ngu has been broadly accepted as the national script and employed in all levels of education. These scripts are associated variously with spoken languages, Classical Chinese and later Mandarin Chinese, with various vintages of standard northern and also later central and southern varieties of spoken Vietnamese, and with French. In effect a DLC extended in time and was asymmetrically acquired by educated administrators, court officials and literati, and others, but not beyond these elite circles. The patterns of communication under colonial rule, Chinese or French, were always diglossic with Vietnamese forms as L and the imposed foreign spoken languages H. Within this however the historical record shows an orthographic nested arrangement (a “script cluster”) of the 3 different writing systems (Lo Bianco 2001), that at various stages of history have occupied usage in different areas of society, and have served distinctive and often antagonistic socio-political, religious and cultural interests. Chu Han was used by the social elite and occupied a hegemonic position for matters of politics and administration (De Francis 1977). When Chu Nom was developed from the Chu Han by the Vietnamese in the thirteenth century, it arose as a dissident script in repudiation of the use of Chu Han in the political domain, and thus a form of resistance to Chinese-modelled aristocracy that used it. Chu Nom was thereafter established as the standard script for both Vietnamese literature and vernacular communication. For many years after its development various political and military leaders promoted the deposition of Chu Han by Chu Nom from the political arena, attempts that were resisted by social elites loyal to the Confucian heritage inherited from the Chinese dynasties that had established aristocratic governance at Hanoi. In this manner, for a period of 800 years, there existed a dynamic but stable digraphia between Chu Han and Chu Nom. The DLC evolving in this remarkable script manifestation of an underlying nationalist and independence-oriented political struggle stretches a millennium of time and is instantiated in specific smaller struggles of schools, teachers, cultural figures, writers and is even expressible at the level of national aspiration. The latter was articulated most clearly by revolutionary communist and independence leader

50

J. Lo Bianco

Ho Chi Minh whose demands for political and cultural unification came to accept a pragmatic role for romanized writing in mass literacy as an aim of unification dependent on defeat of French colonial domination. In this very broad diachronic lens we can identify a sustained or continuous narrative for cultural distinctiveness alongside political autonomy, taking form at different historical junctures and events in radically different ways. As alliances with diverse social groupings and geographical areas of the country were required at different times a complex array of local accommodations were forged. Yet for the period from the evolution of Quoc Ngu as a guide for language study by missionaries to its standardization and adoption in schooling, reinforced by roman alphabet based French education, the three scripts were essential, present and part of a unitary national orthographic cluster. In DLC terms we can represent script as a core element of a DLC as follows at all times administrative elites required Vietnamese + the Contested Colonial Other language, usually both, Chinese or French [and its script] for designated functions. At another historical junctures some elements changed while others were retained. Over all periods the need for knowledge of Vietnamese as the language of governance remained, for nationalists it was obviously central to public mobilisation against colonial rule and generally a national unifying language. However, the other component of the formula shifts from Chinese to French and from Chinese characters to Roman alphabetic orthographic conventions. Vietnamese itself also changes since it is imagined by earlier advocates of cultural autonomy as being the repudiation of Chinese language and characters via a Chinese based writing system (Che Nom) but French is repudiated via a Romanised writing system (Quoc Ngu). Over the long course of struggle various configurations of language/script were proffered by rival interests. While the number of royalists attached to Classical Chinese and more particularly to Character writing had dwindled markedly during the establishment of French rule from the nineteenth century, many remained passionately attached to a Chinese-linked independence for Vietnam. Similarly, adherents for French were also only ever a small elite and over time they too lost ground. A smaller political grouping passionately attached to a French/Vietnamese compact of cultural connection with smaller or greater degrees of political autonomy also emerged. Each of these groupings emphasized variations on the pattern of the three languages and three scripts available in the sociolinguistic and orthographic scope of Vietnam. The struggle for political independence was reflected in how different elites shifted between DLC languages to characterise the national unity that was being forged in various ways. It is clear with the benefit of observing a process unfolding over an extended time that the clusters of writing, what I am calling here the script cluster embedded in the Vietnamese DLC, carry traces of past linguistic arrangements, or ‘regimes’ (Kroskrity 2000). When a new nation is being formed among the possible communicative orders that gain adherents, however realistic or unrealistic, must be considered part of the ideological envelope in which a DLC is manifested. It is important to recognise here that large numbers of Vietnamese were multilingual, speakers of

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

51

Vietnamese with French, Chinese, for the brief period of Japanese occupation during the second world war, of Japanese as well, and later and very strongly today, of English. Vietnam is however an extensively multilingual population. The sub-­ national languages of hill tribes and urban minorities are absent from the depictions of the national DLC, underscoring the political and nation-consolidating functions that the struggled over languages and their scripts represented. During the late nineteenth century and early decades of the twentieth century as French consolidation of its rule over all the territory of Vietnam was secure politically and militarily, but strongly contested culturally, a small number of individuals charting the future cultural fortunes of the entire population generated a DLC that contained a unitary Chinese, French and Vietnamese cluster of languages tied to the script equivalents. These coexisted in a nested trigraphia, a structuring of three orthographies attached to the national language, and two rival languages with their exclusive forms of writing. The domains in which the DLC was enacted included individuals and households, religious sites of Confucian, Buddhist and Catholic adherence, schools, printing presses and their associated small circles of literary or political collaborators, and public administration. Quoc Ngu prevailed effectively with the politics of literacy of the first post-independence era, after the decisive Battle of Dien Bien Phu, part of the First Indochina War (1946–1954), which pitted the French colonial forces against the Viet Minh (both communist and nationalist forces), but prior to full reunification of the country within its currently recognized borders, with the defeat of the United States in 1975 (Lo Bianco 2001). A decisive cultural force were the mass literacy campaigns that extended into the second independence phased, buttressed by a total national effort for universal literacy, women and girls, rural and urban populations, native Vietnamese and ethnic minorities. Via a series of injunctions calling for and largely achieving total mobilisation the communist leadership imposed a moral duty of literacy acquisition and the decisive selection of Quoc Ngu for these purposes both entrenched its presence in the literate life of the country but assisted in overcoming its foreign origins problem. As the literate were implored to educate the non-literate in acts of political solidarity and mobilisation at home, workplace and in public settings, in villages and towns as well as cities, Quoc Ngu supplanted its rivals. In effect this process characterises a kind of dynamic DLC in which it isn’t entire languages that are invoked, but scripts and aspects of functional literacy that are clustered according to the political and cultural interests of diverse groups, constituted according to their social position at different historical junctures. Culturally the independence movement in Vietnam had to resolve the colonial associations of two components of its nested trigraphia. Regarding romanised writing it had to overcome its association with a dominant then a repudiated colonial French overlord class, but also ultimately with Vietnamese ‘collaborators’, and a Catholic-loyal component of both European and native Vietnamese converts. Regarding Chu Nom and Chu Han the cultural struggles were similar in some ways and different in others. Chy Nom was associated with a seditious kind of Chinese language and Chu Han with a Confucian-loyal class of Vietnamese literati tied to classical Chinese culture as a superstrate dominant language. Chu Nom was at least

52

J. Lo Bianco

tied to Vietnamese language but Chu Han relegated the national language to non-­ cultural or minor administrative functions. The script trigraphia complemented but also exposed and highlighted the unique Dominant Language Constellation of Chinese, French and Vietnamese in their struggle to function as national vehicular languages. Injunctions from leaders, cultural and political, can be seen as serving a kind of language planning via discourse and persuasion. I have given this extremely abbreviated account of an extended historical experience of establishing a national language regime to argue that scripts should be included in an enriched and expanded DLC. In its current formulations the DLC concept as it exists in the literature focuses on spoken language. As an assemblage of languages a DLC usually suggests an individual’s construction of the personally useful cluster of national/official language, auxiliary languages and personal preferences or locally desirable and useful languages. In most post-colonial settings pragmatic management of public life has involved compromises involving continued use of current and former colonial states’ languages and their demands to be granted continuing status in education and political/administrative and economic life, struggles around finding an ‘honourable’ or ‘dignified’ place for national language and at local levels, languages of tribal groups, hill populations or minority urban immigrant clusters. While struggle around script is not unique to Vietnam, among the main nations that inherited Chinese Character writing Vietnam is unique as the only nation to fully and permanently romanize its writing. Korea devised the unique Hankul orthography, while Japan retained Chinese Characters supplemented by its indigenous Hiragana/Katakana syllabaries. The fluctuation in the DLC revealed by reflecting on the Vietnamese experience therefore ties to political conditions, and associated political ideologies. However, this does not exhaust the considerations. We can draw a parallel with the process of romanization of Turkish in 1928. Here romanization replaced Arabic script as part of the political and cultural modernisation project associated with Atatürk İnkılâpları (Atatürk’s Reforms). The Ottoman general then republican leader and founding president of Turkey, Mustafa Kemal (known as Atatürk after 1934) was “the guiding spirit of the language reform” (Aytürk 2008, p.  16). Atatürk “cultivated a great interest in linguistics, devoting much of his valuable time between 1932 and 1936 to the creation of a nationalist language policy and a new Turkish free of foreign material” (p. 14) and invested in this process a national modernization aspiration succeeded, in contrast to the failure of the more or less contemporary efforts of a similar now hardly remembered aim to romanize Hebrew, explained by Aytürk (2010) with the notion of “script charisma”. According to Aytürk the presence of script charisma in Turkey but its absence among Jews accounts for the success of language reform in Atatürk İnkılâpları but its opposite with respect to Hebrew. The Vietnamese instance shows that while pure functionality and ease of mass literacy education were motivating forces in the selection of Quoc Ngu and the decisive reasons for its ultimate adoption it required and helped generate a deeper cultural affiliation as part of Fishman’s ‘nationing’ concept. Quoc Ngu Vietnamese became the identificatory

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

53

script of a new kind of political citizenship overcoming its cultural limitations, in conditions of non-democratic politics. Deep rooted social, intellectual and cultural consequences flow and political repercussions threaten from such substantive reforms, so that even the “catastrophic success” (Lewis 1999) of the Turkish reform of 1928–1929, completely enacted, resonates today in new politics of language in Turkey. In Dominant Language Constellations, therefore, we can see clearly in the Vietnamese case that these can be both further differentiated by script, or brought together by script, so that Vietnamese can be linked to Chinese by its use of an admittedly dissident script (Chu Nom) and then distanced from Chinese and linked closer to French by romanization. This in turn represented a repudiation of spoken and written French in domains of education and public life in independent Vietnam. The structure of the three language three script DLC pertaining here is tied together by an often unnoticed or barely remarked script politics and policy determination related to mass national literacy and its social consequences. More research is warranted to applying the DLC and script in contexts of regular sociolinguistics of language use and also in policy settings.

6  In Conclusion The clusters of variables discussed above, and the dynamic relations among them supply a strong additional value to the DLC which is the only characterisation of multilingualism in contemporary sociolinguistic theorising that lends itself to such productive analysis of real world policy/political questions. Most multilingualism discussions are not differentiated according to the clusters of language choices that are typically the focus of policy making and political debate. The various ways to analyse multilingualism discussed earlier are analytically revealing and provide empirically robust accounts. They make available for deeper attention of educators and scholars the lived realities of multiple languages and their presence in the lives of citizens and the realities of cities and societies in the more super-diverse environment of contemporary culture. However, in language planning settings, accounts of language diversity, of repertoires of communication varieties, and of ever more complex kinds of multilingualism tend to have less traction in decision making and resource distribution settings. A part of the reason for this is that public policy settings are essentially problem-solving domains in which the purpose of deliberation is to make and execute choices which then determine funding allocations, program design and implementation and all the associated administrative action. Problems are not neutral however and the pre-policy phase inevitably involves contest about what will be decided as the problem to be addressed and by whom and what action will be taken. In the extended case of politics and policy I have discussed in relation to script reform history in Vietnam we see struggle for national formation and the bitter divergence of culturally antagonistic interests. The key language problem that attracted political intervention of the highest political

54

J. Lo Bianco

authorities in revolutionary times and after state consolidation by institutions was how to intervene to direct the script dimensions of the DLC to support the political ideology of socialist nationalism. The kind of policy making discussed above can be described as policy as argument and policy as political action. It reveals two sets of DLCs in dynamic tension: the policy makers’ intended ‘ideal state’ and the policy makers’ perceived ‘problem state’. The problem state essentially refers to what powerful interests judge to be the deficiencies of what socio-linguistically empirically exists in the demography, the extant condition or the current communication order, judged to need changing. The ideal state refers to the problem state transformed through policy action into the intended, resolved, linguistic dispensation that powerful interests want to produce. The contest between the various prescriptions of the problem base and the idealised future state is conducted between various interests capable of effecting change, and so varies according to their perception, political ideology, level of information and interests. The idealised end point is achieved through disruption to the current DLC, in effect to current actual speakers’ and communities’ DLCs, and the associated economic, social, ethnic and linguistic capabilities and ideologies. Language planning and the DLC merge in the present discussion in formal public policy of states engaged in struggles for national independence, in the vita activa and praxis of enlightened individuals seeking linguistic justice and continuity or demanding on behalf of interests they represent. However, the accounts of scholars seeking to explain and theorize the multiplicity of languages now sharing social space all over the world rarely interact with the policy domain. The Dominant Language Constellation is a promising conceptual innovation because it fosters more productive dialogue between academic accounts of language diversity and the realm of policy. If these domains can be aligned through a shared body of concepts they can become mutually comprehensible, and this will be more likely if academics furnish accounts of demo-linguistics that are persuasive and politically tractable.

References Agnihotri, R.  K. (2014). Multilinguality, education and harmony. International Journal of Multilingualism, 11(3), 364–379. Arendt, H. (1958). The human condition: A study of the central dilemmas facing modern man. Chicago: University of Chicago Press. Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in Educational Settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2007). Current multilingualism as a new linguistic world order. CLCS Occasional Paper #67. Dublin: Trinity College. Aronin, L. (2016). Multi-competence and Dominant Language Constellation). In V. Cook & L. Wei (Eds.), The Cambridge Handbook of Linguistic Multicompetence (pp. 142–163). Cambridge: Cambridge University Press.

A Meeting of Concepts and Praxis: Multilingualism, Language Policy and…

55

Aronin, L. (2017). Conceptualizations of Multilingualism: An Affordances Perspective. Critical Multilingualism Studies. 5, 1(2017), 7–42. http://cms.arizona.edu/index.php/multilingual/ article/view/109/163 Aronin, L., & Singleton, D. (2008). Multilingualism as a New Linguistic Dispensation. International Journal of Multilingualism, 5(1), 1–16. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Aytürk, I. (2008). Politics and language reform in Turkey: The ‘Academy’ debate. Wiener Zeitschrift für die Kunde des Morgenlandes, 98, 13–30. Aytürk, I. (2010). Script Charisma in Hebrew and Turkish: A comparative framework for explaining success and failure of Romanization. Journal of World History, 21(1), 97–130. https://doi. org/10.1353/jwh.0.0106. Borg, C., Buttigieg, J.  A., & Mayo, P. (2002). Gramsci and education. Lanham: Rowan and Littlefield. Busch, B. (2012). The linguistic repertoire revisited. Applied Linguistics, 2012, 1–22. Chik, A., Benson, P., & Moloney, R. (2019). Multilingual Sydney. New York: Routledge. Choi, J., & Ollerhead, S. (Eds.). (2017). Plurilingualism in teaching and learning, complexities across contexts. New York: Taylor and Francis, Routledge. CoE, Council of Europe, (2007) From linguistic diversity to Plurilingual education: Guide for the development of language education policies in Europe. Main Version. Strasbourg, CoE. Cohen, R., & Fischer, C. (Eds.). (2018). Routledge handbook of diaspora studies. London: Taylor and Francis, Routledge. Connell, R. (2007). Southern theory: The global dynamics of knowledge in social science. Sydney: Allen and Unwin. Conteh, J., & Meier, G. (2014). The multilingual turn in languages education: Opportunities and challenges. Bristol: Multilingual Matters. Coronel-Molina, S. M., & McCarty, T. L. (Eds.). (2016). Indigenous language revitalization in the Americas. New York: Routledge. De Francis, J. (1977). Colonialism and language policy in Viet Nam. The Hague: Mouton Eades, D. (2013). Aboriginal ways of using English. Canberra: Aboriginal Studies Press. Elden, S. (2013). The birth of territory. Chicago: University of Chicago Press. Evans, N. (2017). Ngûrrahmalkwonawoniyan: Listening here. The Journal of the Australian Academy of the Humanities, 8, 34–44. Fishman, J. A. (1971). The sociology of language: An interdisciplinary social science approach to language in society. In J. A. Fishman (Ed.), Advances in the sociology of language (Vol. I, pp. 217–104). The Hague: Mouton. Fishman, J. A. (2009). Language policy and language shift. In T. Ricento (Ed.), An introduction to language policy: Theory and method (pp. 311–329). Malden: Wiley. Gottlieb, N., & Chen, P. (Eds.). (2001). Language planning and language policy: East Asian perspectives. Richmond: Curzon. Gramling, D. (2016). The invention of monolingualism. London: Bloomsbury. Gumperz, J.  J. (1964). Linguistic and social interaction in two communities. American Anthropologist, 66(6, Part 2), 137–153. Heugh, K. (2018). Multilingualism, diversity and equitable learning: Towards crossing the ‘abyss’. In P. Van Avermaet, S. Slembrouck, K. Van Gorp, S. Sierens, & K. Marijns (Eds.), The multilingual edge of education (pp. 341–367). London: Palgrave. Heugh, K., & Skutnabb-Kangas, T. (Eds.). (2010). Multilingual education worlds: From the periphery to the centre. Hyderabad: Orient BlackSwan. Ives, P. (2004). Language and hegemony in Gramsci. London: Pluto Press. King, L., & Carson, L. (2016). The multilingual city: Vitality, conflict and change. Bristol: Multilingual Matters. Kroskrity, P. V. (Ed.). (2000). Regimes of language: Ideologies, polities, and identities. Santa Fe: School for Advanced Research Press. Lewis, G. (1999). The Turkish language reform: A catastrophic success. Oxford: Oxford University Press.

56

J. Lo Bianco

Little, D., Leung, C., & Avermaet, P. V. (2014). Managing diversity in education, languages, policies, pedagogies. Bristol: Multilingual Matters. Lo Bianco, J. (2001). Viet Nam: Quoc ngu, colonialism and language policy. In N.  Gottlieb & P. Chen (Eds.), Language planning and language policy: East Asian perspectives (pp. 159–206). Richmond: Curzon. Lo Bianco, J. (2015). Multilingual education across Oceania. In W.  E. Wright, S.  Boun, & O.  Garcia (Eds.), The handbook of bilingual and multilingual education (pp.  604–617). Chichester: Wiley. Lo Bianco, J. (2016). Conflict, language rights, and education: Building peace by solving language problems in Southeast Asia (Language Policy Research Network (LPREN) Brief). Washington, DC: Center for Applied Linguistics. Lo Bianco, J. (2019). Uncompromising talk, linguistic grievance, and language policy: Thailand’s Deep South conflict zone. In M.  Kelly, H.  Footitt, & M.  Salama-Carr (Eds.), Handbook on languages at war. London: Palgrave Macmillan. Lo Bianco, J., & Bal, A. (2016). Learning from difference: Comparative accounts of multicultural education. Dordrecth: Springer. Lo Bianco, J., & Wickert, R. (Eds.). (2001). Australian policy activism in language and literacy. Melbourne: Language Australia Publications. May, S. (2014). The multilingual turn: Implications for SLA, TESOL and bilingual education. New York: Routledge. Oldfield, J., & Lo Bianco, J. (2019). A long unfinished struggle: Literacy and indigenous cultural and language rights. In J. Rennie & H. Harper (Eds.), Literacy education and indigenous australians – Theory, research and practice. Singapore: Springer. Otsuji, E., & Pennycook, A. (2015). Metrolingualism: Language in the city. London: Routledge. Piller, I. (2016). Linguistic diversity and social justice: An introduction to applied sociolinguistics. Oxford, UK: Oxford University Press. Rizvi, F., Louie, K., & Evans, J. (2016). Australia’s diaspora advantage: Realising the potential for building transnational business networks with Asia. Report for the Australian Council of Learned Academies. www.acola.org.au. Accessed 2 Mar 2019. Rosa, J., & Flores, N. (2017). Unsettling race and language: Toward a raciolinguistic perspective. Language in Society, 46(5), 621–647. Ruiz, R. (2010). Reorienting language-as-resource. In J. Petrovic (Ed.), International perspectives on bilingual education: Policy, practice, and controversy (pp. 155–172). Charlotte: Information Age Publishing. Rymes, B. (2013). Communicating beyond language: Everyday encounters with diversity. New York: Routledge. Salzburg. (2017). Statement for a multilingual world. Salzburg Global Forum. http://www.salzburgglobal.org/topics/article/fellows-co-create-salzburg-statement-for-a-multilingual-world. html. Accessed 10 Jan 2020. Singleton, D., Fishman, J. A., Aronin, L., & Ó Laoire, M. (Eds.). (2013). In Current Multilingualism: A New Linguistic Dispensation. Berlin: Mouton de Gruyter. Skutnabb-Kangas, T., Phillipson, R., Mohanty, A., & Panda, M. (Eds.). (2009). Social justice through multilingual education. Bristol: Multilingual Matters. Spolsky, B. (2012). The Cambridge handbook of language policy. Cambridge: Cambridge University Press. Stroud, C., & Mpendukana, S. (2009). Towards a material ethnography of linguistic landscape: Multilingualism, mobility and space in a South African township. Journal of Sociolinguistics, 13(3), 363–386. Tölölyan, K. (2012). Diaspora studies: Past, present and promise (International Migration Institute, Paper 55). Department of International Development, University of Oxford, UK. UNESCO. (2009). Investing in cultural diversity and intercultural dialogue: UNESCO world report, United Nations Educational, Science and Culture Organization. Paris: UNESCO. Vertovec, S. (2007). Super-diversity and its implications. Ethnic and Racial Studies, 30(6), 1024–1054. https://doi.org/10.1080/01419870701599465.

Where DLC Meets Multilingual Syntactic Development Éva Fernández-Berkes and Suzanne Flynn

Abstract  Multilingualism has grown as a natural extension of first and then second language acquisition, therefore the general understanding what some fundamental concepts, such as the initial state (S0), mean in multilingual developmental research has grown blurry over the years. Apart from meta-linguistic knowledge and strategies and other aspects, such as age, motivation, environmental factors, previously learnt languages might influence or even determine the success of a language learner’s endeavor to learn a next language. In this paper, we reflect on what syntactic knowledge multilingual learners have when starting to learn a next language and what role accumulated linguistic knowledge plays in the development of a subsequent language over time. In other words, here we attempt to define what S0 for multilingual acquisition is. The paper concludes that the temporal sequencing of full-fledged language grammars in the mind of a learner prior to the acquisition of Ln loses its validity as the fundament to model syntactic development in language acquisition. It follows that Aronin’s (Dominant language constellation as a method of research. Contribution to the thematic symposium: multi-competence and dominant language constellations. Paper presented at the 10th International Conference on L3 Acquisition and Multilingualism, Vienna, Austria, 2016) Dominant Language Constellation (DLC) can be used as a tool to evaluate what syntactic knowledge learners can draw upon, which would imply that developmental research is in the position to look at learners from the multilingual perspective, i.e. it may focus on to discover how languages might be connected in the mind of a learner. Keywords  Multilingualism · DLC · Initial state · Syntactic development

É. Fernández-Berkes (*) University of Applied Sciences Burgenland, Eisenstadt, Austria e-mail: [email protected] S. Flynn Massachusetts Institute of Technology, Cambridge, MA, USA © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_4

57

58

É. Fernández-Berkes and S. Flynn

1  Introduction Many would define a multilingual individual as a person with a high, practically native, level of competence in more than two languages. In fact, as Cook and Newson expressed it: “most people, or indeed all people, have multiple grammars in their minds” (Cook and Newson 2007: 223). We have also argued earlier (e.g. Flynn and Berkes 2017) that the increased pace of globalization has contributed to the fact that most people use multiple languages on a daily basis, therefore multilingualism has to be necessarily regarded as the most common and thus standard norm. In spite of the advances made in discovering the nature of the implied competences in multilingual development (Cabrelli Amaro et  al. 2012; Aronin and Singleton 2012; Bhatia and Ritchie 2013; Cook and Wei 2016), linguistic studies seem to be wanting a principled understanding of what exactly these implied competences are and the level a learner needs to have in order to be called multilingual. The Commission of the European Communities elaborated an indicator to measure proficiency based on appraising competence level in the four language skills: reading, writing, speaking and listening (Commission of the European Communities 2005). However valid and objective the resulting rubric might seem, we all seem to believe that knowing a language implies a much wider scope of linguistic competences including domains of cognition, such as sensory-motor, conceptual-­ intentional, etc. Therefore, the endeavor to study multilingualism must lead linguistic disciplines to define their own scope where each can apply their own tools to disclose a part of the puzzle that reveals what multilingualism is about. In this chapter, we focus on what syntactic knowledge learners have when starting to learn their next language, i.e. we attempt to shed light on what the initial state of language development is. We will quote some already published studies to support our claims and assume that the results can be extended to a large population of individuals. We believe that the resulting image on our puzzle piece fits into the great picture of a multilingual learner. Our reflections on the initial state will lead us to use the idea of Aronin’s (2016) Dominant Language Constellation (DLC) as a tool to evaluate what knowledge learners can draw upon when acquiring their next target language. The studies we introduce here involve learners of English whose background languages are similar or differ only minimally, i.e. participants who use the same languages on a daily basis are explicitly grouped together and subcategorized according to their level of English measured by an independent standardized test. Each study investigates learners’ production on a specific feature of English. Although the original studies meant to draw conclusions on what patterns emerge in L2/L3 development, here we focus on what the results can tell us about the initial state for multilingual acquisition.1 By comparing these patterns to the ones produced by other learners of English, 1  For the sake of simplicity, multilingual acquisition will be often referred to as L3 acquisition, regardless whether the new target language is acquired as the third, fourth, etc. in time by the learner.

Where DLC Meets Multilingual Syntactic Development

59

be it English as L1 or L2, we believe we will be able to provide important insights into the role accumulated linguistic knowledge plays in the development of a subsequent language over time.

2  Research Agenda for Multilingualism Asking researchers to confine the scope of their research as we have just proposed might seem to be counterproductive at first sight as it might reduce applicability of results. Nonetheless, multilingualism has grown as a natural extension of first and then second language acquisition and, given the proliferation of work done in the field, it appears to be necessary to distinctly determine the limits each field can deal with to clarify the fundamental concepts that are employed and the ultimate questions the research tries to provide answers for. Firstly and most importantly, developmental research in multilingual acquisition must define what the nature of the initial state (S0) in multilingual learners is. This is the foundation based on which essential assumptions are made that guide individual studies. In the next step, the individual fields of multilingual research must describe what they mean by development. This is where their scope is defined and confined, precisely because each field must state clearly the object of its investigation, i.e. it must define what is developed and how this development occurs. One related facet to this specific phase of developmental research is to look at what particular aspects of the L1 or other languages known effect or stimulate subsequent language learning. The phenomenon of copying or rather introducing previously known linguistic elements into the learning process of the next language is broadly known as transfer or cross-­ linguistic influence (CLI). Needless to say that the complexity of studying CLI in multilingual acquisition increases exponentially with the amount of languages a learner knows prior to his or her new linguistic experience. In fact, most recent research has focussed primarily on the aspect of CLI (e.g. Cenoz et  al. 2001; Hammarberg 2001; Angelovska and Hahn 2017 among others) and have identified patterns that largely enrich the knowledge we have of a multilingual learner. Most of the cases identified as CLI are, nonetheless, descriptive in nature and, rather frequently, such phenomena seem to be explainable assuming non-linguistic cognitive knowledge and skills. Consequently, there is an urgent need to investigate the linguistic nature of these phenomena and come up with genuine proposals concerning their role and what they indicate, i.e. how clear signs of CLI extended to a wide range of population effect development. Finally, one must define, prior to research, what the end state of language knowledge is in the specific field of linguistic investigation. Roughly speaking, researchers must know in which direction knowledge of a language learner must evolve in order to be able to establish phases of accomplishment. The end state, therefore, does not necessarily correspond in the different areas of developmental linguistic research, i.e. the learner’s language competence in the end state according to studies

60

É. Fernández-Berkes and S. Flynn

investigating syntactic structure development will most probably not match the language competence a learner has when he or she reaches the end state according to phonological studies. In the followings we focus on how the first phase of the research agenda, the S0, needs to be formulated in the field of developmental syntax where our research interest lies. However, before we elaborate the matter, we need to present the assumptions that lie behind our understanding of human language in general and its development in an individual in particular. The discussion of these assumptions is the topic of the next chapter.

3  Some Assumptions Before we proceed, we need to lay out some of the assumptions we have regarding language and language learning. All these assumptions are essential to understand what we mean by S0 in multilingual research. Firstly, we accept and work with the generative view on language, according to which there is essentially only one human language, or as Friederici et al. (2017) have formulated it most recently, language is a “biologically determined computational cognitive mechanism”. Hence, the cognitive and computational underpinnings for human language are universal across the species and unique to humans. This suggests that human languages can be envisaged as mapped on a spectrum and the individual languages only differ from each other in essentially a minute number of ways or degrees, thus applying the same computational choices to an array of lexical entities would result in “languages” that are fundamentally equal. This is not only true to languages from the same so-called language family, e.g. the Romance languages, but also to languages that are considered typologically non-related, like e.g. Japanese and Hungarian. It is important to emphasize here that the choices we alluded to refer to the computation of structures, be it a morpheme or a greater assembly of grammatical units that determines, e.g., head-directionality. Secondly, this assumption also implies that what has to be acquired when learning any new target language is miniscule in comparison to what we already know about language just because we are human. This knowledge is obviously unconscious and implicit as humans are not aware of its existence the same way they are of their body parts, but it can be recorded by, e.g., FMRI or MEG measurements (see recent studies, e.g. Berwick et al. 2013; Friederici et al. 2017). The assumption that knowledge from having acquired the first language remains intact and active for subsequent learning is not uncontroversial but plausible (see short overview on this matter in Berkes and Flynn 2013). A natural consequence of the above listed assumptions is to regard multilingualism as the natural state of the human mind or brain. Given appropriate and requisite input, the capacity of a person to learn a language is unlimited precisely because the mechanism to compute this input is biologically endowed. On the other hand, we know from experience that not everybody is a successful language learner and the

Where DLC Meets Multilingual Syntactic Development

61

degrees of multilingualism very much vary across individuals. Nonetheless, this state of the affairs does not contradict the theory of an innate cognitive capacity for language, as the lack of a product does not contradict the existence of tools that could create that product. This is precisely the point where generative conception differs from behaviorist and Skinnerian hypotheses. In terms of language development, the latter theories hypothesize that the S0 for learning L1 or L2 is an empty brain, i.e. no knowledge that could be helpful to facilitate the task of acquiring a next language. The construct of an empty mind was also the basis for the early genesis of Lado (1957) and Fries’s (1945) Contrastive Analysis approach to next language acquisition.

4  The Initial State in Multilingual Language Development 4.1  Some Historical Background There were outstanding attempts within the generative framework for linguistic theory to determine the character of the language faculty and its deployment in language learning. First-language (L1) acquisition studies provided important insights into how the mind and language work. The pioneering work of Braine (1963), followed by that of Brown (1973), argued for the child developing its grammar using its own rules independent of those of an adult speaker, an assumption that paved the way to instituting L1 acquisition as a research field in its own right. L1 acquisition, however, encounters serious difficulties when attempts are made to explore the interaction of general cognitive and specifically linguistic processes in the learner’s mind, because limitations deriving from general developmental, that is, maturational deficits bear upon the language learning process of children. In order to understand language and language learning as a crucial part of human cognitive processes, linguistic theory had to turn to accommodate the results provided by second-language (L2) acquisition studies, where L2 refers to a language acquired after infancy, for here maturational issues play practically no role in the process of language development. L2 acquisition research, then, provides an opportunity to examine language development independent of the influence of other developing cognitive processes, a necessary step in setting the basis for an adequate theory of language acquisition and competence (Flynn et al. 1998; Flynn and O’Neil 1988; Martohardjono 1993; Epstein et al. 1996; Guasti 2002; Rizzi 2004). According to the Chomskyan theory of language acquisition, it is the individual’s language faculty, a linguistic autonomous organ, which generates knowledge of language by responding to language input. Chomsky called the theory of this language faculty Universal Grammar (UG) and defined it as “the children’s pre-linguistic initial state” (Chomsky 1981: 7). Consequently, L2 research had to meet the challenge of describing theoretically how L2 learners access UG in their language development of the L2. Some of the

62

É. Fernández-Berkes and S. Flynn

most representative theories were developed in Bley-Vroman (1989), Clahsen and Muysken (1986), Johnson (1988), Johnson and Newport (1989), Schwartz and Sprouse (1996), Hawkins and Chan (1997), Epstein et al. (1996), Flynn (1983 and subsequent work); see Eubank (1991) for arguments and counterarguments on these theories. For a detailed review on the different hypotheses about access to UG in L2 acquisition, see e.g. White (1989, 1998, 2003), Cook and Newson (2007). For general treatments of the field of second language, see the chapters in Ritchie and Bhatia (1996, 2009). Two principal models were developed to capture possible ways in which UG may be represented in L2 language acquisition. The first, the Maturation Model, which was based on the Maturation Hypothesis (Borer and Wexler 1987; Felix 1984), claims that for L2 acquisition, UG is available to the L2 learner only in the form of a languagespecific grammar viz., the L1, which is supposed to be the beginning basis, the initial state S0, for the development of the L2 grammar. Thus, under this hypothesis, new language knowledge is acquired through L1 alone via some yet to be defined transfer mechanism. The Strong Continuity Model (Flynn and Lust 2002; also in Flynn 2009 as “Constant Model”) is based on the Strong Continuity Hypothesis as articulated for the study of L1 acquisition (Lust 1999, 2006; Boser et  al. 1992; for examples see Boser et al. 1995; Whitman et al. 1991). This model claims that it is not UG which changes over time but the theory building capacity of the L2 learner. According to this model, UG remains distinct from the developing language-specific grammar and is continuously available for the L2 learner, as implicit principles that guide and constrain language acquisition. Learners with the help of their UG, the actual S0, which triggers a language-specific grammar in view of new language input, map the language-specific grammar by dissociating and integrating grammatical components to construct their new language-specific grammars, a process referred to as “grammatical mapping” (Flynn and Lust 2002; Flynn et al. 2005, also in Lust 2006). Accepting the Strong Continuity Model of UG does not mean denying obvious differences between the processes of L1 and L2 acquisition, but, rather, claiming that these differences are not due to a change in UG. Furthermore, new language-­ specific knowledge is not accessed only through L1, but benefits from the knowledge UG may provide. Given this model, the S0 is not absolute, rather “the state of the mind/brain prior to experience with particular data and a particular new acquisition task” (Flynn and Lust 2002: 114) or in Chomsky’s words: the standard idealized model of language acquisition takes the initial state S0 to be a function mapping experience (primary linguistic data, PLD) to a language. UG is concerned with the invariant principles of S0 and the range of permissible variation. (Chomsky, 1995: 169)

The testing of hypotheses about the S0, defined as above, in L2 acquisition is crucial to the development of an understanding of the L2 learner’s representation of grammar at the point of encountering the L2 input for the first time. (White 1989, 1998, 2003; see related discussions on the nature of the L1 in the L2 grammar in Gass and Selinker 1983, 1992). Having said that, we must look at the implications such a definition of the S0 has on the explicative power of the two key models of UG access in L2 acquisition.

Where DLC Meets Multilingual Syntactic Development

63

Given the Maturation Model, it is assumed that learners’ UG has developed into the grammar of their L1, thus the S0 for L2 cannot be other than L1. For this reason, development is supposed to be controlled by UG but only via the learner’s L1. It is not clear then how this model predicts the syntactic development of the new target language (see extended discussion in Epstein et al. 1996). In contrast, the Strong Continuity Model hypothesizes that it is UG that guides the acquisition of an L2, as it does in L1 acquisition, due to the fact that UG in its entirety is available to the learner. Consequently, non-linguistic strategies such as translating bits of language and fitting them into the new target grammar cannot be the force that drives development, rather the learners’ syntactic ability to apply unconscious knowledge of grammar to their increasing experience in the target language. It follows, that the Strong Continuity Model treats the learners as syntacticians who, albeit unconsciously, construct their new target grammar. All in all, the Strong Continuity Model rejects transfer on the level of fundamental computational mechanisms, as a result of understanding language learning not as a maturational, but rather a computational process. However, the Strong Continuity Model does recognize the role of other languages known in terms of subsequent language development as will be discussed in more detail below.

4.2  Multilingual Acquisition Models and the Initial State It is widely accepted that multilingual learners approach language learning somewhat differently from how L2 learners do in certain respects. It is an empirical fact that the acquired knowledge of at least one L2 plays a role in subsequent acquisition. L3 learners may also rely on certain meta-linguistic strategies they became familiar with during the course of learning previous languages. The development of cognitive science in the last 50 years has contributed greatly to uncovering a long series of aspects, such as age, motivation, environmental factors, and other meta-­ linguistic knowledge, that influence or even determine the success of a language learner’s endeavor. Thus, the danger, when designing a study, is to try to account for all factors that bear on the acquisition of a new target language and postulate an initial knowledge of vast complexity. Even leaving non-linguistic factors aside, multilingual acquisition presents a higher degree of complexity than L2 acquisition. Therefore, complexity needs to be controlled and sufficiently limited to be able to differentiate the effect from the cause. Our goal is then to define what linguistic knowledge S0 in multilingual acquisition contains, before we venture to make any proposition regarding its nature. The suggestion that multilingual learners transfer their existing linguistic knowledge, i.e. their linguistic experience gained in the course of L1 to Ln−1 acquisition, to learn Ln is an obvious one, although of a very abstract nature. In the course of learning a language, learners end up acquiring Ln, albeit at varying levels, thus the exact content of what has been transferred needs to be investigated applying rigorous research methods. In fact, the question how multilingual acquisition is

64

É. Fernández-Berkes and S. Flynn

influenced by previously learnt languages, i.e. how CLI is manifested in the representation of a new target language, has been the target of passionate debates. Numerous studies and volumes in recent years are dedicated to the exploration of how L1 and/or any L2 bear upon the acquisition of subsequent language(s) (see e.g. Cenoz et  al. 2001; De Angelis 2007; Gabrys-Barker 2006; Kidd et  al. 2007; Hammarberg 2009; Odlin et  al. 2006; Leung 2005, 2009; Angelovska and Hahn 2017). However, despite the vast interest in CLI in all fields of language acquisition, there have been relatively few experiments to explore CLI at a syntactic level. It is true that syntactic CLI is best noticed in the case of negative transfer, i.e. when learners’ production does not coincide with target-like constructions, but emphasizing the aspect of negative transfer, or interference, would be a conceptual error, because language acquisition occurs in the great majority of cases if conditions are favorable. Hence, serious attempts are to be made to discover how development occurs and reveal the real nature of any influence that has seemingly been caused by transfer. The research on syntactic CLI within the framework of generative morphosyntax has inspired numerous empirical studies necessary to explore further how language learning occurs. Based on the results of these studies, several influential models for multilingual acquisition were stipulated in the last two decades, which in turn gave birth to further experiments to validate them. All of these models coincide in assuming that multilingual learners have access to UG, nonetheless that they process their L3 under altered conditions from those of their L2. This is precisely the raison d’être of the study of multilingual development, since L3 research may render answers that the study of L2 acquisition alone cannot. Given the generally assumed fact that the learners’ S0 contains previously attained linguistic knowledge, L3 acquisition models provide insights about the role of previously learnt languages. Indeed, one of the issues that differentiates these models is the privileged role the given model bestows on L1 or on any subsequent L2 when learning L3, as will be explicated later. The study of multilingual acquisition also raises issues concerning the Maturation Model itself on the grounds that an L3 model needs to address the question of what is available to the L3 learner in terms of UG. Therefore, multilingual acquisition models are to account for L3 acquisition and in doing so they delineate the nature of the S0 for all subsequent language learning. The L2 Status Factor (Bardel and Falk 2007) and the Typological Primacy Model (Rothman and Cabrelli Amaro 2010) were initially set up to model the initial phases of L3 development emphasizing the role of previous linguistic experience as the main drive for initial language acquisition, thus also the base for CLI which, according to these models, guides further development. The Cumulative Enhancement Model for language acquisition (Flynn et al. 2004), as its name implies, was set up with the objective to account for early L3 acquisition as well as multilingual language development. Two recent attempts followed to refine previous models, the Linguistic Proximity Model (Westergaard et  al. 2016) and the Scalpel model (Slabakova 2016). In the following, we will briefly introduce the mentioned models in light of the research focus adopted here.

Where DLC Meets Multilingual Syntactic Development

65

As its name suggests, the L2 Status Factor (L2SF) focuses on the learners’ tendency to activate the last-learned language rather than their L1, especially in the initial phases of L3 acquisition and predicts a general tendency to activate and transfer linguistic knowledge and/or elements of the last-learned language into subsequent L3 language acquisition. In other words, they maintain that when it comes to recounting the variables that influence syntactic transfer, the last-learned language plays a primary role in determining the L3 S0 due to the higher degree of sociolinguistic and cognitive congruence surrounding the acquisition of L2, L3, … Ln. These factors include age of onset, proficiency, exposure and use of the new target language as well as learning strategies, metalinguistic knowledge, etc. Finally, Bardel and Falk maintain that, due to the resulting cognitive load, L1 in fact does not seem to be accessible for multilingual learners in the course of acquiring a next language. Additionally, Bardel and Falk propose that the L2SF guides syntactic transfer as well; that is, they posit that syntactic development in subsequent acquisition is also affected by the specific syntactic features of the last-learned language. According to this hypothesis, learners are driven by their cognition-based learning strategies acquired in the course of L2 acquisition, independent of any typological similarities among the involved languages. Consequently, the L2SF contradicts the Typological Primacy Model (TPM), which does not reject L1 transfer. The model seeks to predict which set of language properties a multilingual learner is likely to transfer when learning a subsequent language Ln. Based on the theory of the (psycho)typologically-motivated transfer (Kellerman 1983), the TPM suggests that the learner may choose, with the help of an internal parser, the typologically more similar system for multilingual transfer to facilitate the acquisition of Ln. The real typological proximity or the typological proximity as perceived by the learner is to guide the learner in selecting the system to be transferred to facilitate the acquisition of Ln. To sum up, both the L2SF and the TPM discussed above are primordially models of the S02 as both models agree that it is precisely the already-existing grammars in the mind of the learner at the beginning of the acquisition process that make L3 language learning essentially and qualitatively different from L1 or L2 acquisition. Hence, the fundamental question these two models seek to investigate is the role of previously attained language-specific morphosyntactic knowledge at the S0 of L3 in attempts to tease apart the variables that guide perceived syntactic transfer. The Cumulative Enhancement Model for Language Acquisition (CEM) (Flynn et al. 2004) hypothesizes that language learning is cumulative, i.e. all previously known languages are available to the learner to constructively enhance subsequent language learning. According to the CEM the mind does not redundantly represent syntactic structures, therefore any prior language can be strategically drawn upon in subsequent acquisition and may enhance the acquisition of further languages. This claim was based on a series of studies which focused on the acquisition of relative

2  Attempts were made to extend findings to the process of subsequent interlanguage development (see Falk and Bardel 2010; Rothman 2010).

66

É. Fernández-Berkes and S. Flynn

clauses in English (Flynn and Lust 1980; Flynn 1983, 1987, 1989). Flynn et  al. (2004) concluded, on the one hand, that the universal knowledge underlying the free relative seems to be fully available for the learner and, thus, acts as a developmental precursor in the acquisition of English.3 On the other hand, previously acquired knowledge of CP structure appears to influence the development of target-specific CP structure. Learners whose deep-level syntactic experience already contained the English-like CP setup seemed to benefit from this knowledge, at least with respect to developing the target-like CP. This claim contrasts with deficit models (e.g., interference, negative transfer). Performance errors may negatively influence production data or the learner may licitly opt for using the strategy of translation and thus transferring elements of grammar from one language to the other, but such errors are not necessarily an indication of the level of competence at the deepest level, at the level where the construction of language-specific grammar takes place. Similar to the CEM, the recently elaborated Linguistic Proximity Model and the Scalpel Model also recognize the importance of variables such as recent linguistic input or typological affinity, as they provide a legitimate basis for strategies a multilingual learner has access to and may apply in order to proceed. Both of these models postulate that multilingual learners accumulate linguistic knowledge and that they may have access to any part of this knowledge during development of a next target language. Based on such understanding of the S0 for multilingual acquisition, these models attempt to account for syntactic language development. Both the Linguistic Proximity Model and the Scalpel Model find that it is CLI that drives language development on a property by property basis. In other words, multilingual learners develop the next language by continuously analyzing and comparing, albeit unconsciously, grammatical properties of the target grammar with latent properties in their previously accumulated and activated linguistic knowledge. The CEM, on the other hand, postulates that multilingual learners instead of transferring syntactic elements, they construct the grammar of a specific target language observing constraints imposed by UG. The fundamental prediction the model makes regarding language development is that the increasing linguistic representation of the target language is non-redundant, thus language acquisition is cumulative and this enhanced knowledge serves as S0 for the next language to be learnt. Summing up, the L2SF and the TPM models aim to describe the S0 of L3 acquisition; they attempt to tease apart the intervening variables in order to determine the most likely source of syntactic transfer for L3 acquisition, whereas the Linguistic Proximity Model and the Scalpel Model attempt to describe how CLI stimulates language development without being explicit about what the S0 de facto consists of. The CEM pursuits the task to reveal the nature of language development by trying to discover the representations projected by multilingual learners and at the same time informs us about the S0 for subsequent language learning. 3  Indeed, it seems to be a general finding in L1 acquisition that L1 learners build upon knowledge of the free relative to construct lexically headed relative clauses (see Hamburger 1980, Flynn and Lust 1980 for English; Packard 1988 for Mandarin; Lee 1991, Lee et al. 1990 for Korean; Murasugi 1991 for Japanese; Foley 1996 for French; Somashekar 1999 for Tulu; Mróz 2010 for Polish; Flynn et al. 2004 for Kazakh).

Where DLC Meets Multilingual Syntactic Development

67

5  The Initial State of Multilingual Acquisition: A Proposal As we have mentioned above and in accordance with many studies within the framework of cognitive science, we assume that the mind of a multilingual learner is not empty. We believe that the knowledge of previously known languages along with all manner of meta-linguistic knowledge and strategies fill the mind of a learner. Therefore, we propose that the S0 refers to the state of the mind or brain prior to experience with specific target language data and a particular new acquisition task, or as Flynn and Lust (2002) puts it, the state of the learner’s mind is “initial with regard to a particular language acquisition endeavor”. This also means that two persons learning e.g. German, the first as L2 and the other as L3, are in an S0 with respect to German at the moment of initiating their studies of German. Such a definition has the advantage to unify all language acquisition fields, L1, L2, L3… Ln, as it defines the starting stage based on which further development can be studied applying the specialized tools of each specific field. In addition, this definition allows for varying content at this state, i.e. the knowledge of language the learner starts out with. Nevertheless, the general belief, according to which the number of known languages relates proportionately to the amount of knowledge of language a person has, is not supported by developmental studies, thus it is precisely this linguistic content that needs to be defined. As an illustration, let us consider a Chinese learner of English L2 and another Chinese learner of English L3 who acquired Indonesian as his or her L2. English manifests overt Wh-movements (1.a), although constrained, as exemplified in (1.b); whereas, in Indonesian and Chinese there is no such overt Wh-movement (see Martohardjono and Flynn 1995 and references given there). (1)

a. Whati did Tom fix ei [that the man had broken]? b. *Whoi did Tom fix the door [that ei had broken]?

Knowledge of the additional Indonesian L2 does not provide a Chinese learner of English any knowledge about functional features that allow Wh-movement or constraint such movement. Therefore, we believe that a bilingual Chinese L1/ English L2 and a multilingual Chinese L1/Indonesian L2/English L3 learner have the same knowledge regarding Wh-movement in the S0 of their learning English. What is it then that makes up the content of the S0? In line with the CEM, we propose that a multilingual learner in the S0 has implicit knowledge of language specific grammars, i.e. grammar representations of previously learnt languages. As we have argued above, such an understanding is consistent with other models (see also González Alonso and Rothman 2016), however, only the CEM makes the explicit claim that the representations of these grammars are non-redundant, not prioritized with respect to each other and to some degree undifferentiated. In order to support such an argument, we cite some fundamental findings of a series of language development studies focusing on the acquisition of relative clauses in English. Flynn (1983, 1987) investigated the English L2 acquisition of adult Japanese L1 and Spanish L1 speakers, Flynn et  al. (2004) the English L3 acquisition of adult Kazakh L1 speakers who acquired Russian as their L2, and

68

É. Fernández-Berkes and S. Flynn

Berkes and Flynn (2012) contrasted results from the English L2 acquisition of adult German L1 and the English L3 acquisition of adult Hungarian L1 speakers who acquired German as their L2. All of these studies used the same test sentences, an adoption of the original study of Flynn and Lust (1980) of monolingual children acquiring English, and employed the same procedure. Kazakh and Japanese share the characteristic of being head-final languages whereas Spanish, Russian, German and English count as head-initial. According to current linguistic theory, the feature responsible for head-directionality is on the complementizer phrase (CP), therefore learners could have easily transferred their knowledge of CP if CP in any of the previously known languages matched the English CP.  However, results of these studies revealed that learners might have simply transferred elements of grammar to tackle the acquisition task, but actual development happened as a result of the learners having constructed their target grammar from elementary to more complex structures. All studies showed that free relatives appeared to be developmental precursors to headed relatives in the process of building a full-fledged, language-­ specific CP architecture, even in the case of the acquisition study of English L2 by German L1 learners where the two languages match as regards CP directionality. The Hungarian L1/German L2 and the Kazakh L1/Russian L2 groups learning English L3, as well as the Spanish L1 group learning English L2 seemed to skip the more primitive level in their development, i.e. production data did not reveal significant difference in learners’ imitation of headed vs. free relatives. This fact implied that learners’ knowledge in their S0 with respect to English contained knowledge they could draw upon. This knowledge these learners appeared to have access to was at a deep structural level, which they used to construct the new target language. Thus, such knowledge cannot be either prioritized or differentiated. To illustrate what is meant by non-redundancy, let us take the study presented in Berkes and Flynn (2012) of a group of native German speakers learning English L2 and a group of native Hungarian speakers with German L2 learning English L3. As we mentioned earlier, Hungarian, German and English match as regards head-­ directionality. This match, however, is not absolute because the German word order within the subordinate relative clause is SOV, which contrasts with the SVO word order in English. Hungarian, similar to English, usually manifests an SVO word order in subordinate relative clauses, but this language is considered to be organized around the concept of topic and focus rather than the concept of subject (Li 1976). This implies that the featural setups for CP in the three languages do not match completely and grammatical structures cannot be simply transferred from any of the previously learnt languages. We assume then that the S0 of learners in the German L1/English L2 group contained a CP with a feature for head-directionality they could draw upon but not the correct value for subordinate word order. Yet, the S0 of the multilingual learners with respect to English already comprehended the additional knowledge that the featural setup for CP optionally allows for different word orders in the subordinate relative clause. The statistical analysis of the results of the L2 study group revealed significant differences between correct production of lexically-headed and free relatives at the low and mid levels of their acquisition of English L2, therefore we concluded that

Where DLC Meets Multilingual Syntactic Development

69

the feature responsible for constituent word order within subordinate clauses may play a role in the acquisition process of subsequent target language grammars. The statistical analysis of the results of the L3 study group suggested, on the contrary, that Hungarian L1/German L2 learners of L3 English did not need to fall back onto the more primitive level of using free relatives in order to build up the target-like CP, since this group showed no evidence that they found the imitation of free relatives easier than that of headed ones. The learners’ implicit cognition was manifested by the active use of CPs correctly selected according to the syntactic requirement of English L3. With other words, the Hungarian L1/German L2 learners had implicit computational knowledge regarding CP they could draw upon when acquiring the English CP, and thus skipped the more primitive stage of having to construct the new grammar from scratch. Indeed, the L3 group did not at any stage exhibit a significant preference for free relatives, whereas German L1 learners of English L2 did. Hence, this study seems to support the claim that learners’ implicit knowledge of specific grammars, such as that of Hungarian and German, is an interconnected conglomerate which may be prompted to choose the adequate featural setup for the target CP by having marked CP in the Lexicon as subject for variation. If representations of Hungarian and German in the learners’ mind were separate, they would have manifested a clear preference for free relatives, due to the fact that the new target grammar, their English L3, would have had to be constructed. Summing up, learners did not appear to have redundant representations of each grammar they became familiar with but their German L2 prompted them to be aware of the variability that was provided by different featural setups. In conclusion, although the actual learning process might take time, the addition of a sole feature to the Lexicon, out of which lexical and functional categories emerge, keeps processing load at a minimum, which seems to be a desirable outcome also supported by current syntactic theories. It is important to emphasize at this point that in developmental syntax the focus is not on the development of e.g., meta- or non-linguistic knowledge as it relates to language development, but rather on the development of implicit and unconscious computational knowledge in the mind of a learner. It is true though that language acquisition may be more or less accelerated as a result of the working of certain non-linguistic strategies, but these need to be disentangled and brought to light. All in all, in order to pursue the objectives of both developmental syntax and syntactic CLI research, it is fundamental to discern and reveal the computational knowledge learners can draw upon in constructing new target grammars, i.e. the content of the S0 with regard to the acquisition of the next language. A logical extension of this is to define briefly what is meant by end state of language development in syntactic development. In contrast to the general belief, being in the end state does not make the learner a native or near-native speaker of the target language. It implies, however, that the learner’s implicit representation of the fundamental architecture of the language specific grammar has reached a state where no further increase in deep-level syntactic knowledge is necessary to cope with the target language. This implicit actualization of the language serves as a supporting structure based on which further increase of competence may occur. Most

70

É. Fernández-Berkes and S. Flynn

surprisingly, reaching the end state of syntactic development occurs at an early stage of acquisition. Returning to the original purpose of this chapter, it follows that defining the DLC of the learner or groups of learners is a necessary step in the planning phase of a study. On the one hand, if what ultimately counts is the implicit cumulative linguistic experience, then the order in which a multilingual learner has acquired his or her language repertoire plays a minor if any role provided this repertoire contains full-­fledged, but not necessarily conscious, knowledge of the structural setup of the different languages. This implies that the temporal sequencing of L1…Ln loses its validity as the fundament to model syntactic development in language acquisition. One of the main arguments for such a claim comes precisely from the amount of reported studies in favor of the influence of L1 in L3 development compared to the amount that found clear L2 influence in the acquisition of the next language (see our review of the L2SF and the TPM models above). These models received vast support from a series of wellgrounded and rigorously performed studies, therefore we must accept that both theories have captured an aspect of language learning: the importance of acquired linguistic knowledge prior to next target language. Nonetheless, increasing amount of evidence is found that the developmental stages of the acquisition of a language coincide across learners with different DLCs, thus working with learners with different DLCs gains in significance. An immediate conclusion is that we do not need to focus on relating each non-target-like structure in the learner’s production of the target language to one of his or her known languages as the source for CLI. Instead, we may allow for stages of deep-level structural development in the acquisition process as a result of a computational process unconsciously done by the learner. As a result, individual non-linguistic or meta-linguistic strategies to process new target-like elements such as memory, using translation, etc., automatically fall out. As a conclusion, it may be stated that taking the different DLCs as the linguistic background for learners to acquire a next language, rather than focusing on the temporal sequencing of previous linguistic input, places a developmental study in the position to look at learners from the multilingual perspective, i.e. it aims at discovering how languages might connect in their minds affecting all known languages and resists the temptation to compare their achievement with monolinguals (see premises of multi-competence in Cook and Wei 2016).

References Angelovska, T., & Hahn, A. (Eds.). (2017). L3 syntactic transfer: Models, new developments and implications. Amsterdam: John Benjamins. Aronin, L. (2016). Dominant language constellation as a method of research. Contribution to the thematic symposium: Multi-competence and dominant language constellations. Paper presented at the 10th International Conference on L3 Acquisition and Multilingualism, Vienna, Austria. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins Publishing Company.

Where DLC Meets Multilingual Syntactic Development

71

Bardel, C., & Falk, Y. (2007). The role of the second language in third language acquisition: The case of Germanic syntax. Second Language Research, 23(4), 459–484. Berkes, É., & Flynn, S. (2012). Further evidence in support of the cumulative-enhancement model: CP structure development. In J. Cabrelli Amaro, S. Flynn, & J. Rothman (Eds.), Third language acquisition in adulthood (pp. 143–164). Philadelphia/Amsterdam: John Benjamins Publishing Company. Berkes, É., & Flynn, S. (2013). Multilingualism: New perspectives on syntactic development. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of bilingualism and multilingualism, second edition (pp. 137–167). Chichester: Wiley. Berwick, R.  C., Friederici, A.  D., Chomsky, N., & Bolhuis, J.  J. (2013). Evolution, brain, and the nature of language. Trends in Cognitive Sciences, 17(2), 89–98. https://doi.org/10.1016/j. tics.2012.12.002. Bhatia, T. K., & Ritchie, W. C. (Eds.). (2013). The handbook of bilingualism and multilingualism (2nd ed.). Chichester: Wiley. Bley-Vroman, R. (1989). What is the logical problem of foreign language learning? In S. M. Gass & J.  Schachter (Eds.), Linguistic perspectives on second language acquisition (pp.  41–68). New York: CUP. Borer, H., & Wexler, K. (1987). The maturation of syntax. In T.  Roeper & E.  Williams (Eds.), Parameter setting (pp. 123–172). Dordrecht: Reidel. Boser, K., Lust, B., Santelmann, L., & Whitman, J. (1992). The syntax of CP and V-2 in early German child grammar: The strong continuity hypothesis. In Proceedings of the North Eastern Linguistics Association (Vol. 22, pp. 51–66). Amherst: University of Massachusetts. Boser, K., Santelmann, L.  M., Barbier, I., & Lust, B. (1995). Grammatical mapping from UG to language specific grammars: Deriving variation in the acquisition of German, Dutch and Swedish. In Proceedings of the 19th Annual Boston University Conference on language development. Boston: Cascadilla Press. Braine, M.  D. S. (1963). The ontogeny of English phrase structure: The first phase. Language, 39(1), 13. Brown, R. W. (1973). A first language: The early stages. London: Allen and Unwin. Cabrelli Amaro, J., Flynn, S., & Rothman, J. (Eds.). (2012). Third language acquisition in adulthood. Amsterdam: John Benjamins. Cenoz, J., Hufeisen, B., & Jessner, U. (2001). Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives. Clevedon: Multilingual Matters. Chomsky, N. (1981). Lectures on government and binding. Dordrecht/Cinnaminson: Foris Publications. Chomsky, N. (1995). The minimalist program. Boston: MIT. Clahsen, H., & Muysken, P. (1986). The availability of universal grammar to adult and child learners: A study of the acquisition of German word order. Second Language Research, 2, 93–119. Commission of the European Communities. (2005). Communication from the Commission to the European Parliament and the Council: The European indicator of language competence (Reprinted from: http://www.europarl.europa.eu/RegistreWeb/search/simple. htm?reference=COM_COM(2005)0356 (Oct 2, 2017). Cook, V., & Newson, M. (2007). Chomsky’s universal grammar: An introduction (3rd ed.). Oxford: Wiley-Blackwell. Cook, V., & Wei, L. (Eds.). (2016). Handbook of linguistic multi-competence. Cambridge: Cambridge University Press. De Angelis, G. (2007). Third or additional language acquisition. Clevedon: Multilingual Matters. Epstein, S. D., Flynn, S., & Martohardjono, G. (1996). Second language acquisition: Theoretical and experimental issues in contemporary research. Behavioral and Brain Sciences, 1, 677–758. Eubank, L. (Ed.). (1991). Point counterpoint: Universal grammar in the second language. Amsterdam: John Benjamins Publishing Company. Falk, Y., & Bardel, C. (2010). Object pronouns in German L3 syntax: Evidence for the L2 status factor. Second Language Research, 27(1), 59–82.

72

É. Fernández-Berkes and S. Flynn

Felix, S. W. (1984). Two problems of language acquisition: On the interaction of universal grammar and language growth. Passau: Universität Passau. Flynn, S. (1983). A study of the effects of principal branching direction in second language acquisition: The generalization of a parameter of universal grammar from first to second language acquisition. Ithaca: Cornell University dissertation. Flynn, S. (1987). A parameter-setting model of L2 acquisition: Experimental studies in anaphora. Dordrecht: Reidel. Flynn, S. (1989). Spanish, Japanese and Chinese speakers’ acquisition of English relative clauses: New evidence for the head-direction parameter. In K.  Hyltenstam & L.  K. Obler (Eds.), Bilingualism across the lifespan: Aspects of acquisition, maturity, and loss (pp.  116–131). Cambridge: Cambridge University Press. Flynn, S. (2009). UG and L3 acquisition: New insights and more questions. In Y.-k. I. Leung (Ed.), Third language acquisition and universal grammar (pp. 71–88). Bristol: Multilingual Matters. Flynn, S., & Berkes, É. (2017). Toward a new understanding of syntactic CLI: Evidence from L2 and L3 acquisition. In T. Angelovska & A. Hahn (Eds.), L3 syntactic transfer: Models, new developments and implications (pp. 35–61). Amsterdam: John Benjamins. Flynn, S., & Lust, B. (1980). Acquisition of relative clauses in English: Developmental changes in their heads. In W. Harbert & J. Herschensohn (Eds.), Cornell working papers in linguistics 1 (pp. 33–42). Ithaca: Department of Modern Languages and Linguistics, Cornell University. Flynn, S., & Lust, B. (2002). A minimalist approach to L2 solves a dilemma of UG. In V. Cook (Ed.), Portraits of the L2 user (pp. 95–120). Clevedon: Multilingual Matters. Flynn, S., & O’Neil, W. (Eds.). (1988). Linguistic theory in second language acquisition. Dordrecht: Kluwer Academic Publishers. Flynn, S., Martohardjono, G., & O’Neil, W. (Eds.). (1998). The generative study of second language acquisition. Mahwah: Lawrence Erlbaum Associates, Inc. Flynn, S., Foley, C., & Vinnitskaya, I. (2004). The cumulative-enhancement model for language acquisition: Comparing adults’ and childrens’ patterns of development. International Journal of Multilingualism, 1(1), 3–17. Flynn, S., Foley, C., Gair, J., & Lust, B. (2005). Developmental primacy of free relatives in first, second and third language acquisition: Implications for their syntax and semantics. Paper presented at the Linguistic Association of Great Britain, Cambridge University. Foley, C. (1996). Knowledge of the syntax of operators in the initial state: The acquisition of relative clauses in French and English. Ithaca: Cornell University dissertation. Friederici, A.  D., Chomsky, N., Berwick, R.  C., Moro, A., & Bolhuis, J.  J. (2017). Language, mind and brain. Nature Human Behaviour, 1(10), 713–722. https://doi.org/10.1038/ s41562-017-0184-4. Fries, C. C. (1945). Teaching and learning English as a foreign language. Ann Arbor: University of Michigan Press. Gabryś-Barker, D. (2006). The interaction of languages in the lexical search of multilingual language users. In J.  Arabski (Ed.), Cross-linguistic influences in the second language lexicon (pp. 144–166). Clevedon: Multilingual Matters. Gass, S. M., & Selinker, L. (1983). Language transfer in language learning. Rowley: Newbury House Publishers. Gass, S. M., & Selinker, L. (1992). Language transfer in language learning. Amsterdam: John Benjamins. González Alonso, J., & Rothman, J. (2016). Coming of age in L3 initial stages transfer models: Deriving developmental predictions and looking towards the future. International Journal of Bilingualism, 0(0), 1367006916649265. https://doi.org/10.1177/1367006916649265. Guasti, M. T. (2002). Language acquisition: The growth of grammar. Cambridge, MA: Bradford Books, MIT Press. Hamburger, H. (1980). A deletion ahead of its time. Cognition, 8, 389–416. Hammarberg, B. (2001). Roles of L1 and L2  in L3 production and acquisition. In J.  Cenoz, B.  Hufeisen, & U.  Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 21–41). Clevedon: Multilingual Matters.

Where DLC Meets Multilingual Syntactic Development

73

Hammarberg, B. (Ed.). (2009). Processes in third language acquisition. Edinburgh: Edinburgh University Press. Hawkins, R., & Chan, C.  Y.-h. (1997). The partial availability of universal grammar in second language acquisition: The ‘failed functional features hypothesis’. Second Language Research, 13, 187–226. Johnson, J. S. (1988). Critical period effects on universal properties of language: The status of Subjacency in the acquisition of a second language. Urbana: University of Illinois dissertation. Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60–99. Kellerman, E. (1983). Now you see it, now you don’t. In S. M. Gass & L. Selinker (Eds.), Language transfer in language learning (pp. 112–134). Rowley: Newbury House. Kidd, E., Brandt, S., Lieven, E., & Tomasello, M. (2007). Object relatives made easy: A cross-­ linguistic comparison of the constraints influencing young children’s processing of relative clauses. Language & Cognitive Processes, 22(6), 860–897. Lado, R. (1957). Linguistics across cultures. Ann Arbor: University of Michigan Press. Lee, K.-O. (1991). On the first language acquisition of relative clauses in Korean: The universal structure of COMP. Ithaca: Cornell University dissertation. Lee, K.-O., Lust, B., & Whitman, J. (1990). On functional categories in Korean: A study of the first language acquisition of Korean relative clauses. In E.-J.  Baek (Ed.), Papers from the Seventh International Conference on Korean linguistics (pp.  312–333). Toronto: University of Toronto Press. Leung, Y.-k. I. (2005). Second vs. third language acquisition of tense and agreement in French by Vietnamese monolinguals and Cantonese-English bilinguals. In J. Cohen, K. T. McAlister, K. Rolstad, & J. MacSwan (Eds.), ISB4: Proceedings of the 4th International Symposium on bilingualism (pp. 1344–1352). Sommerville: Cascadilla Press. Leung, Y.-k. I. (Ed.). (2009). Third language acquisition and universal grammar. Bristol: Multilingual Matters. Li, C. N. (Ed.). (1976). Subject and topic. New York: Academic. Lust, B. (1999). Universal grammar: The strong continuity hypothesis in first language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of child language acquisition (pp. 111–155). San Diego/London: Academic. Lust, B. (2006). Child language: Acquisition and growth. Cambridge: Cambridge University Press. Martohardjono, G. (1993). Wh-movement in the acquisition of a second language. Cross-linguistic studies of three languages with and without syntactic movement. Ithaka: Cornell University dissertation. Martohardjono, G., & Flynn, S. (1995). Language transfer. In L.  Eubank, L.  Selinker, & M. Sharwood Smith (Eds.), The current state of interlanguage (pp. 205–218). Philadelphia: John Benjamins Publishing Company. Mróz, M. (2010). Bilingual language acquisition: Focus on relative clauses in Polish and English. Warsaw: Wydawnictwa Uniwersytetu Warszawskiego. Murasugi, K. (1991). Noun phrases in Japanese and English: A study in syntax, learnability and acquisition. Storrs: University of Connecticut dissertation. Odlin, T., Alonso Alonso, R., & Alonso-Vázquez, C. (2006). Fossilization in L2 and L3. In Z. Han & T. Odlin (Eds.), Studies of fossilization in second language acquisition (pp. 83–99). Clevedon: Multilingual Matters. Packard, J. L. (1988). The first language acquisition of prenominal modification with de in mandarin. Journal of Chinese Linguistics, 16(1), 31–53. Ritchie, W.  C., & Bhatia, T.  K. (Eds.). (1996). Handbook of second language acquisition. San Diego: Academic. Ritchie, W. C., & Bhatia, T. K. (Eds.). (2009). The new handbook of second language acquisition. Bingley: Emerald Group Publishing Ltd. Rizzi, L. (2004). On the study of the language faculty: Results, developments and perspectives. Linguistic Review, 21, 323–344.

74

É. Fernández-Berkes and S. Flynn

Rothman, J. (2010). L3 syntactic transfer selectivity and typological determinacy: The typological primacy model. Second Language Research, 27(1), 107–127. Rothman, J., & Cabrelli Amaro, J. (2010). What variables condition syntactic transfer? A look at the L3 initial state. Second Language Research, 26(2), 189–218. Schwartz, B.  D., & Sprouse, R.  A. (1996). L2 cognitive states and the full transfer/full access model. Second Language Research, 12, 40–77. Slabakova, R. (2016). The scalpel model of third language acquisition. International Journal of Bilingualism, 21, 651–665. https://doi.org/10.1177/1367006916655413. Somashekar, S. (1999). Developmental trends in the acquisition of relative clauses: Cross-­ linguistic experimental study of Tulu. Ithaca: Cornell University dissertation. Westergaard, M., Mitrofanova, N., Mykhaylyk, R., & Rodina, Y. (2016). Crosslinguistic influence in the acquisition of a third language: The linguistic proximity model. International Journal of Bilingualism, 21, 666–682. https://doi.org/10.1177/1367006916648859. White, L. (1989). Universal grammar and second language acquisition. Amsterdam: John Benjamins Publishing Company. White, L. (1998). UG access in L2 acquisition: Reassessing the question NFLRC Netwok #9 (Vol. HTML document). Honolulu: University of Hawaii, Second Language Teaching and Curriculum Center. (Reprinted from: www.nflrc.hawaii.edu/Networks/NW09. 18 Nov 2011). White, L. (2003). Second language acquisition and universal grammar. Cambridge: Cambridge University Press. Whitman, J., Lee, K.-O., & Lust, B. (1991). Continuity of the principles of universal grammar in first language acquisition: The issue of functional categories. In NELS Proceedings of the North Eastern Linguistic Society Annual Meeting (Vol. 21, pp. 383–397). Amherst: University of Quebec at Montreal, University of Massachusetts.

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual Contexts: African Perspectives Felix Banda

Abstract  The chapter explores the notion of dominant language constellations (DLCs) in multilingual contexts of Zambia where 72 indigenous languages are spoken and English is used as the main language of education. I use data from migration and sociolinguistic literature, national census, online news media and call-in radio programmes to illustrate the multi-layered DLCs operating across individual/ household, community, regional and national boundaries. I argue that the different tiers of DLCs are a consequence of years of trans-regional and transnational migration as well as intense language and interethnic contact. The official language zoning, in which seven indigenous languages have been assigned regional language status for local government administration and early primary school education in ten provinces, and the colonial legacy of English as the language of national government business, secondary and higher education and socioeconomic mobility, have only added to the complexity and dynamism of DLCs. I show how the language practice of blending linguistic features from different languages results in discourses that reflects the DLCs operating in time and space. I conclude that the fluid multilingual contexts illustrated in this chapter suggest different DLCs operating simultaneously in and across individuals/households, communities and regions, and that the languages constituting these DLCs are not fixed leading to complex configurations and multiple strata of DLCs. Keywords  Africa · Zambia · Multilingualism · Dominant language constellation · Bemba · Nyanja · Ethnolinguistic diversity

F. Banda (*) Linguistics Department, University of the Western Cape, Cape Town, South Africa e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_5

75

76

F. Banda

1  Introduction Africans tend to be at least bilingual: in a local African language, in a regional African language and in at least one of the colonial languages (Banda 2016a, b; Kashoki 1975). Multilingualism and the language contact phenomena in Africa has often been linked to recent developments in mobile technology, colonialism and urbanisation (‘modernity’). However, sociolinguistic and cultural evidence suggests that there has been linguistic and ethno-cultural co-existence in Africa before explorers, missionaries and colonialists set foot on the continent. The use of multiple languages by Africans has also been associated with what is referred to in the literature as African youth language. However, having lived in a world without borders before the onset of colonialism in 1880, the language contact phenomenon is not entirely new to Africa. In their edited volume on globalisation on language vitality in Africa Vigouroux and Mufwene (2008) have argued that both globalisation and colonisation are not new phenomena, as Africans have been moving from one part of the continent to another for centuries before the arrival of Arabs, missionaries and European colonialists. The agriculturalists and pastoralists inhabited land used by hunter-gatherers, whom they also dominated. Whereas the missionaries learnt African languages and preferred to conduct their work in indigenous languages, the European colonialists and settlers that followed them after the partition of Africa were more interested in exploiting the economic resources than actively impose a European language on local populations. As for education, the first 3–5 years of primary education was in an African, with English or other European language introduced later, and by that time the majority of Africans had dropped out of school, leaving only a few elites to continue with education in English (Banda 1996). The status quo has not shifted much in contemporary Africa, where English or other colonial languages are main languages of post-primary education and national government business, while a few African languages have been proclaimed languages of regional administration and basic primary education. In everyday interactions, Africans mainly use African languages and sometimes, in combination with a European language. There is therefore a dynamic social structuring or division of labour in which languages, including ‘minority’ ones, co-exist (Banda and Jimaima 2017; Jimaima 2016; Mambwe 2014). Banda and Jimaima (2017), Jimaima (2016) and Mambwe (2014) show how non-official and so-called minority languages are found in dispersed linguistic landscapes outside their regions, and used in domains associated with English or the seven official Zambian languages. In addition, non-official and so-called minority languages have increasingly found a niche in high impact television advertisements, and plays aired on national radio and television, as well as in national online and print media (Banda 2016a). Studies on the co-existence and language use in Africa (Vigouroux and Mufwene 2008; Mc Laughlin, 2008, 2009) and Zambia in particular (Moody 1985; Banda 1996, 2016a, b; Mambwe 2014; Jimaima 2016; Banda and Jimaima 2017) suggest a fluid, not static, and complex multilingual linguistic dispensation (Aronin and Singleton 2008) in urbanising

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

77

Africa, where all languages minor or major, official and non-official, have roles to play. Therefore, considering that Africa still has the largest concentration of languages, numbering more than 2000 (Batibo 2005), and with a considerable number being in close proximity, either through historical coincidence or migration, makes the continent a rich site for research on what is called Dominant Language Constellation (DLC).  Aronin and Singleton (2012: 59) define DLC as “a set of languages on which the life of an individual or a society is dependent.” In this chapter, it is suggested that an individual or a community can have more than one set of DLCs. The aim of this paper to show that given Africa’s fluid multilingualism and intense language contact in which related and unrelated African languages, and colonial languages are in place, DLCs can shift in time and place dependent on the interlocutors and sociolinguistic context and languages in use. In particular, I use migration and sociolinguistic literature, national census data (Zambia Stats 2010), online news sites and two national call-in radio programmes to describe the shifting DLCs in the different data sets as indicative of home/individualised, community, regional and the wider national configurations of sets of DLCs. Considering that Zambia has 72 languages and seven of them have been designated ‘official’ regional/ provincial languages and used for initial literacy and local administration makes the study of DLCs very interesting. How are DLCs configured in multilingual homes, communities and the different regions/provinces? Are there overlaps and/or different layered DLCs across homes, communities and regions (or provinces as they are called in Zambia) considering the mobility and fluidity of language in multilingual contexts of Africa – especially considering domains of language use are not strictly applied as in the West (cf. Banda 1996, 2016a, b; Mambwe 2014)? A DLC is a group of one’s most important languages that operate as a holistic resource and thus enables an individual to meet all needs in a multilingual environment. Unlike linguistic repertoire, which embraces all the skills and registers of all the acquired languages, the DLC includes only the most expedient languages for a person (Aronin 2016). However, in fluid multilingual contexts in which several languages are spoken in homes and communities, the most expedient languages may change in time and space and in tandem with the linguistic/performative identities of the interlocutors. Communal DLCs are manifestations of how societal multilingualism occurs through linguistic practices. Individuals in a household and community in multilingual society may have different DLCs depending on whom they interact with in their daily routines. Aronin and Singleton (2012) argue that the notion of DLC relates to Gumperz’s (1968) conceptualisation of a speech community, defined as sharing a language variety. However, they are quick to point out that the monolingual orientation of the speech community notion as well as the fact that speakers can be spread over community and national boundaries makes it difficult to apply the concept in the fluid world of cultural flows (Appadurai 1996). DLCs can be constituted at individual, community, national and international sites of interaction. According to Aronin and Singleton (2012), for the Russian speaking community in Israel, their DLC is often constituted by Russian, effectively the ‘home’ language, Hebrew, the official

78

F. Banda

language of the country, and English the main language of higher education. The three languages can be said to form a consortium of the most important languages enabling a Russian migrant to meet the communication needs. It is possible that the individual may have other languages in their repertoire such as Latvian, but which may not necessarily be used frequently (Aronin and Singleton 2012, Aronin 2016). The DLC therefore becomes the mainstay of multilingual linguistic dispensation as the languages constituting it meet the everyday communication requirements of an individual or community. In conceptualising DLC, Aronin (2016) compares and contrasts DLC with the notion of linguistic repertoire, thanks to recent interest in the concept by academics such as Busch (2012) and Blommaert and Backus (2013). Aronin (2016) characterises linguistic repertoire as focusing on an individual’s linguistic skills in countable languages. Linguistic repertoire also focuses on the sum total of the languages or varieties that an individual or community of speakers is able to deploy in communication of various kinds. DLC on the other hand focuses on the selection of languages and skills deemed important to achieve communicative or interaction goals in specified contexts. Aronin illustrates the difference between linguistic repertoire and DLC by giving an example of Rose, whose repertoire is constituted by eight (8) languages: Spoken Arabic, formal Arabic, Hebrew, English, Spanish, Brazilian, Portuguese, Italian and Latin. However, only three (3) languages: English, spoken Arabic and Hebrew, constitute her DLC. Hebrew and Arabic are the official languages of Israel. She uses Arabic at home with family and relatives. English is the main language of higher education, and the language she uses at work and to source information online. The languages constituting a DLC can be conceived to function together as a language entity (Aronin 2016). This means they can be said to reflect languaging or translanguaging discourse in which the linguistic features of different languages are used simultaneously. In multilingual contexts of Africa, for example, it is not always possible to distinguish the different languages as first, second, third, etc. languages, as multiple languages are often acquired simultaneously (Banda 2009, 2010, 2016a). It is often the case that other than the former colonial language (English), in the case of urbanizing Zambia, and the official local/regional (Zambian) language, two or more other languages are often spoken in communities. In interaction, the languages are often deployed as a translanguaged unit. It would therefore be interesting to explore the linguistic features constituting the DLCs of such translanguaged discourses, and how these relate to individual and societal DLCs in place.

2  Early Migration and Multilingualism in Zambia Large-scale migration into a geographical area now called Zambia is said to have commenced around 1500  AD and lasted way into the late nineteenth century (Roberts 1976). Wotela (2010) identifies three waves of migration. The first wave

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

79

migrated straight from the Great Lakes region, with one group settling in what is now part of northern-east Zambia near Lake Tanganyika, while the other in the South-central part. The first group comprised the Mambwe-Lungu language cluster, while the second comprised the Tonga-Ila group of languages (Wotela 2010). The second wave of migrants into Zambia, were also from the Great Lakes region, but had originally settled in the Luba-Lunda Kingdoms situated in today’s Democratic Republic of Congo (DRC) before settling in the north-west and north-­ eastern Zambioa. These include the majority of ethnic groups in Zambia, such as the Bemba who settled in the northern part in the mid-eighteenth century and only to be joined by the Lunda of Luapula Province around 1740 (Richards 1940; Cunnison 1959 both cited in Wotela 2010). The third group comprised two main groups  – the Ngoni and the Sotho-Lozi groups, whose movement was triggered by events related to Shaka’s rise to power, and wars and plunder of conquered ‘tribes’ that followed. The movement is called Mfecane ‘crushing’ in Zulu, or Difaqane in Sotho which means forced removal or migration or scattering. This took place between 1815 and 1840. Zwangendaba’s Nguni (later called Ngoni) group settled in Eastern Zambia (Northern Malawi and parts of Tanzania), while the Sotho speaking Kololo displaced by the Difaqane wandered north and settled in today’s Western Zambia after defeating the Luyana and other ‘tribes’ in the area. Lozi language developed from the interaction of Kololo (Sotho), Luyana and other languages spoken the vicinity. However, I want to add the fourth wave of migration inspired by colonisation and the money economy that was thrust on the Africans. The colonial government and its agents required cheap labour for administration, while their agents needed labourers to work on farms, mines, homes and other business concerns. In Zambia, the British colonial government introduced what was called the poll tax, which forced young men to migrate to emerging urban areas to look for paid jobs. The towns and places of employment became sites at which different languages were spoken and ‘mixed’. However, colonialism also brought with it English as a language of education, trade, modernity and aspiration. As Banda (1996) and Banda and Mwanza (2017) observe, in colonial times those Zambians who spoke a little English were ‘rewarded’ with clerical work or supervising other Africans tasked to do manual and the hard work on farms and the mines. They were also paid better than those who actually did the manual job. The high status enjoyed by English language has not waned and in fact it has prospered even more than before in postcolonial Zambia (Banda and Mwanza 2017). The effect of migration is that as people move they not only transform the languages in their repertoire, but also in those of whom they interact. As outlined elsewhere a typical Zambian has at least three languages in their repertoire: two indigenous languages (the regional and community languages) and English (the colonial and official language of education and business). In relation to the conceptualisation of DLCs, it means the constitution of DLCs is not only complex but also dynamic.

80

F. Banda

Therefore, it can be said that migration in Africa and Zambia in particular, is not new; and more significantly, it can also be said that multilingualism is not a new phenomenon in Zambia. The description above also gives credence to Marten and Kula’s (2007: 18) argument that “the present-day language situation in Zambia is not so much a product of the colonial era, but is instead based on a dynamic system of multilingualism which has developed over several centuries. A number of the languages which play a part in the contemporary linguistic environment have been spoken  – in older forms  – in the area since the middle of the last millennium”. Banda (2016b) adds that people in present day Zambia are migrants who were brought together as a ‘nation’ by the colonial dictation, but the multilingual linguistic dispensation was set in motion way before Europeans set foot on Africa.

3  The Current Language Situation in Zambia Sociolinguistic literature suggests there are 72 languages spoken in Zambia (Kashoki 1978; Banda 2005, 2016a). However, only seven of them have been designated official (regional) languages. English is the main language of government, business and education at national level. The seven Zambian languages, Bemba, Nyanja, Lozi, Luvale, Tonga, Kaonde and Lunda have been ‘zoned’ to particular regions or provinces, where they are used for local government administration. The ten provinces, (with official regional languages in brackets) are: Lusaka (Nyanja), Central (Bemba), Copperbelt (Bemba), Eastern (Nyanja), Western (Lozi), Southern (Tonga), Muchinga (Bemba), North-Western (Kaonde, Lunda and Luvale), Luapula (Bemba) and Northern (Bemba). The pigeon-holing of languages into the ten language zones is in part a reflection of the colonial government’s and missionaries’ activities, which sought to bring order to what was considered the ‘tower of babel’ by selecting a few languages to be used for administration and limited primary education for a few Africans (Banda and Mwanza 2017). The colonial government selected Bemba, Nyanja, Tonga and Lozi as local languages of administration. The colonial government also thought it would be too expensive to print material for schools and government in more than these four languages. It is not clear what criteria was used to select these languages, but it became apparent to local people that they needed to speak these languages for wider communication as well as if they were to have a chance to get government employment (Banda and Mwanza 2017; Banda 1996; Wotela 2010). After Zambia’s independence in 1964, the emerging African government increased the number of languages to seven, with the addition of Lunda, Luvale and Kaonde. However, this suggests that speakers of the other languages have to use one or more of these regional languages in addition to their own in education and other official contexts of social interactions in their everyday lives. The latest Zambian census data shows that Bemba is the most widely spoken language followed by Nyanja (Zambia Stats 2010). It is also the case that despite the language zoning, languages are spoken outside their regions (Banda and Jimaima

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

81

2017). In their analysis of the vitality of several Zambian languages, using the linguistic landscape approach, Banda and Jimaima (2017) show that both regional and minority languages being used in the landscapes and in interactions outside the regions they originate. The rapid spread of languages and cultures across so-called regional and ethnolinguistic boundaries can be said to have been accelerated by the emergent African government of Kenneth Kaunda, Zambia’s first president after Independence on 24 October 1964. He tried to counter the colonial government’s policy of encouraging and even constructing ‘tribal/ethnic’-based communities in which to operate the so-called ‘divide-and-rule’. Kaunda, instead, encouraged inter-­ ethnic marriages and free movement of people across ‘tribal/ethnic’ community boundaries. Kaunda’s policy of encouraging inter-ethnic marriages coupled with that of transferring civil servants to work in territories other than their ‘tribal’ region also left its mark, as couples and children speaking two or more languages in the home is common (Banda 2016b). Increased inter-ethnic marriages and contact and movement of people escalate chances and forms of language contact, cross-­linguistic influences and multilingual linguistic dispensations, all leading to complex sets of DLCs operating in a community. Given the linguistic contexts outlined above, it is not surprising that sociolinguistic research shows that multilingualism in Zambia is not confined to urban areas, rural areas have also experienced forms of multilingual existence as communities speaking different languages and practicing different cultures live side-by-side, inter-marrying and criss-crossing in their daily interactions (Banda 2005, 2016b; Marten and Kula 2007). These people would go to the same markets and churches, while their children go to the same schools. Moody (1985) in his analysis of 25 conversations in various contexts; Banda (2016a) in his analysis of online newspapers and commentaries, and Mambwe’s (2014) study of radio and TV discourses, all suggest that typically Zambians speak at least four languages. This not only alludes to the fluid multilingual existence; it also suggests multiple and differently layered DLCs in operation in time and space, as one moves from one site of interaction to another in social contexts.

4  D  LCs, Language Zoning and Provincial Languages of Wider Communication The national census statistics (Zambia Stats 2010) show that both non-official and the seven official Zambian languages are spoken in varying amounts in other regions than where they originate. For instance Bemba is not an official language in Lusaka Province but is spoken by 17.3% of the population in the province (Zambia Stats 2010). Table  1 also shows that Tonga is the official language in Southern Province (74.7%) is also widely spoken in Central Province (15.5%) and Lusaka Province (4.3%). It is interesting that Nyanja is only natively spoken by 52% in the Eastern Province where it originates but by 61.9% in Lusaka Province

Total 33.5 1.2 1.8 5.5 11.4 20.3.8 3.0 2.6 1.3 1.9 1.5 1.9 1.7 4.7

Central 31.8 10.4 2.3 1.0 15.5 9.3 0.4 0.2 0.2 0.1 0.2 0.9 0.7 6.1

Copperbelt 83.9 0.1 9.2 0.3 0.8 0.8 0.1 0.2 0.1 0.3 0.3 0.7 2.1 0.5

Eastern 0.6 0 0 0 0.1 52 21.4 16.5 0 0 0 0 0.2 3.3

Luapula 71.3 0 0 0 0.1 0.1 0 0 0 0 0 0 0.1 16.3

Lusaka 17.3 0.6 0.1 1.3 4.3 63.1 1.6 0.4 0.3 0.1 0.2 0.2 6.2 3.1

Muchinga 46.9 0 0 0 0.1 0.3 0 8.2 0.5 0 0 0 0.1 3.8

Source: Adapted from Zambia Stats (2010). The languages in bold-font type are official regional languages

Language of communication Bemba Lenje Lamba Lozi Tonga Nyanja/Chewa Nsenga Tumbuka Mambwe Lunda Luvale Kaonde English Other language

Table 1  Languages of wider communication North-Western 4.9 0 0 0.7 0.3 0.6 0 0 0 33.8 19.5 29.6 0.6 7.4

Western 0.5 0 1.7 69.6 0.2 0.6 0 0 0 0.3 5.1 0.3 0.1 13.2

Southern 2.8 0.1 0 4.0 74.7 7.2 0.2 0.1 0 0.1 0.4 0.1 0.1 1.2

82 F. Banda

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

83

where it has been adopted as the official regional language. Lamba, Lenje, Tumbuka, Mambwe and Nsenga are not official languages but are spoken in many parts of the country. From the table, we get a glimpse of typical DLCs operating in each region/provinces, but which may differ at district and community levels, as well as on the urban-rural cline. It seems a typical DLC in each (urbanized areas) of the provinces will include the official indigenous language and English, and perhaps one or more other local languages. However, in rural areas, the DLC may contain fewer languages, which could include the mother tongue (and/or the regional language if the latter is not the mother tongue) and English especially for those that go to school. In the urban areas Central Province, a DLC could include the official regional language Bemba (31.8%), Nyanja (9.3%), Tonga (15.5%), Lenje (10.4%) and one or two other languages in various combinations. At one time the author taught English at a secondary school in the Central Province and would also typically speak Nyanja, Nsenga, English and Bemba at home, with neighbours and fellow teachers. The author’s DLC was slightly different while in the Eastern Province, where he did his primary and secondary school. He spoke Nyanja, Nsenga and English at home, and also spoke Tumbuka a very common language in the community he grew up. English was the medium of instruction at both primary and secondary school, while Nyanja the official regional language was a subject. At secondary school he also took French as a subject. During his primary education a typical DLC for the author would include Nyanja, Nsenga, Tumbuka and English, while at secondary school French was added to the languages constituting his DLCs.

5  The Impact of Language Zoning on DLC Considering the language zoning in place in Zambia, DLCs may differ depending on the province and also at personal levels. Moreover, since there are so many languages spoken in the different regions or provinces, the proclamation that a particular language is the ‘official’ language does not mean that other languages are not used in everyday interactions. It is with this in mind that Table 1 and the impact of language zoning on DLCs should be interpreted. Zambia Stats (2010) shows that although Bemba is the official language of Central Province, only 31.8% indicated they use it for communication; for the Copperbelt it was 83.9%; Luapula 71.3%; Muchinga 46.9% and Northern 69.2%. Similarly, although Nyanja is the official indigenous language in the Eastern Province, Zambia Stats (2010) shows that only 52% speak it while in Lusaka Province it is spoken by 63.1%. Further, Lunda is widely spoken by 33.8% of the population in North Western Province, Luvale 19.5% and Kaonde 29.6%. In Southern Province 74.7% speak Tonga while 69.6% speak Lozi in Western Province. The sociolinguistic picture that emerges from these data indicate the regional languages are not necessarily used by all as languages of wider communication in their designated regions. In essence, other local languages

84

F. Banda

that are not necessarily official may constitute very critical components of DLCs. It is also the case that the same family and community may have different DLCs. In other words, official regional languages are one component among other languages constituting multi-layered DLCs. Typically in the Eastern Province of Zambia one would expect a DLC constituted by English, the main language of government business and education; Nyanja the regional official language and language of initial literacy in schools, and Ngoni, Nsenga, Tumbuka and other languages spoken by local people. For instance, in the Lundazi District in the Eastern Province local people use Tumbuka as the main language of communication; English at work and in schools and Nyanja as official regional indigenous language. A Tumbuka child will use Tumbuka to communicate with parents and relatives at home, and also with neighbours and members of the community; she will need the regional language Nyanja and English for school purposes. Her father and mother, on the other hand, may have Nyanja in their linguistic repertoire, but do not necessarily need it in their DLC, as Tumbuka and English will cater for all their linguistic needs. However, supposing the child moves from the rural District of Lundazi to Chipata City District, in the same Eastern Province, but which has different languages in use: she will use Tumbuka at home, English and Nyanja at school, but likely Ngoni/ Nsenga and with neighbours and for social interactions, as Ngoni is the lingua franca in Chipata. The same DLC may obtain if the Tumbuka home language child were to move to Lusaka, or if she lives in Lusaka where Nyanja is the designated regional language. On the other hand, a child who speaks Tumbuka at home on the Copperbelt Province will have Bemba as language for wider communication in the community and for initial literacy in education, and English as the main language of education. A child coming from a Bemba speaking home on the Copperbelt Province has Bemba as language of wider communication in the community and for initial literacy in education, and English as the main language of education, government business and to watch both local and international television channels, which almost exclusively broadcast in English. This will be similar to a child from a Nyanja speaking home in Lusaka Province; the difference being in Nyanja replacing Bemba as the for wider communication in the community and for initial literacy in education. It could be argued that the patterns of 3 languages or more constituting DLCs are true of other provinces, but the constitution of the configuration of the languages may differ individually, at home and regional spheres of interaction. The fluid and complex configurations of DLCs are confounded by the situational and socio-cultural (and even religious) variables that might add to the number of languages constituting the DLC. For instance, the Copperbelt Province and Lusaka Province are originally Lamba and Soli speaking areas, respectively. Bemba speakers originally came from Luapula and Northern Provinces to come and work on the Copper mines, and their language ended up gaining prominence as local Lamba people refused to work for white mine owners for a pittance. Similarly, Nyanja speakers were originally brought in from Malawi as ‘policemen’ to enforce the colonial laws among the growing influx of black people from different parts of the country to the urbanizing capital city of Lusaka. However, the association of Bemba

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

85

and Nyanja with socio-economic status means they are often used as first, second, third or fourth languages not just on the Copperbelt and Lusaka Provinces, but other provinces as well, as illustrated in the national census data (Zambia stats 2010). Moreover, a Zambian Muslim requires Arabic for prayers, and would have the language as part of their DLC. Similarly, for cultural reasons those in inter-ethnic marriages may need to have the language spoken by the in-laws in their DLC.  For instance, the author currently uses English, Nyanja and Bemba at home; Ngoni and Nsenga to nephews/nieces, and some relatives to mother’s side; Nyanja to brothers and sisters and to relatives on father’s side; Bemba to the older in-laws and Nyanja, Bemba and English to nephews and nieces on wife’s side; Tumbuka with child-hood friends and certain social networks; and English (and some times Afrikaans and Xhosa in explanations and examples) for teaching at a university in Cape Town. Before moving to Cape Town, and residing in Lusaka, the author’s DLCs were different. For instance, while living in a block of flats in Lusaka (Zambia), the author would use Nyanja and English to the spouse, while the spouse would respond in Bemba, Nyanja or English in the home. In everyday interactions outside the home, the author would use Bemba, Nyanja and English to the spouse’ relatives, but Nyanja, Nsenga and Ngoni to his relatives. The spouse would use Bemba to her relatives both in Lusaka and in another region, Nyanja with the spouse’ family members and Nyanja with two neighbours on the same floor and Bemba with two neighbours on a different floor. With the neighbours on the same floor, the author would use Nyanja, Nsenga, Tumbuka and English, but Nyanja and English on neighbours on a different floor. At the local markets and for successful interactions in multilingual Lusaka generally, the author and spouse require skills in Nyanja, Bemba, and to some extent Nsenga and Ngoni in addition to English the language of education and national business. This illustrates that two people in the home can have multiple, parallel and overlapping configurations of DLCs in operation with different significant others in homes, regional and national levels.

6  DLCs and Print and Online Media Table 1 suggests that owing to their official regional language status Bemba, Nyanja, Tonga, Lozi, Lunda, Luvale and Kaonde  – are the more important languages of wider communication. This means each respective region will have English (the main language of education official government business) and at least one of these local languages in the DLCs. Since Bemba and Nyanja are the most widely spoken and more distributed languages across the country as is evident in Table 1, many everyday interactions in the different provinces may involve one of these two languages in addition to the official regional Zambian language of the province and English. The dominance of English, Bemba and Nyanja are reflected in media and Zambian music lyrics. Bemba and Nyanja are dominant in popular Zambian music (Mambwe 2014). There appears to be an assumption by the media in print, TV or online that a typical Zambian consumer speaks/understands English, Bemba and

86

F. Banda

Nyanja. Thus, for everyday interactions with the different media platforms these languages need to be part of a Zambian’s DLCs, as the media including English language media often carry extended extracts of Bemba and Nyanja without translations. Consider the following headline from The Post daily newspaper. Extract 1 TODAY’S EDITORIAL COMMENT: Kutumpa Source: The Post newspaper (4 November 2015)

The phrase Kutumpa means (to be) ‘stupid’ in Bemba. Although it is an English language newspaper, The Post does not provide an English gloss. This was a quote from the president who was castigating civil servants for campaigning for the opposition party in the coming presidential and parliament elections. The main story of the editorial read in part as follows. …But before I go back to Lusaka, I want the civil service to change, maka maka ba teacher baku Chingola. Let me take this advantage (sic) to warn those teachers: I will sort you out, you are supposed to serve the people of Zambia, but you are using government resources to campaign for the opposition. Nganamitamfya kutumpa kwenu. [maka maka ba teacher baku Chingola ‘especially teachers in Chongola’. Nganamitamfya kutumpa kwenu ‘If I fire you, blame your stupidity’.]

The translaguaged discourse comprising Bemba and English reflects typical languages in the DLC on the Copperbelt Province, where the President made the speech. However, The Post, which has since closed, was a national daily newspaper. The paper does not translate the Bemba parts of the discourse, which could be taken to mean that the editor makes the assumption that Bemba is typically part of the Zambians’ DLC, otherwise a gloss in English would have been provided. Consider the following headlines and news stories, one in Nsenga and Nyanja, and the other in Bemba in two online newspapers. The headline below appeared in the English online newspaper Zambian Watchdog on19 March 2015: Extract 2 Zambia si nyoko AB Tells Kambwili

Sinyoko means ‘it’s not your mother’ in English, but in an insulting way. Si- is a negation prefix in both Nsenga and Nyanja, while –nyoko is the root meaning ‘mother’ in Nsenga. But this gloss is not provided. The rest of the accompanying story is written in translanguaged Nsenga-Nyanja and English and with no English gloss proved: Andrew Banda (AB) says Chishimba Kambwili should shut up and allow citizens to discuss the President’s health. And Andrew says Zambians are yet to see the worst of the PF government. “Chishimba Kambwili should just shut up; we all have a stake in this country. If things go wrong, we must criticise. Kambwili cannot stop us from debating the President’s health. Zambia si dziko ya anyina olo awisi Kambwili, koma ni ya tonse (Zambia does not belong to Kambwili’s mother or father; it’s for all of us) …. Source: Zambian Watchdog (19 March 2015).

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

87

Another English online paper Zambia Reports on 23 August 2015 used a similar sub-headline but in Bemba. Extract 3 Flashback: Zambian Watchdog says GBM Is Corrupt; “Te ndalama Shanoko…”

The selection of the language, Bemba works in the same way as illustrated above. Te- is a negation prefix in Bemba equivalent to the Nsenga-Nyanja sidescribed above. Thus, Teshanoko (te-sha-noko) in Bemba is Sizanyoko (si-zanyoko) in Nsenga-Nyanja, which literary translates into ‘not of your mother.’ Note the difference in concord: in the above example, dziko ‘country’ has ya ‘of’ while ndalama ‘money’ has za, which is sha in Bemba. In short, like in the above example the choice of the choice of shanoko is pejorative as the Bemba honorific marker ba- has been dropped. The polite way would be sha-ba-noko (Banda 2016a). There is a sense not just from the census data but also from online daily newspapers that Nyanja and Bemba are perceived to be critical languages of wider communication. This makes the two languages candidates for DLCs regardless of a Zambian’s province of residence. The argument is that regardless of province of origin and residence, one requires English, Bemba and Nyanja to access and also to partake in the online commentaries. Banda (2016a) in his characterisation of the Internet as a platform for counter hegemonic discourse shows that Zambians’ commentaries on stories that are in English, are often in Bemba or Nyanja or a blend of the languages together with English.

7  DLCs and National Radio Broadcasts I indicated earlier that the Zambia nation census data (Zambia Stats 2010) suggest more than 64% of Zambians speak a language in the Bemba or the Nyanja language clusters as first or additional languages. Popular Zambian music and entertainment programmes on the national television broadcaster often have Bemba and/or Nyanja as critical languages for accessing the message or the entertainment value (Banda 2016a; Mambwe 2014). This is not to mean that other local languages are not used; but it seems there is awareness among musicians that they would reach a bigger market if their lyrics have vocabulary from the two languages. Thus, for one to appreciate the message in national radio and TV programmes and even English daily and online newspapers, one needs to have Bemba and Nyanja in the DLCs. The data below recorded from an ‘English’ radio station in Lusaka shows three languages in use: English the national language, Nyanja the regional official language and Bemba a non-official regional language.

88

F. Banda

Extract 4: ‘What to Call a Toyota Hilux Pickup’ Turn 1 2 3 4

Speaker Host: Caller: Host: Caller:

5

Host:

6 7

Caller: Host:

8 9 10

Caller: Host: Caller:

Dialogue Good morning! Who I am speaking to? Hello! Ndine Zulu[I am Zulu] Please, go ahead. Ingankale Toyota RB yaku farm. [It can be Toyota RB for the farm’ (referring to former president of Zambia, Rupiah Banda who came from retirement at his farm to become the country’s president)] Ahah mwayambo nyonza manje mwaona ka? […you are now teasing/offending, you see] Awe tukamba [no I am speaking’ (meaning he is not teasing)] Awe takamba kudala kuti Toyota RB niya bantu bobeula in the beginning of the edition [we have already stated that Toyota RB is for wealth people in the beginning of the edition] Awe neo ndine farmer [no I am a farmer] Ok ndimwe bafarmer! (Laughter) [Ok you are a farmer!] hahahahaha (ends call)

Source: Mambwe (2014: 154–155) Note: the English forms are in bold-type

The radio host begins with a greeting in English: good morning. Who am I speaking to? (turn 1) and maintains the language in turn 3: please go ahead, despite the caller switching to a hybrid Nyanja-English form in turn 2 in hello ndine Zulu ‘hello I am Zulu’. Here it is needs to be noted that in Zambia, a person using one language such as Bemba (or English) and another using Nyanja and continuing with the conversation is commonplace. After turn 3, the host abandons the official language policy of English, and follows the caller’s lead in using hybrid forms involving English-Nyanja and Bemba in the rest of the conversation. Let us look at the composition of the blends in more detail. Turn 4 Ingankale Toyota RB yaku is in Nyanja while farm is in English. Turn 5 is entirely in Nyanja, while turn 6 is made of morphemes from Nyanja and Bemba: awe tukamba ‘no we are speaking’, where awe ‘no’ and tu ‘we’ (plural subject marker) are Bemba and the root –kamba ‘speak’ is Nyanja (Mambwe 2014). The subject marker morpheme for ‘we’ in Nyanja is ti. It is interesting that, perhaps to try to capture the formal aspects of radio programme, the caller chooses to use the more formal and honorific tu ‘we’ instead of the informal ni ‘I’. In turn 7, the host deploys hybrid language involving Nyanja, Bemba, Nsenga and English: awe is Bemba; takamba kudala kuti Toyota RB is Nyanja; niya can be said to Nsenga; bantu bobeula is Town Nyanja/Town Bemba but with morpho-­phonology of Bemba and Nyanja. Ba- is noun class 2 for plural/honorific subject marking in Bemba; in Nyanja it would be a-; while bo- in bo-beula is the Bemba concord marker of the noun class. Note that whereas the [b] in bo- is pronounced as a voiced bilabial fricative, the one in –beula is pronounced as a voiced bilabial stop. The rest of the sentence is in English. Turns 8 comprise awe ‘no’ in Bemba; neo in Nsenga, ndine in Nyanja and farmer is in English. Turn 9 continues with hybrid language comprising English morphemes from Nyanja and Bemba: Ok is in English; ndimwe is in Nyanja and bafarmer is made up of plural morpheme marker ba- and the English ‘farmer’.

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

89

The morphemes and linguistic features from the languages Nyanja, Bemba and English (and Nsenga) reflect those found in online newspapers. Consider also the following extract from a popular radio programme on Zambia’s national broadcaster beamed from Lusaka city: Extract 5: ‘Police and You’ Turn Speaker 1. Host:

2. 3.

Discussant [Jere]: Host:

4.

Discussant [Jere]:

5.

Host:

6.

Caller:

7.

Discussant [Jere]: Caller:

8.

Dialogue Tikulandilani ku program yathu ya police and you. Ine ndine Miriam Zulu. Muno mustudio ndili ndi a Jere. A Jere patsani moni omvetsela. [We welcome you all to our program, police and you. In the studio, I am with Mr. Jere. Mr. Jere greet the listeners]. Ndiku patsani moni onse amene mvetsela kuwailesi. Ambuye amdalitseni…. [we greet everyone listening to the wireless radio. God bless you]. Lelo tifuna ticezelane panjila za utambwali kapena kuti uconman. Omvetsela tumani lamya pa number iyi 251,681. A Jere kansi uconman ndiye kuti bwanji? [Alright. Today we want to discuss modern theft or con. Listeners the number to call us on is 251681. Mr. Jere what is the act of con?]. Zikomo amai pafunso. Uconman niukawalala woipistisa ngako ndiye wabwela manje mu Lusaka….uconman ndi ukawala wakuba zinthu kapena ndalama kupitlila munjila zautambwali olo kapena tikambe kuti theft by false pretenses…kunamidwa pokumbelani. [Thank you madam for the question. Conning is a very bad type of modern theft which has become common in Lusaka. It is theft through crooked ways or theft of property or money through trickery or false pretences] [Phone rings] Conde a Jere tiyankhe phone. Hello, muli kumvestela ku police and you… Culani dzina lanu nakwamene mtumila. [Please Mr. Jere, let us answer this phone call. Hello you are listening to police and you. Tell us your name and the place you are calling from.] Ndine Bwalya ku Matero. Mulishani ba Jere.[I am Bwalya from Matero. How are you Mr. Jere?]. Ndili bwino a Bwalya. Muli Bwanji…kuli bwanji ku Matero? [I am alright Mr. Bwalya. How are you and how is Matero?] Ndifye bwino mukwai. Ku Matero kulifye bwino. Ba Jere ilyo lyashi mwaleta lilisana important pantu ubu buconman nabucilamo sana. Ine I encountered buconman one day lintu nalefwaya ing’anda…. [I am aright thank you. Matero is fine. Mr. Jere the topic you have brought today is very important because conning has become very rampant. I encountered it one day when I was looking for a house….]

Source: Mambwe (2014: 158–159)

The first noticeable aspect is that although this is a Nyanja programme, the title ‘Police and You’ is in English. Mambwe (2014) notes that although both the host (Miriam Zulu) and the discussant (Mr. Jere) use Nyanja with bits and pieces of linguistic features from English, the caller responds mostly in Bemba interspaced with English linguistic features. In essence the entire interaction takes with the host and discussant using (mostly) Nyanja while the caller uses (mostly) Bemba. In turn 1, the host welcomes the listeners in Nyanja with a few English words ‘programme’

90

F. Banda

and ‘Police and You.’ The discussant in turn 2 uses only Nyanja. The host uses mostly Nyanja but draws on English to explain the word utambwali, which she realises most Nyanja speaker will not understand, as it is uncommon, thus opts for the more common u-conman. This is a (urban) Nyanja term, which combines the rural or ‘standard’ Nyanja prefix morpheme for ‘state of [being]’ u- and ‘conman’ from English. Mr. Jere, in turn 4 uses mostly Nyanja and legalese in English ‘theft by false pretences’, which would be difficult to translate in Nyanja without losing meaning. As implied above, the caller’s language choice is interesting. In turn 6, Mr. Bwalya introduces himself in Nyanja, but from here on, he responds in Bemba interspaced with linguistic features from English. This does not mean that he does not speak or understand Nyanja, as he is responding to a discussion, which is taking place in Nyanja. In any case, Matero, where Mr. Bwalya resides is in Lusaka, and Nyanja is the official regional language. Nyanja is the main language of interaction and local business in Matero. Thus, Nyanja, Bemba and English constitute Mr. Bwalya’s DLC, as well as the DLCs of the host and discussant. It is seems clear that in spite of the languages and linguistic features chosen the three interlocutors are able to converse and understand each other (cf. Mambwe 2014). It is also the case that the three languages constituting the DLC are being used as a unitary linguistic entity (Aronin 2016) rather than as bounded and linguistically differentiated systems. The three languages are thus used as semiotic resources for meaning-making in social contexts. This supports Heller’s (2007: 1) argument against the view that, “languages are objectively speaking whole bounded, systems, and for the notion that speakers draw on linguistic resources which are organised in ways that make sense under specific social conditions”. The linguistic choices are not random or fixed. As Mambwe (2014) argues, speakers in the complex multilingual contexts of Zambia, do not always combine the different languages, or choose the same language combinations and linguistic features. The choices of language or language blends are a function of the social and cultural contexts of interaction. Similarly, DLCs can be said to operate across different social contexts. The radio presenters in Extracts 4 and 5 need English, Bemba, Nyanja and Nsenga as part of the DLCs to enable them field and answer questions and interact with callers as part of their job. Moreover, considering that Extracts 4 and 5 are broadcast from Lusaka Province to a national audience (all 10 provinces), suggests that DLCs operate across community and regional (and even national) levels. This means successful access to the national broadcaster requires one to have a DLC that includes English, Bemba and Nyanja. At regional level, in Zambia, there are some radio stations called ‘community radio stations’ designed to broadcast news and programmes with a local community flavour. In the Western Province where Lozi is the regional official language, Lozi and English constitute part of the languages of the DLCs, while in Southern Province Tonga being the official regional language; Tonga and English constitute part of the languages of the DLCs. In the Eastern and Lusaka Provinces, where Nyanja is the official regional language, Nyanja and English are part of languages constituting the DLC.  On the Copperbelt Belt Province, Northern, Muchinga, Central and Luapula Provinces where Bemba is the regional language, Bemba and

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

91

English constitute part of the languages of the DLCs. In all the regions the other languages that are used as mother tongues, and for other everyday critical interaction and communication needs, constitute the rest of the languages in the DLCs. For instance, in the Eastern Province non-official languages such as Nsenga, Tumbuka and even Bemba (an official language in another province) are very critical for everyday communication depending on which part of the province you reside, thus may be part of DLCs.

8  Summary and Conclusion There is no community or region/province in Zambia that is completely monolingual, as can be attested from the national census data (Zambia stats 2010) and Table 1. The seven languages that have been given official regional language status have their sociolinguistic profiles elevated in their regions of influence, and for some languages such as Bemba and Nyanja, their influence is noticeable even in regions where they are not official regional languages. The number of speakers of the regional official languages is boosted by the fact that they are spoken as second, third, fourth, etc., languages by speakers of other languages, which makes the regional official language ready candidates for DLCs. Considering the complex multilingual situation discussed in this chapter, there is a sense that many combinations of DLCs are at play, as one traverses home, community, regional and national boundaries of linguistic intercourse. Thus, continued migration and the propensity for inter-cross ethnic marriages and interactions means that DLCs which operate at home levels may reflect or run parallel to those that operate at community, regional and national levels. The examples I have used in this chapter outline the linguistic diversity and complex multilingual linguistic dispensation in place. The examples also showed that in the linguistic diverse contexts of Zambia, the languages constituting the DLC are blended and reflected in the discourses resulting from the language practices. Given the complex linguistic situation in Zambia spawned by years of migration and diversity, DLCs are dynamic and keep revolving in time and space over an individual’s lifespan. The complex multilingual contexts, cultural and linguistic diversity and density means different interlocking and diversifying DLCs may operate within the same locality and interactional order. For instance, a person from a Tonga speaking home in Lusaka, needs Nyanja the language of education and wider communication, which is also the official regional language and English the main language of official government communication and business. Someone from a Soli speaking home will have a slightly different DLC in Lusaka, as she will not need Soli to interact with relatives, and Nyanja and English. The official zoning of seven (7) indigenous languages in ten (10) provinces means that the DLCs operating in the different province are different. Each of the seven (7) Zambian languages and English will be part of the DLC in a province where it has official status. However, the situation is much more complex as people

92

F. Banda

speaking different ethnic languages often live side-by-side in Zambia. Thus depending on home language and range of inter- or cross-linguistic/ethnic contact, many DLCs are in operation in each of the ten provinces. For instance in the Eastern Province where English is the main language of business and education and Nyanja is the local lingua franca and official regional language, a Tumbuka speaking family in Lundazi (a district in the province), where Tumbuka is widely spoken still needs Nyanja the official regional language and English, the main language of education and government business. It is clear that the sociolinguistic situation in Zambia is complex, and so are the multilingual practices and linguistic dispensations in place. Similarly, the constitution and configurations of the DLCs is also complex, which may differ from (and within the) home, community, region to national spheres depending on the official regional language and the multitude of other non-official languages spoken. It seems therefore that the DLCs may differ from one person to another in the same household, community, region or nation. It also seems the case that a person can have more than DLC, but since some DLCs are used more than others, it can also be said that some DLCs are more dominant than others from an individual, community, regional and national perspectives. However, from the data it seems the DLCs in Zambia will typically have the regional language, the languages used in the home and community and English. Since Bemba and Nyanja tend to be spoken outside the official zones, these two languages usually become part of the DLCs for many Zambians regardless of region. In this regard, at provincial level, each of the seven (7) indigenous languages plus English constitutes the DLCs, which could include two sometimes more ‘home’ languages spoken locally but not necessarily recognised as official. Essentially different DLCs may operate at home, community, regional and national levels, and are reflected in the translingual practices and discourses of speakers in time and space.

References Appadurai, A. (1996). Modernity at large: Cultural dimensions of globalization. Minneapolis: University of Minnesota Press. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L., & Singleton, D. (2008). Multilingualism as a new linguistic dispensation. International Journal of Multilingualism, 5(1), 1–16. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Banda, F. (1996). The scope and categorization of African English: Some sociolinguistic considerations. English World-Wide, 17(1), 63–75. Banda, F. (2005). Analysing social identity in casual Zambian/English conversation: A systemic functional linguistic approach. Southern African Linguistics and Applied Language Studies, 23(3), 217–231. Banda, F. (2009). Critical perspectives on language planning and policy in Africa: Accounting for the notion of multilingualism. Stellenbosch Papers in Linguistics, 38, 1–11.

Shifting and Multi-layered Dominant Language Constellations in Dynamic Multilingual…

93

Banda, F. (2010). Defying monolingual education: Alternative bilingual discourse practices in selected Cape Town schools. Journal of Multilingual and Multicultural Development, 31(3), 221–235. Banda, F. (2016a). Towards a democratisation of new media spaces in multilingual/multicultural Africa. Stellenbosch Papers in Linguistics Plus, 49, 105–127. Banda, F. (2016b). Language policy and orthographic harmonization across linguistic, ethnic and national boundaries in southern Africa. Language Policy, 15, 257–275. Banda, F., & Jimaima, H. (2017). Linguistic landscapes and the sociolinguistics of language vitality in multilingual contexts of Zambia. Multilingua, 36(5), 595–625. Banda, F., & Mwanza, D. (2017). Language in-education policy and linguistic diversity in Zambia: An alternative explanation to low reading levels among primary school pupils. In M. K. Banja (Ed.), Selected readings in education (pp. 109–132). Lusaka: Litovia Limited. Batibo, H. (2005). Language decline and death in Africa: Causes, consequences, and challenges. Clevedon: Multilingual Matters. Blommaert, J., & Backus, A. (2013). Superdiverse repertoires and the individual. In I.  Saint-­ Georges & J.-J.  Weber (Eds.), Multilingualism and multimodality (pp.  11–32). Rotterdam: Sense Publishers. Busch, B. (2012). The linguistic repertoire revisited. Applied Linguistics, 33(5), 503–523. Cunnison, I. (1959). The Luapula peoples of northern Rhodesia: Customs and history in tribal politics. Manchester: Manchester University Press. Gumperz, J. (1968). Types of linguistic communities. In J.  A. Fishman (Ed.), Readings in the sociology of language (pp. 460–472). The Hague: Mouton. Heller, M. (2007). Bilingualism: A social approach. London: Palgrave Macmillan. Jimaima, H. (2016). Social structuring of language and the mobility of semiotic resources across the linguistic landscapes of Zambia: A multimodal analysis. Unpublished PhD thesis, Linguistics Department. University of the Western Cape. Kashoki, M. (1975). Migration and language change: The inter-action of town and country. In D.  Parkin (Ed.), Town and country in central and eastern Africa (pp.  228–249). London: Clarke, Doble and Brendon, Ltd.. Kashoki, M. (1978). The language situation in Zambia. In S. Ohannessian & M. E. Kashoki (Eds.), Language in Zambia (pp. 9–46). London: International African Institute. Mambwe, K. (2014). Mobility, transformation and localisation of language in multilingual contexts of urban lusaka. Unpublished PhD thesis, University of Western Cape. Marten, L., & Kula, N. (2007). Zambia: ‘One Zambia, one nation, many languages’. In A. Simpson (Ed.), Language and national identity in Africa (pp. 291–313). Oxford: Oxford University Press. Mc Laughlin, F. (2008). The ascent of Wolof as an urban vernacular and national lingua franca. In C. B. Vigouroux & S. S. Mufwene (Eds.), Globalization and language vitality: Perspectives from Africa (pp. 142–170). London: Continuum. Mc Laughlin, F. (2009). Senegal’s early cities and the making of an urban language. In F.  Mc Laughlin (Ed.), The languages of urban Africa (pp. 71–85). London: Continuum. Moody, J. (1985). Zambians talking: Twenty-five English. Lusaka: UNZA: Institute of African Studies. Richards, A.  I. (1940). The political system of the Bemba tribe of North-Eastern Rhodesia. In M.  Fortes & E.  E. Evans-Pritchard (Eds.), African political systems (pp.  83–120). London: Oxford University Press. Roberts, A. (1976). A history of Zambia. New York: Africana Publishing. Vigouroux, C. B., & Mufwene, S. S. e. (2008). Globalization and language vitality: Perspectives from Africa. London: Continuum. Wotela, K. (2010). Deriving ethno-geographical clusters for comparing ethnic differentials in Zambia. World Cultures eJournal, 17(2), 1–31. https://escholarship.org/uc/item/0j58j1nz. Zambia Census [Zambia stats] 2010. (2010). National census report. Republic of Zambia. Lusaka: Central Statistics Office.

Part II

Institutional Expressions of DLC

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present State and Challenges for the Future Siv Björklund, Mikaela Björklund, and Kaj Sjöholm

Abstract  Teachers in officially bilingual Finland are, according to the national curricular guidelines, expected to promote the development of language awareness and multilingual competence among their students. In order to do so, they need a well-­ developed multilingual awareness themselves. The concept Dominant Language Constellation could be used as a tool to describe the domains and functions of different languages and to promote multilingual awareness among student teachers. Thus the aim of this chapter is to describe what patterns of Dominant Language Constellations emerge at the societal and individual levels within the domain of education in Finland and to compare how these societal and individual patterns match each other. The research approach is qualitative and the data consist of the objectives and assessment descriptors specified for languages in the last two national curricula (2004, 2014) and data from two pre-service student teachers in Swedish-­ medium teacher education and two Swedish immersion student teachers. The chapter ends with a discussion of the usefulness of Dominant Language Constellation as a tool for describing multilingual language patterns at the societal and individual levels as well as for developing multilingual awareness among student teachers in Finland. Keywords  Dominant Language Constellation · Multilingual awareness · Teacher education · Student teachers · Finland

S. Björklund (*) · M. Björklund · K. Sjöholm Åbo Akademi University, Vaasa, Finland e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_6

97

98

S. Björklund et al.

1  Introduction The primary aim of this chapter is to describe what patterns of Dominant Language Constellations (DLC) emerge at the societal and individual levels within the domain of education in Finland and to compare how these societal and individual patterns match each other. We also endeavor to discuss briefly the usefulness of DLC as a tool for describing multilingual language patterns at the societal and individual levels as well as for developing multilingual awareness among student teachers in Finland. The concept DLC was originally proposed and used by Aronin (2006) to refer to a “person’s most expedient languages, functioning as an entire unit and enabling an individual to meet all his/her needs in a multilingual environment” (Aronin 2019, p. 21). Thus, the definition of DLC aims at highlighting the way human beings use their languages instead of a more static view of what languages we have or master (cf. language repertoire) (Aronin 2019). It should be noted that Aronin’s definition entails a relationship between the most expedient languages of an individual and a multilingual environment. The interdependence between these two together shapes the dynamics of DLC, and therefore individual DLC and societal DLC constantly influence each other. We find that Finland, a bilingual nation with two national languages, Finnish and Swedish, provides an interesting case where the inclusion of more than two languages in the language education program is natural and thus provides a good starting point for the investigation of societal DLC, and in this case, within the field of education. Contrastingly, officially bilingual Finland is compared with other Nordic countries by far the least multilingual and multicultural with no more than 7.5% speaking other languages than the two national languages Finnish (87.3%) and Swedish (5.2%)  (Official Statistics Finland 2020). In Sweden, for instance, the percentage of people with a foreign background is 25.5% (SCB 2020). The population of Finland is relatively small (5.5 million), but there has been a steady growth of the population during the last hundred years. A current trend is that the natural growth (due to the number of births exceeding the number of deaths) has come to an end during the last few years. Thus the population growth is today almost completely due to immigration. As the speakers of indigenous languages in Finland have tended to decrease, the influx of immigrants has made Finland more culturally and linguistically diverse than before. The largest groups of immigrants are the Russian speakers (81.600) and the Estonians (49.400). The Arabic speakers (31.900) have increased rapidly during the last few years and have exceeded the number of English  speakers (22.000) and Somali speakers  (21.900) (Official Statistics Finland 2020). Even though many countries are much more multilingual, we believe that Finland may provide a great potential for the development of successful policies and practices for multilingual education (cf. Björklund et  al. 2013). Does, for instance, Finland’s history as an officially bilingual country and its experiences of managing official bilingualism prove useful as linguistic diversity increases? (cf. Latomaa and Suni 2011). In practice, Finland’s bilingual status implies that there are two separate school systems (a Finnish and a Swedish one) with a similar curriculum. In bilingual areas where there exist Finnish- and Swedish-medium schools side by side,

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present…

99

there is occasionally some co-operation across the language boundaries (e.g. exchange of students and the like). Although multilingualism is considered to be an attractive aim at all levels of education and is mentioned in many official documents and national strategies, English has emerged as an indispensable global lingua franca, especially after Finland entered the European Union in 1995. A dichotomous situation seems to have arisen between a view that it is enough to know English versus a view that other languages and insights into multilingualism are also needed. The indispensability of English tends to be supported for instance at the university level, because the application for external funding of international research projects has to be done in English, also when it comes to the dissemination of important research results in international journals, English is the natural choice (cf. Heimonen and Ylönen 2017). At the same time, Swedish has lost ground as a national language, partly due to the eagerness of the 2015–2019 government to centralize, and partly due to the anti-­ Swedish policy of the Finns’ party. Some Swedish institutions have been closed down or been merged with Finnish institutions. Similar trends have been noticed among the national minorities (e.g. Sámi people in Finnish Lapland). But today, after Brexit and after EU having lost some of its stability, there is a trend that the Nordic dimension may be reinstated which implies that the importance of Swedish in Finland may be restored. It has also been noticed that some immigrants choose Swedish (instead of Finnish) as a secondary language as it is seen as a bridge to the cultures and languages of other Nordic countries.

2  L  anguages in National Curricula and Pre-service Teacher Education Teacher education programmes in Finland generally hold high standard, almost all kinds of teachers are required to hold a master’s degree. It is also noteworthy that Finland has achieved top results in the Pisa evaluations of students’ educational attainment at the school level (Programme for International Student Assessment 2018), especially so in reading, but also in science; in fact considerably better than in the other Nordic countries. Though schools are mostly separated into schools with either Finnish or Swedish as the language of instruction, there are also some good examples of cross-language co-operation between the two systems as in the pre-service education of immersion teachers (Swedish immersion) for the Finnish-­ medium schools. Together with Canadian pre-service programs, this program represents the most highly developed education of language immersion teachers worldwide (see Peltoniemi 2015). The data collected in this study comprises of two student teachers who in the future will act as class teachers in grade levels 1–6 (7–12 year-old pupils) in Swedish immersion, and two student teachers who prepare for work as class teachers for the corresponding age groups in Swedish-medium schools in Finland. Both student teacher-groups aim for a master’s degree. However, the student teachers in Swedish-medium schools have education  as their major

100

S. Björklund et al.

subject, while the major subject of the student teachers in Swedish immersion is the Swedish language that entails courses in Swedish language and culture as well as on immersion education and bi- and multilingualism. The focus on language in the latter program is due to the fact that the program predominantly addresses Finnish-­ speaking students who enter the program with an independent user level (B2), in terms of the Common European Framework of Reference for Languages (hereafter CEFR, Council of Europe 2001), in the other national language Swedish, but who in their profession will use mainly Swedish as the language of instruction in class. The two target groups represented by four student teacher voices are well suited for investigating possible DLCs. Student teachers in Swedish immersion prepare for operating daily in Swedish, Finnish and other languages in school. Similarly, student teachers for Swedish-medium schools have to attend to both Swedish and Finnish as many of them will work in regions where Swedish is a de facto minority language and Swedish-medium classes may consist of two thirds of bilingual Swedish-Finnish pupils. Thus, both types of student teachers can be expected to have developed some language awareness, because of the active language related choice of the immersion student teachers, and the national language status, but de facto minority position, of Swedish in Finland. Finland introduced a new core curriculum for basic education in 2014, which has been implemented from August 2016 (Finnish National Board of Education 2015). It also seems that the 2014 national curriculum exhibits some aspirations to adjust the classroom practices to match the increasing cultural and linguistic diversity by expecting teachers to promote the development of language awareness and multilingual competence which is here illustrated with the following target descriptions related to cultural diversity and language awareness. The relevant core areas from the point of view of this article are highlighted in bold face. T1 supervise the pupil to notice the linguistic and cultural diversity in the closest environment and in the world, as well as the position of English as a global language T2 motivate the student to appreciate her/his own linguistic and cultural background and the linguistic and cultural diversity in the world, as well as to treat people without prejudice T3 guide the pupil to notice what unites and separates different languages, as well as support the development of her/his linguistic power of deduction (Finnish National Board of Education 2015, pp. 246–247) The targets described above are general in kind, and require all teachers to face up to the challenge of supporting language and cultural awareness among their pupils, as well as fostering a sound linguistic and cultural identity and pride of one’s own linguistic resources. In relation to the explicit language syllabi, it is worth pointing out that there are different syllabi for the first languages Finnish, Swedish, Sámi, Romani, Sign language in the two latest national curricula from 2004 and 2014. The 2014 national curriculum (Finnish National Board of Education 2015), presents six syllabi for the second national language subject: an advanced syllabus in Swedish and Finnish, an intermediate syllabus in Swedish and Finnish, and the syllabus in

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present…

101

native-level Swedish and Finnish. In addition to that, seven different syllabi have been defined for the foreign language subject: advanced syllabi in English, another foreign language, and Sámi; an intermediate syllabus in an unspecified foreign language; and basic syllabi in an unspecified foreign language, Sámi, and Latin.

3  Research Questions Our study addresses the use of DLCs as a methodological tool for describing and analysing multiple language use in the domain of education in Finland. The analysis consists of two parts, where the first part aims to study how Finnish society positions the use of different languages in the school context and the second part aims to investigate multiple language dynamics among individuals in teacher education. In the first part, we scrutinize the two recent national core curricula for basic education to investigate how assessment descriptors of different languages are described and prescribed. In the latter part of the empirical analysis, we highlight how student teachers conceive their use of languages and their multilingual identity in officially bilingual Finland. The focus is on looking into two groups of student teachers, Swedish L1 speakers and Finnish L1 speakers. It is believed that a DLC approach might be a useful methodological tool in a study like this. Unlike a language repertoire, a DCL comprises the languages, which, together, perform the most vital functions of a language (Aronin 2017). Whereas language  repertoire  relates to the totality of an individual’s skills and resources where each language is dealt with separately, DLC is concerned with the vehicle languages standing out as being of prime importance. Thus, DLC is seen as an emergent entity, a form of social action and linguistic practice, a process rather than a product (Aronin 2017). Our empiric research questions are: • What pattern of DLCs emerge in the two latest national core curricula for basic education in Finland? • What patterns of DLCs emerge among bi-/multilingual student teachers in Finland? We expect these operationalized research questions to provide us with a platform for discussing the following questions: • What can be suggested regarding the usefulness of DLC as a tool for describing multilingual language patterns at the societal and individual levels? • What is suggested for developing multilingual awareness among student teachers in Finland?

102

S. Björklund et al.

4  Data and Method As mentioned, the focus is set on the educational domain and more specifically on basic education and the approach is qualitative, hermeneutic-phenomenological (Geanellos 2000). Thus, the method could be described as directed interpretive content analysis (cf. Armat et al. 2018; Geanellos 2000). As the source of data at the societal level the two most recent Finnish national core curricula for basic education were chosen and explicitly the sections where the objectives, core content and assessment criteria for mother tongue/first language (L1) Swedish, Finnish and Sámi, the advanced syllabi for second national languages (SNL) Swedish, Finnish, and the advanced syllabi for foreign languages (FL) English and other unspecified foreign languages for pupils in basic education (grades 1–6) are described (Finnish National Board of Education 2004, pp.  44–83, 118–147;  Finnish National Board of Education 2015, pp. 160–177, 196–205, 220–228). The reader ought to note that both the Swedish and the English versions have been made use of in the analysis of the 2004 national curriculum texts, whereas mainly the Swedish version was used for analysing the texts from the 2014 national curriculum. The English version of the latter turned out to be less faithful to the original versions in the national languages, which is why the authors’ own translations to English are used for the 2014 national curriculum. Regarding the curricular texts, the aim was to identify patterns related to language as a vehicle for communicative purposes and as expression of identity (cf. Aronin 2019) of the above mentioned languages, in order to identify societal DLCs in both curricula. Hence, there was a degree of directedness in the analysis based on previous knowledge and the preconceptions of the researchers. The analysis then pended between part and whole in order to validate the results and keep them as close to the lived experience as possible. The analysis was challenged by the fact that the categorization of objectives has developed since 2004. In 2004 the objectives and assessment descriptors were categorized into: Reading/Text comprehension, speaking, listening comprehension and writing, whereas the categories used in 2014 identifies communication, understanding and production. In addition to this, culturally related aims are also referred to in both curricula, in 2004 under the heading Cultural skills and in 2014 as Growing into cultural diversity and language awareness and thus makes it impossible to compare some aspects on a one-to-­one basis. The data comprising the student teacher voices were gathered during the years 2015–2016 through recorded, self-directed group discussions with a set of questions on language identity, attitudes towards their languages and language use within different domains. At the time when the group discussions were made the focus of the research team was not on individual DLCs among student teachers but on their perceptions about language use in order to learn how aware the student teachers from the two target groups are of their own multiple language use and language identity. Hence, the question of what languages constitute dominant languages of the individual student teachers was never explicitly addressed during the discussions, but during the discussions it was quite apparent that language dominance emerges in different ways, i.e. in how student teachers define themselves linguistically, in what domains and how frequently they use their languages. The data of the four student

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present… 103

teachers have been analysed in accordance with the above described method, and with a focus on data that can be used for comparisons with the curricular texts of the study. In the selection of individual student voices, two individuals from each group (Swedish-medium student teachers and Swedish immersion student teachers), representing different profiles in terms of languages as vehicle for different communicative purposes and languages as expressions of identity, were chosen. All participants highlighted in this study have been informed about the use of the data for research purposes and have approved to participate in the study.

5  Results The first part of the results presents data related to the two latest national curricula and hence attempts to provide some answers regarding the communicative purposes and identity expressions of different languages at a society level. The second part presents results related to individual student teachers, their language use and language identities.

5.1  Curricula The results presented below illustrate and describe variation in the prescribed communicative purposes and identity aspects of some of the languages presented in the national curricula from 2014 and 2004. Competence levels are used to illustrate the expected growth in communicative functions. For the sake of clarity of presentation and economy of space, this chapter only presents results concerning the first language (L1) syllabi Swedish (SWE), Finnish (FI), and Sámi, as well as second national language (SNL) advanced level syllabi for Finnish and Swedish and foreign language (FL) advanced level syllabi for English and other foreign languages (commencing for beginner pupils in grades 1–6). 5.1.1  Communicative Purposes The communicative functions are illustrated with examples for the language learning assessment descriptors in the 2014 curriculum (Finnish National Board of Education 2015). Even though the studied curriculum does not apply the levels of competence identified in the CEFR (Council of Europe 2001) for L1-syllabi, an interpretation of the verbal descriptors used in the curriculum has been made in relation to the CEFR-levels with the result described below. Firstly, Table 1 shows that quite high competence expectations are identified for all three L1s, which indicates a very broad spectrum of communicative purposes. Secondly, differences are also noted regarding the competence levels in the different

104

S. Björklund et al.

Table 1  Interpretation of the expected competence levels for good competence in L1 Swedish, Finnish, and Sámi by grade 6  in the  2014 national curriculum (Finnish National Board of Education 2015) Language L1 Swedish L1 Finnish L1 Sámi

Communication B2-C1 B2-C1 B2-

Interpretation B2-C1 B2-C1 B2

Production B2 B2 B2-C1

Table 2  Description of good performance at the end of grade 6 Language SNL FI SNL SWE FL ENG FL Other

Communication 2004 and 2014 n.a./A2.1 n.a./A1.3

Interpreting 2004 and 2014 A1.3 (listen), A2.1 (read)/A2.1 A1.3 (listen), A1.3 (read)/A1.3

Production 2004 and 2014 A1.3 (speak), A1.3 (write)/A1.3 A1.2 (speak), A1.2 (write)/A1.2

n.a./A2.1 n.a./A1.3

A2.1 (listen), A2.1 (read)/A2.1 A1.3 (listen), A1.3 (read)/A1.3

A1.3 (speak), A1.3 (write)/A2.1 A1.2 (speak), A1.2 (write)/A1.2

L1s presented in the curriculum. Both Finnish and Swedish as first languages appear to follow the typical pattern of expecting interactional and receptive language uses to develop faster than the productive uses. Interestingly enough, the L1 Sámi-­ descriptors represent an almost opposite pattern, where higher expectations are set for the productive uses than for interpretation and communication. This can be interpreted as an indication of differences in expected communicative purposes and functions (see Sect. 5.1.2 below) of L1 Finnish and Swedish over the other official minority languages identified in the 2014 national curriculum. In order to follow the consistency of expected competence level patterns of different languages over time, the assessment criteria for the two SNLs Swedish and Finnish as well as the FL English and other FL in the two last national curricula were compared. The results are shown in Tables 2 and 4, where the levels and descriptors defining a “good performance” for grade 6 are presented. As mentioned, the assessment criteria, or “descriptions of good performance” as they are called (Finnish National Board of Education 2004, 2015), for language education in the two last national curricula are based upon the levels of competence identified in the CEFR (Council of Europe 2001). However, the areas of competence identified have changed to some extent between the curricular reforms, which means that the results presented are our interpretations of the identified areas of competence. In the 2004 curriculum the CEFR-competence levels were divided into the four traditional areas of language competence: listening comprehension, text comprehension, speech, and writing, whereas the 2014 curriculum uses the division communication, interpretation, and production. In the 2004 curriculum explicit criteria for the assessment of communicative and interactional aspects of language competence were not included, even though communication strategies is identified as one of the key content areas, as indicated in Table  2. It is also worth noting that for SNL Finnish in the 2004

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present… 105

curriculum, the communicative focus can only be found implicitly among the objectives. An explicit development has thus taken place regarding the communicative aspect of language performance. This is relevant for all listed languages, but most prominent for SNL Finnish, indicating that there has been a strengthened focus on the communicative and interactional functions of the languages included over time. There has also been an increase in the competence levels expected for SNL Finnish and FL English between the two latest curricular reforms, whereas the competence levels identified for SNL Swedish and Other FLs have remained stable over time. Contrary to the L1-results, this pattern seems to indicate that SNL Swedish is to be regarded more as a “national foreign language” than  a second national language with a wide variety of communicative functions. The competence levels stated for English as FL point in the direction that English is moving towards being a “third national language” (c.f. Leppänen et al. 2008) in terms of communicative purposes. Even though it is not part of the actual data presented, it might be worth noting, that the advanced level syllabus for SNL Swedish included in the table above is studied by a minority of Finnish-speaking pupils who commence their studies of SNL Swedish in grades 1–6, whereas a majority of Finnish-speaking students in Finland start their studies of Swedish in grade 6. A majority of Swedish-speaking pupils, on the other hand, commence their studies in the SNL Finnish along the advanced syllabus. 5.1.2  Functions and Identity Regarding the objectives described for different L1s in the 2014 national curriculum (Finnish National Board of Education 2015), the cultural objectives provide a clear variation, which seems to indicate differences both regarding functions and identity. As shown in Table 3 below, the most obvious difference is related to Sámi, which in this descriptor is clearly viewed as a language for maintenance of an indigenous Table 3  Assessment criteria related to language, cultural awareness, knowledge and identity for L1 Swedish, Finnish and Sámi in grade 6 in the 2014 national curriculum (Finnish National Board of Education (2015) Language, cultural knowledge, awareness and identity in the 2014 national Language curriculum L1 The pupil has knowledge of her/his own as well as the languages and cultures of Swedish others, and can describe and reflect upon similarities and differences between different languages and cultures. (p. 166) L1 The pupil is able to describe noticed similarities and differences between different Finnish languages and cultures. The pupil can talk about media- and cultural offerings of their own interest/of interest to them. (p. 171) L1 Sámi The pupil can describe the main characteristics of the Sámi way of life and name the most common manifestations/expressions of Sámi culture. (p. 177)

106

S. Björklund et al.

minority culture (cf. Sarivaara 2016). The difference to both L1 Finnish and Swedish is remarkable, since the descriptors for both of these clearly point towards language and cultures of contact and the importance of interlingual awareness. There are, however, also differences in the descriptors for L1 Swedish and Finnish. The descriptor for Swedish departs from the pupil’s own language and culture and contrasting that to other languages and cultures, whereas that point of departure is not mentioned in the descriptor for Finnish. One could speculate that it might be an indication of the will to highlight the pupils’ ownership of Swedish, whereas the majority position of Finnish has not made such an identity marker an explicit issue. Also, knowledge as well as an ability to describe and reflect upon cultural and linguistic differences and similarities are stressed for Swedish. For Finnish these cognitively demanding aspects are not mentioned, but rather the ability to talk about the linguistic and cultural similarities and differences noticed by the pupil. Furthermore, the Finnish descriptor refers to consuming different media products, a dimension missing entirely in the cultural descriptors for the other two L1s, even though literature and other cultural and multimedia products are mentioned elsewhere e.g. in the curricular objectives and/or assessment descriptors. Hence, the interactional function and the relation to current media are stressed for Finnish rather than aspects of widening ones frames of reference and deepening the ability for reflection. These culturally related descriptors seem to indicate that Sámi is taught mainly to strengthen cultural and linguistic identity, whereas the descriptors for the other L1s indicate that the languages are seen as a useful point of reference in relation to other languages., i.e. strengthening language awareness and a broader sense of language identity. For Swedish the functions related to cognitive development and reflection become clear, whereas mainly the interactional aspects in relation to other people and media products are stressed for Finnish. On the basis of these descriptors one could argue that Swedish as a second national language is to be seen as a language of education, with little connection to contemporary media products. However, that is too narrow a view, since other parts of the 2014 national curriculum, as mentioned, do balance the lack of reference to media-products in the culturally related assessment criteria. What can be claimed, however, is that the culture-­transmitting function is more connected to current media and culture products for L1 Finnish than for L1 Swedish and Sámi. In Table  4 we will continue with a look at the culturally related assessment descriptors for the SNLs and FLs. These assessment descriptors related to cultural skills and awareness show that the curricular scope has become both broader and more explicit between the two national curricular reforms. This developmental pattern can be noted regarding all the languages specified in Table 4. For instance, in the 2014 descriptors a language awareness perspective is present in the descriptors for all the listed languages, and hence also an evident expectation of multilingual language communities, whereas the 2004 descriptors focus on contrasting the pupil‘s own culture and the target language culture. Especially for SNL Finnish one can see a progression from more nationally focused, realia-centred assessment criteria to a broader multilingual and reflective

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present… 107 Table 4  Description of good cultural skills in SNLs and FLs at the end of grade 6 in 2004 and 2014 national curricula (Finnish National Board of Education 2004, 2015) Language 2004 national curriculum (Eng) SNL FI “The pupils have to be able to state some of the most important differences between Finnish and Swedish, and have a command of basic information, such as names, localities, places and manners in Finland.” (p. 130) NB! Not explicitly highlighted as cultural skills. SNL “The pupils will know the main SWE contents of, and differences between, their own culture and Swedish-language culture. The pupils will be able to interact with Swedish-speaking individuals in simple everyday situations.” (p. 120)

2014 national curriculum (Swe) The pupil is able to give reasons why Swedish and Finnish are spoken in Finland. The pupil perceives similarities and differences related to e.g. grammar, vocabulary and meanings between Finnish and the mother tongue or other languages the pupil knows. The pupil can describe in what situations/connections one can read and hear Finnish. (pp. 200–201)

The pupil is able to provide reasons to why Swedish is spoken in Finland and describe the linguistic diversity in the Nordic countries. The pupil is able to observe/notice differences and similarities related to grammar, vocabulary or other aspects between Swedish and the mother tongue or other languages the pupil knows. The pupil is able to tell/explain where Swedish can be seen or heard. (p. 204) FL ENG “The pupil will know the main The pupil is able to generally describe what kind contents of, and key similarities of languages there are in their own surroundings, what languages are most commonly spoken in the and differences between the world and how wide-spread English is. The pupil culture of their own language and the target language, be able perceives similarities and differences related to to interact with speakers of the e.g. grammar, vocabulary and meanings between English and the mother tongue or other languages target language in simple the pupil knows. (p. 223) everyday situations.” (p. 141) FL Other “The pupil will know the main The pupil knows that the languages of the world contents of, and key similarities are divided into language families and can describe to which one the target language belongs. and differences between the The pupil can account for the languages spoken in culture of their own language and the target language, be able her/his surroundings. The pupil perceives to interact with speakers of the similarities and differences related to e.g. grammar, vocabulary and meanings between the target language in simple target language and the mother tongue or other everyday situations.” (p. 141) languages the pupil knows. The pupil can describe where the target language can be read and heard. (p. 227)

scope. Contrary to the differences related to cognitive development and reflection in relation to the L1s, the descriptors for SNL Swedish and Finnish are quite similar, with the exception that the Nordic dimension is mentioned explicitly for SNL Swedish and that similarities and differences regarding meanings are mentioned for SNL Finnish and the FLs, but not for SNL Swedish. A strong resemblance is thus

108

S. Björklund et al.

evident also for FL English, where the most marked difference is that pupils are not required to explain where English can be seen and heard, presumably because it is taken for granted that English is present in the surroundings of even young pupils. English is brought forth as a language with a global function, whereas the dimension that different languages might be more or less related is stressed for other foreign languages, drawing a picture of a quite different identity relations and scope of functions for the different FLs.

5.2  Individual Student Teacher Voices The excerpts chosen to  highlight individual student voices on  the dynamics and communicative purposes of their multiple languages and languages as expression of identity comprises four female student teachers. Thus, the data presentation follows the same categorization as the one used in the previous section on languages perspectives in Finnish national core curricula 2004 and 2014. The individual students were anonymized by naming the Swedish-medium student teachers Frida and Dana and the Swedish immersion student teachers Peeta and Opri. 5.2.1  Communicative Purposes Though the definition of multicompetence has been modified during the years (see e.g. Cook 1991, 2003, 2012), it clearly aims at treating the diverse languages as a coherent whole, rather than selecting one language separately. In the voices of the Swedish-medium student teachers this coherence between the separate languages is expressed by explicit comparisons but also by using different verbal markers to intricately group languages together referring to the competence level (see examples 1–2). 1. “Swedish is my mother tongue, Finnish almost as my mother tongue and I know English and French.” (Frida) 2. “I know Finnish, English, I ought to be able to say I know German, French, Spanish. I knew them, then. The Nordic languages one understands acceptably, except Icelandic” (Dana). In her list of languages Frida chooses to start by mentioning Swedish as her mother tongue, whereas Dana excludes Swedish in her first list of languages even though she later during the discussion confirms that she regards Swedish as her mother tongue. Frida groups English and French together as languages she “knows”, whereas Dana groups Finnish and English together and then inserts a brief comment on her “ought-to-have” knowledge of German. Following her remark about her

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present… 109

competence in German, Dana adds French and Spanish, and the discourse indicates that she is now listing languages she associates with a previous learning context (probably school) where she has studied the languages but now feels that her competence has decreased. The excerpt ends with a shift towards a more general remark that Nordic languages, except Icelandic, are in general comprehensible for Swedish-speakers. The reflections by the Swedish immersion student teachers Peeta and Opri show a similar point of departure as found for Dana; i.e. the mother tongue is taken for granted and is therefore not included when reflecting about multiple language use and communicative purposes. Also the discourse centered on language identity supports the notion that the Finnish language is very much at the core of the languages used by Peeta and Opri (se next section). For their other languages Peeta and Opri do not refer to skills and competences but rather to domains of learning and usage. All four student teachers express a positive view towards  strengthening the less active languages in their language repertoire or enlarging their present repertoire with new languages. Opri would like to maintain and gain more fluency in the languages she has already learnt, especially French, a language that also Dana would like to learn better. Frida would like to learn Spanish and Italian and expresses the view on oral interaction as an important communicative purpose, since she states that she would be most successful in this effort if given the opportunity to spend time abroad. Opri  too underlines the importance of regular oral interaction with native speakers and concludes that the best way to learn languages is to live in the country. Peeta endeavors to refresh her skills in French and Spanish and would like to include German and Russian in her language repertoire. 5.2.2  Functions and Identity The national languages of Finland position themselves as languages that are part of the early years in two of the four student teachers lives. Frida states that she did not master Finnish during her early years even though her mother is Finnish-speaking. She maintains that she actually did not learn Finnish until she moved to the capital area of Finland. At home Opri’s mother spoke Swedish with the children who answered in Finnish. Whereas Frida and Opri were raised in families with both languages, Dana and Peeta have similar trajectories when it comes to learning the other national language. They did not encounter the other national language until later in life. Dana was raised in a small Swedish-speaking village and she really had to struggle to learn Finnish, while Peeta says she learnt Swedish when she began her university studies. It is noteworthy that in the discourse of the two Swedish immersion student teachers there are no words indicating that the learning of the other national

110

S. Björklund et al.

language (Swedish) has been specifically important or stressful. It seems like Opri as a child did not find it relevant to use Swedish at home since her family’s shared language was Finnish, whereas Peeta finds university studies in the Swedish language to be her entry point of learning the language though she has had several years of Swedish lessons in school before university. In the same vein, both Swedish-­ medium student teachers have had Finnish at school, but both underline that they come from very Swedish-dominant surroundings and that it required active engagement to learn Finnish. Dana explicitly says that she has struggled with learning Finnish but that many today think that she is a Swedish-Finnish bilingual. This discourse shows the different conditions under which the two national languages of Finland are being used in daily life. For Swedish-dominant speakers in Finland it is essential to be able to use both languages, whereas Finnish-dominant speakers do not experience the same need to master both languages. The voices of the student teachers thus clearly show how societal patterns influence individual patterns. Consequently, Opri and Peeta do not specify contexts or domains where Swedish would be more predominant than other languages and they just note that Finnish is used at home. Contrastingly, Frida and Dana estimate that part of their daily activities are in Swedish, part in Finnish. Furthermore, their wordings also show that there is a self-­initiated responsibility to motivate the use or non-use of Swedish: 3. “Swedish 60–70% at work, Finnish 30–40% in the family and in society. This is probably due to convenience and conformism.” (Frida) 4. “Nowadays more Swedish, earlier both oral and written Finnish.” (Dana) For the other languages of Frida, Dana, Opri and Peeta, the domains for learning the languages are mainly school and abroad. Frida learnt English at school and French at school and abroad. Dana does not specify domains for her different languages (see previous section) except for the English language which she uses for literary reading. Opri learnt English while studying in Great Britain where she also learnt Norwegian and Danish. She has spent a year in Germany as an exchange student, has travelled frequently in Sweden because of her hobby and has taken a course in French. Likewise, Peeta has spent time in Sweden where she has had language courses in French and Spanish. She mentions that she has almost forgotten all her Danish but it remains unclear where she has learnt Danish. She learns more English when skyping in English with friends in Europe. When asked about the linguistic identity (perceptions about being mono-, bi- or multilingual), all four student teachers add an explanation to their answer and the question opens for a lengthy discussion and even shifts in perspectives. 5. “I am bilingual, earlier monolingual. When I was young English did not have the same position in society as it has now.” (Frida)

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present…

111

6. “…but I have had to struggle with Finnish, so I see myself as being monolingual Swedish, but if multilingual means to be able to use many languages so OK, but not bilingual if bilingual means to get two languages from home.” (Dana) 7. “I am monolingual but mentally bilingual…Inside of me I have a very strong Swedish identity but to produce Swedish, it is not at all as easy as Finnish… I would perhaps say that I am multilingual. I know a little bit of every language… Now perhaps monolingual or multilingual, and yes, bilingual…But when you are bilingual, it feels as you should be as comfortable as when using Finnish” (Opri) 8. I am... sort of monolingual Finnish. … It feels like you should have got it [the other language] at home. … Right now I would classify myself as a monolingual who feels at home in Finnish, Swedish and English… Whenever I get an opportunity to use English I drive there and use it! [Another discussion partner asks: But why are you not multilingual then?] Maybe it is something in the Finnish mindset, that I cannot allow myself.” (Peeta) As shown in the excerpts, there are arguments that support identifications of monolingual, bilingual or multilingual states among all four student teachers. To identify as a bilingual person seems to be strongly associated with being raised bilingually at home and poses challenges to Dana and Peeta, whereas Opri underlines the importance of being at ease/being fluent in using both languages. All of them hesitate to readily define themselves as multilinguals, but indicate that they are more inclined to define themselves as multilinguals in a pragmatic sense (cf. Dana and Opri above). Only Frida and Peeta explicitly mention English in the language identity discussion, it is the only other language besides Finnish and Swedish that shows up in this discourse. Frida’s argument expresses almost a defense for why English is not as visible among her languages as generally expected in Finland, whereas Peeta emotionally claims that she is ready to drive to places where she has an opportunity to use English.

6  Conclusions and Implications for the Future We have used two different sets of data to show how multicompetence and DLCs can be identified and analysed in the domain of education in Finland both societally and individually. For the purpose of our study, we have separated a societal level, where the focus is on a document analysis of the two most recent national curricula, and an individual level, represented by group discussions with four student teachers. In both sets of data we have focused on language as vehicle for communicative purposes and language as expression of identity. Even though the two sets of data are different, they show to what extent the DLCs described in the steering documents and the DLCs of student teachers match, and if there is a potential to develop towards multicompetence. To sum up the results at the society level, we ask what patterns of DLCs emerge in the two latest national core curricula for basic education in Finland. A distinct development towards a more explicit recognition of diversity among students and in society is clearly visible. This is linked with an expected growth of language

112

S. Björklund et al.

awareness, mainly connected to the assessment descriptors used for Swedish and Finnish as L1s and SNLs and FL English. The descriptor for Sámi as L1 does not contain that perspective, stressing Sámi as a separate entity, which is interesting, considering the fact that  there are several different Sámi languages present in Finland and the whole Northern region (cf. Lainio et al. 2017). The general path of development is thus in the direction towards viewing languages as interconnected and useful tools, but there is still has a long way to go to reach a consistent view on multicompetence (cf. Cook 2012). This process then has to become teacher-driven, which sets requirements for consistent efforts within teacher education. Regarding the internal relations between the studied language descriptors, the pattern to emerge is thus one where the communicative uses and identity-­ related functions of L1 Sámi (and other minority languages) is more limited than those of the two national languages as L1. On the other hand, the language-specific linguistic and cultural identity aspects are much more explicit for the minority languages, even though they are mentioned more explicitly also for L1 Swedish than L1 Finnish. The expected competence patterns differ between the national languages and Sámi, which we take as an indication of the more limited communicative scope expected for the Sámi language. Another internal relation relevant for our discussion is the fact that there are higher competence demands set for SNL Finnish (advanced syllabus for Swedishmedium schools) than for  SNL Swedish (advanced syllabus for Finnish-­medium schools), which seems to indicate that the communicative uses of the SNL is to be regarded as more crucial for Swedish-speakers than for Finnish-speakers. What the empirical data does not illustrate, but which further strengthens that assumption, is the fact that a majority of Finnish-speaking pupils in Finland do not study Swedish in accordance with the advanced syllabus, but follow an intermediate syllabus commencing in grade 6 as mentioned earlier. The internal relation patterns between the SNLs and FL English is interesting, since the competence descriptors contain higher expectations for English than the SNLs. This can be seen as a clear indication of a move away from English as a foreign language towards a second language, i.e. a language of wide communicative uses. Also, the culturally related descriptor indicates that English is a language of relevance in the local community as well as a language for international communication. Hence, the conclusion can be drawn that English is definitely to be included in the expected DLC at a society level. This interpretation is further strengthened by the fact that the national curricula do not include descriptors for other specified foreign languages, only a common for all other possible foreign languages and thus not part of the expected societal DLCs. To conclude, we thus claim that the DLCs at the society level indicate that the expected DLCs are slightly different, depending on whether one attends the Swedish- or Finnish-medium school. For pupils in the Swedish-medium schools the expected DLC consists of the first language Swedish, followed by both Finnish and English as equally important, with complementary communicative uses and identity  functions. For pupils in the Finnish-medium schools the DLC is expected to consist of L1 Finnish, followed by English for both domestic and

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present…

113

international communicative functions. On the basis of our findings, we claim that Swedish is included in that constellation, although with a slightly more narrow scope of communicative uses and weaker identity claims than English. The inclusion of Swedish in the DLC is relevant when the language is studied at an advanced level, but not necessarily for pupils who commence their studies later. The analysis of the four student teacher voices show that communicative purposes, use and function indicate what languages constitute DLCs. Education, home and abroad stand out as three domains of prime importance for language learning and use. Furthermore, it is not possible to identify present DLC constellations among the individual student teachers without including several measures on language communicative purposes  and identity. The language identity examples show that student teachers view themselves as monolingual, bilingual and multilingual. The bilingualism perspective seems to be more associated with equally high competence in both national languages (cf. two monolingualisms referred to as the two-­solitudes assumption by e.g. Cummins 2007) and the multilingualism perspective with pragmatic views (knowledge of and ability to use several languages). The discussion further shows, that the individual definitions are dynamic and change in line with different perspectives and during different time periods throughout an individual’s life. Finnish is the language that the two Swedish immersion student teachers think that fulfill most of their communicative purposes, whereas the Swedish-medium student teachers also stress Finnish as an important language for several communicative purposes after Swedish. They also expresses a responsibility for making active use of Swedish. When more languages than Finnish and Swedish are mentioned, Swedish and English are grouped together as an equal pair for both communicative purposes and use by the two immersion student teachers. The Swedish-medium student teachers are somewhat more oriented towards the simultaneous use of two languages but English is implicitly referred to as their third dominant language. In the light of the four student teachers’ reported multiple language use it is a matter of definition if two or three languages belong to their dominant languages, but the result of the analysis clearly show that there are three languages (Finnish, Swedish and English) that function as the three most dominant languages in their present constellations. Even if English can be classified as a “foreign” language in the sense that there are not so many native speakers of English in Finland, its use and frequency has a clear position and status not comparable with those other languages mentioned in the language repertoire of all four student teachers. We assume that Swedish would not have been present in the two Swedish immersion student teachers’ DLC if they had not chosen to enter Swedish immersion teacher education. Though they use Swedish daily during their teacher training they tend to group Swedish and English together with equal status for both competence and use. A comparison between the DLCs emerging in the national core curricula and the four student teachers shows a clear picture of the presence of three languages (Finnish, Swedish and English). Among these three languages the balance between Swedish and English seems to be specifically delicate and subject to change. In addition, the analysis also revealed that there is little evidence of familiarity with regional/national minority languages. There are many languages of newcomers in

114

S. Björklund et al.

Finland neither mentioned nor included in the languages the student teachers mention they would like to learn in the future. The Russian language is the only language mentioned in the language bucket list of the four student teachers (at present Russian, Estonian and Arabic are the three biggest languages in terms of first language speakers in Finland). In relation to teacher education, one has to assume that new student teachers will come with only slightly different views on expected DLCs and the majority will probably need support to raise their awareness of the communicative and functional value of the variety of minority languages presently less visible in curricula and individual DLCs. The explicit use of the concept DLC may be a way to raise the language awareness of student teachers and a holistic understanding of the variation, dynamics and didactic opportunities of multilingual classrooms. The constellation charts (cf. e.g. Aronin 2019) used to illustrate the DLCs and other languages present in the dynamic language use of an individual or institution used as support for think-aloud protocols, individual or focus group interviews has turned out to scaffold language awareness and support reflection related to language identity, multilingual language use, as well as status and functions of different languages (cf. Björklund and Björklund forthcoming). It would not have been possible to arrive at the cross-language comparisons needed to display DLCs if we had not had the whole picture of languages mentioned in our two data sets. On the other hand, we think there is a need to refine analyses to be able to present the nature of the DLCs in even more holistic and dynamic ways. In both our data sets, languages are separated as isolated entities. We find that this separation does not optimally illustrate the multifaceted and intertwined ways in which languages are used simultaneously and overlapping in situated daily language use. We need to continue to search for ways to best highlight the real nature of multicompetence.

References Armat, M., Assarroudi, A., Rad, M., Sharifi, H. & Heydari, A. (2018). Inductive and deductive: ambiguous labels in qualitative content analysis. The Qualitative Report, 23(1), 219–221. Retrieved from: https://nsuworks.nova.edu/tqr/vol23/iss1/16 Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–163). Hohengehren: Schneider Publications. Aronin, L. (2017). Dominant language constellations (DLC) as an approach to studying multilingual practices. Paper presented at the International Symposium on Bilingualism (ISB11), University of Limerick. Ireland on 11–15th June 2017. Aronin, L. (2019). What is multilingualism? In D. Singleton & L. Aronin (Eds.), Twelve lectures on multilingualism (pp. 3–34). Bristol: Multilingual Matters. Björklund, M., & Björklund, S. (forthcoming) Embracing multilingualism in teaching practicum in Finland? DLC as a tool for uncovering individual and institutional multilingualism. In E. Vetter & L. Aronin (Eds.), Dominant Language Constellations Approach in Education and Language Acquisition. Springer.

Societal Versus Individual Patterns of DLCs in a Finnish Educational Context – Present…

115

Björklund, M., Björklund, S., & Sjöholm, K. (2013). Multilingual policies and multilingual education in the Nordic countries. International Electronic Journal of Elementary Education, 6(1), 1–21. Cook, V. (1991). The poverty-of-the-stimulus argument and multi-competence. Second Language Research, 7(2), 103–117. Cook, V. (Ed.). (2003). Effects on the L2 on the L1. Clevedon: Multilingual Matters. Cook, V. (2012). Multi-competence. In C. Chapelle (Ed.), The encyclopedia of applied linguistics (pp. 3768–3774). New York: Wiley-Blackwell. Council of Europe. (2001). Common european framework of reference for languages: learning, teaching, assessment. Cambridge: Cambridge University Press. Cummins, J. (2007). Rethinking monolingual instructional strategies in multilingual classrooms. Canadian Journal of Applied Linguistics/Revue canadienne de linguistique appliquee, 10(2), 221–240. Finnish National Board of Education. (2004). National core curriculum for basic education 2004. Helsinki: Finnish National Board of Education. Finnish National Board of Education. (2015). Grunderna för läroplanen för den grundläggande utbildningen 2014. Helsinki: National Board of Education. Geanellos, R. (2000). Exploring Ricoeur’s hermeneutic theory of interpretation as a method of analysing research texts. Nursing Inquiry, 7, 112–119. Heimonen, E., & Ylönen, S. (2017). Monikielisyys vai “English only”?Yliopistojen henkilökunnan asenteet eri kielten käyttöä kohtaan akateemisessa ympäristössä. In S. Latomaa, E. Luukka, & N. Lilja (Eds.), Kielitietoisuus eri-arvoistuvassa yhteiskunnassa -Language awareness in an increasingly unequal society (AFinLAn vuosikirja 2017. Suomen soveltavan kieliyhdistyksen julkaisuja n:o 75) (pp. 72–91). Jyväskylä: Suomen soveltava kieliyhdistys. Lainio, J., Pesonen, S., & Nordin, M. (2017). Nationella minoritetsspråk i skolan – förbättrade förutsättningar till undervisning och revitalisering (SOU 2017:91). Stockholm: SOU. Latomaa, S., & Suni, M. (2011). Multilingualism in Finnish schools: Policies and practices. ESUKA-JEFUL, 2(2), 111–136. Leppänen, S., Nikula, T., & Kääntä, L. (Eds.). (2008). Kolmas kotimainen – lähikuvia englannin käytöstä Suomessa. Helsinki: Suomalaisen kirjallisuuden Seura. Official Statistics Finland. (2020). Population structure [e-publication]. Helsinki: Statistics Finland. Retrieved from: http://www.stat.fi/til/vaerak/index_en.html Peltoniemi, A. (2015). Kansainvälinen näkökulma kielikylpyopettaja-koulutukseen (Vaasan yliopiston julkaisuja ja raportteja 200). Vaasa: Vaasan Yliopisto. Programme for International Student Assessment. (2018). Pisa 2015. Results in focus. Sine loco: OECD. Retrieved from: https://www.oecd.org/pisa/pisa-2015-results-in-focus.pdf Sarivaara, E.  K. (2016). Emergent Sámi identities: From assimilation towards revitalisation. In R.  Toivanen & J.  Saarikivi (Eds.), Linguistic genocide or superdiversity? New and old language diversities. Bristol: Multilingual Matters. SCB (2020). Befolkningsstatistik i sammandrag 1960–2019. Retrieved from: https://www.scb.se/hittastatistik/statistik-efter-amne/befolkning/befolkningens-sammansattning/befolkningsstatistik/ pong/tabell-ochdiagram/elarsstatistik–riket/befolkningsstatistik-i-ammandrag/

Language Background Profiling at Canadian Elementary Schools and Dominant Language Constellations Nikolay Slavkov

Abstract  In a global world of unprecedented migrations and superdiversity, awareness and encouragement of bilingual and multilingual practices are also on the rise. In many contexts, children’s language repertoires can be varied and complex and the notion of a Dominant Language Constellation, a core set of languages highly relevant in an individual’s daily life, can apply from an early age. This chapter focuses on language background profiling at elementary schools in Canada. Such profiling is typically done at initial school entry to inform schools as well as larger regional and provincial educational authorities about the potential diversity, use of and proficiency in different languages by the incoming cohorts (and their families). School registration forms thus typically include questions about children’s first, home, primary, additional, etc. languages. The formulations, number, and combinations of questions may vary to a large degree from one school district to another, and also from province to province. Analysis of such questions can be revealing in terms of educational and societal assumptions and orientations to bi/multilingualism, and in terms of general views of language. I offer a discussion of the notion of a native speaker and certain monolingual norms that ensue from it and can be seen in language background profiling practices at schools. Then, building on previous work and incorporating new data, I provide a general summary of the questions used in a large sample of registration forms from various schools and boards in five Canadian provinces. I situate these within the context of recent multilingual frameworks such as plurilingualism, multicompetence and translanguaging, and then focus specifically on the notion of a Dominant Language Constellation, arguing that its premises may entail benefits for the conceptualization of language background profiling. Keywords  Language background profiling · Native speaker · First language/ mother tongue/native language · Dominant Language Constellation (DLC)

N. Slavkov (*) Official Languages and Bilingualism Institute, University of Ottawa, Ottawa, Canada e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_7

117

118

N. Slavkov

1  Introduction Canada is a country with two official languages (English and French), with a high rate of immigration, and with Indigenous languages and cultures. Census data indicate that between 1961 and 2016 the bilingualism rate in the two official languages rose from 12.2% to 17.9%.1 Additionally, non-official languages are an important part of Canadian society and are spoken by more than 20% of the population (sometimes alone but mostly in addition to one or both official languages), according to the 2016 census (Statistics Canada 2017a, b).2 Multiculturalism is officially recognized and valorized through federal legislation (i.e. Multiculturalism Act), and multilingualism, although typically not addressed thoroughly at the federal level, is widely recognized and frequently mentioned in educational contexts. As such, Canada represents an interesting case for studying Dominant Language Constellations (DLCs), that is, subsets of individual repertoires that contain the most commonly used and relevant languages for a given person. Depending on various geographic, educational, professional, personal and other circumstances, in the context of two official languages and large ethnic and linguistic diversity, children can develop various DLCs from an early age. In this chapter I relate DLCs to the educational system and investigate how children entering elementary school are profiled in terms of their languages. I begin by providing background on the educational system and on language profiling practices in general. Then I present a study that analyses the questions about incoming student’s languages asked on school registration forms in several Canadian provinces. In discussing the data, I adopt a DLC lens and highlight the potential benefits of this approach in terms of understanding and capturing linguistic diversity in novel and interesting ways.

1  Bilingualism in census questionnaires is operationalized as the ability to conduct a conversation in both English and French. 2  Note that the term non-official language can be problematic as it may imply lesser value associated with languages spoken by Indigenous people and immigrants in Canada. I use the terms official and non-official as reported by Statistics Canada while also emphasizing the view that all languages are valuable regardless of legal status.

Language Background Profiling at Canadian Elementary Schools and Dominant…

119

2  Background 2.1  Overview of the Educational System The school system in Canada is publicly funded and provincially mandated.3 Typically, provincial authorities organize schools in regional clusters, commonly referred to as district school boards (terminology may vary slightly across provinces). While some smaller provinces may have only a few school boards, larger provinces have multiple districts. For example, Ontario, the province where the highest proportion of the Canadian population resides (approx. 13.5  million or 38.3% in 2016), has 82 boards and 10 educational authorities with close to 5000 schools serving close to 2 million students, according to data from the 2015–2016 school year (Ontario Ministry of Education 2017). During the same school year, the province’s total investment in education was valued at $22.6 billion. Ontario also has Canada’s largest and most multicultural urban centre, Toronto, and the province routinely claims in curricular and other documents to have “some of the most multilingual student populations in the world” (Ontario Ministry of Education 2016, p. 100). In contrast to Ontario, Canada’s smallest province, Prince Edward Island, has a population of about 143,000, and only two school boards serving about 20,000 students in total. Official language minorities in Canada generally have the right to education in their language, as ensured by both federal and provincial legislation. Thus, francophone populations in English-speaking provinces have the right to education entirely in French, offered through francophone school boards; conversely, anglophones in the French-speaking province of Quebec have the right to education entirely in English. The current study discusses only provinces where English is the majority language and data from francophone school boards is excluded.4 However, the study includes data from French immersion programs, which provide education (partially) in French and which are offered by anglophone school boards because they generally target non-francophone students (i.e. learners of French). The provinces on which this chapter focuses are Alberta, Ontario, British Columbia, Manitoba and Prince Edward Island. This covers a large geographic area from the Atlantic to the Pacific coast of Canada.

3  This research does not focus on private schools, which also exist in various forms across the country. 4  Data from francophone school boards in Quebec and in other Canadian provinces is currently being analyzed and will be reported in future publications.

120

N. Slavkov

2.2  Language Background Profiling Language background profiling at public schools is a relatively unexplored domain from a research perspective. Apart from the studies by Slavkov (2016, 2018) on which this chapter draws, to the best of my knowledge no other scholarly work in Canada has explored this topic (for a perspective from the US, however, see Bailey and Kelly 2013). Language background profiling is usually done through a school questionnaire and sometimes also through a school intake interview. For practical reasons, the former is the more common option and also the option that I focus on in this study. Language background questions vary in number, content and wording, and are typically included as a short subsection of most school registration forms. These forms are generally the same for all schools within a school district but differ from one district board to another, and thus the profiling done in one region or urban centre of a given province may include different questions from those used in another region or urban centre; consequently, the profiling may differ, sometimes minimally and sometimes to a larger degree, within a single province.5 Since education is a provincial prerogative, and policies and practices are not uniform, profiling from one province to the next also differs. In terms of content, registration forms may ask questions regarding the child’s first language, home language(s), additional languages, whether the child is an English Language Learner (ELL), etc. (see below for a list of sample questions); forms sometimes also collect information about parents’ ability to understand the school language, language of previous school (if a student is transferring to a new school), known speech-language impairments or special needs, etc. As some of these examples of questions imply, language background profiling can serve multiple purposes. It is generally a tool that provides a basic snapshot of a child’s languages, which helps the school, the board, and the province “get to know” their incoming cohorts of students. It is not uncommon for schools’ websites or promotional materials to boast linguistic diversity by reporting the number of different languages spoken by their students (sometimes dozens of languages for a single school). Similar claims can also be found at the school board and provincial levels, not only in promotional materials but also in curricular documents, as already indicated in the introduction section above. In addition to providing data about linguistic diversity, language background profiling can serve the purpose of (at least preliminary) identification of students’ needs in terms of support in learning the school’s language, and this may help educational authorities allocate additional support or resources to certain students, schools or boards. While language background profiling may seem to be a straightforward and matter-­of-fact process, it is important to recognize its scope and significance. Over a typical 12 year educational cycle, millions of children are profiled for their language backgrounds across Canada’s public school systems. This has the potential of 5  Sometimes single schools may also have their individual registration forms, which contributes to even greater variation in profiling questions in a province.

Language Background Profiling at Canadian Elementary Schools and Dominant…

121

creating robust and informative databases that can be useful not only for educational authorities, but also for local communities and for society in general. However, because of the high degree of variability on the local level (i.e. registration forms within a single province have a different number of questions, use different formulations and may refer to different aspects of a child’s language repertoire) it is not always clear how data coming from one school board compares to others within the province, and how the overall data coming from different boards is compiled at the provincial level in a reliable and uniform way. Slavkov (2016, 2018) recommends some degree of standardization in the wording of various questions as well as in the number and types of questions asked, in order to improve the quality and consistency of the data collected from an entire province. It should also be recognized that it is challenging to create a set of questions that tap into bilingual or multilingual children’s complex linguistic repertoires within a limited space on a school registration form (where language background profiling usually represents only a brief section and the rest of the space is dedicated to other important registration information). For example, a seemingly straightforward question such as What is the student’s first language? may be ambiguous: if we interpret it in a chronological sense, the answer may be the language that the child learned first in life, i.e. the native language; however, if the child was exposed to one language from birth and then later on acquired a different language, for example, through immigration, then the second language may become stronger and thus may be reported as the child’s “first” language. In other words it is not always clear whether “first” refers to chronological order of acquisition or current ability in a given language repertoire. To take another example of a commonly asked question, What is the home language?, one may also interpret this in various ways due to the complexities of home language use: while a child can be exposed to a minority language at home (e.g. spoken by the parents to the child), often times that child may respond in the majority language or use mixed utterances; thus it may be difficult to identify a single home language unless the questionnaire asks specifically about what the most frequently spoken language by the child or about differences in receptive versus productive abilities (so far only one form that uses such questions has been found).

2.3  Native Speakerness and Monolingual Norms The idea that “The first language a human being learns to speak is his native language; he is a native speaker of this language” (Bloomfield 1933, p. 43) and the concept of perfect abstract competence or infallible intuitions that speakers have about their native language (Chomsky 1957 and subsequent work) have had powerful impacts in scientific, social and educational domains. The pervasiveness of such a definition of a native speaker, sometimes called biodevelopmental (Davies 1991) or an unalterable historical fact (Cook 1999), has arguably advanced and reinforced monolingual norms in domains outside of theoretical linguistics. For example,

122

N. Slavkov

language learners may aspire (or be expected by educators or employers to aspire) to what is often an unattainable native speaker standard, rather than embracing and celebrating their multicompetence (Cook 1991, 1999, 2013) and bi/multilingual linguistic and cultural repertoires. Language teachers may also be impacted by covert and overt expectations to converge with and enforce monolingual standards or identities associated with the target language (for a recent overview see Slavkov 2016; Wernicke 2017a, among others). As Wernicke (2017b) aptly points out, debates about the native speaker norm in the literature have “alternatively proclaimed the native speaker as ‘dead’ (Paikeday 1985) and as ‘alive and kicking’ (Phillipson 1992), and many agree that the notion should ultimately be ‘displaced’ (Rampton 1990) by more productive and relevant ideas of language expertise.” In language background profiling, monolingual targets based on native speaker norms can also be found. For example, registration forms that focus exclusively on questions about incoming students’ first language may reinforce the idea that a person’s status as a native speaker of a given language is the most important and relevant piece of language-related information necessary for school entry (Slavkov 2016, for example found that some registration forms in the province of Ontario do indeed show such a preoccupation with students’ first language). Profiling students based on nativist assumptions undoubtedly represents a genuine attempt to find out about language learner status and potentially be in a position to offer special support in the learning of the school language. Furthermore, in some cases, such questions may also be tied to provincial legislation and thus are needed to confirm eligibility for additional language support funding. However, categorizing students based on their first language also runs the risk of discursive othering and entrenchment into invariable monolingual norms that may not be fully aligned with the premises of inclusive education. This underlying contrast is exemplified by curricular documents that expressly celebrate multilingualism but also dwell on English language learners as “students whose first language is a language other than English, or is a variety of English that is significantly different from the variety used for instruction” (Ontario Ministry of Education 2016, p. 100). While no explicit reference to the correctness or prestige of the language variety used for instruction is made, potential reinforcement of a lower status for other varieties, such as Englishes from the outer circle (Kachru 1983; Kachru and Smith 2008) and marginalization of Indian, Caribbean, African, etc. speakers of English is possible, in addition to marginalization of speakers of other languages. Such statements may imply linguistic acculturation pressures, may be seen as condescending to speakers of languages or varieties that are not the ones spoken by the school, and may reinforce native speaker norms (i.e. speakers of a given language are superior). This may be especially relevant in contexts where educational authorities focus only on the school language and try to keep other languages that may be part of students’ linguistic repertoires at bay, outside of the school (García 2009; among others).

Language Background Profiling at Canadian Elementary Schools and Dominant…

123

2.4  Profiling Orientations Slavkov (2016, 2018) develops an analytical framework that distinguishes between a chronological-nativist and a synchronic-functional approach to language background profiling. The former, as the term suggests, is based on an underlying objective of the profiler (i.e. the institution that designs the language background questionnaire) to find out information about the native/first/mother tongue of the child. Under such an approach, the historical or biodevelopmental language facts about the child are foregrounded and information about their current use of other languages is not seen as necessary. Conversely, under the latter approach, the focus is on the current language repertoire of the child and questions about current proficiency in different languages, domains and frequency of use, etc. are foregrounded, while historical or biodevelopmental information is not requested. Thus, a language background profile may lean towards one or the other end of the spectrum, depending on the questions asked. Arguably, a practice to be encouraged is to design questionnaires that collect information from both perspectives in order to obtain a solid snapshot of a child’s repertoire, including past and current language(s) and abilities.

2.5  D  ominant Language Constellations and Other Multilingual Frameworks It is important to situate language background profiling in current multilingualism frameworks. In this section, I offer a brief overview of some key concepts in plurilingualism, multicompetence and translanguaging, and then focus in more detail on Dominant Language Constellations. The notion of plurilingualism (Coste et al. 1997, 2009; Coste and Simon 2009; Canagarajah 2007, 2009; Dagenais and Moore 2008; Marshall and Moore 2013; Taylor and Snoddon 2013; among others) has attracted the attention of scholars, educators, policy makers and various stakeholders around the world and in Canada. Plurilingualism emphasizes the idea of a language repertoire in which an individual’s different languages are integrated and viewed as a whole, rather than separately; at the same time, plurilingualism considers the use of different languages or varieties for different purposes and in different domains. Equal proficiencies in multiple languages (e.g. monolingual targets in two or more languages) are not assumed or expected. Furthermore, language competence is not treated as fixed and in isolation but as a form of a fluid and integrated linguistic, social and intercultural practice, in an increasingly transnational world. Plurilingual values are linked to the Common European Framework of Reference for Languages (Council of Europe 2001) and are increasingly recognized around the world. They are also related to the notion of multicompetence (Cook 1991, 1999, 2013) that has contributed to important conceptual and practical advances in terms dispensing with monolingual assumptions and norms in the field of language teaching and learning and in applied linguistics

124

N. Slavkov

in general (recall also the importance of the notion of multicompetence in relation to the concept of the native speaker discussed earlier in this chapter). Another influential framework that has gained a high level of recognition is translanguaging. The term was originally used in the context of bilingual education in Wales (Williams 1994, 2002) and referred to planned and systematic use of two languages in a classroom. This would involve providing students with input in one language (e.g. English) and requiring output in another (e.g. Welsh) and vice versa, as a language learning/support activity (see also Lewis et al. 2012). However, the term was subsequently expanded to cover a wide set of meanings, including spontaneous/organic use of more than one language in the family, community, educational and institutional contexts (García 2009 and subsequent work). In recent work, García and colleagues summarize a holistic view of bilinguals and multilinguals, highlighting creativity, critical construction and selection of language features, coming from the (combined) resources of more than one language (e.g. García et al. 2017). Thus, translanguaging goes beyond the idea of standard, prestigious, codified (national) varieties, which may in some cases marginalize bi/multilingual and bi/multicultural speakers. That is, translanguaging draws on a conceptualization of language as a general communicative tool with hybrid properties and such a view is intended to transcend monolingually-oriented norms reproduced and reenforced in various educational and societal contexts. Plurilingualism, multicompetence and translanguaging offer interesting insights into how language background profiling may be conceived of and possibly modified across the five provinces mentioned in this study, and potentially in other Canadian and global contexts. However, a more detailed discussion of these frameworks falls outside of the scope of this chapter where I would like to focus on a multilingual approach that is especially relevant to the issues arising in this study: Dominant Language Constellations (DLC). The Dominant Language Constellation (Aronin 2006, 2016, 2019) is a concept related to and complementary with the language repertoire in that it refers to a given individual’s multiple languages. However, DLC is a narrower term that focuses only on a subset of languages from the larger linguistic repertoire. As Aronin (2019) points out, a DLC encompasses only what she calls the most expedient or vehicle languages of an individual. Even though in a globalized world a person may be exposed to various languages through multiple additional channels (e.g. classroom language learning, travel, migration, geographic proximity to other speech communities, linguistic heritage, media, the Internet, etc.), only a subset of these languages may be put to use in our daily lives and those are the languages included in an individual constellation. For Aronin, the repertoire is a wider pool of language resources that may remain dormant or unused for a period of time, while the DLC consists of the languages that are actively and frequently used. As bi/multilingualism are dynamic phenomena, changes may occur and languages may shift from the repertoire to the DLC and vice versa. For example, due to social or personal circumstances (e.g. migration) a language may become less relevant for a given individual and leave the DLC to enter the repertoire; conversely, when new circumstances render a language from the repertoire necessary or relevant to an individual’s current functioning in life, the

Language Background Profiling at Canadian Elementary Schools and Dominant…

125

language will enter the DLC.  Overall, while the repertoire is seen as a resource repository, the DLC is about active usage and specific observable language behaviors. Another important aspect of the DLC concept is that it treats languages as a unit (similarly to other multilingual frameworks) and strict separation is thus not required or necessary. Aronin (2019) addresses the question of a native language with regard to a DLC and argues that in a DLC framework there is no immediate need to identify a (single) native language within an individual’s constellation. As a specific example, she cites Canagarajah (2016, p. 446) who recalls starting to speak English, Tamil and Sinhala around the same time in his childhood; he also reports speaking more often in Tamil but writing more often in English, and thus further questions the concept of a (single) native language. Finally, the DLC approach involves a non-linear conceptualization and uses visualisations in the form of constellation maps where dominant languages are clustered in a circle and denoted with five-point stars. Languages in one’s repertoire that are not encompassed by the DLC boundaries are languages that are mastered less well or used less frequently. In addition, other languages that a person has been exposed to, has some knowledge of, or understands to some degree are placed in more remote locations on the map and denoted by four-point stars.

3  The Study 3.1  Objectives and Methodology Keeping the above background in mind, one of the objectives of this study is to continue the previous line of inquiry in examining language background profiling at elementary schools across Canada and integrate previous data from samples of registration forms in three Canadian provinces (Ontario, Alberta and British Columbia) reported in Slavkov (2018) with newly collected data from two additional provinces (Manitoba and Prince Edward Island). This objective is achieved through research questions 1–3 below. A second objective is to provide a discussion of language background profiling in light of recent frameworks in multilingualism, focusing specifically on Dominant Language Constellations. This objective is achieved through research question four below. R1. How much variation is there in the number, type, combination patterns and level of detail of language profiling across different school districts and across provincial boundaries? R2. Do language profiling questions asked on school registration forms reflect a chronological-nativist orientation (i.e. focus on first language), a synchronic-­ functional orientation (i.e. focus on current language(s) spoken), or a mixture of both?

126

N. Slavkov

R3. To what extent are the complexities of bilingualism and multilingualism captured by the (combinations) of language background questions on school registration forms across the five provinces? R4. How do the data obtained relate to existing multilingual frameworks, and more specifically Dominant Language Constellations? The total number of sample registration forms collected from the five provinces was 123. Each language-related question on the forms was extracted and entered into a database. The total number of questions extracted was 283 for the five provinces. Identical questions from different forms were extracted but assigned the same code in the database. Questions with different formulations, even if minimal differences existed (e.g. first language of student vs. the student’s first language), were assigned distinct codes. This procedure resulted in 53 distinct codes for Ontario, 44 for Alberta, 27 for British Columbia, 20 for Manitoba and 2 for Prince Edward Island. Even though Prince Edward Island is the smallest Canadian province and has only one English language school board with a single registration form representing just two questions, this data was considered important as it applies to an entire province. After the procedure for individual codes uniquely identifying distinct question formulations was applied, questions that had similar meaning were collapsed into larger categories and assigned additional codes corresponding to these categories. Eleven such categories were established and are illustrated in Table  7.1 below. Examples of the types of questions that were included in each of these categories (a non-exhaustive list) are included in the table. The current study focuses only on categories 1–5 as they are most relevant to the research questions identified. Categories 6–11 are excluded from the results presented in the next section, as they refer to a child’s educational history, health status, parent’s abilities to communicate with the school, and legal eligibility to participate in francophone minority or ESL programming.6 Also note that a certain degree of overlap exists in the categories. For example, categories 2 and 3 both refer to the home language of the student; however, since category 3 emphasizes the concept of a primary language (assuming more than one language may be spoken in a household), it was considered different from category 2. Categories 3 and 4 also overlap to an extent, as the language spoken most frequently and fluently may also be the primary language; however, these two categories were kept as distinct because one

6  Note that even though categories 7 and 11 both refer to ESL, the former focuses on previous instruction received while the latter refers to legal eligibility for ESL support. As such they were kept separate. Also note that questions from categories 10 and 11, whose purpose is to determine legal eligibility for francophone education or ESL support, often asked about the student’s first language or primary (home) language, etc. Even though such questions are of primary interest in this article, they were not collected for the purpose of language profiling but for eligibility determinations, and as such are not included in the analysis, as mentioned above. This is only a technical point since the forms also repeated those questions in a separate section not related to legal eligibility criteria, and in that case such questions were incorporated into categories 1–5, as appropriate.

Language Background Profiling at Canadian Elementary Schools and Dominant…

127

Table 7.1  Question categories and sample questions across five provinces (non-exhaustive list) Category 1. First language

Sample questions first language, first language spoken, my child’s first language learned (specify), birth language, etc. 2. Home language home language(s), language spoken at home, language at home, language student speaks at home, student is a bilingual learner (e.g. speaks both English and another language in the home), etc. 3. Primary main language spoken by the student at home, primary language spoken language at home most often at home, what language is mainly spoken at home?, etc. language spoken most often, primary language in which student is most 4. Language fluent, language most used, etc. spoken most often/fluently 5. (Other) other language(s) spoken, second language spoken, languages known/ languages spoken, etc. previous school language, language of instruction at previous school, 6. Previous previous instruction language: English or French, instructional language at language of previous school, etc. instruction 7. Previous ESL has the student been receiving English as a second language (ESL) assistance assistance: Yes/No?, etc. severe delay involving language, past assistance: speech/language, student 8. Language/ health information: speech/language problems? student uses ASL or a speech/hearing signed language as their first language and develops English, as their impairment second language through reading and writing, etc. 9. Parents’ ability parent/guardian information: speaks school language yes/no?, interpreter to speak English required: yes/no?, etc. 10. Francophone are you eligible to have your child receive a French first language (francophone) education? education eligibility 11. ESL English as a second language (ESL) eligibility: ESL students are identified eligibility as Canadian-born or foreign-born students. A Canadian student is eligible for ESL support when the primary language spoken at home is a language other than English and the student meets eligibility requirements after assessment. Is your child within this category?

of them restricts the domain of language use to the home, while the other one does not. Finally, note that within the categories, certain questions involved the plural use of the word language (e.g. first language vs. first languages; home language vs. home languages, etc.), which is one indicator of whether the questionnaire allows for or assumes bi/multilingualism or not. An analysis in this regard will also be presented in the results section below.

3.2  Results The number of language background profiling questions asked on the registration forms samples collected from the five provinces varied, as indicated in Table 7.2.

128

N. Slavkov

Table 7.2  Number of language-related questions per registration form Number of language-­ related questions asked 0 1 2 3 4 5 6 7 8 Total:

Number & % of forms ONTARIO 2 5% 8 18% 11 25% 10 23% 6 14% 5 11% 1 2% 0 0% 1 2% 44 100%

Number & % of forms ALBERTA 0 0% 2 7% 12 43% 10 36% 4 14% 0 0% 0 0% 0 0% 0 0% 28 100%

Number & % of forms BRITISH COLUMBIA 1 4% 8 33% 7 29% 6 25% 0 0% 2 8% 0 0% 0 0% 0 0% 24 100%

Number & % of forms MANITOBA 0 0 21 81% 3 12% 1 4% 0 0% 1 4% 0 0% 0 0% 0 0% 26 100%

Number & % of forms PRINCE EDWARD ISLAND TOTAL 0 0 3 2% 0 0% 39 32% 1 100% 34 28% 0 0% 27 22% 0 0% 10 8% 0 0% 8 7% 0 0% 1 1% 0 0% 0 0% 0 0% 1 1% 1 100% 123 100%

Fig. 7.1  Question categories for the five provinces

As the data indicate, most forms asked between one and three questions (72% of the total sample), however, there were forms that asked as many as eight language background questions and also forms that asked none. In terms of the general categories of questions, there was also a high degree of variation, as indicated in Fig. 7.1. The single most frequently asked type of question in Ontario comes from category one, first language (asked by 86% of the registration forms in the Ontario sample). In Alberta, British Columbia, Manitoba and Prince Edward Island, the percentage of forms that contain a question from this category is lower (43%, 33%, 4% and 0%, respectively). British Columbia and Manitoba forms most commonly

Language Background Profiling at Canadian Elementary Schools and Dominant…

129

ask questions from category two, home language (75% and 81%, of the two provinces’ samples, respectively). In Alberta and Prince Edward Island the most commonly asked questions are from category three, primary language at home (64% and 100% respectively) and also category five, other languages, for Prince Edward Island only (100%).7 Overall, these data not only highlight the variability in the language background profiling questions asked by the different provinces, but also suggest different orientations with regard to the chronological-nativist versus synchronic-­functional continuum discussed earlier. As a preliminary generalization, more forms in Ontario seem to display a chronological-nativist orientation, as there is a higher emphasis on category 1, first language, questions in that province. However, this idea should be approached with caution as most forms do not ask a single profiling question and thus after exploring the combination patterns of the forms, the orientation may switch from chronological-nativist to mixed (e.g. if questions from both category 1 and another category are asked). The question combination patterns observed in the five provinces are reported in Table 7.3. The Table 7.3 data is organized in simple (single question) and complex (two or more questions) patterns of questions included in the registration forms sampled across the five provinces. This table further reinforces the finding that a high degree of variability exists both within and across provinces in terms of not only the types of questions chosen but also in terms of how they combine on the individual forms. It also indicates that there are forms with exclusively chronological-nativist orientation as they focus only on incoming students’ first language (pattern 3) while others forms have a synchronic-functional orientation as they ask questions only about students’ current languages spoken (patterns 1–2, 4–5, 15–18). Still others show a mixed orientation in that they ask questions both about the incoming students’ native languages and about their current language abilities (patterns 6–14). Overall, in the total sample from these five provinces, 9% of the forms had a chronological-­ nativist, 47% had a synchronic-functional, and 45% had a mixed orientation to language background profiling. As a final step of the analysis, the question formulations in all forms were examined with regard to plural -s on the word language. This analysis was done in order to determine whether the profiling allowed for multiple responses to categories 1 (first language), 2 (home language) and 5 (other language); categories 3 (primary language at home) and 4 (language spoken most often/fluently) were excluded from this analysis as they ask about primary or strongest language, which are by definition concepts in the singular. The results from this analysis are presented in Table 7.4. As the data indicate, 97% of the forms that ask a question about incoming students’ first language (category 1) only allow a single answer to that question. That is, such forms do not allow for the possibility of a child being a simultaneous 7  Recall that the data from Prince Edward Island, the smallest Canadian province, is based only on one registration form that covers the schools for the English school board. As such, it is useful to keep in mind that the percentages generated in Fig. 7.1 are based on a single question in category three and category five. Nonetheless, the data are important as they represent an entire educational provincial authority, albeit in a small province.

130

N. Slavkov

Table 7.3  Question combination patterns

Pattern # Pattern 1: Simple Pattern 2: Simple

Pattern 3: Simple Pattern 4: Simple Pattern 5: Simple Pattern 6: Complex Pattern 7: Complex Pattern 8: Complex Pattern 9: Complex Pattern 10: Complex Pattern 11: Complex Pattern 12: Complex Pattern 13: Complex

# of forms in each province ON = 2, AB = 5, BC = 7, Total = 14 ON = 2, AB = 5, BC = 6, MB = 18, Total = 31 ON = 8, AB = 1, MB = 1, Total = 10 BC = 1, Total = 1

1. First language

2. Home language(s)

4. Language 3. Primary spoken most language at often / fluently home √





√ √

MB = 2, Total = 2 ON = 15, AB = 1, BC = 4, Total = 20 ON = 3, AB = 9, Total = 12 ON = 3, BC = 2, Total = 5 ON = 1, BC = 4, Total = 5 ON = 2, AB = 1, Total = 3 ON = 1, AB = 1, Total = 2 ON = 1, AB = 1, Total = 2 ON = 2, Total = 2

5. Other language(s)





















√ √ √











√ √ √

(continued)

Language Background Profiling at Canadian Elementary Schools and Dominant…

131

Table 7.3 (continued)

Pattern # Pattern 14: Complex Pattern 15: Complex Pattern 16: Complex Pattern 17: Complex Pattern 18: Complex

2. Home language(s)

4. Language 3. Primary spoken most language at often / fluently home √

ON = 1, Total = 1





BC = 1, MB = 2 Total = 3 PEI = 1 Total = 1



# of forms in each province ON = 1, Total = 1

1. First language √

5. Other language(s)

√ √

√ √

MB = 1 Total = 1



Table 7.4  Singular vs plural use of the word ‘Language’ for question categories 1, 2, 5 Question category Category 1: First language(s)

Category 2: Home language(s)

Category 5: Other language(s)

Province Ontario Alberta British Columbia Manitoba Prince Edward Island Total Ontario Alberta British Columbia Manitoba Prince Edward Island Total Ontario Alberta British Columbia Manitoba Prince Edward Island Total

Singular 36 95% 12 100% 8 100% 1 100% 0 0% 57 97% 11 42% 3 50% 18 100% 11 52% 0 0% 43 61% 2 33% 1 100% 1 100% 1 20% 1 100% 6 43%

Plural 2 0 0 0 0 2 15 3 0 10 0 28 4 0 0 4 0 8

5% 0% 0% 0% 0% 3% 58% 50% 0% 48% 0% 39% 67% 0% 0% 80% 0% 57%

132

N. Slavkov

bilingual (for an overview of the phenomenon of bilingual first language a­ cquisition, BFLA, see De Houwer 2009).8 With regard to category 2, home language, 39% of the forms sampled in the five provinces use a plural formulation and thus allow for more than one home language to be listed. The rest of the sample for this category assumes that children have a single home language. Finally, category 5, other language, is the one where the most plural use was found, representing 57% of the forms that ask this question across the five provinces. It should be noted, however, that this category was mostly used in a combination with a question from another category (e.g. first or home language) and thus, even if used in the singular, the concept of other language already allowed for a bilingual or multilingual linguistic profile.

4  Discussion 4.1  Research Questions 1–3 To return to the research questions identified earlier, the results augment previous findings (Slavkov 2016, 2018) with regard to the large degree of variation of language background profiling questionnaires both within and across provinces (R1). This variation was seen in multiple respects: the number of questions asked per form, the specific formulations used, the general categories within which the questions fell, and the combination patterns of questions on forms that included more than one category. On the one hand, this is an indication of a high degree of autonomy of local school districts and in some cases individual schools. Such an autonomy is perhaps welcome in the sense that language background profiling can be adapted to some degree to the particular demographic make-up of the populations in different districts within and across provinces. On the other hand, some degree of consistency and uniformity of the questionnaires must be required by the provinces because the existing picture raises questions of how data from the local level is treated and inputted in larger provincial databases in a uniform and reliable manner. This concern warrants a recommendation to provincial authorities to consider creating a standardized pool of questions with identical formulations and combination patterns that must be used consistently throughout a given provincial educational system, in addition to any optional questions that local schools or boards may wish to develop and use in their specific forms. With regard to the second research question (R2), the data indicated that 9% of the forms collected across the five provinces had a chronological-nativist orientation to language background profiling, 47% had a synchronic-functional 8  It is worth noting that since this data were first collected, some forms in Ontario have changed and currently allow up to three first languages to be listed. This suggests a welcome change in progress from monolingual to multilingual norms and warrants further investigation and potential reanalysis in future work.

Language Background Profiling at Canadian Elementary Schools and Dominant…

133

orientation, and 45% a mixed orientation. A purely chronological-nativist approach is problematic as it exemplifies a monolingual norm. That is, focusing only on a child’s first/native language and also assuming that a child can only have a single native language (97% of the forms in the entire sample did not allow for simultaneous bilingualism) is a clear indication of monolingual assumptions. This is a surprising finding in a country like Canada, where official bilingualism exists on the federal level (and variably at provincial and local levels), where awareness of bilingualism is expected to be high, and where official minority and bilingual/immersion education generally exists nationwide. This finding is also an indication of how pervasive (implicit) monolingual norms may be in society, even within the larger context of official bilingualism, high immigration rates, and in fact, a recent increase in persons declaring more than one mother tongue (i.e. simultaneous bilingualism or multilingualism) in the 2016 Canadian census (Statistics Canada 2017a). The third research question in this study (R3) aimed to address to what degree the complexities of bilingualism and multilingualism are captured by the existing language background profiling forms across the five provinces. There is no doubt that educational authorities are aware that they serve bilingual and multilingual populations, and this awareness is visible through the language background profiling questions themselves. That is, asking questions about children’s first, home, primary, and additional languages by itself is an indication of educational systems that are trying to understand and adapt to diverse populations. At the same time, in addition to the issues with regard to variability, consistency, and orientation already discussed with R1 and R2, the existing picture of language background profiling in the five provinces included in this study is subject to further recommendations. Bilingualism and multilingualism are not static and straightforward phenomena and cannot be captured easily by asking simple questions such as “What is the child’s first language,” “What is the child’s home language” or a combination thereof. Such profiles run the risk of being static and incomplete. Studies in the field of family language policy (Curdt-Christiansen 2013; King et  al. 2008; King and Fogle 2013; among others) have revealed multiple intricacies in home language use. For example, bilingual children in a minority context may become dominant in the majority language and respond to their parents in the majority language, even if the parents consistently address the children in the minority language (De Houwer 2009). Thus, a single home language cannot be identified when parents tend to use one language and children another, or when different adults speak different languages in the household (these are well-known phenomena in immigrant, minority, or intercultural/mixed families). In addition, bi/multilingual speakers often engage in various types of code-mixing, to various degrees and with various interlocutors in their daily lives. In cases of such dynamic and fluid language use patterns, a seemingly straightforward question about a child’s first language or generally about the home language may, in fact, perplex parents who are trying to fill out a school registration form. There is no easy solution to this issue, and since school language background profiling is often coupled with collecting other important registration information, it is understandable that little space is available on a single form, and thus a long list

134

N. Slavkov

of complex questions is not practical. Nonetheless, questions explicitly asking about a child’s productive abilities, allowing multiple responses for first language, and tapping into a child’s dominant or primary language may be more appropriate for bi/ multilingual populations. Questions invoking the concept of a primary language were found in some registration forms but were in the minority (recall Fig.  7.1, category 4, and to some extent category 3); furthermore, such questions did not indicate they referred specifically to the child, so in the case of the primary language at home, an answer could be ambiguous between the primary language of the household in general, versus the child’s primary language specifically. Some preliminary recommendations with regard to improved language profiling were offered in previous work (Slavkov 2016, 2018) and will thus not be discussed here. Building on earlier work and on the current study, in this chapter I highlight a few suggested formulations that seem both economical and relevant, such as: “What is the child’s first/native language (list more than one, if applicable),” “What language(s) does the child currently speak most often: _______________ at home; _______________ outside home (list more than one, if applicable, starting with the language the child speaks most frequently),” “What is the child’s strongest language? (list more than one if the child seems equally proficient in more than one language)”. While these are still preliminary and may need further elaboration or potential reformulation, I believe they can serve as a point of reference for educational authorities who are interested in updating their language background profiling practices.

4.2  R  esearch Question 4: Dominant Language Constellations and Other Multilingual Frameworks Other than presenting specific findings about language background profiling across five Canadian provinces through research questions 1–3, an important additional research question identified earlier was how the overall findings relate to Dominant Language Constellations. A DLC framework offers some conceptual and practical benefits. From a conceptual perspective, the idea that the dominant language constellation contains only the core, most expedient and actively used languages in an individual repertoire translates well into the idea of creating a language background profile for a child at school entry. To tap into the most relevant language abilities of the child, that profile needs to focus on the dominant languages that bi/multilingual children bring with them to the school environment. As mentioned earlier, language background profiles rarely distinguish between receptive and productive skills, and thus under umbrella terms such as home language or other languages, parents may list languages that a child understands but does not speak. Consistent with the assumptions of a DLC approach, these are not the core languages that a language profile needs to focus on, and while such information may be collected, it needs to remain in the constellation periphery and also be properly annotated in terms of receptive versus productive abilities.

Language Background Profiling at Canadian Elementary Schools and Dominant…

135

Also in line with a DLC framework, information on native language may be collected, but an individual may have two or more mother tongues, and as such the concept of the language background profile must be flexible and allow the idea of multiple responses in this category. This idea is also widely recognized in various bilingual and multilingual frameworks, but the transition to it from traditional, monolingual views of language is slow and painstaking. The DLC framework offers appealing visualization options, which can be both informative and convenient in creating individual students’ language background profiles for school entry purposes. Such visualizations help re-conceptualize traditional lists of questions and answers, or skills and abilities, which tend to be linear and seemingly static, into multidirectional and fluid notions, making apt analogies with outer space and astronomical phenomena. While DLC maps are not necessarily practical for inclusion in registration forms due to space limitations, they may be beneficial in sensitizing local, regional and provincial school authorities to the open and dynamic representational possibilities of the complex bi/multilingual realities that incoming students may be bringing into schools. Furthermore, as mentioned before, apart from school registration forms, some schools use additional tools for language background profiling, such as intake interviews, which are not discussed in this study. Nonetheless, it is worth mentioning that DLC maps may be used both as a conceptual and a practical tool in such interviews, where the interviewer can mark an incoming child’s languages on an adapted version of a map. In terms of practical implications related to registration form design, the DLC framework can be linked to specific types of language background profiling questions. For example, concepts such as dominant or primary language are important to both the DLC framework and language background profiling. As mentioned earlier, such questions were relatively rare in the sample of forms across the five Canadian provinces studied, even though they are insightful in terms of a child’s language profile. The concept of a dominant and primary languages is, of course, complex as well. For example, different individuals may speak in one language more often but read and write in another. While recognizing this added complexity, it should be pointed out that children enter Canadian schools generally at the age of 4 or 5 years old, and as such they are mostly preliterate; thus, questions about a dominant or primary language can be based on spoken language. Overall, the DLC framework is not only compatible with but also enhances the specific question formulations tentatively proposed in this chapter (recall the end of the previous section).

5  Conclusion Language background profiling in Canada, and more generally around the world, is an understudied but important phenomenon, not only because it involves large populations but also because it reflects societal views and (mis)conceptions of language. This study confirmed previous accounts of language background profiling in Canada, suggesting that profiling can be made more meaningful and accurate. The

136

N. Slavkov

variability in formulations, number, type and combinations of questions, as well as the conceptual underpinnings of profiling practices indicate a need for further improvement and elaboration. The analysis of registration forms revealed an overall lack of explicit recognition of simultaneous bilingualism, a surprising finding in a country with two official languages; furthermore, issues with regard to concepts of language dominance, receptive vs. productive skills, and monolingual norms in general were identified. While Canada is widely known for strong publicly-funded education, research-based opportunities for continuous improvement continue to exist. Insights from multilingual frameworks such as plurilingualism, translanguaging, multicompetence, and Dominant Language Constellations, to which this study accorded a special role, represent fruitful avenues to pursue further in terms of both conceptual and practical advances in understanding and capturing linguistic diversity in open and inclusive ways. Acknowledgements  This research has received funding from the Social Sciences and Humanities Research Council of Canada (Insight Development Grant # 430-2017-00558) and from the University of Ottawa. Many thanks are due to the following students for assistance with data collection: Stephanie Krulicki, Xiaoyang Zhang, Stephanie Marshall, and Linlan Zhang.

References Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L. (2019). Dominant language constellation as a method of research. In E.  Vetter & U.  Jessner (Eds.), International research on multilingualism breaking with the monolingual perspective (pp.13–26). Cham: Springer. Bailey, A. L., & Kelly, K. R. (2013). Home language survey practices in the initial identification of English learners in the United States. Educational Policy, 27(5), 770–804. Bloomfield, L. (1933). Language. New York: Holt, Rinehart, & Winston. Canagarajah, S. (2007). Lingua Franca English, multilingual communities, and language acquisition. Modern Language Journal, 91(5), 921–937. Canagarajah, S. (2009). The plurilingual tradition and the English language in South Asia. AILA Review, 22(1), 5–22. Canagarajah, S. (2016). Crossing borders, addressing diversity. Language Teaching, 49(3), 438–454. https://doi.org/10.1017/S0261444816000069. Chomsky, N. (1957). Syntactic structures. The Hague: Mouton. Cook, V. J. (1991). The poverty-of-the-stimulus argument and multicompetence. Second Language Research, 7(2), 103–117. Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33(2), 185–209. Cook, V. J. (2013). Multi-competence. In C. A. Chapelle (Ed.), The encyclopedia of applied linguistics (pp. 3768–3774). New York: Wiley-Blackwell.

Language Background Profiling at Canadian Elementary Schools and Dominant…

137

Coste, D., & Simon, D.-L. (2009). The plurilingual social actor: Language, citizenship and education. International Journal of Multilingualism, 6, 168–185. https://doi. org/10.1080/14790710902846723. Coste, D., Moore, D., & Zarate, G. (1997). Competence plurilingue et pluriculturelle. Vers un cadre europeen commun de reference pour l’enseignement et l’apprentissage des langues vivantes: Etudes preparatoires [Plurilingual and pluricultural competence. Toward a common European reference framework for teaching and learning living languages: Preparatory studies]. Strasbourg: Editions du Conseil de l’Europe. Coste, D., Moore, D., & Zarate, G. (2009). Plurilingual and pluricultural competence. Strasbourg: Council of Europe. Retrieved from www.coe.int/t/dg4/linguistic/ publications_en.asp?toprint=yes&-40 Council of Europe. (2001). The common European framework of reference for languages. Cambridge: Cambridge University Press. Curdt-Christiansen, X. L. (2013). Family language policy: Sociopolitical reality versus linguistic continuity. Language Policy, 12(1), 1–6. Dagenais, D., & Moore, D. (2008). Représentations des littératies plurilingues, de l’immersion en français et des dynamiques identitaires chez des parents en chinois. Revue canadienne des langues vivantes, 65(1), 11–31. Davies, A. (1991). The native speaker in applied linguistics. Edinburgh: Edinburgh University Press. De Houwer, A. (2009). Bilingual first language acquisition. Tonawanda: Multilingual Matters. García, O. (2009). Bilingual education in the 21st century: A global perspective. Malden: Basil/ Blackwell. García, O., Ibarra Johnson, S., & Seltzer, J. (2017). The translanguaging classroom: Leveraging student bilingualism for learning. Philadelphia: Caslon. Kachru, B. B. (1983). The Indianization of english: The english language in India. Oxford: Oxford University Press. Kachru, Y., & Smith, L. E. (2008). Cultures, contexts, and world englishes. New York: Routledge. King, K. A., & Fogle, L. W. (2013). Family language policy and bilingual parenting. Language Teaching, 46(2), 1–13. King, K.  A., Fogle, L., & Logan-Terry, A. (2008). Family language policy. Language and Linguistics Compass, 2(5), 907–922. Lewis, G., Jones, B., & Baker, C. (2012). Translanguaging: Origins and development from school to street and beyond. Educational Research and Evaluation, 18(7), 641–654. Marshall, S., & Moore, D. (2013). 2B or not 2B plurilingual? Navigating languages literacies, and plurilingual competence in postsecondary education in Canada. TESOL Quarterly, 47(3), 472–499. Ontario Ministry of Education. (2016). The Kindergarten program. Retrieved from: https://files. ontario.ca/books/edu_the_kindergarten_program_english_aoda_web_oct7.pdf Ontario Ministry of Education. (2017). Education facts: Schools and school boards, 2015–16. Retrieved from: http://www.edu.gov.on.ca/eng/educationFacts.html Paikeday, T. (1985). The native speaker is dead! Toronto: Paikeday Publishing. Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford University Press. Rampton, M. B. H. (1990). Displacing the “native speaker”: Expertise, affiliation, and inheritance. ELT Journal, 44(2), 97–101. https://doi.org/10.1093/eltj/44.2.97. Slavkov, N. (2016). Search of the right questions: Language background profiling at Ontario public schools. The Canadian Journal of Applied Linguistics, 19(1), 22–45. Slavkov, N. (2018). What is your ‘first’ language in bilingual Canada? A study of language background profiling at publicly funded elementary schools across three provinces. International Journal of Bilingual Education and Bilingualism, 21(1), 20–37. Statistics Canada. (2017a) Census in brief: Linguistic diversity and multilingualism in Canadian homes. Retrieved from http://www12.statcan.gc.ca/census-recensement/2016/as-sa/98-200x/2016010/98-200-x2016010-eng.cfm

138

N. Slavkov

Statistics Canada. (2017b). Census in brief: English-French bilingualism reaches new heights. Retrieved from http://www12.statcan.gc.ca/census-recensement/2016/as-sa/98-200x/2016009/98-200-x2016009-eng.cfm Taylor, S. K., & Snoddon, K. (2013). Plurilingualism in TESOL: Promising controversies. TESOL Quarterly, 47(3), 439–445. Wernicke, M. (2017a). Navigating native-speaker ideologies as FSL teacher. Canadian Modern Language Review, 73(2), 208–369. Wernicke, M. (2017b, November 22). The native speaker in L2 education: What is the impact on teacher identity? Paper presented in the Educlang Seminar Series, Faculty of Education, University of Ottawa, Ottawa, ON, Canada. Williams, C. (1994). Arfarniad o ddulliau dysgu ac addysgu yng nghyd-destun addysg uwchradd ddwyieithog [An evaluation of teaching and learning methods in the context of bilingual secondary education]. Unpublished doctoral dissertation, University of Wales, Bangor. Williams, C. (2002). A language gained: A study of language immersion at 11–16 years of age. Bangor: University of Wales Bangor, School of Education.

Dominant Language Constellations in the Language Repertoires of Multilingual South African Students Susan Coetzee-Van Rooy

Abstract  There is a growing research focus on the nature of multilingualism at the societal and individual levels as a current phenomenon. A variety of approaches are followed by scholars of multilingualism. In this chapter, the concept of the dominant language constellation (DLC) of multilingual people coined by Aronin and colleagues (Aronin L, Singleton D, Multilingualism. John Benjamins, Amsterdam, 2012; Aronin L, Ó Laoire M, International Journal of Multilingualism 10:225–235, 2013) is applied to the language repertoire survey work done on urban present-day South African multilingualism. The focus is on Southern Sotho and Zulu home language participants and the factors that motivated their learning and current use of the languages included in their repertoires. This investigation is important, because more work is needed to understand the relationship between the DLC and the language repertoire; and there is a paucity of data from Africa in the DLC framework. The main finding of the exploration is that there is a clear complementary relationship between the language repertoire and the DLC.  Additional findings are that claims made about the size of a typical DLC (to include three languages) are supported empirically by the data reported in the chapter. Furthermore, the findings hold implications  for language in education practices for urban multilingual students in South Africa. Keywords  Language repertoire · Dominant language constellation · South African urban multilingualism · Southern Sotho · Zulu · Education

1  Introduction The current focus on the phenomenon of multilingualism has been documented well by several researchers (Aronin and Singleton 2012; Edwards 2002; Pavlenko 2011; Todeva and Cenoz 2009; Van der Walt 2013). The phenomenon is studied from a variety of perspectives (Aronin and Hufeisen 2009: 104; Aronin and Jessner-Schmid S. Coetzee-Van Rooy (*) North-West University, Vanderbijlpark, South Africa e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_8

139

140

S. Coetzee-Van Rooy

2015: 274). The list with examples of work done on multilingualism that follows is by no means exhaustive of this growing field. For example, multilingualism is studied from psycholinguistic perspectives (Herdina and Jessner 2002; Jessner 1999, 2008). The relationship between multilingualism and emotion is studied by scholars (Pavlenko 2007; Dewaele 2016). Multilingualism scholars also focus on the issue of multilingualism and identity at personal (Kramsch 2004; Todeva and Cenoz 2009; Norton 2013) and societal levels (Laitin 2007; Lo Bianco 2010; Duchêne and Heller 2012). The field of multilingualism and education research is vibrant (Hornberger 2009; Cenoz and Gorter 2011; Canagarajah and Ashraf 2013). A variety of new and traditional methodologies are used to study multilingualism (Aronin and Hufeisen 2009). Scholars use language repertoire surveys (Benor 2010; Sharma 2011; Coetzee-Van Rooy 2012), language history interviews (Coetzee-Van Rooy 2016), multilingual portraits (Busch 2010; Van Zyl 2016), and social network analyses (Botha 2017), among many others. Recently, scholars like Aronin and colleagues (Aronin 2006; Aronin and Singleton 2012; Aronin and Ó Laoire 2013; Aronin 2016) introduced the concept, the “dominant language constellation” (DLC) as “a method for research in multi-­ competence” (Aronin 2016: 156). The DLC is defined as “the group of his/her most important languages that, functioning as an entire unit, enable him/her to act in a multilingual environment in such a way as to meet all of his/her needs” (Aronin and Singleton 2012: 59). Aronin and Singleton (2012: 62) state that the language repertoire “relates to the totality of an individual’s or a community’s linguistic skills” (Aronin and Singleton 2012: 62), and that the DLC is the: group of one’s most important (vehicle) languages, functioning as an entire unit, and enabling an individual to meet all his or her needs in a multilingual environment … The DLC includes only the most expedient languages for a person, rather than all the languages known to them, as would be the case in a language repertoire. Unlike a repertoire, a DLC comprises the languages which, together, perform the most vital functions of language (Aronin 2016: 146-147).

It is clear that Aronin and her associates accept a complementary relationship between the DLC and the language repertoire. To date, research methodologies with which DLCs have been described are not presented in an extensive way. The example of a DLC presented in Aronin and Singleton (2012: 61-62) and the discussion about methodology in Aronin and Jessner-Schmid (2015: 64–66) are conducted at a macro-level. In other words, there is no description of how the information that leads to the presentations of the DLCs (the figures) is gathered. The identification of the DLCs seem to be intuitive and based on the vast experience of the researchers in the multilingual contexts that they study, informed by interviews with participants. There is reference to the fact that the DLC is not a “tangible phenomenon” but a “mental construct” (Aronin and Jessner-Schmid 2015: 65); and that it serves as an “abstract model” (Aronin and Jessner-Schmid 2015: 65) and a “lens” (Aronin and Jessner-Schmid 2015: 66) that directs multilingualism research. Without a clear description of how DLCs should be mapped, it would remain difficult to gather comparable data that would lead to a deeper understanding of current urban multilingualism. One of the aims of this chapter would be to explore how the language

Dominant Language Constellations in the Language Repertoires of Multilingual South…

141

repertoire survey instrument used in the study could potentially assist in the important goal of mapping DLCs across the world to contribute to a global understanding of present-day multilingualism. Against this introduction, the aim of the chapter is to explore what language repertoire survey data reveal about the nature of DLCs in the multilingual minds of South African students. This chapter hopes to contribute information that deepens our understanding of the relationship between the language repertoire and the DLCs that lodge in these repertoires. A secondary aim of the chapter is to explore the usefulness of language repertoire surveys as a research tool to study DLCs empirically. A third aim is to contribute South African data to the discussion of multilingual repertoires and DLCs because there is a paucity of work from deeply multilingual urban contexts in Africa and India (Todeva and Cenoz 2009).

2  Comparing a Language Repertoire and DLC View In their work on multilingualism, Aronin and colleagues (Aronin and Singleton 2012: 62; Aronin 2016: 144) maintain that the concept language repertoire is related to, but not the same as, the idea of the DLC of a multilingual individual. There are two problems that they highlight with the notion of a language repertoire, which they believe are solved when one uses the notion of the DLC. The first problem that they observe, is that views of a person’s language repertoire relate to the “totality of an individual’s or a community’s linguistic skills” (Aronin and Singleton 2012: 62). It is argued that, “Repertoire includes every language or skill that an individual or a group possesses” (Aronin 2016: 144). In their view, this is problematic because not all languages in the repertoire are essential in a person’s daily life (Aronin and Singleton 2012: 63). In other words, information about the totality of language skills gathered in a repertoire approach towards multilingualism hides the fact that some languages are used more frequently and for specific purposes than others in the repertoire. Ultimately, the argument could be that a language repertoire approach misrepresents the nature of the multilingualism maintained at individual and societal levels, because it does not take into account that not all languages in the multilingual repertoire are actively used to the same extent regularly. It is argued that the concept of the DLC corrects this problem, because it directs the research focus to the most important languages in the multilingual language repertoire that are used to perform important functions daily. In other words, rendering a more precise view of the nature of the multilingualism lodged in individual minds and societies. Although I agree with the important qualification of the nature of multilingual repertoires added by the DLC concept, I see a close and important relationship between the details represented in a description of the complete language repertoire and the specific DLC within complete multilingual language repertoires. I believe that the language repertoire approach followed in this chapter caters specifically to describe complex multilingualism and in this way provides an important empirical baseline for the identification and description of the DLC within complex

142

S. Coetzee-Van Rooy

multilingual repertoires. Although I agree with the criticism of some language repertoire approaches as expressed by Aronin and colleagues, I believe that a multilingual language repertoire approach should be considered when DLCs are studied. If the aim of a language repertoire survey is to simply create a list of all the languages known by participants and the self-reported perceptions of the language skills they possess in these languages, then language repertoire work would indeed not render detailed information about the co-existence of languages in the mind of a multilingual person and how languages co-perform communication functions. However, if a language repertoire survey instrument (Coetzee-Van Rooy 2012) takes care to gather detailed information about: • what participants view as their strongest language (and the respective ranking of their second, third, fourth and fifth strongest languages, for example); • when they learnt their strongest language; • why they learnt their strongest language; • what their perceptions of their proficiency in specific skills in their strongest language are; • their beliefs towards their strongest language in terms of liking to use the language, their confidence in using the language, the importance of being good in the language and their perception about the relationship between the language and their identity; • which factors contributed to the learning of their strongest language; • which factors contribute to the current use of their strongest language; and gather the same information for the participants’ second, third, fourth and fifth strongest languages, then the relative importance and prominence of the languages in the repertoires of multilingual participants are documented carefully. In fact, the detail gathered by this type of language repertoire survey would enable a description of the DLCs in the multilingual repertoires of the participants to a great extent, addressing the second problem with a repertoire approach observed by Aronin and colleagues. The second problem with a language repertoire approach as raised by Aronin and colleagues, is that language repertoire work do not provide exact information of the languages known by a multilingual person that perform the “most vital functions of language” (Aronin 2016: 147). In other words, a focus on “all the languages known to them, as would be the case in a language repertoire” (Aronin 2016; 146–147) study, does not provide information about “the languages which, together, perform the most vital functions of languages” (Aronin 2016: 147) or the DLC. If the language repertoire survey approach implemented elsewhere (Coetzee-Van Rooy 2012) and described above is followed, detailed information about the DLCs are gathered because the instrument enables a detailed description of the perceived strongest languages in the repertoire as well as detailed information about when, how and why each of these languages were added to the repertoire as well as which factors contribute to the current use of the languages and the attitudes towards the languages. In fact, if this language repertoire survey approach is used to gather baseline information about the language repertoire and DLCs, and is further used in

Dominant Language Constellations in the Language Repertoires of Multilingual South…

143

combination with methodologies like that of language history interviews (CoetzeeVan Rooy 2016), language portraits (Busch 2010) and language use in social networks (Botha 2017), a clear picture of the complexities lodged in the multilingual mind would become more visible and would advance our understanding of multilingual repertoires and the DLCs within them markedly. In my view, the language repertoire survey approach used elsewhere and described in this chapter, is not only related to the idea of the DLC, it forms the baseline data set from which a systematic and more exact study of the DLCs in multilingual minds could be undertaken. In addition to this, as already stated by Aronin and colleagues, the notion of the DLC within multilingual repertoires provides an even sharper focus on the complexities related to multilingualism and how these should be taken into consideration when language in education policies or social policies related to multilingualism and social cohesion are formulated and implemented (Aronin 2016: 159). In the rest of the chapter, selected information from the specific language repertoire survey approach used elsewhere is presented to explore the potential of the baseline language repertoire data gathered to deepen understanding of the DLCs of Southern Sotho and Zulu home language participants in a larger study.

3  Empirical Study The data reported in this chapter form part of a longitudinal study of the language repertoires of multilingual students at the Vaal Campus of the North-West University (NWU) of South Africa. The main aim of the larger study is to document and trace the vitality of multilingualism in this population and region over time. The language repertoire survey is conducted with first year students at the NWU Vaal Campus every 5 years. The first year of data gathering was 2010. For the purpose of this chapter, two questions from the language repertoire surveys conducted in 2010 and 2015 will be analysed with a view to explore how the data provide detailed information about the DLCs in the minds of the Southern Sotho and Zulu home language participants. In the rest of this section, the South African language context will be described, as well as the participants, instrument and data gathering and analysis processes relevant to the chapter.

3.1  Contextualization The language repertoires of speakers of African languages are complex (Bamgbose 2011; Banda 2009; Bokamba 2015; Prah 2010). Oane and Glanz (2010:4) provide a recent and succinct description of the multilingual realities in Africa where “multilingualism is normality”. They highlight the estimations of the vast number of languages used on the African continent (numbers vary between 1000 and 2500

144

S. Coetzee-Van Rooy

languages); and that “monolingual states are non-existent” in Africa (Oane and Glanz 2010: 8). In this context, children ordinarily grow up with more than one home language and several family and neighbourhood languages and this context complicates, for example, the language in education decisions made in Africa (Oane and Glanz 2010:4, 14). In South Africa, eleven languages have been declared as official. The 2011 Census data in South Africa indicate that Zulu (with 11,587,374 speakers or 23% of the population), Xhosa (with 8,154,257 speakers or 16% of the population) and Afrikaans (with 6,855,082 speakers or 14% of the population) are the biggest home languages in the country. English, Northern Sotho, Tswana and Southern Sotho have around 4 million home language speakers each (or about 7–9% of the population used one of these four languages as home languages in 2011). Tsonga, Swati, Venda and Ndebele have between 1 and 2 million speakers (or about 2–4% of the population used one of these four languages as home languages in 2011). In the Vaal Triangle region in South Africa, which is located about 70 kilometres South of Johannesburg in the Gauteng province, there are 903,748 people. The 2011 Census data indicated that 422,198 (or 47%) speakers in the Vaal Triangle region use Southern Sotho as a home language, 144,229 (or 16%) use Zulu as a home language and 136,990 (or 15%) use Afrikaans as a home language. All the other languages used in the region have between 3000 and 65,000 users. The Vaal Triangle region is therefore a typical urban diverse South African context where multilingualism is the norm.

3.2  Participants The participants in the larger study (n = 1933) include all first year students that attended the welcoming and orientation programme of the NWU on the Vaal Campus on 23 January 2010 (n = 1011) and 21–23 January 2015 (n = 922). There were 1253 (out of 19231 or 65.15%) female participants and 670 (out of 1923 or 34.84%) male participants. The average age of the participants was 19 years. Both the gender and age distribution are typical for South African first year university students. The participants reported to know an average of 3.18 languages (n = 1922, minimum = 1 and maximum = 20; standard deviation = 1,742). This is an important finding, in that it seems to correspond with the general finding reported by Aronin and colleagues (Aronin and Singleton 2012: 63; Aronin 2016: 151) that a typical DLC, based on their research to date, seems to include three languages. In the rest of the data analysis, the notion that three languages usually forms the DLC within a broader multilingual repertoire will be considered. The home language distribution of the participants (see Table 8.1) reflects the distribution of prominent languages in the region as reported in the Census 2011

 10 participants in 2015 did not provide answers to the gender question.

1

Dominant Language Constellations in the Language Repertoires of Multilingual South… Table 8.1 Home language distribution of the participants in the larger study (2010 and 2015)

Home language Southern Sotho Afrikaans Zulu English Tswana Northern Sotho Xhosa Venda Ndebele Swati Tsonga / Shangaan No answer Shona 2 home languages Other German 3 home languages Ndonga Portuguese Spanish Total

145

Frequency Percentage 680 35.18 393 20.33 252 13.04 163 8.43 158 8.17 102 5.28 89 4.60 23 1.19 14 0.72 14 0.72 14 0.72 9 0.47 6 0.31 5 0.26 4 0.21 2 0.10 2 0.10 1 0.05 1 0.05 1 0.05 1933 100.00

data in the contextualization section. Southern Sotho, Afrikaans and Zulu are the most prominent languages in the region and are also the most prominent home languages of the participants in the larger study. For the purpose of this chapter, the analysis will focus on the Southern Sotho home language and Zulu home language participants as these are the prominent home languages in the region. For the purpose of the chapter, the strongest, second strongest, third strongest, fourth strongest and fifth strongest languages for the 2010 and 2015 participants in the larger study are also important as indicators of the size and shape of the complete language repertoires, and the identification of the DLCs in these repertoires. An additional strength of the nature of the language repertoire survey instrument, is that one could make very fine distinctions between data sets. For the purpose of this chapter, I am, for example, only going to include the data sets of participants in the project who reported that Southern Sotho or Zulu is their home language and strongest language; and that English is their second strongest language. In other words, instead of clustering the data of 680 Southern Sotho home language participants or the 252 Zulu home language participants in the larger study in one analysis, a much more fine-grained analysis is conducted by selecting a very specific type of language repertoire for analysis: that of Southern Sotho and Zulu home language users who also perceive Southern Sotho and Zulu as their strongest languages and English as their second strongest language. This is an important distinction to make. In both data sets (2010 and 2015) about 20% of the participants in the larger study report that Southern Sotho or Zulu is their home language, but that English is perceived as their strongest language; and

146

S. Coetzee-Van Rooy

Southern Sotho or Zulu is regarded as their second strongest language. These subtle differences in the complete language repertoires should be taken into consideration in any analysis, because a perceived difference between home language and strongest language surely holds important implications for the nature of the DLC of participants. For example, in the case where a group of participants report that Southern Sotho or Zulu is their home language, but English is their strongest language and Southern Sotho or Zulu is their second strongest language, one would need to carefully consider the use of these languages in the domain of education, specifically for literacy functions. For the purpose of this chapter, only the cases where the Southern Sotho or Zulu was regarded as home language and the strongest language of the participants in the larger study are included in the analysis. At least three additional future DLC analyses should be done from the data gathered in the larger study. One DLC to consider separately would be the case where participants perceive the home language as second strongest language and English as strongest language. The DLCs of language participants in the larger study (and in the region), that use minority African home languages (see Table 8.1 for home languages with less than 150 participants, for example) as strongest languages would also be important to investigate as these repertoires and DLCs could be different from users of dominant languages. The already observed Afrikaans-English bilingual DLCs of Afrikaans home language participants (Coetzee-Van Rooy 2013) and the potential English dominant DLCs of monolingual speakers should also be investigated separately. These analyses would provide a complete picture of the multilingual repertoires and DLCs in this context. For the participants in this study (2010 and 2015), a total of 352 (out of 1933 or 18.21%) of the Southern Sotho home language participants regarded Southern Sotho as their home language and strongest language and English as their second strongest language. In the case of the Zulu participants, 121 (out of 1933 or 6.26%) regarded Zulu as their home language and strongest language and English as their second strongest language. The third strongest languages of the selected participants are reported in Table 8.2. From Table 8.2, it is clear that Zulu is the perceived as the most prominent third strongest language for the Southern Sotho home language participants and that Southern Sotho is perceived as the most prominent third strongest language for the Zulu home language participants in this specific analysis. This is not an unexpected finding when one takes the Census 2011 data for the region into account. The three most prominent languages in the Vaal Triangle region are Southern Sotho, Zulu and Afrikaans. Furthermore, in a multilingual urban context like South Africa, it makes a lot of sense for home language users of a Sotho language to acquire an Nguni language and vice versa. The Sotho languages in South Africa include Southern Sotho, Northern Sotho (or Pedi) and Tswana; and communication in this language family is mutually comprehensible for speakers of the three Sotho languages. The Nguni languages (which are also mutually comprehensible) include Zulu, Xhosa, Ndebele and Swati. Knowledge of one Sotho and one Nguni language gives South Africans access to seven of the official languages to some extent. The relative high numbers reported for Tswana as third strongest language for the Southern Sotho

Dominant Language Constellations in the Language Repertoires of Multilingual South…

147

Table 8.2  Frequencies for perceived third strongest languages in the repertoire for Southern Sotho and Zulu home language and strongest language participants Southern Sotho home language and strongest Zulu home language and strongest language language and English second strongest language and English second strongest language 3rd strongest 3rd strongest language Frequency Percentage language Frequency Percentage Zulu 116 32.95 Southern Sotho 44 36.36 No answer 105 29.83 No answer 35 28.93 Afrikaans 52 14.77 Xhosa 18 14.88 Tswana 40 11.36 Afrikaans 8 6.61 Xhosa 23 6.53 Swati 7 5.79 Northern Sotho 14 3.98 Northern Sotho 4 3.31 Italian 1 0.28 Tswana 2 1.65 Venda 1 0.28 Ndebele 1 0.83 Tsonga/Shangaan 1 0.83 Venda 1 0.83 Total 352 100.00 Total 121 100.00

home language participants and Xhosa as third strongest language for the Zulu home language participants is therefore not unexpected, because Tswana is closely related to Southern Sotho and Xhosa is closely related to Zulu. To add these languages to the repertoires would not be difficult for the Southern Sotho and Zulu home language participants in this specific analysis. The frequencies for perceived fourth strongest languages are reported in Table 8.3. What should also be noted in Table 8.2, is that 29.83% of the Southern Sotho home language participants and 28.93% of the Zulu home language participants do not report the use of a third strongest language. This percentage increases in the case of the fourth strongest languages, reported in Table  8.3 where 66.19% of the Southern Sotho home language participants and 56.20% of the Zulu home language participants do not report the use of a fourth strongest language. In the case of perceived fifth strongest languages for these participants, 80% of the participants report to not use a fifth strongest language (see Table 8.4). This is an important finding in the context of the assumptions made by Aronin and colleagues concerning the typical size of a DLC (Aronin and Singleton 2012: 63; Aronin 2016: 151). It seems that the South African data for multilingual urban students that participated in the larger study, and the participants selected for this specific analysis, provide empirical evidence for this assumption. In the larger study, participants report an average of 3.18 languages that form part of their repertoires; and in the selected group of Southern Sotho and Zulu home language participants that are discussed in this chapter, it is clear that 70% report to include a third strongest language in their repertoires; 40% included a fourth strongest language and 20% reported the inclusion of a fifth strongest language in their repertoires. Even in the context of a deeply multilingual urban society like South Africa where there is widespread societal and individual multilingualism, it seems that

148

S. Coetzee-Van Rooy

Table 8.3  Frequencies for perceived fourth strongest languages in the repertoire for Southern Sotho and Zulu home language and strongest language participants Southern Sotho home language and strongest Zulu home language and strongest language language and English second strongest language and English second strongest language 4th Strongest 4th Strongest language Frequency Percentage language Frequency Percentage No answer 233 66.19 No answer 68 56.20 Zulu 46 13.07 Xhosa 17 14.05 Tswana 25 7.10 Áfrikaans 14 11.57 Afrikaans 18 5.11 Southern Sotho 10 8.26 Xhosa 14 3.98 Swati 5 4.13 Northern Sotho 11 3.13 Tswana 4 3.31 Venda 3 0.85 Ndebele 2 1.65 French 1 0.28 Portuguese 1 0.83 Tsonga / Shangaan 1 0.28 Total 352 100.00 Total 121 100.00 Table 8.4  Frequencies for perceived fifth strongest languages in the repertoire for Southern Sotho and Zulu home language and strongest language participants Southern Sotho home language and strongest Zulu home language and strongest language language and English second strongest language and English second strongest language 5th Strongest 5th Strongest language Frequency Percentage language Frequency Percentage No answer 287 81.53 No answers 96 79.34 Tswana 16 4.55 Swati 5 4.13 Afrikaans 13 3.69 Afrikaans 4 3.31 Xhosa 12 3.41 Southern Sotho 4 3.31 Northern Sotho 9 2.56 Xhosa 4 3.31 Zulu 8 2.27 Ndebele 2 1.65 French 1 0.28 Northern Sotho 2 1.65 Italian 1 0.28 Tswana 2 1.65 Ndebele 1 0.28 Tsonga / Shangaan 1 0.83 Shona 1 0.28 Venda 1 0.83 Tsonga/Shangaan 1 0.28 Venda 1 0.28 Other 1 0.28 Total 352 100.00 Total 121 100.00

most of the participants selected for the analysis in this chapter (70%) hold three languages (also evidenced by the mean number of languages known by participants in the larger study) in their repertoires and the strongest three languages reported in the repertoire could well be conceived of as the DLCs for the selected participants in this study. Based on these baseline findings, the rest of the analysis in this chapter will be conducted with the data for strongest, second strongest and third strongest languages reported by the selected Southern Sotho and Zulu participants.

Dominant Language Constellations in the Language Repertoires of Multilingual South…

149

3.3  Language Repertoire Survey Instrument The language repertoire survey instrument was compiled by using existing language survey instruments in the public domain (with acknowledgement and references to the authors) and self-compiled questions. Section A of the questionnaire consists of questions that relate to typical biographical information: gender, age, population group, highest qualifications of parents and jobs of parents. The questions in Section B focus on the self-identification of home languages by the participants, the number of languages known by the participants and the order in which the participants have learnt the languages that they know. This question is included in the survey instrument to clearly distinguish between the historical order in which languages in the repertoires of the participant were acquired or learned; and the current strength of languages in the repertoires of the participants which form the focus of the next section of the survey instrument. I therefore agree with Slavkov (see Chap. Societal versus individual patterns of DLCs in a Finnish educational context. Present state and challenges for the future in this volume) that both these pieces of information are necessary when one studies the language repertoires of urban multilingual people and that these pieces of information should not be conflated. Section C contains the main questions of the questionnaire. These questions focus the attention of participants on their strongest, second strongest, third strongest, fourth strongest and fifth strongest languages. One’s “strongest language” is defined as “the language that is the easiest for you to use to express yourself and people who know your strongest language understand what you want to communicate with ease in this language. This is the language in which you are the most proficient in your opinion”. Once the “strongest language” is identified, participants answer a set of questions referring to their “strongest language”. These questions focus on when participants started to learn their “strongest language”, why they learnt the language, their perceptions about their proficiency in the language for typical language skills, four questions that relate to their attitude towards the language, a set of questions about the contribution of factors towards the learning of the language and a set of questions about the contribution of the same factors that motivate the current use of the language. This set of questions is repeated for the second strongest language, third strongest language, fourth strongest language and the fifth strongest language. One could add as many strongest languages as one wants to the questionnaire. Based on the researcher’s understanding of the nature of the multilingualism of South Africans, informed by literature, it was decided to limit the data to five languages to ensure that the scope of the data set does not become impossible to comprehend. In additional case study projects, questions for more languages were added to accommodate the language repertoires of individuals with, for example, 13 languages. This is not difficult to administer in individual case studies. Section D includes three questions that assess participants’ views about the necessity of multilingualism in the South African context. As should be clear from the description of the language repertoire survey instrument in this section, it provides a lot of information that would enable the study of

150

S. Coetzee-Van Rooy

the DLCs of multilingual participants. The mere notion of asking participants to identify their strongest, second strongest etc. languages provide an indication of languages that are more and less prominent in the DLCs. The detail about motivations for learning specific languages in the repertoire, and details about factors that influenced the learning of specific languages and the current use of specific language hold the potential to describe the DLCs in multilingual repertoires with a lot of detail. In this chapter, I want to focus on two questions in Section C of the questionnaire that provide information about the factors that contributed to the learning of languages in the participants’ repertoires and an assessment of the same factors in the current use of languages in the repertoire (see Appendix for the specific questions). An anonymous reviewer noted that apart from the factors listed in the Appendix in the survey instrument, there are numerous others that also influence language acquisition, for example, social media use, economic activities (like marketing), political discourse, religious experiences and many more. This observation is absolutely correct and it indicates the need to update classic instruments like those first proposed by Fishman (1965: 81) in his iconic article “Who speaks what language to whom and when?” and the factors proposed by Marian et al. (2007) which are also used in the survey instrument in the study reported on in this chapter. This observation points to an important issue to address in the design and update of survey instruments used in language repertoire and DLC research. Unfortunately, I did not include these factors in the survey instrument used for this longitudinal study where data were collected in 2010 and 2015; and a new data set will be collected in 2020. In the interest of comparing the data sets across time, I also foresee that I will not adjust the instrument by adding more factors that influence the learning and current use of languages in the repertoires of the participants. However, this issue could be addressed in surveys of new studies that are not related to this longitudinal project in this region and in interview, language portrait and language history studies. Responses to the questions that were asked about the factors that contributed to the learning and the current use of the languages in the participants’ repertoires could contribute to our understanding of the DLCs of the multilingual South African participants in the study in terms of providing details about which languages are used for at least these functions.

3.4  Data Gathering and Analysis The data were gathered in 2010 and 2015 with the help of research assistants. During the welcoming and orientation programme for first years on the Vaal Campus of the NWU, time was scheduled to inform the first years about the project and to get their informed consent to participate in the project by completing the language repertoire survey. The participants who agreed to participate spend 30–45 min to complete the questionnaire. The research assistants remained in the rooms to assist with questions.

Dominant Language Constellations in the Language Repertoires of Multilingual South…

151

In the larger project, the main method of analysis includes frequency analyses, and appropriate inferential statistics to answer specific research questions. For example, t-tests for independent samples are used to determine if the size of the multilingual repertoires between 2010 and 2015 differ in a statistically significant way. For the purpose of this chapter, I use correspondence analysis of the data of the relevant questions (see Appendix) to investigate how the languages included in the repertoires of the selected participants relate to the factors that contributed to them learning the languages that they know and their current use of the languages that they know. A correspondence analysis aims “to explore the structure of categorical variables included in [a] table” (Hill and Lewicki 2006: 127). For the purpose of the chapter, I compiled tables that included the means for the strongest, second strongest and third strongest languages reported by the Southern Sotho and Zulu home language participants selected for the analysis from the broader study for each of the factors that contributed to their learning of the languages in their repertoire and to their current use of the languages in their repertoire. The correspondence analysis provides a visual representation of two issues represented in the data. First of all, the relationship between row (or column) variables are plotted in relation to a point of origin (where the 0 of the X- and Y-axis meet on the plot). In the plots presented in this chapter, the row co-ordinates represent the factors that contributed to the learning of the languages in the repertoires of the participants; and the column co-­ ordinates represent the languages in the repertoires of the participants (e.g. Southern Sotho perceived as strongest language, English perceived as second strongest language and Zulu perceived as third strongest language). Row (or column variables) that are presented close to one another on the plots are associated. The second issue to note in the interpretation of a correspondence analysis, is that the association between row and column co-ordinates is represented by the length of the lines from the origin and the sharpness of the angle when a row and column variable is connected to the origin. The sharper the angle and the longer the lines from the origin, the stronger the association. With the correspondence analysis that I present in this chapter, I would therefore like to determine first of all if there are associations between the factors that contributed to the learning and current use of the languages in the repertoires of participants; and second of all, I would like to investigate the strength of associations between factors that contribute to the learning and current use of the languages and the strongest, second strongest and third strongest languages in the repertoires. The findings of the correspondence analysis are presented in the next section.

4  Findings The main idea with the analysis presented in this section is to explore how data gathered via the language repertoire survey instrument as described in this chapter could cast light on the DLCs in the complete repertoires of the participants. As already reported above, the notion that three languages potentially form a DLC in

152

S. Coetzee-Van Rooy

these multilingual repertoires have been confirmed. The attempt to link the contribution of factors towards the learning and current use of the strongest, second strongest and third strongest languages in the multilingual repertoires of the selected Southern Sotho and Zulu participants aims to explore if this language repertoire information could provide a clearer view of how these languages function when they are conceived of as a DLC that forms part of a multilingual repertoire.

4.1  T  he DLC s of the selected Southern Sotho Home language and Strongest Language Participants When one considers the prominent means in Table 8.5, it is clear that family contributed the most as a factor to the learning of Southern Sotho as a strongest language for the selected participants. The factors that contributed the most to the learning of English as a second strongest language and Zulu as a third strongest language are reading, the school, watching TV and listening to the radio. These findings are not surprising. Southern Sotho is the most prominent home language in the Vaal Triangle region (as reported by the Census 2011 data) and one would expect that the family is the most important factor that contributes to the learning of Southern Sotho by participants that perceive the language as their home language and strongest language. The finding that education and the media drive the learning of English perceived as a second strongest language is also not unexpected. English is the dominant language of learning and teaching in South African schools and in the media. The finding that education and the media drive the learning of Zulu is surprising to some extent, but can be explained. Zulu is the most prominent home language in South Africa and in the Gauteng province, Zulu is fast becoming a prominent language in diverse urban settings. The South African Broadcasting Corporation (SABC) would cater for Zulu speakers to a large extent and schools in urban settings like Gauteng would also enable the learning of Zulu as a school

Table 8.5  Mean scores for selected Southern Sotho home language and strongest language participants and factors that contributed to the learning of the languages in the DLC Factors that contributed to the learning of languages in the dominant language constellation Friends Family Reading School Watching TV Listening to the radio Self-instruction / tapes

Southern Sotho SL 5.86 6.69 5.38 5.01 4.18 4.77 4.57

English 2nd SL 5.54 4.06 6.60 6.79 6.57 6.20 5.52

Zulu 3rd SL 5.55 3.81 6.59 6.82 6.61 6.20 5.34

Note: SL = Strongest Language; Scale: 1 = did not contribute at all; 7 = contributed the most

Dominant Language Constellations in the Language Repertoires of Multilingual South…

Dimens ion 2; Eigenvalue: .00003 (.1334% of Inertia)

0.010

153

2D Plot of Row and Column Coordinates; Dimension: 1 x 2 Input Table (Rows x Columns): 7 x 3 Standardization: Row and column profiles FRIENDS-LEARNING

0.008

Zulu 3rd SL

0.006 0.004 0.002 0.000

SCHOOL-LEARNING READ-LEARNING RADIO-LEARNING WATCHTV-LEARNING

Southern Sotho SL

-0.002FAMILY-LEARNING -0.004 English 2nd SL

-0.006 -0.008 -0.010

SELF-LEARNING

-0.012 -0.014 -0.4

-0.3

-0.2

-0.1

0.0

0.1

Dimension 1; Eigenvalue: .01944 (99.87% of Inertia)

0.2

Row.Coords Col.Coords

Fig. 8.1  Correspondence analysis of factors that contributed to the selected participants’ learning of Southern Sotho as a home and strongest language, English as a second strongest language and Zulu as a third strongest language

subject and as a language of learning and teaching in the first 3 years of school in the environment. Figure 8.1 presents the correspondence analysis of the factors and languages presented in Table  8.5. The first finding inferred from Fig.  8.1 is that the factors interaction with friends, interaction with family and self-instruction are not similar in the data set, because they are not presented in close proximation on the plot. The factors school, read, listening to the radio and watching TV are in close proximity on the plot and therefore are similar. When interaction with family and Southern Sotho as a strongest language is connected from the origin, the sharpness of the angle indicates that these elements in the data set are associated; and the length of the lines from the origin indicate a strong association. Similarly, there is a reasonably small angle between interaction with friends and the learning of Zulu as a third strongest language which indicates an association; although the shorter lines from the origin indicate a weaker association. There is an association between English perceived as a second strongest language and self-learning on the plot. The angle between the factors is small and the lines are long which indicate a strong association. There is a large angle and a short line between the education and entertainment factors and the acquisition of English perceived as second strongest language in the repertoires of the participants. In other words, there is a weak association between

154

S. Coetzee-Van Rooy

Table 8.6  Mean scores for selected Southern Sotho home language and strongest language participants and factors that contributed to current use of the languages in the DLC Factors that contributed to the current use of languages in the DLC Friends Family Reading School Watching TV Listening to the radio Self-instruction / tapes

Southern Sotho SL 5.37 6.49 4.67 4.24 4.00 4.50 4.09

English 2nd SL 5.64 4.04 6.40 6.46 6.40 6.24 5.51

Zulu 3rd SL 5.60 3.89 6.33 6.49 6.42 6.21 5.37

Note: SL = Strongest Language; Scale: 1 = did not contribute at all; 7 = contributed the most

these factors in the data set. The correspondence analysis indicates that in this DLC, the family is closely associated with the learning of Southern Sotho as home language and strongest language, English as a second strongest language is associated in some way with self-instruction or learning the language via tapes; and Zulu as a third strongest language is associated with the learning of the language to use with friends. The analysis of the data that indicates the perceptions of the participants towards the contribution of the same factors towards their current use of the languages in their DLC indicates that the same factors that contributed to the learning of the languages for the selected group of participants continue to contribute to the current use of the languages (see Table 8.6 and Fig. 8.2). In Table 8.6, the family remains the factor that is perceived to contribute the most towards the current use of Southern Sotho as home language and strongest language; education and the media remain the most prominent factors that contribute to the current use of English as a second strongest language and Zulu as a third strongest language. The correspondence analysis in Fig. 8.2 presents a visual representation of these findings. The family remains closely associated with the current use of Southern Sotho as home and strongest language (see the long lines from the origin and the sharpness of the angle). English as a second strongest language remains associated in some way with self-instruction or learning the language via tapes (see the long lines from the origin and the sharpness of the angle). Zulu perceived as a third strongest language is associated most strongly with the learning of the language to use in school in the context of the current use of the language in the repertoires of the participants. This finding is different from the findings reported for Fig. 8.1. The functional distribution of languages that are present in the DLC of the selected Southern Sotho participants for this analysis is fairly clear: Southern Sotho as strongest language is associated with the family and could be regarded as the language that marks the cultural identity of the participants. English as second strongest is associated with self-instruction. Zulu as third strongest language is associated with interaction with friends in the context of learning the language; and for use in school in the context of current use of the language.

Dominant Language Constellations in the Language Repertoires of Multilingual South…

155

2D Plot of Row and Column Coordinates; Dimension: 1 x 2 Input Table (Rows x Columns): 7 x 3 Standardization: Row and column profiles

Dimension 2; Eigenvalue: .000 01 (.0662% of Inertia )

0.006

Zulu 3rd SL SCHOOL-CURRENT

0.004

FRIENDS-CURRENT WATCHTV-CURRENT 0.002 FAMILY-CURRENT Southern Sotho SL 0.000

RADIO-CURRENT READ-CURRENT

-0.002 -0.004

English 2nd SL

-0.006 SELF-CURRENT

-0.008 -0.010 -0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

Dimension 1; Eigenvalue: .02069 (99.93% of Inertia)

Row.Coords Col.Coords

Fig. 8.2  Correspondence analysis of factors that contributed to the selected participants’ current use of Southern Sotho as a home and strongest language, English as a second strongest language and Zulu as a third strongest language

4.2  T  he DLC s of the Selected Zulu Home language and Strongest Language Participants When one considers the mean scores for the factors that contributed to the learning of languages in the repertoires of the selected Zulu home and strongest language participants in Table 8.7, it is clear that the family contributed the most to the learning of Zulu; education and the media contributed the most to the learning of English as a second strongest language; and friends contributed the most to the learning of Southern Sotho as third strongest language. These findings are not surprising, as akin to the Southern Sotho home language participants, the strongest language that is also regarded as the home language, is mainly learnt via the family as indicated by the selected Zulu participants for this analysis. For the selected Zulu home language speakers, English remains the language that is important for the school and in the media – also similar to the Southern Sotho home language data. The learning of Southern Sotho as a third strongest language is different for the Zulu home language participants, which is also understandable keeping the language environment in the Vaal Triangle region in mind where Southern Sotho is the most prominent home language.

156

S. Coetzee-Van Rooy

Table 8.7  Mean scores for selected Zulu home language and strongest language participants and factors that contributed to the learning of the languages in the DLC Factors that contributed to the learning of languages in the dominant language constellation Friends Family Reading School Watching TV Listening to the radio Self-instruction / tapes

Zulu as SL 5.51 6.83 5.43 4.38 4.65 4.80 4.66

English 2nd SL 5.71 3.98 6.62 6.87 6.69 6.20 5.54

Southern Sotho 3rd SL 6.08 3.28 2.46 2.62 3.94 3.77 2.87

Note: SL = Strongest Language; Scale: 1 = did not contribute at all; 7 = contributed the most

The correspondence analysis data presented in Fig. 8.3 indicate that for the participants selected for this analysis, the learning of Zulu as a home and strongest language is associated with the family (see the long lines and sharpness of the angle when these row and column variables are connected to the origin). The learning of Southern Sotho as a third strongest language is associated with friends (see the long lines and the sharpness of the angle when these row and column variables are connected to the origin). The learning of English perceived as second strongest language in the repertoires of the participants is associated most strongly with learning English for use in the school (see the long lines and sharpness of the angle when these row and column variables are connected to the origin). The mean scores in Table 8.8 indicate that the family contributes the most to the current use of Zulu as a home and strongest language; education and the media continue to contribute the most towards the current use of English as a second strongest language; and friends contribute to most towards the current use of Southern Sotho as a third strongest language. The correspondence analysis data presented in Fig.  8.4 indicate that for the selected participants, the learning of Zulu as a home and strongest language is associated with the family; and the learning of Southern Sotho as a third strongest language is associated with friends. The learning of English is associated with the education and media sectors and self-instruction. The functional distribution of languages that are present in the DLC of the selected Zulu participants for this analysis is fairly clear (as is evidenced by the long lines and the sharpness of the angles when these row and column variables are connected to the origin): Zulu as strongest language is associated with the family and could be regarded as the language that marks the cultural identity of the participants. English as second strongest is associated with education; and Southern Sotho as third strongest languages is associated with friends.

Dominant Language Constellations in the Language Repertoires of Multilingual South…

2D Plot of Row and Column Coordinates; Dimension: 1 x 2 Input Table (Rows x Columns): 7 x 3 Standardization: Row and column profiles

0.3

Dimens ion 2; Eigenv alue: .01400 (40.62% of Inertia)

157

FAMILY-LEARNING 0.2 Zulu SL READ-LEARNING

0.1

SELF-LEARNING 0.0

-0.1

SCHOOL-LEARNING RADIO-LEARNING English 2nd SL WATCHTV-LEARNING FRIENDS-LEARNING SS 3rd SL

-0.2 -0.3

-0.2

-0.1

0.0

0.1

0.2

Dimension 1; Eigenvalue: .02048 (59.38% of Inertia)

0.3

Row.Coords Col.Coords

Fig. 8.3  Correspondence analysis of factors that contributed to the selected participants’ learning of Zulu as a home and strongest language, English as a second strongest language and Southern Sotho as a third strongest language Table 8.8  Mean scores for selected Zulu home language and strongest language participants and factors that contributed to current use of the languages in the DLC Factors that contributed to the current use of languages in the DLC Friends Family Reading School Watching TV Listening to the radio Self-instruction / tapes

Zulu SL 5.18 6.71 4.56 3.65 4.15 4.30 3.96

English 2nd SL 5.93 4.15 6.59 6.57 6.42 6.08 5.47

Southern Sotho 3rd SL 5.95 3.17 2.79 2.80 3.78 3.93 2.84

Note: SL = Strongest Language; Scale: 1 = did not contribute at all; 7 = contributed the most

5  Discussion of Findings The aims of the chapter were to explore the complementary co-existence of the DLC within the broader language repertoires of multilingual South Africans; the potential of an existing language repertoire survey instrument to provide detailed information about the DLCs of multilingual participants; and to present some

158

S. Coetzee-Van Rooy

2D Plot of Row and Column Coordinates; Dimension: 1 x 2

Dimension 2; Eigenval ue: .01255 (38.64% of Inertia)

Input Table (Rows x Columns): 7 x 3 Standardization: Row and column profiles

0.15

READ-CURRENT 0.10 0.05 0.00 -0.05

FAMILY-CURRENT Zulu SL

SCHOOL-CURRENT SELF-CURRENT English 2nd SL

WATCHTV-CURRENT RADIO-CURRENT

-0.10 -0.15 Southern Sotho 3rd SL FRIENDS-CURRENT

-0.20 -0.25 -0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

Dimension 1; Eigenvalue: .01993 (61.36% of Inertia)

0.4

Row.Coords Col.Coords

Fig. 8.4  Correspondence analysis of factors that contributed to the selected participants’ current use of Zulu as a home and strongest language, English as a second strongest language and Southern Sotho as a third strongest language

empirical data about the complete language repertoires and the specific DLCs of urban Southern Sotho and Zulu home language South Africans. Related to the first aim of the chapter, is the inference that it is clear that it is not productive to work towards distinctions between a complete language repertoire and a specific DLC within the repertoire. The data in this analysis indicate that the deepening of our understanding of the operation of multilingualism in the minds of people will be best served from a perspective where specific DLCs are identified working from a detailed description of complete language repertoires. In my view, the DLC is a specific configuration of languages situated and emergent from within the broader language repertoire. The identification of the DLC cannot be made without a detailed description of the complete multilingual repertoire in my view. If scholars want to study language repertoires and DLCs within those repertoires in a systematic and consistent way across contexts, the current language repertoire survey instrument provides a useful starting point that could be adjusted as and when more types of research questions emerge in the research community. From this analysis and in my view, multilingualism studies should not depart from a language repertoire or a DLC view; but should rather clearly understand the relationships between the two concepts and actively use information about the language

Dominant Language Constellations in the Language Repertoires of Multilingual South…

159

repertoire and the DLC to advance our understanding of contemporary urban multilingualism across the world. Concerning the second aim of the chapter, three findings should be discussed. First of all, the existing language repertoire survey instrument clearly provides empirical data about the complete language repertoires of multilingual people that could enable comparison across diverse contexts. It enables researchers to get information about the constitution of the complete language repertoires (number of languages in the repertoire and which languages are included in the repertoire); and in addition to this information, the differentiation between strongest, second strongest, third strongest, fourth strongest and fifth strongest languages in the repertoires provide a simple and direct way to assist with the determination or identification of the DLC within the broader repertoire. Research questions like which languages comprise the DLC could be answered with more confidence via the use of the language repertoire survey instrument, because the frequencies related to the strongest, second strongest, third strongest, fourth strongest and fifth strongest languages provide an empirical baseline from which arguments could be made about the in- or exclusion of languages from the repertoire in the DLC. Related to this observation, it is exciting to report that from the analysis of selected data from the larger project discussed in this chapter, it is clear that there might be an empirical basis for the claims made by Aronin and colleagues (Aronin and Singleton 2012: 63; Aronin 2016: 151, 156) that a typical DLC includes three languages. If this is true for the complex multilingual repertoires of urban South Africans, as well as for the selected individual European repertoires discussed by Aronin and colleagues (Aronin and Singleton 2012; Aronin 2016), it might contribute significantly to direct studies about multilingualism and cognition and it would surely have vast implications for language in education policies and dispensations in multilingual contexts. Much more data from different contexts should be gathered with the language repertoire survey instrument before these claims could be tested in a more systematic and comprehensive way. Concerning the third aim of the chapter, to contribute information about the DLCs of urban multilingual South Africans, the chapter contributes more information about the nature of the language repertoires reported from the larger project elsewhere by identifying the DLCs for the selected groups of Southern Sotho and Zulu participants that were included in the analysis. The emergence of a DLC that included Southern Sotho, English and Zulu in different configurations for the selected participants forcefully re-directs research attention to theoretical and policy matters. The implications of the finding that there is a DLC that emerged from the data analysis for the selected Southern Sotho and Zulu home language participants in the chapter is important in an education context. The current struggle in the South African broader education context to optimize an effective language in education policy have been documented thoroughly (Antia and Dyers 2016; Beukes 2010; Desai 2016; Kerfoot and Simon-Vandenbergen 2015; Madiba and Mabiletja 2008; Plüddemann 2015; Probyn 2015; Webb 2012; Van der Walt 2013; Hibbert and Van der Walt 2014; Turner and Wildsmith-Cromarty 2014). A deeper understanding of the nature of multilingual repertoires and the specific DLCs in those repertoires

160

S. Coetzee-Van Rooy

could support the formulation of sensible multilingual language policies and implementation plans markedly. In this context, the use of Southern Sotho, Zulu and English present in the DLC of the first year selected participants in this study, seem to be the languages that should be included in multilingual pedagogies aimed at advancing the academic success of these students. At the very least, the findings of this study point to language-in-education approaches that take the three languages in the DLCs into account in higher education. It is beyond the scope of this chapter to tease out all the potential theoretical implications of the identification of the DLC for these participants. One theoretical implication to consider is the power of the DLC concept as a further conceptual extension of the notion of “diglossia”. The concept “diglossia” was coined by Ferguson (1959: 325) to describe the situation where “two varieties of a language exist side by side throughout the community, with each having a definite role to play”. Fishman (1967) was the first to extend the meaning of the concept. Fishman (1967:29) maintains that “diglossia” can also be used to describe The use of several separate codes within a single society (and their stable maintenance rather than the displacement of one by the other over time) was found to be dependent on each code’s serving functions distinct from those considered appropriate for the other.

More recently, Aronin and Singleton (2012: 43) stated: sets of languages, rather than single languages, now often perform the essential functions of communication, cognition and identity for individuals and communities.

Based on the empirical evidence presented in this chapter, I propose that the current expression of the notion that languages can co-exist in the minds of individuals and societies, especially if they perform distinct functions, is expressed best by the notion that “sets of languages”, represented by the DLC within the broader multilingual repertoire, perform functions today.

6  Conclusions The most important conclusion to consider, is the observation proposed in this chapter that multilingualism scholars should approach their work from a language repertoire and a DLC view. Language repertoire survey instruments developed with the specific aim to document all (or as many as possible) of the languages in the repertoires of multilingual people complement the DLC approach in an important way: it provides an instrument that is able to deliver empirical and comparable baseline data for the identification of the DLC in complex multilingual repertoires across different contexts. The observations of the anonymous reviewer about the updating of these instruments to include current factors that are influential in driving the acquisition and learning of languages (for example, social media) should be taken up by the research community. Using updated language repertoire survey instruments could be seen as one way of broadening the use of methodologies to advance

Dominant Language Constellations in the Language Repertoires of Multilingual South…

161

our understanding of the complexity of current urban multilingualism, as expressed by Aronin and Hufeisen (2009: 103) in their review of methodologies that are appropriate for multilingualism research: We argue that multilingualism studies could significantly benefit from the still less widespread, emerging methods introduced in addition to the remarkably broad array of traditional methods of research.

The potential of combining methodologies like language repertoire surveys, with a focus on the DLCs in the repertoires, language portraits of the repertoires and the DLCs, language history interviews and social network data to deepen our understanding of the phenomenon of current multilingualism seem endless. It is the combination of quantitative research data (for example from language repertoire surveys) combined with qualitative research data (for example from language portrait research) which I predict would deepen our understanding of the complexity of the nature of multilingual urban language repertoires and the DLCs in them even more. The second important conclusion is that data from this study of urban multilingual Southern Sotho and Zulu home language students in Gauteng in South Africa seem to support the notion that a DLC often includes three languages, situated in a broader multilingual repertoire. If this finding is replicated in a variety of diverse contexts, this would constitute an important finding in the context of understanding the main characteristics of current urban multilingualism across the world. It would hold important implications for language in education policy and planning; as well as planning for multilingualism and social cohesion in super-diverse contexts. Detailed analyses from more DLCs in the larger study reported on in this chapter should be done to present a more complete understanding of the nature of urban South African multilingualism. The third conclusion, of potential theoretical importance, is that the notion of the DLCs as “sets of languages” co-existing within the multilingual repertoires of speakers perform functions in current contexts. The potential of DLCs for separate domains as indicators of the stability of multilingual repertoires should be explored further. In conclusion, I want to communicate my delight in the results of this chapter that connect the DLC for the first time to empirical data represented with a correspondence analysis. To date, representations of the DLCs of multilingual people were created without a direct link to empirical data. The representations of the DLCs reported as results from a correspondence analysis included in this chapter present the first attempt to render a DLC embedded in empirical data. To me, this is an exciting step to take in the hopefully long journey that multilingualism scholars will continue to make in the discovery and documentation of the plethora of DLCs of urban multilingual people across the world and across their lifespans.

162

S. Coetzee-Van Rooy

Acknowledgements  I want to thank Larissa Aronin for approaching me to consider the relationship between my language repertoire research and the emerging concept of the DLC; and to be part of the broader research community interested in understanding current urban multilingualism. I hope that the data from South Africa could contribute to a small extent to make the marvel of African multilingualism more visible in the research community. I want to thank Bertus van Rooy and Theresa Louw who assisted me to gather the 2015 data. This work is based on research supported in part by the National Research Foundation (NRF) of South Africa (Grant Number 103477). The views stated in the chapter should not be attributed to the NRF.

Appendix  lease indicate how much the following factors contributed P to you LEARNING your STRONGEST language. Indicate your response by putting a tick (√) or cross (X) in the appropriate block below

Factor to consider

Contributed Average Contributed Did not Contributed Contributed contribu- to a large contribute Contributed to a small extent a lot the most tion extent at all very little

Interacting with friends

1

2

3

4

5

6

7

Interacting with family

1

2

3

4

5

6

7

Reading

1

2

3

4

5

6

7

Language tapes/ 1 self-instruction

2

3

4

5

6

7

Watching TV

1

2

3

4

5

6

7

Listening to the radio/music

1

2

3

4

5

6

7

Used at school/ for education

1

2

3

4

5

6

7

Dominant Language Constellations in the Language Repertoires of Multilingual South…

163

 lease indicate how much the following factors contribute P to you CURRENTLY USING your STRONGEST language. Indicate your response by putting a tick (√) or cross (X) in the appropriate block below

Factor to consider

Contributed Average Contributed Did not Contributed Contributed contribu- to a large contribute Contributed to a small extent a lot the most tion extent at all very little

Interacting with friends

1

2

3

4

5

6

7

Interacting with family

1

2

3

4

5

6

7

Reading

1

2

3

4

5

6

7

Language tapes/ self-instruction

1

2

3

4

5

6

7

Watching TV

1

2

3

4

5

6

7

Listening to the radio/music

1

2

3

4

5

6

7

Used at school/ for education

1

2

3

4

5

6

7

Source for questions in Appendix: Marian et al. (2007)

References Antia, B., & Dyers, C. (2016). Epistemological access through lecture materials in multiple modes and language varieties: The role of ideologies and multilingual literacy practices in student evaluations of such materials at a South African University. Language Policy, 15, 525–545. Aronin, L. (2006). DLCs: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L., & Hufeisen, B. (2009). Methods of research in multilingualism studies: Reading a comprehensive perspective. In L. Aronin & B. Hufeisen (Eds.), The exploration of multilingualism: Development of research on L3, multilingualism and multiple language acquisition (pp. 103–120). Amsterdam: John Benjamins. Aronin, L., & Jessner-Schmid, U. (2015). Understanding current multilingualism: What can the butterfly tell us? In U. Jessner-Schmid & C. Kramsch (Eds.), The multilingual challenge: Cross-­ disciplinary perspectives (Vol. 16, pp. 271–292). Boston: Walter de Gruyter GmbH & Co KG. Aronin, L., & Ó Laoire, M. (2013). The material culture of multilingualism: Moving beyond the linguistic landscape. International Journal of Multilingualism, 10(3), 225–235. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Bamgbose, A. (2011). African languages today: The challenge of and prospects for empowerment under globalization. In E. G. Bokamba (Ed.), Selected proceedings of the 40th annual

164

S. Coetzee-Van Rooy

conference on African linguistics (pp. 1–14). Somerville: Cascadilla Proceedings. https://pdfs. semanticscholar.org/9622/ef8b7829c7785930cb141f6a7592eaafba49.pdf. Banda, F. (2009). Critical perspectives on language planning and policy in Africa: Accounting for the notion of multilingualism. Stellenbosch Chapters in Linguistics (SPIL) PLUS, 38, 1–11. Benor, S. B. (2010). Ethnolinguistic repertoire: Shifting the analytic focus in language and ethnicity. Journal of SocioLinguistics, 14(2), 159–183. Beukes, A.-M. (2010). ‘Opening the doors of education’: Language policy at the University of Johannesburg. Language Matters, 41(2), 193–213. Bokamba, E. (2015). Multilingualism as a sociolinguistic phenomenon: Evidence from Africa. In E. C. Zsiga, O. Tlale Boyer, & R. Kramer (Eds.), Languages in Africa: Multilingualism, language policy and education (pp. 21–48). Georgetown: Georgetown University Press. Botha, W. (2017). The use of English in the social network of a student in South China. English Today, In Press. Busch, B. (2010). School language profiles: Valorizing linguistic resources in heteroglossic situations in South Africa. Language and Education, 24(4), 283–294. Canagarajah, S., & Ashraf, H. (2013). Multilingualism and education in South Asia: Resolving policy/practice dilemmas. Annual Review of Applied Linguistics, 33, 258–285. Cenoz, J., & Gorter, D. (2011). A holistic approach to multilingual education: Introduction. The Modern Language Journal, 95(3), 339–343. Coetzee-Van Rooy, S. (2012). Flourishing functional multilingualism: Evidence from language repertoires in the Vaal triangle region. International Journal of the Sociology of Language, 218, 87–119. Coetzee-Van Rooy, S. (2013). Afrikaans in contact with English: Endangered language or case of exceptional bilingualism? International Journal of the Sociology of Language, 224, 179–207. Coetzee-Van Rooy, S. (2016). The language repertoire of a Venda home language speaker: Reflections on methodology. Language Matters, 47(2), 269–296. Desai, Z. (2016). Learning through the medium of English in multilingual South Africa: Enabling or disabling learners from low income contexts? Comparative Education, 52(3), 343–358. Dewaele, J.-M. (2016). Multi-competence and emotion. In L.  Wei & V.  Cook (Eds.), The Cambridge handbook of linguistic multi-competence (pp. 461–477). Cambridge: Cambridge University Press. Duchêne, A., & Heller, M. (Eds.). (2012). Language in late capitalism: Pride and profit (Vol. 1). New York: Routledge. Edwards, J. (2002). Multilingualism. London: Routledge. Ferguson, C. (1959). Diglossia. WORD, 15(2), 325–340. https://doi.org/10.1080/00437956.195 9.11659702. Fishman, J. A. (1965). Who speaks what language to whom and when? La linguistique, 1(Fasc. 2), 67–88. Fishman, J. A. (1967). Bilingualism with and without diglossia; diglossia with and without bilingualism. Journal of Social Issues, 23(2), 29–38. Herdina, P., & Jessner, U. (2002). A dynamic model of multilingualism: Perspectives of change in psycholinguistics (Vol. 121). Clevedon: Multilingual Matters. Hibbert, L., & Van der Walt, C. (Eds.). (2014). Multilingual universities in South Africa: Reflecting society in higher education. Bristol: Multilingual Matters. Hill, T., & Lewicki, P. (2006). Statistics: Methods and applications: A comprehensive reference for science, industry, and data mining. Tulsa: StatSoft, Inc. Hornberger, N. H. (2009). Multilingual education policy and practice: Ten certainties (grounded in indigenous experience). Language Teaching, 42(2), 197–211. Jessner, U. (1999). Metalinguistic awareness in Multilinguals: Cognitive aspects of third language learning. Language Awareness, 8(3–4), 201–209. Jessner, U. (2008). A DST model of multilingualism and the role of metalinguistic awareness. The Modern Language Journal, 92(ii), 270–283.

Dominant Language Constellations in the Language Repertoires of Multilingual South…

165

Kerfoot, C., & Simon-Vandenbergen, A.-M. (2015). Language in epistemic access: Mobilising multilingualism and literacy development for more equitable education in South Africa. Language and Education, 29(3), 177–185. Kramsch, C. (2004). The multilingual experience: Insights from language memoirs. Transit, 1(1). https://escholarship.org/uc/item/9h79g172. Laitin, D. D. (2007). Language repertoires and state construction in Africa. Cambridge: Cambridge University Press. Lo Bianco, J. (2010). The importance of language policies and multilingualism for cultural diversity. International Social Science Journal, 61(199), 37–67. Madiba, M., & Mabiletja, M. (2008). An evaluation of the implementation of the new Language-­ in-­Education Policy (LiEP) in selected secondary schools of Limpopo Province. Language Matters, 39(2), 204–229. Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The language experience and proficiency questionnaire (LEAP-Q): Assessing language profiles in bilinguals and multilinguals. Journal of Speech, Language and Hearing Research, 50(4), 940–967. Norton, B. (2013). Identity and language learning: Extending the conversation. Bristol: Multilingual matters. Oane, A., & Glanz, C. (2010). Why and how Africa should invest in African languages and multilingual education: An evidence- and practice-based policy advocacy brief. Hamburg: UNESCO Institute for Lifelong Learning. Pavlenko, A. (2007). Emotions and multilingualism. Cambridge: Cambridge University Press. Pavlenko, A. (Ed.). (2011). Thinking and speaking in two languages (Vol. 77). Bristol: Multilingual matters. Plüddemann, P. (2015). Unlocking the grid: Language-in-education policy realisation in post-­ apartheid South Africa. Language and Education, 29(3), 186–199. Prah, K. K. (2010). Multilingualism in urban Africa: Bane or blessing. Journal of Multicultural Discourses, 5(2), 169–182. Probyn, M. (2015). Pedagogical translanguaging: Bridging discourses in south African science classrooms. Language and Education, 29(3), 218–234. Sharma, D. (2011). Style repertoire and social change in British Asian English. Journal of SocioLinguistics, 15(4), 464–492. South African Census. (2011). Census in brief. http://www.statssa.gov.za/census/census_2011/ census_products/Census_2011_Census_in_brief.pdf. Accessed on 2 June 2016. Todeva, E., & Cenoz, J. (Eds.). (2009). The multiple realities of multilingualism: Personal narratives and researchers’ perspectives (Vol. 3). Berlin: Walter de Gruyter. Turner, N., & Wildsmith-Cromarty, R. (2014). Challenges to the implementation of bilingual/ multilingual language policies at tertiary institutions in South Africa (1995–2012). Language Matters, 45(3), 295–312. Van der Walt, C. (2013). Multilingual higher education: Beyond English medium orientations. Bristol: Multilingual Matters. Van Zyl, A. J. M. (2016). The native-speaker debate: The case of the Afrikaans-English teacher’s identity in Thailand. Potchefstroom: Potchefstroom University. unpublished MA dissertation. Webb, V. (2012). Managing multilingualism in higher education in post-1994 South Africa. Language Matters, 43(2), 202–220.

Part III

Personal and Group Experiences with DLC

The Evolution of Personal Dominant Language Constellations Based on the Amount of Usage of the Languages Sarasi Kannangara

Abstract  The central characteristic of a Dominant Language Constellation (DLC) is its particularity and specificity regardless of whether it exists in an individual or within a community. The uniqueness of a DLC may vary depending on its linguistic, societal, political, or geographical context. These variations can be observed in interrelated factors, such as the prominence of the language, its level of proficiency, and the functions allocated to it. A detailed analysis of my individual Dominant Language Constellations during important life events offers a better understanding of the interdependent relationship between prominent outside factors -both geographical and societal- and the development and evolution of my unique DLC. For this study, eight different graphs are developed to represent the divergence of my personal DLC using important life events as benchmarks. The personal DLC consist the most important languages in my day to day communication based on their amount of usage. The graphs also investigate my language repertoire while closely looking at the languages that are generally considered insignificant in their usage within the DLC. The study will not only provide an insight into my non-traditional childhood and adult multilingualism but also shed light on multilingualism in Sri Lankan Society. It closely examines alterations in my DLC provoked by changing schools, making new acquaintances, and geographical changes such as studying in another Country. Furthermore, it reveals that despite other changes within my DLC over time, the prominent role of English remains consistent. This could be explained not simply by the utilization of English as the lingua franca but also in relation to my attitude towards English as an elite language in Sri Lanka. The analysis also manifests the insignificance of the Tamil language in my personal DLC, despite its state recognition as an official language alongside Sinhala in Sri Lanka. In the light of this information obtained through the analysis of my personal DLCs, it would be interesting to focus moreover on how an individual DLC exist within a DLC or DLCs of multilingual, multicultural and multi-ethnic societies such as Sri Lanka.

S. Kannangara (*) Technical University of Darmstadt, Darmstadt, Germany © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_9

169

170

S. Kannangara

Keywords  DLC · Individual DLC · Multilingualism in Sri Lanka

1  Introduction Simply explained, a Dominant Language Constellation (DLC) is an entity of different languages that are used by an individual or shared by a community on a day to day basis. “For any individual, the Dominant Language Constellation is the group of his/her most important languages that, functioning as an entire unit, enables him/ her to act in a multilingual environment in such a way as to meet all of his/her needs” (Aronin and Singleton 2012, p. 59). The central characteristic of a DLC is its particularity and specificity, regardless of whether it is the DLC of an individual or a community. Even two DLCs consisting of the same languages can differ significantly from each other depending on its linguistic, societal, political, or geographical significance for the individual concerned or the social context in which the languages are used. According to Aronin and Singleton (2012), most of the differences lie in the prominence of the language, the level of proficiency the speaker has in that language and on the function allocated to each language. (pp. 64–65). A closer examination of the different factors that create the uniqueness of a Dominant Language Constellation reveals that they are interconnected. For example, one’s place of residence, which is the geographical context, is connected to the functions allocated to a certain language and the amount of functions allocated to a particular language influence the amount of usage and eventually the proficiency in that language. On these grounds, analysing an individual Dominant Constellation may illuminate the relationship between the outside factors and the constitution of a unique DLC that presents an individual with a thorough understanding of his or her own linguistic identity. Furthermore, this paper discusses the multilingual status of Sri Lanka for the purpose of recognizing the political and societal influences on a personal DLC.

2  Multilingualism in Sri Lanka The personal DLC of an individual is inevitably linked with the DLC of his or her language community. Therefore, it is essential to carefully examine the current multilingualism in Sri Lanka and its development through the years. The state of multilingualism is quite complex in Sri Lankan society and has had a very turbulent past. Throughout the history, biased language policies contributed immensely to the decades-long civil war and the prevailing hostility between the two major ethnicities of the country: the majority Sinhalese and the minority Tamils: Language is at the root of the ethnic conflict in Sri Lanka, the South Asian island that has come close to being torn asunder in almost two decades of ferocious fighting between majority Sinhalese and minority Tamils, costing at least 50,000 lives by conservative estimates (Canagarajah 2005, p. 420).

The Evolution of Personal Dominant Language Constellations Based on the Amount…

171

2.1  National Language Policy Sri Lanka has experienced many external language influences throughout its long history. Its geographic proximity to India has shaped its linguistic development through successive invasions from India, compounded by three distinct phases of European colonial rule, Portuguese, Dutch, and, most significantly and recently, British. But, in the current paper, I would like to focus more on the period after British colonization, which has shaped Sri Lankan multilingualism to its current status. Until 1956, 8 years after independence from British rule, English was the official language of Sri Lanka (then called Ceylon), when the controversial ‘Sinhala Only’ Act took effect. This was legitimately passed by a Sinhala-dominated parliament, which forced a drastic change upon the bilingual community of Sri Lanka, vastly discriminating against the minority of Tamil speaking nationals. “The local regime has not been slow to legislate on language. Language is a sensitive issue for the community since it was the ‘Sinhala Only Act’ of 1956, and language-based discrimination in education and employment thereafter, that spawned the present conflict.” (Canagarajah 1995, p. 188). Sinhala remained the sole officially designated language of Sri Lanka for almost three decades despite the inappropriateness of a single language being the only state-approved form of communication in a multilingual community. However, Act No 28 of 1958, assigned Tamil as a language of administration in the North and Eastern Provinces. The Sri Lankan government then tried to treat the language conflict with another temporary solution by recognizing Tamil as a national language, while maintaining the status of Sinhala as the only ‘official’ language according to Chapter IV of the Sri Lankan Constitution in 1978. (Ranjit and Weerasekara 2013, pp.  74–75) The most significant modification to the language policy, namely the 13th amendment took place in 1987, which recognized Sinhala and Tamil both as official languages and the 16th amendment establishing an equal position for both Languages in administration and legislation: According to the 16th amendment passed in 1988 that the Sinhalese and Tamil are the languages of administration of Sri Lanka, and Tamil shall be used in Northern and Eastern provinces. All laws and subordinate legislation shall be enacted or made and published in Sinhala and Tamil together with a translation of in English (Ranjit and Weerasekara 2013, p. 75).

The 13th amendment also declares English language as the link language. No further explanation is given either to the definition “link language” and or to its role as a link language. Should it be a link between the different ethnic groups of Sri Lanka or a link between Sri Lanka and the international community is unclear. (Fernando 2017) However, the official website of the department of official languages publishes the current National language policy as follows. The fundamental law pertaining to the languages in Sri Lanka is enshrined in Chapter IV of the Constitution of the Democratic Socialist Republic of Sri Lanka (1978). The provisions thereof have been amended by the 13th and 16th Amendments to the Constitution respectively in 1987 and 1988. Moreover, Article 12 (2) of Chapter III of the Constitution

172

S. Kannangara

r­ ecognizes the right to language as a fundamental right. The Sinhala and Tamil languages are both Official and national languages in Sri Lanka while English is the Link Language. Apart from the statutory provisions for the implementation of the Official Languages Policy, administrative provisions have been made in implementing the same1

Since the end of the ethnic conflict, several alterations were added to the Sri Lankan linguistic policies in rectifying the past errors of language inequality. For example, according to the public administration circular 07/2007, public servants should be able to carry out their functions and duties both in Sinhala and Tamil. According to this circular, every public servant should obtain their language proficiency within 5 years. Likewise, the ministry of education has increased the number of periods to teach the Tamil language in schools (Ranjit and Weerasekara 2013, p. 75). The actuality of the multilingual situation in Sri Lanka doesn’t always correspond with the positive changes happening in the country’s political landscape. There are unfortunate shortcomings in implementing the above-mentioned changes in reality. For example, according to a study conducted by Ranjit and Weerasekara 2013, in the Balangoda division of the Ratnapura district, home to a high density of the Tamil population, 81% of the subjects among clients and 60% of the subjects among public service officers reported of the inability to consecutively attain and provide services in the Tamil language. (Ranjit and Weerasekara 2013, p.  76). Likewise, in listing recommendations for better implementation of the official language policy, the Official Language Commission highlights the fact that only 8.23% of public officials can speak Tamil in a country, where 26% of the population is Tamil-speaking (Perera 2015, p. 62). In the national political arena, both the Sinhala-Buddhist movement and Tamil nationalism have played a hindering role in achieving an ideal multilingual society in the country. For instance, the main theme of the Sri Lankan election in 1956 was driven by linguistic nationalism which resulted in passing the Sinhala Only Act as mentioned earlier. Even today, there are a few political parties that follow the idealism of a Sinhala speaking monolingual country (De Votta 2007, pp.  17–18). Canagarajah claims that it was the emergent militant nationalism that unleashed a Tamil-only or ‘pure Tamil’ movement among the Tamil speaking society in northern Sri Lanka. Such developments have not only caused a decline in the usage of the English language but also created a shift from bilingualism to monolingualism (Canagarjah 1993, pp. 188–189, 1995, p. 604). It was also understood that speaking or learning Sinhala, the language of the foe was not encouraged in the North-Eastern Tamil communities during the time of the military dispute. It is evident that the status of multilingualism in Sri Lanka has been a highly contested subject and associated with conflict over the decades. Government implementation of formal language policies has been heavily deficient and while the country remains multilingual in the three main languages of its population,

1   Department of Official Languages Sri Lanka: http://www.languagesdept.gov.lk/web/index. php?option=com_content&view=article&id=38&Itemid=163&lang=en retrieved 30.04.2019.

The Evolution of Personal Dominant Language Constellations Based on the Amount…

173

English, Sinhala and Tamil, there are sharp demographic divisions across Sri Lankan society and a record of poorly implemented language policy.

2.2  Status of English Close examination of the status of English in Sri Lanka is also crucial when understanding multilingualism in Sri Lanka. English has been and, most certainly, is still considered a prestige language among Sri Lankans. Ever since British colonialization, English has functioned as a valued linguistic capital over the local Sinhala and Tamil languages to provide socioeconomic advantages. Although leftist governments have sworn to raise the status of Sinhala and Tamil over English, it is the English-speaking bilinguals who dominate the professional and social hierarchies of the island, even today. On the other hand, allowing the lower-middle-class access to English as promised by ‘rightist’ governments have failed utterly since their newly-acquired English is marked as a ‘non-prestige’ sub-standard Sri Lankan English (Canagarajah 1995, p.  604). Even after almost eight decades of political independence from the British monarchy, the English speaking elite (this considers both English monolinguals, and Sinhala-English or Tamil-English speaking simultaneous bilinguals who have learnt English as an L1  in a familial environment) enjoy social prestige and economic benefits by having access to more lucrative occupations over monolingual Sinhala or Tamil speakers, but also not excluding the successive bilinguals or in other words late learners of English. “Knowledge of English is uplifting. Why? Because in our society English has certain values attached to it. Hence whether we like it or not, knowing English is prestigious. Whether correct or otherwise, persons who speak English are considered in our society to be more rational, intelligent and knowledgeable than those who don’t “(Cooray 2010). Later in this paper it will be demonstrated that the high social standing of the English language is clearly reflected in the composition of my personal DLC. However, it shouldn’t be discounted that in post-independence the swabhasha movement disregarded English in an attempt to restore the local vernaculars (specially Sinhala) to their previous position. Yet, even after English was removed from its official functions and status it stayed connected to economy, internationalisation and foreign aid (Lo Bianco 2011, pp. 36–37). At present apart from its social prestige English is associated with its economical significance: Today, however, in all these settings, English is totally transfonned. No longer excoriated, its reserve strength in the private sector of the economy (especially the labour market) reinforced by market internationalisation and the modern culture of consumerism, has overwhelmed state, nationalist, and populist campaigns calling for its rejection. English has been transformed into an object of desire and admiration (Lo Bianco 2011, p. 38)

174

S. Kannangara

3  Multilingualism and Education in Sri Lanka The constitution of my personal DLC in the early years is highly influenced by the education system in Sri Lanka. And, for this reason, this paper would be incomplete without a political and sociolinguistic account of the Sri Lankan education system. Before the introduction of colonial European schooling, education in Sri Lanka was reserved mainly for the sons of noblemen and Buddhist monks, leaving the majority of the population illiterate. Even though Sri Lanka’s education system underwent various changes under the colonial administration of Portuguese and Dutch, it would be adequate to say that schooling in Sri Lanka began to take form and shape under the British colonial era. During this period two types of schools were emerged: the higher status schools, in which teaching was done in English medium and provided education from age 5 to 18 and lower status schools or primary schools that imparted knowledge in the local vernaculars for free. Those who received education in the vernacular languages were discriminated against and looked down upon by English-educated Sri Lankans. It was this form of class-bias education prevalent under the British that prompted the bill for free education (introduced to the State council in 1943 and implemented in 1945) in Sri Lanka. A component of the free education move was the utilisation of S and T as languages of instruction with a view to reducing the socio-economic inequalities in the country. As a result, free education in native languages was introduced to primary schools in 1945, secondary schools in 1953 and at the universities in 1960 (Punchi 2001, pp. 366–368). Sri Lankan education is currently marked by parallel monolingualism. In state schools, Sinhala and Tamil, as the two national languages, are used as exclusive mediums of instruction, also resulting in parallel school systems (Nanayakkara 2013, p. 2). In state schools, two national languages, namely Sinhala and Tamil, are used as the medium of instruction. Therefore, there are only two types of schools based on it; Sinhala-medium schools and Tamil-medium schools. Primary education should only be provided either in Sinhala or Tamil (Nanayakkara 2013, p. 2). Within both of these systems English is a compulsory second languages from primary to secondary levels and is offered as the key professional medium at tertiary educational institutes such as technical colleges and universities. Additionally, the two national languages are taught to students from the other community, Sinhala to Tamils and Tamil to Sinhalese, however the time devoted to teaching the second national language is far inferior to the time devoted to English and in general these programs are considered insubstantial (Punchi 2001, pp.  375–376). After more than a decade these deficiencies still prevail in language education as Premarathna et al. (2016) reveal in their study on medium of instruction in general education of Sri Lanka. They report that unfortunately, much attention is focused on developing English education, but Sinhala and Tamil are almost completely neglected. They point out that the education system recognizes a person called ‘English teacher.’ He or she is given a training to execute his or her duties. But a person such as ‘Sinhala (or Tamil) teacher has not been recognized. As a result, Sinhala or Tamil as a second language teacher is deprived of obtaining a systematic training in language teaching

The Evolution of Personal Dominant Language Constellations Based on the Amount…

175

(Premarathne et al. 2016, p. 17). Nevertheless, English language education in Sri Lanka also faces similar challenges. For an example most of the rural schools lack human resources. Some schools are not provided with an English teacher and in others, where a teacher is available, he or she does not have necessary training to perform teaching duties (Premarathne et al. 2016, p. 17). To overcome shortages in English education the education system promotes various programs to uplift working knowledge of English among students. As a direct result of this process bilingual programs were implemented in some government schools (Premarathne et al. 2016, p. 8). In keeping up with these developments and in response to emerging globalisation the government had to reconsider its medium of instruction. In 2000, the state approved the commencement of English medium classes in government schools. This initiative was highly criticized for its lack of planning and preparation (Balakrishnar and Thanaraj 2015, pp.  170–171). In addition, bilingual education was introduced in 2002, where select subjects were taught in the English language from the sixth standard. Ignoring its bilingual features, Sri Lankan society labelled it as ‘English-medium’ education and criticized it for being a threat to the local vernaculars, as in the time of post-colonial education. An opposing group of individuals also demanded that all subjects be taught in English (Nanayakkara 2013, p. 1). As stated above, the education system in Sri Lanka has gone through many significant changes over the years. Unfortunately, none of these changes have resolved the challenge of devising a suitable language education policy to support the average Sri Lankan. Even with the availability of education in all three languages, most Sri Lankans will nevertheless find one language in their personal DLC.

4  Methodology The discussion above provided a brief account of the contested and turbulent history of language education and social multilingualism in the Sri Lankan context. It is in this setting that the author’s personal DLC has been constructed, reflecting key aspects of the wider situation of language in Sri Lankan society, while still being inflected by or responsive to personal life patterns and directions. The main objective of this study is the exploration of a personal DLC. In this case, the personal DLC of the author in order to understand its uniqueness. Therefore, it is on par with the objectives of qualitative social research. Flick (2014) states that one of the main objectives of qualitative social research is to describe a phenomenon in some or greater detail, which can be subjective experiences of a specific individual or a group. (Flick 2014, pp. 5–6). This approach puts the subjective experiences in focus. Section 4 of this paper aims to achieve this by creating graphs of the author’s personal DLCs during different societal and geographical changes. The pie chart in the graphic represents the Dominant Language Constellation, meaning the most important languages in my language repertoire and their percentages of use. Percentages are based on the amount of usage of different languages and were assigned according to the activities that are carried on in that particular

176

S. Kannangara

language. It shouldn’t be misinterpreted as quantitative data but as an instrument to represent how much a language is used in the following domains based on my personal perception. 1 . Personal interaction: communication with family and friends. 2. Professional domain: communication and use of media in education and at work. 3. Media and Entertainment: Usage of TV, newspapers, internet, novels, movie, music, social media, etc. 4. Other: Interactions at supermarkets, doctors, religious activities, etc. The languages which are situated on the orbit around the DLC are those, which are in use, but are not significant enough to be on the DLC, therefore, the languages in one’s language repertoire. A language repertoire relates to the totality of an individual’s linguistic skills (Aronin 2019, p. 21). Their position, depending on whether they are placed closer to or farther from the DLC, represents the amount of their usage and, consequently, their significance. The languages, which are placed outside the orbit, exist in one’s Meta Linguistic Awareness; but active usage of those languages, even on a receptive level, is minimal or non-existent. The Analysis consists of eight graphs starting from birth to the current day. Furthermore, each graph is provided with an extensive description to delineate the personal, social, and geographical circumstances in which each DLC is formed. Thus, this study aims at the expansion of data by producing explanations and interpretations to the formation of my personal DLC rather than reducing data or its complexity (Flick 2014, p. 11).

4.1  Analyzing the Personal DLC I was born to middle-class parents, who were bilingual in a very non-traditional sense. The traditional view of bilingualism expected bilinguals to have a native-like command in two languages (Bloomfield 1933,  pp. 55–56), but currently, most researchers believe in a broader view what constitutes bilingualism without the imposition of an equal proficiency in both languages and increasingly without the requirement that speakers should have native-like features to their language use (Butler 2013, p. 111). They are both native speakers of Sinhala, one of the official languages in Sri Lanka. None of them possessed native speaker like abilities in English, even though they both encounter English in their day to day life. Fernando (1977) classifies Sinhala – English bilinguals into three groups. Group one is the group of bilinguals, which is characterized by their highly anglicized lifestyles. They are at the top and middle of the social scale and speak standard Sri Lankan English. They differ from the group Two depending on their pronunciation and the degree to which they use English in domestic or social intercourse. Group two is generally of a lower-middle or working-class origin. They regard English very much like a foreign language. They have mostly learnt English later in life and do not use it in their personal domains. Group three, according to Fernando, is a group of receiver bilinguals whose fluency in English is very limited. They come

The Evolution of Personal Dominant Language Constellations Based on the Amount…

177

from the same social background as those of Group two. They learn English only because it is compulsory to do so in all Sri Lankan schools, and their use of English is limited to the classroom (Fernando 1977, pp. 348–358). I must say that my parents nor I can be assigned to any of the groups mentioned above. I wouldn’t call myself a member of the upper class highly anglicized English-­ speaking elite, but I do use English in my day to day social as well as professional life. And I differ as well in relation to the mode of acquisition of English: I have learnt English through familial and social interaction, formal education, and through the input received by reading books, watching movies, etc. Moreover, I do not possess the typical characteristics of a childhood simultaneous bilingual, although I was exposed to both Sinhala and English from birth. Fernando’s classification is broadly informative but is also incapable of capturing the full complexity of bilingual situations., reducing rich continuums and varieties to reductive categories. In my specific case the classification would produce an overgeneralized stereotypical uniformity, and more generally, Fernando’s schema simplifies the complexed and unique characteristics of bilingualism itself. Until I was 4 years of age, we were residing in Colombo, the national capital, which increased the chances of encountering English more than anywhere else in Sri Lanka. Having done his tertiary education in English medium and being the son of an English teacher, my father was more fluent in English than my mother, who’s lexical and grammatical knowledge matched the level of an intermediate level (B1) English learner, but social stigma and making mistakes in spoken English prevented her from actually utilizing it. I should also mention that Sri Lanka has its own variety of English, which is called either Sri Lankan English or Ceylonese English (Kachru’s circles of English 1985). Another term that is commonly found in researching Sri Lankan English is “standard Sri Lankan English”. This describes the variety of English used by Sri Lankans who speak English as their first language, or who are bilingual in English and Sinhala or English and Tamil. Although Meyler (2015) claims that standard Sri Lankan English has not yet been adequately codified, the term itself remains more precise than ‘Sri Lankan English’, which is more of an umbrella term covering all the different ways in which the language is used in Sri Lanka (Meyler 2015, p. 181). My parents spoke another variety, which is popularly called “Singlish,” which, incidentally, is also the name of the popular form of English in Singapore, an abbreviation for Singapore English. This art of communication is criticised by the many who fight for linguistic purism of the Sinhala language. Singlish is characterized by adding English lexical to Sinhala sentence structures. English Drink some water

Singlish Water bonna (drink in Sinhala)

Another characteristic of transfer among many bilinguals in Sri Lanka is using English verbs in isolation, but also, mostly, by repeating it twice.

178

S. Kannangara

English come here/come over take a seat

Sri Lankan English Come come Sit sit

According to the classification of bilingual families by Romain (1995), I grew up in a type 6 mixed languages situation, where the parents were bilingual, and code-­ switching was prominent (p. 185). We had a domestic helper, whose native tongue was Tamil; however, I have no recollection of actively or passively using Tamil. For this reason, 90% of my DLC occupied by Sinhala as the main language of communication at home, and 10% is allocated to English due to usage of English in code-­ switching, and Tamil stays even outside of the language repertoire (Fig. 1). I was sent to a kindergarten run by Catholic nuns,2 where the classroom language was English and also to an English elocution class, which was conducted by a L1 Speaker of English is more appropriate. Nevertheless, I remember communicating with my classmates mostly in Sinhala, the language I was more comfortable with, and only spoke to my teachers in English since it was the rule. At home, communication occurred in Sinhala. I must mention, though, that cartoons and children’s programs on television, like Sesame Street, were broadcasted in their original language and also an occasional fairy tale read by my grandmother in English, mostly on school holidays (Fig. 2).

Tamil 10%

Sinhala English

90%

Fig. 1  From birth to age four

2  Even children of Buddhist parents were sent to catholic kindergartens and schools mostly with the intention of exposing their children to the English language and, thereby, creating better opportunities for them.

The Evolution of Personal Dominant Language Constellations Based on the Amount…

179

Tamil

30% Sinhala English 70%

Fig. 2  Kindergarten and preschool

I was later sent to a state school outside of Colombo for my primary education. I attended a State Sinhala medium school, where the language of instruction was exclusively Sinhala. L2 lessons in English took place once a day, and L2 lessons in Tamil took place once a week. This resulted in the curtailment of the usage of the English language by 10% in my DLC. The placement of Tamil language on the orbit can be explained by the formal education provided by school. However, my DLC does not consist of Tamil, since it was only used in a formal classroom environment and never in day to day communication. Neither the school administration nor my parents gave Tamil language education enough importance, as I remember that the school’s inability to provide ongoing Tamil language teaching was not regarded a serious issue. Religious chanting and prayers in Sanskrit and Pali were learned by heart, with their meanings explained in Sinhala, in school and at Sunday School, which I visited for a very short period. These chants were also included in daily prayer. This is how Sanskrit and Pali became a part of my linguistic awareness (Fig. 3). My language repertoire experienced a prominent shift during the time of my secondary and upper school education due to both formal education and societal change. From the ninth standard onwards, I started taking up English literature as a school subject, which required a lot more reading and writing in the English language. During my university entrance examination (A/Ls), I added three further languages to my repertoire, namely German, French and Japanese, which were taught at school and at private institutions or by private teachers (in very traditional grammar-oriented classroom situations). I started leaning all three of them in the same period as subjects for my A/L examination. None of these languages were used for authentic communication; therefore, they did not belong to my DLC. As a result of the inferior status of Tamil throughout my schooling, I did not acquire a

180

S. Kannangara

Sanskrit and Pali

20% Sinhala

Tamil

English

80%

Fig. 3  Primary school

sufficient communicative competency in the language. Therefore, it moves its way out of the orbit and finds its place outside of my language repertoire. The societal shift in my life occurred with the changing of schools. I started attending a popular and privileged state school in Colombo, which resulted in having bilingual classmates and friends, who were more fluent in their English than I. The language of communication among friends in school was bilingual in English and Sinhala, but the dominant form of communication was English. I needed to improve my knowledge of English in order to belong to this new social group. The increasing interest in English music, books, TV series, and movies during this period occurred partly because of the interaction with this social group. This explains the growth of usage of the English language. Nevertheless, Sinhala remains the most dominant language due to its reinforcement at home. At this time, Hindi entered my language repertoire. The inclusion of Hindi arose thorough watching Indian movies with Sinhala subtitles. After doing this for about a year or so, I could understand 90% of the conversation taking place without subtitles due to the linguistic closeness between Hindi and Sinhala, and could communicate with native speakers while holidaying in India (Fig. 4). My Dominant Language Constellation took a new shape during my primary university education at a State University in Sri Lanka. English and Sinhala still share dominant roles as the languages of everyday communication, but Sinhala loses its usage by 10% because all subjects chosen by me at university are taught in English Medium. German marks its place in the DLC because I was chosen to specialize in the German language, which consisted of at least 20  h of formal education in German at the university. Apart from that, I continued learning German at the Goethe Institute in Colombo. It is worth mentioning here that the German education I received both at the university and the Goethe Institute of Colombo was aimed at communicative competence in the language. Therefore, the classes were conducted

The Evolution of Personal Dominant Language Constellations Based on the Amount…

181

Sanskrit and Pali Hindi Tamil

Japanese 40%

Sinhala

French

English

60%

German

Fig. 4  Secondary and upper school

Sanskrit and Pali

20% 40%

French Japanese

40%

Hindi

Sinhala English

Tamil

German

Fig. 5  University education, Bachelor’s

predominantly in German. During this time, I also was exposed to German art, popular culture (football, music, movies, etc.) and literature, which allowed me to spend more time with the language. A scholarship awarded by the Goethe Institute for 2 months of German language classes at the Goethe Institute in Bonn gave me the opportunity to use the language in everyday communicative situations. I took courses at the University to refresh my French, Japanese and Tamil, so they still hold their position in the orbit closer to the DLC, and I kept on improving my Hindi knowledge inductively as I continued watching Hindi movies (Fig. 5). As mentioned earlier, a geographical shift can bring significant and palpable changes in the Dominant Language Constellation. Two and half years of secondary university education in Kassel, Germany, inevitably made significant changes in my

182

S. Kannangara

Sanskrit and Pali

5% 20%

French

Sinhala English

Japanese Tamil

Hindi

German 75%

Fig. 6  University education, Masters in Germany

DLC. As the graphic indicates, the usage of my native tongue has decreased to 5% due to its restriction by being exclusively associated with communication with my family. English still occupies 20%, since it is used for educational purposes as well as to communicate with non-German speaking international students, with friends from Sri Lanka and also not excluding interactions on social media, reading, movies, TV series etc. 75% of language use is in German due to education, day to day interactions and exposure to media. By now, Japanese, French, and Tamil are positioned outside the orbit since they have become redundant due to lack of usage and maintenance required to keep their knowledge active (Fig. 6). After completing my master’s degree in Teaching German as a Foreign/Second Language, I returned to Sri Lanka, which again provoked major changes in my DLC. English becomes the most dominant language since communication at work and among friends takes place predominantly in English. Most of my reading is also done in English, regarding academic material or otherwise. Online usage consists mostly of English, including social media. Sinhala remains the communicative language within the family and also in conversation with monolinguals in various day to day situations. For example, conversations with the non-academic staff or when shopping etc. However, in both cases, code-switching inevitably occurs, depending on your communicative partner. German remains in my DLC as it is used as the teaching and communicative language in class, and when interacting with native German-speaking colleagues At the Goethe Institute, where all colleagues are multilingual, code-switching happens in all three languages. However, a key characteristic here is that most of the technical jargon related to teaching methodology is always spoken in German (Fig. 7). Currently, I am reading for my PhD at the Technical University of Darmstadt and residing in Darmstadt as well. Now, the German language takes prominence over

The Evolution of Personal Dominant Language Constellations Based on the Amount…

183

Sanskrit and Pali

French

25%

Hindi

35%

Japanese

Sinhala English German

Tamil 40%

Fig. 7  Life as a German teacher in Sri Lanka

Sanskrit and Pali

Hindi

5% 25%

French Japanese Tamil

Sinhala English

70%

German

Fig. 8  Current dynamic language constellation

English and my native tongue. It occupies a large part of all four domains such as personal interaction, professional domain, media and entertainment and as well as other day-to-day interactions in supermarket or at the doctors, etc. Similar to the previous time in Germany Sinhala is used to communicate with my family and friends residing in Sri Lanka through audio and video communication methods. The only difference between the DLCs in the period of my master’s and the current period is that the usage of English has increased by 5% due to the increased amount of academic reading done in the English language during this period (Fig. 8).

184

S. Kannangara

5  Conclusions Analysis of the above graphics reveals that my personal Dominant Language Constellation has experienced striking changes over varying stages of my life. Despite its consistence in recent times comprising Sinhala, English, and German, various functions allocated to these languages and consequently, the amount of their perceived usage has varied according to personal, societal, and geographical shifts. The analysis also confirms that new languages can be added to the personal DLC and some languages might gain or lose their significance “weight” during shifts. Most importantly, this demonstrates the trait of DLC for being dynamic and transient. As Aronin describes, DLC does not represent a definite rigid pattern of organisation. It’s rather comparatively stable as it creates and recreates itself by transforming and replacing its components or undergoing continuous changes (Aronin 2019, p.  23). Another striking factor is that, Sinhala (L1) does not play dominant role in my DLC due to its lack of usage during some geographical shifts yet maintains its superiority in communicative competence. This indicates that in spite of its adaptability to an individual’s environment and personal needs DLC also maintains its stability. Most importantly the languages that constitute a DLC shouldn’t be considered as separate units, rather, they function as one workable language unit not only serving an individual’s communicative needs but also representing his or her language identity (Aronin 2019, pp. 24–25). A DLC does not merely represent a collection of languages used in the lives of individuals, institutions or society. Instead, it functions as an analytical unit to understand the complexity and dynamism of multilingualism. It not only empirically analyse behaviours of individuals and groups in constructing their personal DLC comprising of their most important vehicle languages but also discusses the impact of multilingualism on social arrangements. Furthermore, DLC sheds light on how multilingualism works in real-world communication, which could support in constituting an appropriate and just language policy (Lo Bianco 2020, in print). Despite the difference between my personal DLC and the institutional and societal DLC of Sri Lanka, it is evident that the country’s language policies have immensely influenced the making of my DLC. One of the interesting observations regarding this is that English remains a significant part of my DLC throughout the shifts. I firmly believe that this has not only to do with the role of English in modern globalized society as a lingua franca but also its role as a prestige language in Sri Lanka. Moreover, English is given a great significance by the educational and language policymakers. Tamil never makes its way into my DLC, even though it’s an official language in Sri Lanka and therefore is definitely a part of the societal DLC. This also indicates that the reality of Multilingualism can be quite different from what’s on paper. Regardless of the equal status given to Tamil in the official language policy, it is evident that Tamil does not receive the equal amount of political or social acceptance as the majority language: Sinhala, which in turn is reflected in the shortcomings of teaching Tamil as a second language.

The Evolution of Personal Dominant Language Constellations Based on the Amount…

185

References Aronin, L. (2019). What is Multilingualsim? In L. Aronin & D. M. Singleton (Eds.), Twelve lecture on multilingualism (Blue ridge summit) (pp. 3–34). Bristol: Multilingual Matters. Aronin, L., & Singleton, D. M. (2012). Multilingualism. Amsterdam: Benjamins. Balakrishnar, J., & Thanaraj, T. (2015). Instruction in the English medium: A Sri Lankan case study. In H. Coleman (Ed.), Language and social cohesion in the developing world (pp. 166–177). Colombo: British Council/ Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH. Bloomfield, L. (1933) Language. London: Compton Printing Ltd. Butler, Y.  G. (2013). Bilingualism/ multilingualism and second language acquisition. In T.  K. Bhatia & W.  Ritchie (Eds.), The handbook of bilingualism and multilingualism (pp. 109–136). Oxford: Wiley Blackwell. Canagarajah, S. (1995). The political economy of code choice in a “revolutionary society”: Tamil-­ English bilingualism in Jaffna, Sri Lanka. Language in Society, 24(2), 187–212. https://doi. org/10.1017/S0047404500018583. Canagarajah, S. (2005). Dilemmas in planning English/vernacular relations in post-colonial communities. Journal of SocioLinguistics, 9(3), 418–447. Canagarjah, S. (1993). Critical ethnography of a Sri Lankan classroom: Ambiguities in student opposition to reproduction through ESOL. TESOL Quarterly, 27(4), 601–626. http://www. jstor.org/stable/3587398. Cooray, M. (2010) English Education in Sri Lanka: Standard Sri Lankan English and English teaching in the context of International Standard English. http://archives.dailynews.lk/2010/07/02/ fea03.asp De Votta, N. (2007). Sinhalese Buddhist national ideology: Implications for politics and conflict resolution in Sri Lanka. Washington: East-west Center. Department of Official Languages. (2015). Sri Lanka. http://www.languagesdept.gov.lk/web/ index.php?lang=en Fernando, C. (1977). English and Sinhala Bilingualism in Sri Lanka. Language in Society, 6(3), 341–360. JSTOR. www.jstor.org/stable/4166944. Fernando, L. (2017. February 6) Towards a language revolution for reconciliation and development. Sri Lanka Brief. https://srilankabrief.org/2017/02/ towards-a-language-revolution-for-reconciliation-and-development/ Flick, U. (2014). Mapping the field. In U. Flick (Ed.), The SAGE handbook of qualitative data analysis (pp. 3–18). London/California/New Delhi/Washington, DC: Sage. Lo Bianco, J. (2011). A friendly knife? English in the context of Sri Lankan language planning. In L.  Farrell, U.  N. Udaya Singh, & R.  M. Giri (Eds.), English language education South Asia: From policy to pedagogy (pp.  36–60). Bengaluru: Cambridge University Press/ Foundation Books. Lo Bianco, J. (2020). A meeting of concepts and praxis: Multilingualism, language policy and the dominant language constellation. In J. Lo Bianco & L. Aronin (Eds.), Dominant language constellations: A perspective on present-day multilingualism. Springer. In print. Meyler, M. (2015). Sri Lankan English: An appropriate model for the teaching of English in Sri Lanka? In H.  Coleman (Ed.), Language and social cohesion in the developing world (pp.  178–185). Colombo: British Council/ Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). Nanayakkara, P. (2013) Reading bilingual education in Sri Lanka through international context. http://www.moe.gov.lk/english/images/subject_related/Bilingual_Education/reading.pdf Perera, S. (2015). Reflections on issues of language in Sri Lanka: Power, exclusion and inclusion. In H. Coleman (Ed.), Language and social cohesion in the developing world (pp. 55–74). Colombo: British Council/ Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH.

186

S. Kannangara

Premarathne, A., Yogaraja, S.  J., Medawattegedara, V., Senarathne, C.  D., & Abdullah, M.  R. M. (2016). Study on medium of instruction, national and international languages. In General education in Sri Lanka. Nugegoda: National Education Commission. Punchi, L. (2001). Resistance towards the language of globalisation  – The case of Sri Lanka. In International review of education. 47(3–4) (pp.  361–378). Dordrecht: Kluwer Academic Publishers. Ranjit, H., & Weerasekara, S. N. (2013). Language policy practices and national integration of Sri Lanka. International Journal of Arts and Commerce, 2(6), 73–78. https://www.ijac.org.uk/ images/frontImages/gallery/Vol._2_No._6/8.pdf. Romain, S. (1995). Bilingualism. Oxford: Blackwell Publishers.

Dominant Language Constellations of Russian Speakers in Cyprus Sviatlana Karpava

Abstract This study examines Dominant Language Constellations (DLCs) of Russian speakers in Cyprus. Both questionnaires and oral interviews were implemented for data collection. The participants of the study were 30 international Russian-speaking students studying and residing in Cyprus (17–26 years old) and 50 adult females, native speakers of Russian, members of Russian monolingual or mixed-marriage families (31–65 years old). They come from Russia, Ukraine and Moldova and belong to middle socio-economic class. The analysis of the data revealed that female adults have either Russian or mixed (Russian and Cypriot Greek) cultural and linguistic identity, while students have mainly mixed (Russian and English) identity. It was found that members of mixed-marriage families use Russian, English and Greek as their DLC, while for Russian women and students these languages are Russian and English. Russian is an integral DLC component of the three groups and plays the most prominent role in their individual, family and social lives. Immigration, social milieu, integrative and instrumental motivation affect the constitution, configuration and dynamics of their DLCs. As for language maintenance, nearly all female adult participants try to teach their children Russian at home and send their children to Russian lessons, where they learn to write and read in Russian. Keywords  DLC · Russian · Cyprus · Identity

1  Introduction This study is an attempt to examine language identity (Omoniyi and White 2006; Phan Le Ha 2008; Block 2013), language use and transmission by Russian-speaking community in Cyprus, to reveal which factors affect their linguistic repertoires and attitudes; whether there is any difference between adult immigrant groups, monolingual and mixed-marriage, and international students with respect to their language identities and dominant language constellations (DLCs), family language policies and language transmission strategies. S. Karpava (*) University of Cyprus, Nicosia, Cyprus © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_10

187

188

S. Karpava

The relationship between language and identity depends on socio-political, socio-economic and socio-cultural factors (Pavlenko and Blackledge 2004). Language transmits culture and history, and thus language loss can lead to the loss of inherited knowledge. Language vitality depends on such factors as demography, status, prestige, institutional control, the ethnolinguistic group, its distribution and size (Giles et  al. 1977). According to Grin (1990, 2003) and Lo Bianco (2008), language vitality depends on capacity development, opportunity creation and desire. ‘Language transmission by parents is crucial for language maintenance and recovery’ (Lo Bianco 2008: 25). Linguistic situation in Cyprus is complex and unique. It can be described as diglossic, bi(dia)lectal (Grohmann and Leivada 2012) or discrete bilectal (Rowe and Grohmann 2013, 2014; Grohmann et al. 2017) as two varieties are used by the Greek Cypriot population in Cyprus, Standard Modern Greek (SMG) and Cypriot Greek (CG). Newton (1972) distinguished between “village CG” and “urban CG”, while Hadjioannou et  al. (2011) and Tsiplakou (2014) found that a Pancyprian koiné variety has been developing since 1974, or Cypriot Standard Greek (CSG) as it was suggested by Arvaniti (2010), thus, there is a diglossic relationship between CG, low variety, and SMG, high variety. Arvaniti (2010), Grohmann and Leivada (2012); Rowe and Grohmann (2013, 2014); Leivada et  al. (2017) differentiate between “urban CG” or CSG, high variety, that has high prestige in Cyprus society and village CG, low variety, less prestigious or basilect (Goutsos and Karyolemou 2004). The difference between CG and SMG is mainly in terms of phonetics, morpho-­phonology, lexicon and morphosyntax (e.g., Newton 1972; Arvaniti 2010). The island can also be called multilingual as there are Cypriot minorities (e.g., Latin, Maronites) who live in Cyprus, residents of British origin, immigrants from various countries of Eastern Europe, Asia, and especially the former Soviet Union. According to CYSTAT (2010), Hadjioannou et  al. (2011), total population of Cyprus is 885,600: Greek Cypriots (660,300/74.5%), Turkish Cypriots (88,700/ 10%), Armenians (2700/0.3%), Maronites (4800/0.5%), Latins (900/0.1%) and foreign residents (128,200/14.5%). Russian community in Cyprus is the largest foreign language group in Cyprus (approximately 11,000). There are three main types of Russian population in Cyprus: temporary residents, members of mixed-marriage families and members of immigrant families. Tourists or business people as well as students belong to the first group, they use mainly Russian at home and Russian or English or some other European language on a daily basis—but not Greek. Members of mixed-marriage families, mainly with a Russian woman and a Greek Cypriot man and bilingual Russian–Cypriot Greek children, speak Russian and Greek, while members of immigrant families, where mostly both partners are Russian, who aim for long-term residence in Cyprus, speak Russian at home and English or Greek outside (Karpava 2015; Karpava et  al. 2018). Russian has become a new lingua franca in Cyprus (Eracleous 2015) in addition to English, which is widely-spread and used for communication throughout the island (Schneider 2003, 2007; Buschfeld 2013). In Cyprus, in some cases parents (language exposure at home) is the primary (and the only) source of Russian input for children and it is limited. The success of

Dominant Language Constellations of Russian Speakers in Cyprus

189

minority language transmission thus depends on daily language use at home, attitudes towards language use and preservation, efforts for creating opportunities and incentives for language use in and outside the home (Laleko 2013). Since access to the minority language is often limited to the family, the importance of family interactions for language maintenance and transmission cannot be underestimated (King et al. 2008; Spolsky 2012; Fogle 2013). It is important to know how languages are learned and negotiated within individual families (King et al. 2008 p. 907). Family language policy (FLP) is defined as ‘explicit and overt planning in relation to language use within the home among family members and provides an integrated overview of research on how languages are managed, learned and negotiated within families’ (King et al. 2008 p.1). Each family decides about their FLP (Schiffman 1996; Spolsky 2004; Shohamy 2006). Dominant Language Constellations (DLCs) framework (Aronin 2006, 2016, forthcoming) can help to investigate this complex multilingual situation in Cyprus, and in particular Russian community regarding their diverse linguistic repertoires and multicompetence (Cook 1991, 1999, 2013). “Dominant Language Constellation is the constellation of one’s dominant languages, a group of one’s most important (vehicle) languages, functioning as an entire unit, and enabling an individual to meet all needs in a multilingual environment” (Aronin 2006 p.145; Aronin 2016). DLCs have the most expedient languages for a person rather than all the languages known to him/her (Aronin 2006; Aronin and Ó Laoire 2004). Multicompetence is expressed and can be studied via DLCs: their constitution, configuration and dynamics. Time, effort or expenses devoted to each language use or acquisition, interaction, flow and constant change, attitudes, awareness, behaviour and context should be taken into consideration (Aronin 2016). The choice of languages for DLC is determined by societal, political and historical factors (Aronin 2006; Aronin and Singleton 2012; Aronin 2016, 2019). DLCs are normally modified throughout the life trajectory and presuppose complex relationships between languages for individuals and communities (Cook 2016). According to Aronin and Singleton (2012 p. 59), “DLC is the group of his/her most important languages that function as an entire unit-enable him/her to act in a multilingual environment.” DLC is a unit, it functions as unit, thus, it has characteristics beyond the sum of its parts and possesses a new quality. Characteristics, social status, way of life, preferences, language functions, cognition, expression of identity should be taken into consideration (Aronin 2016). Dominant Language Constellation is seen as a method of research (Aronin 2006; Aronin 2016). The researchers should differentiate between repertoire and constellation, emphasizing the shift from monolingual to multilingual perspective and considering such factors as language policy, community, country, typical configurations of DLC, level of mastery of each language and distribution of practical functions. It is important to delineate the most influential factors that define the qualities of DLC, to analyse configuration differences in the same DLC, (different ages, education groups, gender), whether there are similarities within all the groups. DLC relates linguistic phenomena to social, physiological, cultural and material. It can describe unique, dynamic, diverse and fuzzy reality of the multilingual learning environment,

190

S. Karpava

emergent or new quality (Aronin 2018). According to Lo Bianco (2020), DLC is heuristic as it focuses on the saliency and prominence of the dominant languages that are important in order to cover communication needs of an individual. This study approached the issues of language use, maintenance and transmission as well as family language policy, language and culture identity of Russian speakers in Cyprus from a multilingual perspective, employing the concept of DLC and the framework of multilinguality (Aronin and Ó Laoire 2004). It examined both individual factors such as attitudes, preferences, motivation, metalinguistic awareness and social factors such as family and work activities in order to trace the dynamics of DLCs with Russian in Cyprus and identify the turning points of meaningful changes in them, paying attention to such features of multilinguality as complexity, interrelatedness, fluctuation, multifunctionality, self-balance and self-extension. The following research questions were under investigation: 1. Which factors do affect the constitution, configuration and dynamics of DLCs of Russian speakers in Cyprus? 2. Are there differences in DLC configuration of the three groups of Russian community in Cyprus? 3. What are the most prominent languages in their DLCs? 4. Does the constitution of the DLCs influence family language policy, language use, maintenance and transmission?

2  Study 2.1  Participants There were three groups of the participants, who have the same languages in their DLCs but differ with respect to the interaction of languages in the constellation: 30 Russian-speaking students studying and residing in Cyprus (17–26 years old) and 50 adult females (31–65 years old), native speakers of Russian. With respect to Russian-speaking university students, there were 19 female (63.33%) and 11 male (36.67%) participants. Their country of origin: Ukraine (8/27%), Belarus (2/7%), Moldova (1/3%), Georgia (2/7%) and Russia (17/56%); mean age 21.1 years old; mean length of residence in Cyprus 5.54 years, mean age of onset to L2 English (AoO) 17.28 years. Their L1 is Russian, other languages in their linguistic repertoire are Ukrainian, English, German, Greek, Spanish, French, Swedish, Italian, Turkish and Arabic, see Table 1. The second group of the participants were 10 women who are representatives of Russian-speaking families, with both partners of Russian origin, living in Cyprus. Only women were willing to participate in the research as their husbands were either at work or very busy at home at the moment of data collection. Their mean age 40.9 years old; mean length of residence in Cyprus 5.9 years; mean age of onset to L2 Greek 36.2 years. Their socio-economic status (SES) is high, 7 participants

Dominant Language Constellations of Russian Speakers in Cyprus

191

Table 1 Participants Participants N Gender Female Male Age Mean Min. Max. SD LoR Mean Min. Max. SD AoO Mean Min. Max. SD

Russian women 10 10 0 40.9 33 62 9.94 5.9 1 16 5.21 36.2 22 60 11.65

Russian-CG women 40 40 0 37.17 26 55 5.14 11.5 1 19 3.99 25.9 17 43 5.96

Russian students 30 19 11 21.1 17 26 2.48 5.54 0.5 20 5.29 17.28 5 25 4.72

have university undergraduate degree, while 3 of them have university postgraduate degree. They work as accountants, economists, teachers and engineers. Their L1 is Russian, though they know some other languages such as English, Italian, French, Spanish, Greek and Chinese, see Table 1. The third group of the participants were 40 Russian–CG women, representatives of mixed-marriage families in Cyprus (wife Russian and husband CG). Our participants were only women as this is the most common pattern in Cyprus for intermarriage families: an immigrant female and a local male. Their mean age is 37.17 years, mean age of onset to L2 Greek 25.9  years, mean length of residence in Cyprus 11.5 years. Their SES is mid and high, 10 of the participants have only secondary education, 28 university undergraduate and 2 university postgraduate degree. They work as teachers, managers, psychologists, interpreters and economists. Their L1 is Russian, though they come from various countries such as Russia, Moldova, Ukraine, Belarus, Georgia and Latvia. English, Greek, Bulgarian, Ukrainian, Belorussian, Romanian, Turkish, Spanish, Latvian and Georgian comprise their linguistic repertoire, see Table 1.

2.2  Materials and Procedure Both written questionnaires (Otwinowska-Kasztelanic and Karpava 2015) and semi-structured oral interviews as well as ethno-graphic observations were used for data collection. Observations took place in the social setting chosen for the present study, university and home place of the participants (Marshall and Rossman 1989). They enabled the researcher to learn more about the DLCs of the participants, their

192

S. Karpava

linguistic behaviour, to be exposed to or be involved in their everyday routine activities (Schensul et al. 1999), to describe the existing bilingual/multilingual situation, to develop sampling guidelines and interview guides (Dewalt and Dewalt 2002). In total, 10 observations (20–60 min long) of Russian-speaking students took place at university (during, before and after the lessons, in the library, at the cafeteria and in the corridor). Russian families in Cyprus were observed at their home place (20 observations, 20–60 min long). All the participants filled in the questionnaires in the presence of the researcher; there was no time restriction, then interviews were conducted with the participants who were willing to do it. Overall, 36 participants, 10 students, 4 Russian women and 22 Russian–CG women agreed to be interviewed. The interview questions were based on the questionnaire, but the oral mode allowed the speakers to elaborate more on certain issues and to provide more in-depth information, supported by true examples from their live experience. The questions were focused on socio-economic background of the participants, their language and culture identities, DLCs, home language use, maintenance and transmission, family language policy, attitudes and beliefs, reasons for immigration, language change and acculturation, communication with friends and relatives, ties with homeland. The questionnaire had 50 questions, yes-no, multiple choice and open-ended, while in the interview only open-ended questions were used. The researcher visited them at their home place or at university (students). The participants in Cyprus come from various geographical areas, urban and rural. A snowball sampling technique was implemented in order to access the participants. First, an initial group of Russian speakers in Larnaca was selected, then they indicated other potential participants, members of Russian community in Cyprus, in other geographical areas: Nicosia, Limassol, Paphos and Agia Napa. The group of Russian-speaking students was recruited at a private university in Larnaca. The duration of the interviews was 30–50 min. The data was recorded, transcribed, thematically coded based on the concepts of multilinguality and DLC (Aronin 2016; Aronin and Ó Laoire 2004; Aronin and Singleton 2012) and analysed in line with the grounded theory research method (Glaser and Strauss 1967; Willig 2008). Iterative and recursive content analysis of the data was implemented in order to reveal thematic patterns (Patton 1980; Ward and Wolf-Wendel 2004). The data was thoroughly reviewed in order to find repeating themes; the emergent themes were coded with keywords and phrases, such multilingualism, identity, language use, maintenance and transmission, family language policy, attitudes, DLCs, then the codes were grouped into concepts and categories hierarchically.

3  Results The analysis of the data, obtained via questionnaires, interviews and observations, showed that there are various factors, such as language and culture identity, that do affect the constitution, configuration and dynamics of DLCs of Russian speakers in

Dominant Language Constellations of Russian Speakers in Cyprus

193

Table 2  Language identity Language identity Russian Several languages Greek Both Greek and Russian Both Russian and English

Russian women 60% 40% 0% 0% 0%

Russian-CG women 44% 20% 2% 34% 0%

Russian students 41% 16% 0% 0% 43%

Cyprus. Language and culture identities are complex and mutable in diverse, multilingual communities (Aronin 2016). It was found that Russian women identify themselves mainly with Russian language or several languages, whereas Russian– CG women—either with Russian or with both Greek and Russian or several languages. Russian students—either with Russian or both Russian and English, see Table 2. Their views and attitudes depend on their age, length of residence in the country, socio-economic status, linguistic experience, life trajectory and needs, perceptions, motivation, opportunities, self-image and life scenarios in L2 society. Example (1) is an excerpt from a student’s interview regarding her use of Russian and English that are included in her individual constellation and have different functions: 1. — Русский я использую при общении с родственниками, которые здесь живут и со своими друзьями, с родителями и со своей сестрой…Английский мне нужен для учебы…/ Russian I use when I communicate with my relatives, who live here and with my friends, with my parents and my sister…I need English for my studies. Example (2) is an excerpt from an interview with a Russian woman about the use of Russian in her daily life. It seems that for her Russian is the only resource for communication: 2. — Практически постоянно. Это язык общения дома, с большинством друзей, с друзьями в России и так далее…Я не работаю, сижу дома с ребенком, других иностранных языков я не знаю, но хотела бы выучить. / Nearly constantly. This is the language for communication at home, with the majority of friends, with my friends in Russia and so on…I do not work, I stay at home with my child, I do not know any foreign languages but I would like to learn them… Example (3) shows a mixed language identity of a Russian–CG woman, which is affected by her marital status. She is married to a Cypriot; thus, Greek has a prominent role in her DLC, in addition to Russian. Two languages are interconnected: 3. — Русскоязычная, скорее всего. Но вообще, говорят на каком языке думаешь… а у меня получается, смотря с кем…с мужем на греческом, я уже на русском не могу думать… вот раньше, когда я только учила греческий, я думала на русском, а потом переводила на греческий…/ I am more Russian, but as they say it depends on which language you use in order to

194

S. Karpava

think…and in my situation it depends…with my husband in Greek, I cannot think in Russian…before, when I was only learning Greek, I was thinking in Russian and then I translated into Greek… It is obvious that Russian is an integral DLC component of the three groups and plays the most prominent role in their individual, family and social lives. Immigration, social milieu, integrative and instrumental motivation affect the constitution, configuration and dynamics of their DLCs. Russian women and Russian-­ speaking students come from a higher socio-economic background in comparison to Russian–CG women. The former two groups do not have a strong need to find a job, to support their family and to integrate into Cyprus society as the latter one. Thus, only the third group has a prominent role of Greek in their DLCs due to its utility and functionality in order to meet their communicative needs (Lo Bianco 2020). Regarding culture identity, Russian women identify themselves only with Russian, while Russian–CG women mainly with Russian, some of them with Greek or with both Russian and English. Russian students—with Russian or both Russian and English, see Table 3. It should be noted that Russian is ‘a fundamental defining constituent’ (Aronin and Singleton 2012 p. 61) of their language and culture identity and their DLCs. Greek is the language of L2 country. Greek and English have mainly instrumental values. Not all participants extend their knowledge to three languages; some know only Russian or try to learn English as a foreign language. Example (4) is an excerpt from a student’s interview regarding his culture identity, he identifies himself with Russia and not Cyprus, his DLC comprises Russian and English, though he is exposed to Greek and lives in Cyprus: 4. — 80% русский и 20% английский. Страна, конечно, Россия, но не Кипр. / 80% Russian and 20% English. The country, of course, Russia, but not Cyprus. Example (5) is an excerpt from an interview with a Russian–CG woman who has a mixed Russian–CG cultural identity, she has undergone a shift from being monolingual to bilingual, two languages and two cultures are important for her:

Table 3  Culture identity Culture identity Russian Several languages Greek Both Greek and Russian Both Russian and English Ukrainian Russian and Ukrainian Georgian Christian Western

Russian women 100% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Russian-CG women 64% 0% 26% 0% 10% 0% 0% 0% 0% 0%

Russian students 42% 6% 8% 0% 13% 6% 10% 9% 3% 3%

Dominant Language Constellations of Russian Speakers in Cyprus

195

5. — Однозначно к русской… Если я скажу, что отождествляю себя с Россией и совсем не отождествляю себя с Кипром, это будет неправильно. Я себя отождествляю и с Кипром, и с Россией. / Certainly with Russian. If I say that I identify myself only with Russia, it will not be correct. I belong to both Cyprus and Russia. Example (6) is an excerpt from an interview with a Russian woman who identifies herself only with Russian language and culture and she tries not to lose her ties with her motherland: 6. — За культурой ездим пару раз в год в Москву и в Европу, а потом назад. / As for culture, we visit Moscow and Europe and come back several times per year. Russian women use mainly Russian, English and a little bit of Greek in order to watch TV, listen to the radio, read newspapers/internet and books, write emails and long documents. Russian–CG women use Russian, English and Greek, while Russian students use Russian, English, Greek and other languages. Russian women use mainly Russian, English and a little bit of Greek in order to talk about family, to talk to their partner, children and friends and to communicate at work. Russian–CG women use Russian, English and Greek, while the communicative resources of Russian students are Russian, English, Greek and other languages. The analysis of the data revealed complexity, multifunctionality and inequality of functions of languages, Russian, Greek and English, in linguistic repertoires and DLCs of the participants. Example (7) is an excerpt from the interview with a Russian–CG woman regarding her use of Russian, English and Greek, while examples (8) and (9) belong to a Russian woman and a student respectively. Migration has triggered some changes in the use of their language resources, they choose Russian, English or Greek, depending on a particular context or situation. Some languages were not even present in their linguistic repertoires before; now they are in their active use, in their DLCs. 7. — Ну если учесть, что с мужем мы общаемся на английском, то получается примерно 50 на 50. С детьми только на русском. / If I consider my communication with my husband then it is 50 Russian and 50 English. I use only Russian with my kids. — Использование греческого языка? / What about the Greek language? — По принуждению, только со свекровью. / Only when I have to, with my mother-in-law. 8. — Русский язык для меня все, книги, компьютер, общение…песни только иногда английские слушаю по радио. / Russian language is everything for me, books, computer, communication…I can listen to English songs sometimes… 9. — Раньше, до приезда на Кипр, русский был основной язык, а теперь даже не знаю, все поменялось, я учусь в британском университете, пишу, читаю на английском, только с друзьями общаюсь на русском или с родителями по скайпу…/ Before I came to Cyprus, my main language was Russian, now I do not know, everything has changed, I study at a British university, I write and

196

S. Karpava

Table 4  Reasons for coming to Cyprus Reasons for coming to Cyprus Family Politics Business Work Education Other reasons

Russian women 44% 46% 5% 0% 0% 5%

Russian-CG women 49% 5% 16% 15% 0% 15%

Russian students 58% 0% 0% 0% 25% 17%

read in English, I use Russian only in order to communicate with my friends and with parents via skype. Immigration to Cyprus is one of the main life-forming events, which affected the form of linguistic practice of the participants. Russian women came to Cyprus mainly due to family and politics, Russian–CG women for family, business, work and other reasons; Russian students for family, education and other reasons, see Table 4. Examples (10) and (11) are the excerpts from the interviews with Russian women that show the reasons of their coming to Cyprus. These examples illustrate why their DLCs are dynamic. After moving to Cyprus, they had to restructure their DLCs, either adding new languages or choosing other languages from their linguistic repertoire for active use in their DLCs. 10. — Ну, вообще, в основном из-за детей, конечно, чтобы они получили хорошее образование. Кипр—спокойная страна, море, солнце…/ Overall, mainly because of children, of course, so that they get good education. Cyprus is a quiet country; there is sea and sun here. 11. — Работа. Нашла работу и осталась. / Work. I found a job and I stayed in Cyprus. Example (12) is the excerpt from the interview with a Russian–CG woman and example (13) with a Russian-speaking student: 12. — Короче, я ездила до 2002 года, а потом вышла замуж, осталась. / So, I was going here and there until 2002, and then I got married and I stayed in Cyprus. 13. — Ну, конечно же, учеба, а что же еще? Мои родители со временем собираются перебраться на Кипр, здесь спокойно. / Education, of course, what else? My parents are also thinking to move to Cyprus, it is quiet here. Russian and Russian–CG women like Cyprus and prefer to stay here (80%), while only 45% of Russian students would like to live in Cyprus. This is reflected in their patterns of language practice and sets of languages in their DLCs. Greek can become one of the vehicle languages of the prime importance only for those who aim to stay in Cyprus and integrate into Cyprus society. Russian women have friends who speak only Russian (40%), both Russian and English (20%) or several

Dominant Language Constellations of Russian Speakers in Cyprus

197

languages (40%); Russian–CG women have friends, mainly those who speak different languages (82%) or Greek and Russian (18%); Russian students—those who use Russian (17%), both Russian and English (10%), Greek (3%) and different languages (70%). Knowledge of Greek and English contribute to the complexity of language interrelatedness. The ‘inner circle’ (Aronin and Singleton 2012 p. 61) of the participants’ DLCs depends on their social milieu, integrative and instrumental motivation. Example (14) is an excerpt from a student’s interview regarding her communication with friends. She knows Russian and English, but the lack of Greek knowledge somehow prevents her to communicate with the local students. If we look into Cyprus society DLC, it consists of CG, SMG and English, if we do not take minority and immigrant languages into consideration, thus CG is the vehicle language for the local students, they use it in order to communicate with each other or to exclude Russian-speaking international students from their communication. 14. — Вы в основном говорите на русском? / Do you mainly use Russian for communication? — Да, 70% русский и 30% английский. С русскоязычными студентами и друзьями по-русски. / Yes, 70% Russian and 30% English. With Russian-­ speaking students and friends I use Russian. — А с остальными студентами? / And with other students? — На английском, но они между собой общаются по-гречески, мы мало общаемся. / In English, but they communicate in Greek among themselves, we do not communicate a lot. — Вы знаете греческий? / Do you know Greek? — Вообще, нет. / Not at all. Example (15) is an excerpt from the interview with a Russian woman, which shows that her dominant language is Russian, though Greek is slowly emerging in her DLC. 15. — На каком языке вы общаетесь на Кипре? / Which language do you use for communication in Cyprus? — Только на русском, у меня нет других вариантов. Постоянно, 100%. / Only Russian, I do not have any other variants. I use Russian constantly, 100%. — А греческий? / What about Greek? — Ну, могу сказать спасибо, здравствуйте, καλή μέρα и все…/ I can say ‘thank you’, ‘hello’, ‘good morning; and that’s it… Example (16) belongs to a Russian–CG woman, her family and her work were the main reasons to introduce Greek and English into her DLC in order to be able to function in Cyprus society: 16. — Ну, у меня муж грек-киприот, куда мне деваться, нужно говорить на греческом. С детьми по-разному, то по-русски, то по-гречески, иногда даже легче по-гречески. На работе греческий и английский…/ I have a Cypriot husband, I cannot avoid using Greek with him. With my children, it depends, sometimes Russian, sometimes Greek, Greek even easier for me, At work Greek and English...

198

S. Karpava

The majority of the participants, Russian women (70%), Russian–CG women (92%) and Russian students (57%) think that they need to learn Greek, so most of the participants are ready to restructure their DLCs and to add a new, functional language there. Example (17) is an excerpt from a student’s interview regarding his willingness to learn Greek. The purposes of language use are different. English is mainly used in professional and education spheres. The Greek language is seen as an instrument for integration in Cyprus society. Length of residence in the country affects the stance of the participants towards the languages; Greek and English attain more prominent role, while Russian continues to maintain dominant role in their DLCs. It should be noted as well that other languages, such as Moldovan, Ukrainian, Belarussian, Georgian and Latvian become less relevant for these individuals and leave their DLCs. 17. — А хотели бы вы изучать греческий язык? / Would you like to learn Greek? — Вот сейчас начинаю его изучать. / I have started learning Greek. — Вам интересно изучать греческий язык? / Do you find it interesting to learn Greek? — Да, потому что я хочу здесь остаться и считаю, что мне нужен этот язык./ Yes, because I want to stay here [in Cyprus] and I believe that I need this language. Examples (18), (19) are excerpts from the interviews with Russian women and (20), (21) with Russian–CG women, of course, there is differentiation among the participants and not everybody is ready to learn Greek and use it in everyday life. It depends on individual and social factors, some people have integrative or instrumental motivation, the others consider that they do not need any further changes in their DLCs: 18. — Ну, здесь мы жить собираемся, если в дальнейшем получать гражданство, то нужно будет греческий тоже учить. Наверное, это основное, но потом с ребенком надо как-то заниматься греческим. / If we plan to live here, if we would like to apply for citizenship then we need to learn Greek. Perhaps it is the main reason, but we also need to practice Greek with our child. 19. — Знаете ли вы греческий язык? / Do you know Greek? — Нет, мне достаточно моего знания русского и английского на Кипре. / No, my knowledge of Russian and English is enough for me in Cyprus. 20. — Хорошо, а греческий, пытались ли вы его учить? / Have you tried to learn Greek? — Да, мой муж не знает английский. Сначала мы разговаривали на языке жестов. Потом я купила словарь и начала учить греческий. / Yes, my husband does not know English. First, we were using gestures in order to communicate. Then, I bought a dictionary and started to learn Greek. 21. — Когда вы начали изучать греческий язык? / When did you start learning Greek? — Я ходила на курсы 9 месяцев, но они были очень легкие. Наверное, это скорее всего, зависело от тех людей, которые ходили со мной на курсы.

Dominant Language Constellations of Russian Speakers in Cyprus

199

Очень было много вопросов, и я хотела быстрее продвигаться в этом и из-за этого я не пошла на следующий год. Мне было скучно. Но вот потом начала снова с ребенком. / I was attending Greek course for 9 months, but it was very easy. Perhaps, it was because of the people who were attending this course as well. There were too many questions, but I wanted to progress further and that is why I did not go next year. But then I started learning it again with my child. Examples (22) and (23) show why the students do not have Greek in their DLCs, key reasons are their plans of only temporary residence in Cyprus as well as a high popularity of the English language throughout the island. 22. — Не учу греческий, потому что не собираюсь оставаться жить на Кипре…Все говорят по-английски…Везде стараются помочь, если не знаешь греческий. / I do not learn Greek as I am not going to stay in Cyprus… Everybody speaks English… Everywhere they try to help, if you do not know Greek. 23. — Нет интеграции с Кипрским обществом, причина в незнании языка. / There is no integration with Cypriot society and the reason is the lack of language knowledge. Russian women mainly feel that they belong to Cyprus society but there is no full integration; Russian–CG women mainly believe that they belong to Cyprus society; Russian students think that they belong to Cyprus society, but not only, see Table 5. The participants have their own family history and social opportunities. Their status in L2 society, self as a citizen, depends on their social milieu, career activity, educational awareness and lifestyle, emotions, language learning and use. All of the participants (Russian women: 100%, Russian–CG women: 82% and Russian students 91%) believe that Cyprus society is tolerant towards non-native speakers of Greek. They consider that there is no discrimination of multilingualism in Cyprus. This situation in Cyprus allows them to decide whether they want to change the constitution and configuration of their DLCs, they can add only English or both English and Greek or even stay only with Russian. Example (24) is an excerpt from the interview with a Russian woman showing her view about multilingualism in Cyprus: Table 5  Status in L2 Society Status in L2 Society I belong to Russian society I belong to Cyprus society I belong to neither society I belong to Cyprus society but there is no full integration I belong to both Russian and Cyprus society No answer

Russian women 20% 10% 20% 50% 0% 0%

Russian-CG women 5% 36% 2% 25%

Russian students 3% 34% 22% 23%

28% 4%

18% 0%

200

S. Karpava

Table 6  Code-switching at home Code-switching Never Seldom Sometimes Often Very often No answer

Russian women 30% 10% 0% 20% 20% 20%

Russian-CG women 8% 12% 7% 38% 32% 3%

Russian students 16% 14% 14% 17% 16% 23%

24. — Ощущаете ли вы дискриминацию по языковому признаку? / Do you feel discrimination based on language criteria? — Нет, не ощущаю, мне очень нравится здесь. Киприоты ни разу мне не дали понять, что я говорю что-то не так, то есть они очень хорошо воспитаны и ведут себя деликатно. Они очень хорошо знают английский. / No, I do not feel it, I really like here [in Cyprus]. The Cypriots have never told me that something is wrong, they are very polite and behave properly. They have a very good command of English. All the participants work or study with Russian-speaking people and are allowed to use Russian. This helps them to retain Russian in their DLCs. The choice of their profession/studies was not related to the languages they know. The participants do not experience any discrimination at work or university. They never had bad attitudes towards them because they speak other languages. They code-switch quite often at home, at work and university, which is the evidence that they have several active languages in their DLCs, see Table 6. Languages in their DLCs are interconnected and thus it is easy for them to switch from one language to another. Example (25) is an excerpt from the interview with a Russian–CG woman with respect to code-switching and metalinguistic awareness. The participants seem to be aware of self-balance, attrition of the Russian language and fossilization. Various languages coexist in the DLCs, but the central position is taken by Russian, as it helps the participants to preserve their emotional and historical ties. 25. — Смешиваете ли вы два языка или несколько языков? / Do you mix two or several languages? — В разговоре? Бывает. / In oral communication? It can happen. — Часто? / Often? — Часто, потому что как на Кипр приезжаешь, значит идут вперемежку выражения. Во-первых, язык упрощается. Иногда даже элементарные вещи, ну как это на русском, и заменяешь словами более легкими, которые могут быть поняты только местными. / Often, because when you come back to Cyprus, it means that all expressions mix together. First of all, language is simplified. Sometimes, even some simple things, you do not know how to name them in Russian and you substitute them by the words that can be understood only by local people. Russian-speaking students were not asked questions regarding family and children. Russian women and Russian–CG women have either one or two, three

Dominant Language Constellations of Russian Speakers in Cyprus

201

children. Nearly all of them speak and comprehend Russian. Family is one of the most important factors that affect the constitution, configuration and dynamics of their DLCs. Example (26) is an excerpt from the interview with a Russian–CG woman, which shows language choice and family language policy of her family. Parents are responsible for the DLC configuration of their children; they take certain decisions about language practices at home and their education. 26. — Так, хорошо, на каком языке вы разговариваете со своими детьми и почему? / Well, ok which language do you use when you communicate with your children and why? — Только на русском. / Only in Russian. — Вы так решили? / Have you agreed on this? — Да, мы выбрали, папа на греческом, мама на русском, как носитель языка. / Yes, we have chosen this, father will speak Greek, while mother will speak Russian as a native speaker. The parents are satisfied with their level of Russian. Nearly all of their children can read and write in Russian. They all use Russian at home. Only in the case of Russian–CG women, some children (20%) refuse to speak Russian. The participants mainly have not been advised to stop using Russian with their children. Their children do not refuse to speak Greek and Russian. The participants do not think that their children are discriminated at school because they speak Russian. Their children attend extra-curricular activities and various languages are used there. They also attend Russian lessons. Parents try to teach their children Russian language and grammar. Most of the participants try to develop minority language literacy and preserve Russian in the DLCs of their children, independently whether they attend Russian-, Greek- or English-speaking kindergartens or schools. Example (27) is an excerpt from the interview with a Russian woman: 27. — Хорошо, ходит ли ваш ребенок в школу, на каком языке ведется преподавание? / Well, does your child go to school, which language is used for teaching? — На английском. / English. — Хорошо, на каком языке вы разговариваете с вашим ребенком и почему? / Well, what language do you use in order to communicate with your child? — Ну, на русском, иногда на английском, учителя просто советовали, что чаще с ней разговаривать по-английски дома, поскольку она тогда вообще по-английски не говорила. / Russian, sometimes English, it was a teachers’ advise that I need to speak English at home as she was not speaking English then at all. — Вы были согласны с этим? / Did you agree with this? — Я, так не очень, просто я как бы говорила, но не постоянно… / Not fully, I just spoke [English] but not always. Russian families in Cyprus, with both partners of Russian origin, prefer to send their children to English-speaking kindergartens and schools. Sometimes it is very difficult for a child to accommodate to a new educational system and language, see

202

S. Karpava

examples (28) and (29). Parents’ decisions are part of their FLP, which can lead to the changes of constitution, configuration and dynamics of DLCs of their children, second generation of Russian immigrants in Cyprus. 28. — В какой школе учится ваша дочь? / At which school does your daughter study? — В частной английской школе. / At a private English-speaking school. — Ей нравится там учится, она справляется со школьной программой? / Does she like studying there, does she cope with the school curriculum? — Конечно, ей страшно и неуютно: новая страна, новая школа и еще не получается писать. У нее дополнительная неуверенность из-за новизны. / Of course, she is scared and she does not feel comfortable: it is a new country, a new school and she can’t write. She has an extra uncertainty because of the new/unknown situation. 29. — Вы довольны английской школой? / Are you satisfied with the English-­ speaking school? — У меня дочка пошла в английскую школу, не зная английского вообще. Первый класс ей было сложно, но она нашла себе подружку, которая переводила ей то, что она не понимает. Сразу замечу, что дочка очень чувствительная и ей тяжело даются неудачи. Сейчас второй класс и оценки намного лучше. / My daughter went to English school, without any knowledge of English at all. It was very difficult for her in the first grade, but she found a friend, who translated to her everything that she could not comprehend. I need to admit that my daughter is very sensitive and she cannot stand any failures. Now she is in the second grade and her marks are much better. When they choose an English kindergarten, they want to have native speakers of English there, see example (30). They understand that English is important for the future of their children and they try to introduce it to their DLCs as soon as possible. 30. — Почему вы выбрали английский частный садик? / Why have you chosen an English kindergarten? — Моего не забрать оттуда. Воспитатели все англичанки. Профи. Я за ребёнка там спокойна. / My son really likes there. All the educators are native English speakers. I feel safe about my child. — В предыдущем садике мне не особо нравилось, не следят за детьми, как бы я хотела, то не садик, присмотр за детьми и всё. Не занимаются они там с ними особо. И сомневаюсь, что можно там английский выучить. Обе воспиталки-киприотки. / In the previous kindergarten we did not like so much, they were not looking after the children the way I wanted. It is not a kindergarten; it is just a place where they keep the children. They do not have any classes. And I doubt that you can learn English there. Both of the teachers are Cypriots. Very often, there are still some Russian-speaking teachers, educators, experts at kindergartens in order to help the children to accommodate to English system of education, see example (31). Having several languages in a DLC from a very early

Dominant Language Constellations of Russian Speakers in Cyprus

203

age can help a child to develop multilinguality and interrelatedness among the languages, thus DLC stands as a unit. 31. — Почему вы решили отдать своего сына в английский садик? / Why have you decided to send your son to the English kindergarten? — Сыну очень нравиться и ходит с удовольствием. Проблем с языком нет. Через месяц уже начал понимать, что от него хотят. В саду есть русскоговорящая женщина. Если какие проблемы всегда поможет. / My son likes it very much and goes there with pleasure. We do not have any problems with the language. After one month, he started to understand what they want from him. There is a Russian-speaking woman at the kindergarten. If there are any problems, she will always help them. Individual differences of the children should be taken into consideration, see example (32). It can take some time to add a language to a DLC and have it activated. 32. — У ваших детей есть проблемы с английским языком? / Do your children have problems with the English language? — Конечно же зависит от ребёнка. У меня один пошёл с 5 лет. Через два месяца заговорил. А второй в 4 года. Очень трудно учил. Год вообще не шёл на контакт. В этом году заговорил. / Of course, it depends on a child. One of my children went there [English kindergarten] from the age of five. And the other child at the age of four. It was very difficult for him to study. One year he was not willing to have any contact/ to have any communication. Only this year he has started to speak. Children with L1 background often have friends among L1 Russian friends. They even often help each other in the process of adaptation, in case of sequential bilinguals/multilinguals, in order to make English and/or Greek part of their DLCs, core, most expedient and actively used languages, see example (33): 33. — В какой школе учится ваша дочь? / Which school does your daughter attend? — В государственной греческой школе. / She goes to Greek public school. — У нее были проблемы с греческим языком? / Did she have any problems with the Greek language? — Нет, у меня ребенок прекрасно владеет двумя языками, она тут родилась, она греко- и русскоговорящая, и читающая. Она даже помогает другим детям. Пришла новенькая девочка, только с русским языком. Получила от нашей девочки полное попечение. Детское общение творит чудеса в процессе адаптации. / No, my child knows two languages very well, she was born here [in Cyprus], she speaks and reads Russian and Greek. She even helps other children. A new girl, who knows only Russian, came [to school]. She has received help and support from our girl. Child communication makes wonders in the process of adaptation. Russian-speaking population in Cyprus is very heterogeneous. They have different reasons why they decide to move to Cyprus. Some aim for temporary residence, the others for permanent. The choice of the school for their children, whether it is

204

S. Karpava

Table 7  How often do you insist that your child use Russian at home? Children: Use of Russian at home Never Sometimes Often Very often No answer

Russian women 46% 0% 20% 34% 0%

Russian-CG women 28% 8% 25% 23% 16%

going to be Russian, English or Greek, public or private, depends on whether they plan to return back to Russia or move to any other country or stay in Cyprus, see example (34). The parents try to decide which DLC configuration is going to be the most successful for their children in particular environment, beneficial for their social coherence, career development and personal goals. 34. — Что повлияло на ваш выбор школы, от чего это зависит? / What did affect your choice of the school? What does it depend on? — По поводу английской или русской школы, надо расставить приоритеты. Что важнее, если английский язык, то конечно, английская школа. Если всё-таки русское образование, а оно очень сильно отличается от английского, то, наверное, лучше сразу в русскую школу, там английский тоже преподают. / As for English or Russian school, you need to set priorities. What is more important, if it is the English language, then, of course, English school. If it is Russian education, and it is very different from the English one, then perhaps it is better to send your child to the Russian school, and English is also taught there. Russian women insist more often than Russian–CG women that their children use Russian at home and outside, probably they see more value in Russian and if their children practice L1 more, it will never leave their DLCs, see Table 7. Example (35) is the excerpt from the interview with a Russian–CG woman. Even if the parents put effort to maintain languages in the DLCs of their children, societal factors sometimes can be stronger and can affect the dynamics of the DLC. 35. — На каком языке вы общаетесь с сыном? / Which language do you use in order to communicate with your son? — На русском, но муж на греческом. / Russian, but my husband Greek. — На каком языке вы планировали с ним разговаривать? / Which language did you plan to use with your child? — На русском. Но надо сказать, что русский у него ушел в пассив, когда он пошел в греческий садик, греческий потихоньку вытеснил русский. / Russian. It should be noted that he has passive knowledge of Russian now, when he went to Greek kindergarten, Greek slowly substituted Russian. Only in some cases, they have felt bad attitudes towards them from doctors, nurses, teachers, based on the linguistic factor. All the participants consider that

Dominant Language Constellations of Russian Speakers in Cyprus

205

multilingualism is beneficial for their children, which creates the opportunity for them to expand the boundaries of their DLCs by adding new vehicle languages. Example (36) is an excerpt from the interview with a Russian woman who mainly has an instrumental motivation regarding multilingualism and functionality of the languages in the DLC: 36. — Как вы думаете, знание нескольких языков это преимущество или недостаток для вашего ребенка? / What do you think, knowledge of several languages, is it a benefit or a disadvantage for your child? — Преимущество. / It is an advantage. — Почему вы так думаете? / Why do you think so? — Ну, это больше возможностей. / It means more possibilities. To sum up, it seems that there are differences in DLC configuration of the three groups of Russian community in Cyprus. DLCs of Russian students and Russian women consist of Russian and English. It should be noted that their language repertoire is much broader than their DLCs. They know other languages such as Ukrainian, Belorussian, Arabic, Greek, French, Turkish, German, Romanian, Spanish, Latvian, Georgian, Bulgarian, Italian but the languages of their prime importance in Cyprus are Russian and English. The DLC of Russian–CG woman includes not only Russian, English, but also Greek. The constitution, configuration and dynamics of DLCs of Russian speakers in Cyprus depend on various individual and social factors. First of all, on the age of the participants and age of arrival to Cyprus, their willingness to stay here, temporarily or permanently, to integrate into L2 society. Their socio-economic status, professional opportunities, level of education can be the reasons of language learning and use. Russian is the most prominent language in the DLCs of the Russian speakers in Cyprus. The constitution and dynamics of the DLCs play a crucial role in language use, maintenance and transmission and family language policy. Parents try to maintain the Russian language and transmit it to the next generation. It could be a challenging task due to the influence of the dominant languages, Greek and English, language attrition and fossilization as well as the resistance and lack of motivation of their children. Family language policy can lead to the changes and dynamics of DLCs of their children. The DLC of the Cyprus multilingual society with an emergent dominant role of Russian as a new lingua franca, in addition to Cypriot Greek, Standard Modern Greek, English and other foreign languages of minority and migrant groups shapes the DLCs of the members of the Russian community in Cyprus. DLC can be seen as a unit, but simultaneously as a complex system with several constituent parts (Aronin, 2016). It is a useful framework for investigation of multi-competence, multilinguality and multilingualism in Cyprus, and in particular of Russian community.

206

S. Karpava

4  Discussion and Conclusion The analysis of the data revealed that female adults, members of mixed-marriage families, have either Russian or mixed (Russian and Cypriot Greek) cultural and linguistic identity, while the representatives of Russian-speaking families in Cyprus have mainly Russian or mixed (Russian and English) identity. The first group of the participants has come to Cyprus mainly due to family and work reasons; the second group are in Cyprus for education and business respectively. Russian-speaking students are interested in international education in Cyprus. All participants believe that people in Cyprus are tolerant to multilingualism and they have nearly never experienced discrimination or bad attitude towards them due to their L1. Only L1 Russian adult females, members of mixed-marriage families, believe that they need to learn Greek in order to integrate into Cyprus society, to improve their knowledge and to be successful at work. The DLCs of Russian speakers in Cyprus comprise Russian, English and Greek as these languages are the most important for their functioning in Cyprus society in terms of their cognitive, social and emotional needs. The analysis of the data revealed meaningful differences in configuration of Dominant Language Constellations in Russian-speaking student group, mixed-marriage families and Russian-speaking families in Cyprus, as well as in proficiency, and in functions of the DLC languages. Russian is the ‘dominant’ language in the DLCs of Russian speakers in Cyprus in various spheres of their lives, though the use of L1 has decreased since they immigrated to Cyprus. The EFL/ESL status of English (lingua franca) in Cyprus makes it an indispensable tool for Russian speakers in Cyprus. They use it mainly in public sphere, work, business and education. It has an important role in DLC configuration, especially of Russian women and students who do not want or need to learn Greek. Societal linguistic affordances shape individual linguistic affordances (Gibson 1977, 1979; Singleton and Aronin 2007; Aronin and Singleton 2010, 2012). The widespread use of English in Cyprus discourages immigrants to learn Greek as they can easily live and work without knowing the local language. The role/function of Greek (CG/SMG) in DLC is restricted. Only Russian–CG women are willing to learn Greek in order to integrate into CG society and use it in private (family) and public spheres. Overall, it was found that members of mixed-marriage families use Russian, English and Greek as their Dominant Language Constellation (Aronin and Singleton 2012), while for Russian-speaking families and international students these languages are Russian and English. The DLC of Cyprus society, a very complex one, might affect the constitution, configuration and change of DLCs of Russian community in Cyprus. The linguistic repertoires of L1 Russian participants are much broader and varied than their DLCs as the latter include the most important languages that help them to meet the needs and requirements of Cyprus society.

Dominant Language Constellations of Russian Speakers in Cyprus

207

It should be noted that DLCs of Russian speakers in Cyprus are not stable, they can be characterised by dynamic changes. The important life event/benchmark that affects their DLCs is their immigration to Cyprus, change of their country of residence, their linguistic and cultural environment. After moving to Cyprus there was a shift/change in their life and DLC trajectory. They either added new language(s) to their linguistic repertoires and DLCs and/or rebalanced the constitution/configuration of their DLCs and the role/function of each language. DLC is a good analytic tool that can shed light on multilingual and multicultural situation of Cyprus, especially with respect to immigrant population. As for language maintenance, nearly all female adult participants try to teach their children Russian at home and send their children to Russian lessons, where they learn to write and read in Russian. The participants from mixed-marriage families either use ‘one parent, one language approach’ or mix both languages, while communicating with their children, while only Russian is used in monolingual Russian-speaking families in Cyprus. The constitution of the DLCs influence family language policy, language use, maintenance and transmission. More research is needed, quantitative and qualitative, synchronic and diachronic, in order to examine the (socio)-linguistic situation of the Russian-speaking community in Cyprus with respect to constitution, configuration and dynamics of DLCs, minority language acquisition, use and transmission as well as family language policies.

References Aronin, L., & Ó Laoire, M. (2004). Exploring multilingualism in cultural contexts: Towards a notion of multilinguality. In C. Hoffmann & J. Ytsma (Eds.), Trilingualism in family, school and community (pp. 11–29). Clevedon: Multilingual Matters. Aronin, L., & Singleton, D. (2010). Affordances and the diversity of multilingualism. International Journal of the Sociology of Language, 205, 105–129. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Verlag. Aronin, L. (2018). Paradigm shift in language education for the development of multiliterate and plurilingual agencies. In Dominant Language Constellations in education, language teaching and multilingualism. ECSPM Symposium, Darmstadt, 26–27 September, 2018. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L. (2019). Dominant language constellation as a method of research. In E.  Vetter & U.  Jessner (Eds.), International research on multilingualism breaking with the monolingual perspective (pp. 13–26). Cham: Springer. Arvaniti, A. (2010). Linguistic practices in Cyprus and the emergence of Cypriot standard Greek. Mediterranean Language Review, 17, 15–45. Block, D. (2013). Issues in language and identity research in applied linguistics. ELIA (Estudios de lingüística inglesa aplicada), 13, 11–46.

208

S. Karpava

Buschfeld, S. (2013). English in Cyprus or Cyprus English? An empirical investigation of variety status. Amsterdam: John Benjamins. Cook, V. (1991). The poverty-of-the-stimulus argument and multicompetence. Second Language Research, 7, 103–117. Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33, 185–209. Cook, V. (2013). Multi-competence. In C. Chapelle (Ed.), The encyclopedia of applied linguistics (pp. 3768–3774). New York: Wiley-Blackwell. Cook, V. (2016). Premises of multi-competence. In V.  Cook & L.  Wei (Eds.), The Cambridge handbook of linguistic multi-competence (pp. 1–25). Cambridge: Cambridge University Press. CYSTAT (Statistical Service of the Republic of Cyprus). (2010). Population summary data 1995–2009. Retrieved January 30, 2011, from http://www.cyst a t . g o v. c y / m o f / c y s t a t / s t a t i s t i c s . n s f / p o p u l a t i o n c o n d i t i o n _ 2 1 m a i n _ e n / populationcondition_21main_en?OpenForm&sub=1&sel=2 DeWalt, K., & DeWalt, B. (2002). Participant observation: A guide for fieldworkers. Walnut Creek: AltaMira Press. Fogle, L. (2013). Parental ethno-theories and family language policy in transnational adoptive families. Language Policy, 12, 83–102. Gibson, J. (1977). In R. Shaw & J. Bransford (Eds.), The theory of affordances. In Perceiving, acting, and knowing. Toward an ecological psychology (pp. 67–82). Hillsdale: Lawrence Erlbaum Associates. Gibson, J. (1979). The ecological approach to visual perception. Boston: Houghton Mifflin. Giles, H., Bourhis, R., & Taylor, D. (1977). Towards a theory of language in ethnic group relations. In H. Giles (Ed.), Language, ethnicity and intergroup relations (pp. 307–348). London: Academic. Glaser, B., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. New York: Aldine. Grin, F. (1990). The economic approach to minority languages. Journal of Multilingual and Multicultural Development, 11, 153–173. Grin, F. (2003). Language policy evaluation and the European Charter for regional or minority languages. London: Palgrave Macmillan. Grohmann, K., & Leivada, E. (2012). Interface ingredients of dialect design: Bi-x, socio-syntax of development, and the grammar of Cypriot Greek. In A. M. Di Sciullo (Ed.), Towards a biolinguistic understanding of grammar: Essays on interfaces (pp. 239–262). Amsterdam: John Benjamins. Grohmann, K., Papadopoulou, E., & Themistocleous, C. (2017). Acquiring clitic placement in bilectal settings: Interactions between social factors. Frontiers in Communication, 2, 5. https:// doi.org/10.3389/fcomm.2017.00005. Goutsos, D., & Karyolemou, M. (2004). Introduction. International Journal of the Sociology of Language, (168), 1–17. https://doi.org/10.1515/ijsl.2004.031. Hadjioannou, X., Tsiplakou, S., & Kappler, M. (2011). Language policy and language planning. Current Issues in Language Planning, 12, 503–569. https://doi.org/10.1080/1466420 8.2011.629113. Karpava, S., Ringblom, N., & Zabrodskaja, A. (2018). Language ecology in Cyprus, Sweden and Estonia: Bilingual Russian-speaking families in multicultural settings. Journal of the European Second Language Association, 2, 107–117. Karpava, S. (2015). Vulnerable domains for cross-linguistic influence in L2 acquisition of Greek (Inquiries in language learning 16). Frankfurt am Main: Peter Lang. King, K., Fogle, L., & Logan-Terry, A. (2008). Family language policy. Language and Linguistics Compass, 2, 1–16. Kuznetsova-Eracleous, N. (2015). Linguistic landscape of Limassol: Russian presence. Unpublished MA thesis, University of Cyprus.

Dominant Language Constellations of Russian Speakers in Cyprus

209

Laleko, O. (2013). Assessing heritage language vitality: Russian in the United States. Heritage Language Journal, 10, 89–102. Leivada, E., Papadopoulou, E., Kambanaros, M., & Grohmann, K. (2017). The influence of bilectalism and non-standardization on the perception of native grammatical variants. Frontiers in Psychology, 8, 205. https://doi.org/10.3389/fpsyg.2017.00205. Lo Bianco, J. (2008). Policy activity for heritage languages: Connections with representation and citizenship. In D. Brinton, O. Kagan, & S. Bauckus (Eds.), Heritage language education: A new field emerging (pp. 53–69). New York: Routledge. Marshall, C., & Rossman, G. B. (1989). Designing qualitative research. Newbury Park, CA: Sage. Newton, B. (1972). Cypriot Greek: Its phonology and inflections. The Hague: Mouton. Omoniyi, T., & White, G. (2006). The sociolinguistics of identity. London/New York: Continuum. Otwinowska, A., & Karpava, S. (2015). MILD questionnaire: Migration, identity and language discrimination/diversity. University of Warsaw and University of Central Lancashire, Cyprus. Unpublished manuscript. Patton, M. (1980). Qualitative evaluation methods. Beverly Hills: SAGE. Pavlenko, A., & Blackledge, A. (2004). Negotiation of identities in multilingual contexts. Clevedon: Multilingual Matters. Phan Le Ha. (2008). Teaching English as an international language: Identity. In resistance and negotiation. Multilingual Matters: Bristol. Rowe, C., & Grohmann, K. (2013). Discrete bilectalism: Towards co-overt prestige and diglossic shift in Cyprus. International Journal of the Sociology of Language, 224, 119–142. Rowe, C., & Grohmann, K. (2014). Canaries in a coal mine: Native speakerhood and other factors as predictors of moribundity, death, and diglossic shift in Cypriot Greek. Mediterranean Language Review, 21, 121–142. Schensul, S., Schensul, J., & LeCompte, M. (1999). Essential ethnographic methods: Observations, interviews, and questionnaires. Walnut Creek: AltaMira Press. Lo Bianco, J. (2020). A meeting of concepts and praxis: Multilingualism, language policy and the dominant language constellation. In J. L. Bianco & L. Aronin (Eds.), Dominant language constellations: A perspective on present-day multilingualism (pp. 1–17). Dordrecht: Springer. Schiffman, H. (1996). Linguistic culture and language policy. New York: Routledge. Schneider, E. (2003). The dynamics of new Englishes: From identity construction to dialect birth. Language, 79, 233–281. Schneider, E. (2007). Postcolonial English: Varieties around the world. Cambridge: Cambridge University Press. Shohamy, E. (2006). Language policy: Hidden agendas and new approaches. New  York: Routledge. Singleton, D., & Aronin, L. (2007). Multiple language learning in the light of the theory of affordances. Innovation in Language Teaching and Learning, 1, 83–96. Spolsky, B. (2004). Language policy. Cambridge, UK: Cambridge University Press. Spolsky, B. (2012). The Cambridge handbook of language policy. Cambridge: Cambridge University Press. Tsiplakou, S. (2014). How mixed is a ‘mixed’ system? The case of the Cypriot Greek koiné. Linguistic Variation, 14, 161–178. https://doi.org/10.1075/lv.14.1.07tsi. Ward, K., & Wolf-Wendel, L. (2004). Academic motherhood: Managing complex roles in research universities. The Review of Higher Education, 27, 233–257. Willig, C. (2008). Introducing qualitative research in psychology. Adventures in theory and method (2nd ed.). Maidenhead: Open University Press.

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant Language Constellation Stela Letica Krevelj

Abstract  One of the most prominent topics within the research in multilingualism is the interaction between language systems in a multilingual and its effects on language acquisition and use. However, existing insights into the issue are still only fragmental if one does not take into consideration, both from the methodology and design perspectives, the complexity inherent in multilingual behavior and influences from more than one language system at the time. In this chapter I examine the potential of the Dominant Language Constellation (DLC) approach (Aronin L, Dominant language constellations: an approach to multilingualism studies. In: Ó Laoire M (ed) Multilingualism in educational settings. Schneider Publications, Hohengehren, pp  140–159, 2006; Aronin L, Multicompetence and dominant language constellation. In: Cook V, Li Wei (eds) The Cambridge handbook of linguistic multicompetence. Cambridge University Press, Cambridge, pp 142–163, 2016) for studying crosslinguistic influence (CLI) in multilinguals through quantitative research design. Arguments are based on previously collected data and outcomes of a study which looked into simultaneous interaction of languages of multilinguals during a production task in L3 English (Letica Krevelj S, Crosslinguistic interaction in acquiring English as L3: role of psychotypology and L2 status. Unpublished doctoral thesis. University of Zagreb, Faculty of Humanities and Social Sciences. Zagreb, Croatia, 2014). I provide a detailed description of a common DLC of multilingual participants from the same community in Croatia involving Croatian and Italian as official languages of the community and English as a foreign language. The DLC described and analyzed in terms configurations of DLC offered some novel insights. The benefits of DLC, both as a conceptualization and research tool, as well as its complementarity with the view of language use as dynamic, embodied and embedded are discussed. Keywords  Crosslinguistic influence · Dominant language constellations · Quantitative studies · Multilingualism research

S. L. Krevelj (*) Faculty of Humanities and Social Sciences, University of Zagreb, Zagreb, Croatia © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_11

211

212

S. L. Krevelj

1  Introduction Aronin and Jessner (2014) point to the greater complexity of multilingual research compared to bilingual research. This is evident in many aspects of research such as study design and selection of participants. The complexity is not due merely to the quantitative difference in terms of the number of languages, but to multiple interactions between many variables involved. While different variables interact with one another, the nature of these interactions depends also on the intensity with which variables feature in these interdependencies. In order to come up with more comprehensive insight into multilingual behavior, emergentist usage-based approaches emphasize the need for qualitative studies which try to inspect linguistic behavior of individuals, rather than large samples. Additionally it is claimed that variability, rather than patterns and generalizations, has a greater explanatory power. There is no doubt that multilinguality of each individual is unique (Aronin and Ó Laoire 2004) and studied in depth provides us with valuable data. However, I believe that we should aim at looking for patterns and generalizations in a more holistic manner, through the combination of qualitative and quantitative methodology. One of the long-standing conceptual issues in the field of multilingualism is the assumption that we can count languages in peoples’ repertoires. The issue is tied to an even greater theoretical and empirical issue of defining multilingual speakers and it generates two questions I would like to tackle here: (1) Where do we draw a line between languages/idioms? and (2)What is the threshold proficiency level required for the language to be counted in? The problem of defining proficiency in measurable terms was discussed by De Angelis (2017) in the light of methodological problems involved in selection of research participants. She looked at the proficiency as a variable conditioning cross-linguistic influence (CLI) in additional language acquisition which is directly related to the issue of when the knowledge of previous languages starts to matter. As De Angelis problematised in the paper, the lack of clear boundaries about the proficiency level that needs to be achieved in a language we count in as relevant, leads to great variability in the research results. It is exactly in this respect that I see the potential of the DLC. In this chapter I examine the qualities of DLC as a tool in relation to what it may bring to the study of the role of background languages (the first language (L1) and the second language (L2)) in third language (L3) acquisition. For that purpose, I refer to data and results of a quantitative study on CLI in L3 (Letica Krevelj 2014), and in doing so, I address the unresolved issues that go to the definition of a multilingual speaker themselves and surface as problematic in the operationalization of variables and analysis of data in quantitative studies.

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

213

2  The Concept of DLC I first provide a short summary of points made by Aronin (2016) on the concept of DLC which are later considered in the light of previously conducted study on CLI (Letica Krevelj 2014). More specifically, the points listed here refer to the definition of one’s DLC, and the conceptualization of DLC I see as a valuable tool in research design and analysis of data. –– Aronin (2016: 146) defines DLC as “a group of one’s most important (vehicle) languages, functioning as an entire unit, and enabling an individual to meet all his or her needs in a multilingual environment”. –– The need which languages fulfill in the environment is specified in terms of their functions of means of communication, cognition and an expression of identity. –– DLC has “two faces”: ‘one as a unit and the other as a complex system consisting of several components’ (Aronin 2016: 154). –– Using constitution and configuration as categories in analysis of a DLC whereby constitution comprises features of DLC that provide answers to the questions of what/ which, and configuration captures the proportion of constitution features for each language relative to other languages in a DLC (Aronin 2016: 157). –– The DLC approach “can enhance and focus theoretical understanding of multilingualism, specify and organize previously obtained knowledge, and serve as a research methodology” (Aronin 2016: 154). In order to consider the abovementioned points I will present the aim and outcomes of a quantitative study on CLI in L3 which looked into simultaneous interaction of languages in multilinguals with a common DLC. The results of the study will be only briefly summarized here as they are thoroughly elaborated in Letica Krevelj (2014), but they will be re-examined in the light of the DLC model.

3  CLI in L3 Acquisition One of the topics of interest within the field of multilingualism is the role of previous languages in the process of L3 acquisition. Numerous studies on CLI in L3 acquisition have been conducted in the last 30  years. When looking at particular manifestations of CLI, we find great variability in research findings which point to different variables as those with  the greatest explanatory value. It is practically impossible to control for the effect of each of the variables that may affect the frequency and source of CLI in L3 production, and as a consequence, it is rather difficult to isolate one variable from any other, or a set of variables, when explaining particular behavior. Therefore, the studies concentrated only on a limited number of variables examining their relative weight. Results of different studies suggest that CLI occurs from languages in which multilingual speakers are highly proficient as well as from those in which they have

214

S. L. Krevelj

only a low level of proficiency (see De Angelis 2007; Falk and Bardel 2010). It has been also widely attested that when learners have in their repertoires languages that are typologically close, they will transfer the linguistic information which they find useful and relevant in the language task (De Angelis 2005a, b, 2007; Dewaele 1998; Fouser 2001; Letica Krevelj 2014; Ringbom 2001; Williams and Hammarberg 1998). Nevertheless, it is clear from these studies that not all background languages are equally likely to exert influence or to be activated in L3 production. The central question that has emerged in L3 research concerns the reasons for the activation of particular languages in multilingual production or the relative weight of different factors in determining the source language of CLI. A number of explanations have been put forward, and the source language of CLI in L3 production has been explained by various factors, as well as their interaction. Some studies clearly show that L2 transfer predominates over L1 transfer in L3 production (De Angelis 2005a, b; De Angelis and Selinker 2001; Hall and Ecke 2003; Williams and Hammarberg 1998). This finding was one of the strongest arguments for the “difference assumption” (De Angelis 2007), according to which there is a fundamental difference between the acquisition of an L2 and an L3, and L2 status (Hammarberg 2001) was proposed as an important factor that may explain the tendency of an L2 to be activated in L3 production. On the other hand, the factor of perceived language distance (psychotypology) assumes that both L1 and L2 are likely to transfer depending on the speaker’s perception of which language is closer to the target language (L3). The psychotypology and L2 status factor have surfaced as the most important variables, and the debate on the relative weight of each is an ongoing one.

3.1  A Quantitative Study on CLI in L3 (Letica Krevelj 2014) The aim of the study was to examine the relative weight of two factors, psychotypology and the L2 status factor, in determining the frequency and source language of CLI in L3. The relative weight of the two aforementioned factors was examined in a language constellation of three Indo-European languages (Croatian, English and Italian) which belong to different language families (Slavic, Germanic and Romance). Nevertheless, the language constellation provided for a context in which, at the lexical level, one of the participants’ background languages (Italian) was more closely related to the target language (English), and the other (Croatian) was less closely related to it. However, in this study I was interested to see which background language would be the dominant source of CLI in a context where both a closer and a more distant background language provided for the same amount of formal similarity with the L3 at the level of lexis. The same amount of formal similarity between each of the background languages and the target language was manipulated through a task (synonym provision task in L3 English). At the same time, the task was designed in a way that the instances of CLI could be identified as

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

215

such, regardless of whether the outcome of CLI was positive or negative (commonly referred to positive and negative transfer). In order to provide the condition for measuring the effect of the L2 status factor there were two groups of participants with the mirror image of the same two languages in the role of L1 and L2: speakers of Croatian as L1 and Italian as L2 (CroL1 participant group) and speakers of Italian as L1 and Croatian as L2 (ItaL1 participant group). In order to isolate the effects of these two factors from the multitude of other factors important when determining the frequency and source language of CLI, many other variables had to be controlled. Therefore, both participant subsamples were balanced in terms of their proficiency in L1, L2, and L3, approximate age of onset of L1, L2, and L3, and the order of acquisition of the three languages. The results of the study showed that neither the psychotypology factor, operationalized as the perceived similarity between languages at the language system level, nor the L2 status factor was able to account alone for the frequency and source language of CLI in L3. The participants relied on both their L1 and their L2 in the production task in L3. However, each group of participants (CroL1 and ItaL1) relied statistically more frequently on the language in the role of L11 (Letica Krevelj 2014) whereby the same pattern of behavior was established. The study raised a few questions regarding the status of particular languages in participants’ repertoires and the point at which each language starts to matter or play a role in a multilingual’s makeup. Due to the requirements of the design, all the participants were consecutive multilinguals who started acquiring their L2 after the age of three. However, their exposure to the three languages in the community varied (see Sect. 4.1 and 4.2 below). Given that the study was designed in the way that it could capture both positive and negative occurrences of CLI and the extent of activation of both background languages within the same task, it is believed that it has a potential in explaining further the conditions under which the interaction between the languages exerted the most beneficial effect in L3 production (Letica Krevelj 2016b).

1  Two univariate analyses of variance confirmed a statistically significant effect of the language in the status of L1 on the frequency of CLI from Croatian (F (1,180)  =  16.17, p  =  .000) which explained 8.2% of the variance. The effect of psychotypology was not significant (F (1,180) = .54, p = .465), nor was the interactional effect of psychotypology and language status (F (1,180) = 1.53, p = .218) on the frequency of CLI from Croatian. The same is found in the case of CLI from Italian. There was a statistically significant effect of language in the status of L1 on the frequency of CLI from Italian (F (1,180) = 16.17, p = .000) which accounted for 13.1% of the variance. The effect of psychotypology was not significant (F (1,180) = .54, p = .465), nor was the interactional effect of psychotypology and language status (F (1,180) = 1.53, p = .218) on the frequency of CLI from Italian (for further discussion see Letica Krevelj 2014).

216

S. L. Krevelj

3.1.1  The Context of the Study The study was conducted in Istria County, the westernmost county of Croatia which includes the biggest part (89%) of the Istrian peninsula. Due to its turbulent history with constant ethnic, social, and political changes, Istria County is a county of great ethnic and linguistic diversity. In antiquity Istria was populated by Mediterranean peoples. The Romans conquered Istria in 177  BC, and with the arrival of the Slavs in the seventh and the eighth centuries, the Istrian region became a territory of coexistence of Romance and Slavic populations. At the beginning of the second millennium, part of Istria came under the Venetian rule. At the time, Latin was used as a High Variety and other Romance and Slavonic dialects were spoken as Low Varieties. Between 1470 and 1797, all of Istria was under the Venetian rule, and at that time, the Venetian koiné asserted itself as a variety with certain prestige, but with the arrival of new both Slavonic- and Romance-speaking settlers, different ‘mixed’ language varieties developed (Milani Kruljac 1990). After the collapse of the Venetian rule in 1797, Istria came under the Austrian rule, and Croatian, Slovenian, Italian and German were official languages, even though the Venetian koiné was still used as a lingua franca. After the First World War, Istria became part of Italy. Under the fascist rule all non-Romance idioms were banned and all Croatian schools were closed. After WW II, the majority of Istria became part of Yugoslavia, and between 1945 and 1955 approximately 90% of the Italophone population left Istria and the area was repopulated by people from different parts of Yugoslavia. In the Croatian part of Istria, the Croatian language became the official language (Letica Krevelj 2014 based on Bertoša and Matijašić 2005; Jahn 1999; Milani Kruljac 1990). Today, different linguistic varieties are spoken in Istria County: standard Croatian a variety of Croatian dialects, standard Italian, several Romance idioms (the Istrovenetian dialect, the Istroromanian dialect, and the pre-Venetian Istriot dialect). Besides Croatian (the official language of Croatia), in Istria County, Italian is recognized as the official language of the autochtonous National Italian Community. According to the 2011 Croatian census,2 only 6.83% of the population of Istria County reported Italian as their mother tongue, but they are not equally represented in all regions. There are regions in which almost exclusively either Italian or Croatian are used, which creates contexts in which the population is exposed to the two languages to a fairly different extent. Even though the standard Italian is the official language of the National Italian Community, in every-day communication Italian L1 speakers in Istria (usually referred to as Italophone speakers) use the Istrovenetian dialect. Therefore, sociolinguistic studies characterize the bilingualism in the community as functional primarily Croatian3/Istrovenetian bilingualism. However, sociolinguists agree that it

2  Statistical reports1469/2012. Population by mother tongue, by towns/municipalities, 2011 census. http://www.dzs.hr/default_e.htm

 with or wihout Čakavian dialect

3

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

217

is safe to assume that all Italophone speakers are bilingual while those Croatophone speakers are not necessarily so (e.g., Milani Kruljac 1990; Orbanić 1999; Poropat Jeletić 2017). Italian is used by the Italophone speakers in education, politics and media and it is used mainly in written discourse. Due to its marginalization from social interaction in the community, it has not developed jargon or its informal or colloquial varieties (see Poropat Jeletić 2017). Therefore, the Istrovenetian dialect has a greater ethnolinguistic prestige and higher communication status than the Italian standard (Poropat Jeletić 2017). Consequently, the amount of exposure and use of Italian in the community for the speakers of Croatian as L1 is fairly limited (Filipi 1989) and it is claimed that “the acquisition of the Italian language in the community is of instrumental rather than integrative character” (Scotti Jurić 2003: 384). In schools where Croatian is the main language of instruction, the Italian language is taught according to the curriculum for foreign languages, mainly through a functional-communicative approach. The same applies to the English language that was taught in all Croatian schools as a foreign language from the fourth grade of primary school. However, the omnipresence of English in Europe in different spheres of daily life is also felt in Istria County, where English is mostly present through different types of media: Internet, TV, advertising, and forms of popular youth culture and entertainment. Furthermore, in Istria, it is also often used as the language of the tourist industry. Since 2003, there has been a clear trend of choosing English, over Italian, as the first foreign language learned from the first grade of primary school (Medved Krajnović and Letica 2009; Scotti-Jurić and Poropat 2012). This trend seems to be related to the global influence of English that has affected other bilingual communities around Europe in similar ways (see Hoffmann 1998; Ringbom 2011).

4  Looking at the Data Through the Lens of DLC In this sect. I give a portrayal of participant data collected for the purposes of Letica Krevelj (2014) in two towns in Istria (Buje/ Buie and Rovinj /Rovigo). The data is organized and specified through the configurations of a DLC. The major criteria for the selection of participants in my study was their L1 and L2 being either Croatian or Italian whereby some other homogeneous characteristics such as the age of onset of each language, including the English language as L3, was imposed by the national and community educational system requirements. Schools in these two towns were selected as the greatest number speakers of Italian as L1 could be recruited there. While for the study purposes (Letica Krevelj 2014) it was not relevant whether the participants had either dialect or standard as their L1, assuming that we were

4  ‘Lingua che non serve all’atto della comunicazione sociale e il cui apprendimento assume carattere strumentale e non assolutamente integrativo (Scotti Jurić 2003: 38).

218

S. L. Krevelj

tapping into the same psycholinguistic reality,5 this fact will be portrayed in the configurations of the DLC and discussed further below.

4.1  Configuration of DLC of Croatophone Speakers (CroL1) CroL1 speakers are native speakers of Croatian, but not necessarily also the speakers of the Čakavian dialect. Their second language, in terms of the order of acquisition, is the Italian language which they started studying in the second grade of primary school as a ‘community language’. Their third language is English which they started studying as a foreign language at school at the age of 10. CroL1 speakers may function in their community using only Croatian while the other two languages (Italian and English) may be considered as foreign languages studied in a formal context found in any monolingual environment. Indeed, 44% of participants do not use the Italian language outside the education domain and at least 85% of CroL1 speakers use exclusively the Croatian language in day-to-day contacts with members of the community (Table  1). However, from the relative percentages of use of the three languages there is a lot of variability in the use of each language on daily basis. The percentages of use of Italian (L2) do testify to a greater use (or perceived use) of the L2 in their daily life. Besides the fact that they actively study the Italian language in school (3–4 contact hours a week), CroL1 speakers are surrounded by the Italian language (at least though the linguistic landscape and material culture). Some of the participants also chose to be educated through the medium of the Italian language. The choice of Italian as the language of schooling is often made by parents based on their preferences or ideologies, but it is also possibly motivated by other factors such as the fact that Italian-medium schools offer free books, field trips etc. Additionally, some participants reported it as the language most commonly used with friends and neighbors.6 CroL1 speakers study English as a foreign language in school, but English is also widely present in the community, particularly through media. The exposure to the English language is greatly enhanced by the fact that different TV programs and films in the English language are subtitled rather than dubbed on Croatian TV channels. The participants also reported using English in their daily life, but exclusively with either siblings or peers (younger generations). As for the relative proficiency in the three languages, it is safe to assume that Croatian is their dominant language. However, it is also interesting to point out that the participants did not asses their L1 competence with the highest grade. The self-­ assessed proficiency reported in Table 1 is believed to reflect proficiency in more 5  Through the instrument that captured instances of lexical CLI it was made certain that lexical items were the same in both the dialect and the standard language. 6  The use of either Italian or Istrovenetian was not specified. However, a sociolinguistic study conducted in the same geographical area reports Istovenetian dialect as the preferred language in daily communication of 5% of speakers of Croatian as L1 (Poropat Jeletić 2017).

Social domain

Frequency of use (order in terms of frequency of use) Domains of use Private domain (parents, siblings, relatives)

Order of acquisition Competencea (order in terms of relative level of competence) Age of onset Mode of learning

Less than 10% with siblings and relatives

90.25% use exclusively Croatian Classmates 91.3% Friends 88.9% Neighbors 85%

50–100% (1)

4–8 Largely instructed 0–50% (2)

From birth Acquired

Less than 20% with fathers, siblings and relatives Classmates 33.3% Friends 37.1% Neighbors 38.1%

Occasionally used with siblings Occasionally used with classmates and friends

25–50% (2)

4–7 Largely acquired

L2 3.87 (2)

National language

ItaL1 Croatian

0–20% (3)

10 Instructed

L3 3.79 (2)

Foreign language

Community language

L2 3.63 (3)

English

Italian

L1 4.17 (1)

Configurations Constitution CroL1 Genealogically different Croatian languages (Čakavian dialect) Status Mother tongue

Table 1  Constitution and configurations of the DLC

Not used

Occasionally used with friends

80.53% use exclusively Istrovenetian Classmates 66.7% Friends 62.9% Neighbors 61.9%

(continued)

0–10% (3)

10 Instructed

L3 3.56 (3)

Foreign language

English

50–100% (1)

From birth Acquired

Official language of the Italian National Community (Mother tongue) L1 4.32 (1)

Italian (Istrovenetian dialect)

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant… 219

Daily communication Education Media Material culture

Configurations CroL1 Largely Croatian 3–4 contact lessons a week Material culture Education

3–4 contact lessons a week Education Media

ItaL1 4 contact lessons a week Ever-day communication Education Material culture Education Media Material culture (Daily communication)

Largely Italian

3–4 contact lessons a week Education

a

The participants’ school grade at the end of the previous school year, as well as the participants’ self-assessment of general proficiency in each language on a scale of 1 (poor) to 5 (excellent). Due to the fact that the school curriculum imposed different criteria in grading L2s (Croatian and Italian) in the Croatian- and Italian-medium schools, the self-assessment values were taken as the measure of proficiency in the participants’ L2 in both groups. It was believed that self-­ assessment in a context where the language could be put to use outside of the school environment provided for a more realistic picture of their competence in that particular language. On the other hand, both grades and self-assessment values were taken together to represent the measure of participants’ proficiency in their L3 (English)

Domains of exposure

Constitution Education domains

Table 1 (continued)

220 S. L. Krevelj

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

221

relative terms as it is the one that is embedded and socially constructed. On the other hand, for some participants, the stronger language was English (L3), rather than Italian, despite the fairly lower age of onset and greater length of learning of Italian than English. This fact was attributed (see Letica Krevelj 2014) to the greater exposure of the participants to the English language through TV programs on Croatian channels. As far as the proficiency level in English as L3, in terms of the Common European Framework of Reference, the participants were identified by their teachers as being between levels B1 and B2.

4.2  Configuration of DLC of Italophone Speakers (ItaL1) ItaL1 speakers are almost exclusively speakers of Istrovenetian dialect as their L1, and they are mostly educated through the medium of the Italian language. While they are also fairly competent speakers of Croatian, measures of competence in Italian and Croatian show that they are not equally proficient in both languages and absolute proficiency in either was rarely reported. On the other hand, proficiency in Croatian is most commonly higher than their proficiency in English which is assessed by their teachers as ranging from B1 to B2. In the Italian-medium schools, the Croatian language is taught from the first grade of primary school, as the “national language” (domovinski jezik), and according to the same curriculum of the Croatian language (as the mother tongue) taught in Croatian-medium schools. Furthermore, it should be emphasized that the curriculum focuses more on Croatian grammar and literature than on communicative competence, whereby the focus is often on rote learning of language rules and paradigms (Pavličević-Franić 20027). Therefore, while the mode of learning of the Croatian language may be considered largely incidental, it is also instructed. From the data on the frequency of use and exposure to the three languages in the environment (Table 1) it is obvious that the majority uses the Istrovenetian dialect much more than the Croatian language. In the private domain, the exclusive use of Istrovenetian is reported by more than 80% of ItaL1 participants. On the other hand, the Croatian language is used by many up to almost equal extent as Istroventian. As presented in Table  1, higher percentages of use of Croatian fall into the social domains; communication with classmates, friends and neighbors.

7  From personal communication with an advisor for the Italian minority working in the Education and Teacher Training Agency of the Republic of Croatia, as well as with four teachers of Croatian in Italian-medium high schools, it was found that the demands of the Croatian syllabus do not match the Croatian language competence of the L1 Italian speakers. Some of the remarks provided by the teacherson participants’ competence are: “the learners are able to provide correct definitions of grammatical rules of the Croatian language, but these are often provided in incorrect/non-target like Croatian”, “the literature they read in Croatian is often not entirely understood as it is above their level of language competence”.

222

S. L. Krevelj

The English language is rarely used outside of the education domain and we may conclude that greater exposure to the Italian language in the environment and consequently to the media in Italian provides a context of relatively lower exposure to English (when compared to CroL1 sample) as programs in the English language are dubbed in Italian TV media.

5  A  ssessing the Potential of DLC for Studying CLI in Multilinguals In this section I will revisit the main points made by Aronin (2016) and provide arguments for the potential of the DLC, both as a perspective and as a tool, in studying CLI in multilinguals using a quantitative research design.

5.1  Defining a DLC While at first it may seem that defining one’s DLC is more useful for qualitative studies, I would like to argue that it is rather useful even in quantitative studies. Possibly the most obvious fact is that it helps in outlining more clearly the configurations of DLC which capture the sociolinguistic reality. In that respect DLC allows for a more detailed portrayal of multilinguals. More precisely, it allows for a precise analysis of the opportunities awarded by the social environment as well as speakers’ engagement with the environment. From the examples provided by Aronin (2016) to illustrate the compositions of individual’s DLC, it is obvious that defining dominant constellations, even for an individual, is not a straightforward endeavor. When dealing with quantitative studies, where one is faced with the issue of pre-defining the participant selection criteria, it is even more daunting. One aspect of the issue would be how the researcher conceives it for the purpose of using it as a criteria in participant selection (e.g., how much knowledge of the language one has to have in order for the language to be counted in (see De Angelis 2017) and where we draw the lines between two languages, or languages and dialects, which may be very similar. The other aspect would be how a multilingual functions with it (e.g., it may be dependent on the domain in which particular language may function as vehicle one). According to Aronin, DLC is concerned with the vehicle languages which stand out as being of prime importance. We may also raise the question of the ownership of the authority to define a DLC. DLC may be defined based on multilingual‘s perception and it may not be in line with the researcher’s perspective depending on the purpose/aim of the study. What seems to be difficult here is to reconcile the two perspectives: the researcher’s in labeling the groups within a DLC for the purpose of the study, and the multilinguals’ perception of their DLC (e.g., as being of prime

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

223

importance in daily life). The results of the study presented above (Letica Krevelj 2014) showed that within the production task both background languages played the role and mattered to CroL1 participants even though one would not necessarily consider the Italian language as a part of the DLC of CroL1 participants outside of the educational domain. It was shown that both their L2 and L3 mattered, and they served as vehicle languages just by the mere fact that these languages were used on weekly basis, even though only in the formal /instructed context. The DLC of this group of participants is most certainly time-framed as we acknowledge that their DLC may not be the same once the formal education is over. Their Italian may simply fall out from their DLC and feature only as a language in their linguistic repertoire. I would argue that the concept of DLC could be defined further depending on the purpose of study. The fact that the participants in Letica Krevelj (2014) have the three languages in their mind may be the result of various socially imposed circumstances, but in this case we ask the question of what they do with these languages. It is important to distinguish between two different categorizations: the one solely dependent on the immediate need for social interaction, and the one that leads to the co-construction of knowledge with the same languages present in their minds. We are most certainly interested in both biological and social aspect of that knowledge. Therefore, in defining the conditions in which languages may function as vehicle languages, we are able to differentiate between the origins of the co-constructed knowledge and the relationship between the two (biological and social) given that they are studied in context and as a unit. Besides the function of languages as means of communication, Aronin (2016) also mentions their function in expressing identity. In Letica Krevelj (2014) identity features prominently through the identification with the dialect as the true mother tongue in both participant configurations described above (CroL1 and ItaL1). The identity function is also present in cases where some speakers frequently use the dialect of their L2, as well as report it as their preferred code in daily communication (see also Poropat Jeletić 2017). Therefore, these identities are not necessarily easily defined along the lines of ethnic origins or their L1. The use of English with peers outside the education domain also testifies to the possible identification with English as the language of younger generations already reported in some studies in Croatia (see Ćurlin and Letica Krevelj 2016). Furthermore, in the bilingual community portrayed above, the use of English may function as a transgression of the differences imposed through identities assumed, established or imposed by the Croatian /Istrovenetian bilingualism in the community. I believe that this aspect of the function of language that goes beyond its instrumental value should be given more importance when defining a DLC.

224

S. L. Krevelj

5.2  D  LC as a Unit and a Complex System Consisting of Several Components I would like to argue that the most useful aspect of the concept of DLC is its inherent ability to function as a unit and, at the same time, as a complex system of different components. It enables us to take into consideration all of the constellation’s languages simultaneously when trying to understand how they function together in a particular context. Furthermore, it allows us to examine behavior patterns and differences that emerge within the same DLC constitution and across different configurations of the same DLC acknowledging the interaction within. With a precise definition of a DLC we may be closer to the model of reality whereby the differences in the established patterns (of reality) are assumed to be different not only in terms of the degree/intensity of each of the component parts, but also in the nature of the relationships they establish within the DLC. Rather than introducing the participant selection basis, we may define the unit based on the criteria of defining a DLC (Aronin 2016) which seems to be more natural, or circumstantial criteria. While some variables may feature more strongly, the focus is on all the same variables being present and examined in their interaction. Therefore, under the unity perspective, we may be avoiding the abovementioned reductionist/fragmental perspective in quantitative studies. For example, if we are a priori assuming that the low proficiency in one language has no effect on the phenomenon we are studying, we are unable to examine its effect in interaction with other variables. It has to be emphasized that the question of when a particular language has an effect on the other, or when it starts to matter in overall multilingual proficiency, is a complex one as it can have either beneficial or constraining effect. From the perspective of the DLC as a unit we are striving for a more holistic approach in research whereby the interactional effect of both low and high proficiency with additional variables is given due regard. In this way, I believe, we can avoid the problems inherent in the definition of multilinguals (in terms of the number of languages and the assumption that we can determine the boundaries between them) and work with naturally formed, socially-determined or “circumstantial” categories which can be further examined in the light of their indexicalities. I have previously emphasized that the existence of the Istrovenetian dialect as a somewhat confounding variable in Letica Krevelj (2014). Examined through the lens of DLC, it seems more significant than previously assumed. Its existence affects the use and the perception of the communicative need of the Italian language in and out of the immediate social reality. Additionally, due to the fact that the dialect in question has been evolving and changing in contact with the other languages, it is much more easily described along the similarities and differences in its structural description than in the minds of users. It is logical to assume that the relationships established by those who are using it along the standard version, and often have it consciously compared and contrasted (ItaL1), and those who potentially cannot tell the difference between the two (CroL1), would not be the same. So far it was possible to conclude only that the existence of the dialects in the constitution of this

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

225

DLC acted as a confounding variable when studying the nature of perceived similarity between the two officially recognized standard languages and English, as each dialect has some elements of both the Croatian and Italian language (see Letica Krevelj 2016a). The existence of dialects which have a strong identity function and which necessarily ‘diverge’ from the norms of standard languages may be consciously or subconsciously embraced or rejected due to their value in the social environment. The unity perspective allows us to acknowledge that the competences in each of the two (standard language and its dialect), may also interact in complex ways.

5.3  DLC as a Research Tool Some distinctive properties of the two configurations of DLC were isolated in Table 1 above, but they have not been examined in detail (for various reasons) in empirical studies, where such data potentially carries true explanatory power. Through the DLC I believe we are tapping into patterns that may be further exploited to a greater detail depending on what it is that we wish to study. If we assume that the speaker and environment are embodied and embedded within the DLC, through the patterns of behavior we are able to tap into a more holistic perspective of what variables, how many of them and in what configuration they may have an effect on the behavior being studied. If, in turn, the outcome may be defined in terms of desirable or less desirable outcome, as it is the case in Letica Krevelj (2014), we can come up with configuration pattern that is conducive to beneficial or advantageous behavior of multilinguals. What is more, DLC may help us identify the ways the constituent languages work together or where they constrain one another. Therefore, instead of using a pre-determined categories and definitions of multilinguals and examining the effect of separate variables on the phenomena under study, we can look at the constitution of DLC (as a unit). At the same time, given that we are able to look at the distinct configurations, we are able to see how each configuration pattern plays out, and then try to look for explanation in retrospect. For example, comparative measure of exposure to the languages within the DLC, rather than the variable of exposure to each of the three languages, is much more informative in terms of its overall effect on multilingual behavior. Additionally the variability found in the data allows us to examine the conditions in which even very low proficiency in a language or slight exposure matters, and when it is that a greater amount of exposure is necessary for the phenomena to be observable. Drawing from the data in Letica Krevelj (2014), we can see that the similar amount of reliance on L2s by CroL1 and ItaL1 participants in L3 production task was found despite a great variability in the amount of exposure to L2 out of the domain of education. However, different degrees of proficiency in each language relative to the other two may be examined further by creating even smaller configuration patterns as the units of analysis.

226

S. L. Krevelj

Based on the arguments put forward for the application of DLC as a tool in multilingualism research (Aronin 2016), and prior to the analysis of data through the lens of DLC, I believed the greatest value of DLC in quantitative studies may be in its ability to provide an additional socially embedded qualitative perspective to the quantitative data in the study. However, through the description of the DLC and the isolated configuration patterns, numerous still unresolved issues emerged concerning the variables purported to explain the nature of CLI in multilingual behavior. The variables of proficiency, language exposure and mode of learning, when examined in the context of this study through the lens of DLC, seem to question the possibility of ever explaining the phenomenon in a holistic way by looking at the cause and effect relationships. These most certainly require further inspection and, due to the space limitations of this chapter, will be discussed elsewhere. I believe that DLC helped further in determining clusters of variables whose relative weight can be tested through statistical modelling in order to study the phenomenon of CLI. However, I would like to promote the idea of retrodiction (Larsen-Freeman 2009), stemming from epistemological basis of Dynamic Systems Theory, as the approach in further data analysis. Rather than trying to predict the outcome, I propose that we should be taking the outcome, with all of the established tendencies and variability, as a starting point. It is only then that we can try to describe the phenomena found in the production of multilinguals with different configuration patterns to examine the conditions (or even affordances) conducive to a particular behavior. Therefore, studying the configurations of the same DLC allows us to examine all of these variables and their interdependence in context. More specifically, along the lines of Letica Krevelj (2014), it may tell us under which circumstances, measured and captured through different configurations of DLC, particular background language may play a role in additional language acquisition. I would like to argue that it further helps in explaining not only the frequencies and source language of CLI but even the nature of CLI if unexpected outcomes and variability in the data is examined in retrospect against macro and micro configurations of a DLC.

6  S  upplementary and Complementary Aspect of DLC in Studying CLI The study by Letica Krevelj (2014), used to exemplify the arguments in favour of the concept of DLC, aimed to examine the relative weight of specific constructs of interest in studies on CLI (see Sect. 3). However, it was designed in a way that it could capture a simultaneous interaction of different languages in multilinguals through which examples of both benefits and constraints that act upon the performance on the task could be isolated. It is through the description and analysis of data through DLC and novel findings that the potential of the study became obvious.

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

227

I would like to acknowledge that DLC allows for detailed analysis of interconnections and interactions between languages, language users, and the environment (Aronin 2016), and this is where I see the greatest value of DLC concept for the line of study. Looking at three languages simultaneously allows us to: 1. Define the similarities and differences across the constitution and the configurations within a DLC which portrays the social reality to which participants not only adapt, but also interpret in different ways in terms of both realistic and perceived communicative needs. 2. Create more precisely defined patterns of configurations based on the interdependencies within the same constellation which can be further analyzed according to the research aims. I agree with Aronin (2016) that DLC is complementary with holistic approach, and the view of language use as adaptable, dynamic and interdependent. It is also complementary to the call for better integration of sociolinguistic and psycholinguistic aspects in providing a holistic model of multilingual behaviour. A supplementary aspect of the DLC to our field of interest, as argued in the chapter, is a granted possibility, by a definition of DLC, to avoid a reductionist approach to research. The other aspect is the conceptualization through two faces of DLC which allows for a clearer distinction between two opposing perspectives of reality; the one seen as a unity and the other as the diversity of its component parts. This allows for simultaneously the generalizability and insight into particularities of multilingual behavior. Scooping large under the umbrella of DLC we are given the opportunity to study CLI phenomena through quantitative research (and in a holistic way) and analyze under what conditions previous linguistic knowledge matters in the process of L3 acquisition. We are able to take into consideration numerous factors, such as proficiency achieved in one language relative to that achieved in other constituent languages of the DLC.  The performance of multilinguals (tendencies, variability or possibly the data classified in terms of more or less desirable outcomes) should be examined against the configurations of DLC that are conducive to particular behaviour in the process of learning an L3. It is only in retrospect that these variables and their interrelatedness can be inspected. The description of the configurations of the DLC above revealed some divides between the multilinguals in the community. These may be similar across different bilingual/multilingual communities or they may be unique but they may be at the same time important for the phenomena being studied. If we use DLC as a unit, instead of searching for cause and effect relationships, we should be working with patterns that can be further compared to different DLCs and configurations within the DLC. It may give us more precise grounds on which meta-analysis of research data on CLI could be based. For example, the participants in Letica Krevelj (2014) may be similar to those in other studies conducted in the similar bilingual environments where the third language (English) is introduced through the educational system. There is a great possibility that this comparison may bring about a greater explanatory power when examining (maybe only seemingly) contradictory results

228

S. L. Krevelj

in studying multilingual performance. On the other hand, studying CLI, or other practices and outcomes, of a specific DLC has a true ecological validity and may offer direct implications for language policy and language education.

References Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings (pp. 140–159). Hohengehren: Schneider Publications. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence (pp.  142–163). Cambridge: Cambridge University Press. Aronin, L., & Jessner, U. (2014). Methodology in bi- and multilingual studies: From¸ simplification to complexity. AILA Review 27: Research Methods and Approaches in Applied Linguistics: Looking Back and Moving Forward, 27, 56–79. Aronin, L., & Ó Laoire, M. (2004). Exploring multilingualism in cultural contexts: Towards a notion of multilinguality. In C. Hoffmann & J. Ytsma (Eds.), Trilingualism in family, school and community (pp. 11–29). Clevedon: Multilingual Matters. Bertoša, M., & Matijašić, R. (Eds.). (2005). Istarska enciklopedija. Zagreb: Leksikografski zavod Miroslav Krleža. Census 2011 of the Republic of Croatia: http://www.dzs.hr/default_e.htm Ćurlin, P., & Letica Krevelj, S. (2016). About the limits and boundaries: code-switching of English majors in Croatia. Paper presented at the International Conference of The Croatian Association for the Study of English. University of Zadar, November 18–19. De Angelis, G. (2005a). Multilingualism and non-native lexical transfer: An identification problem. International Journal of Multilingualism, 2(1), 1–25. De Angelis, G. (2005b). Interlanguage transfer of function words. Language Learning, 55, 379–414. De Angelis, G. (2007). Third or additional language learning. Clevedon: Multilingual Matters. De Angelis, G. (2017). Dealing with multilingualism in quantitative research. In J.  McKinley & H.  Rose (Eds.), Doing research in applied linguistics (pp.  91–100). London/New York: Routledge. De Angelis, G., & Selinker, L. (2001). Interlanguage transfer and competing linguistic systems in the multilingual mind. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 42–58). Clevedon, UK: Multilingual Matters. Dewaele, J.-M. (1998). Lexical inventions: French interlanguage as L2 versus L3. Applied Linguistics, 19(4), 471–490. Falk, Y., & Bardel, C. (2010). The study of the role of the background language in third language acquisition. The state of the art. International Review of Applied Linguistics in Language Teaching. IRAL, 48(2–3), 185–220. Filipi, G. (1989). Situazione linguistica Istro-quarnerina. Ricerce sociali, 1, 73–82. Fouser, R. J. (2001). Too close for comfort? Sociolinguistic transfer from Japanese into Korean as an L≥3. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp.  149–169). Clevedon: Multilingual Matters. Hall, C.  J., & Ecke, P. (2003). Parasitism as a default mechanism in L3 vocabulary acquisition. In J.  Cenoz, B.  Hufeisen, & U.  Jessner (Eds.), The multilingual lexicon (pp.  71–87). Dordrecht: Kluwer.

Studying Crosslinguistic Interaction in Multilingual Production Through the Dominant…

229

Hammarberg, B. (2001). Roles of L1 and L2  in L3 production and acquisition. In J.  Cenoz, B.  Hufeisen, & U.  Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 21–41). Clevedon: Multilingual Matters. Hoffmann, C. (1998). Luxemburg and the European Schools. In J. Cenoz & F. Genesee (Eds.), Beyond bilingualism: Multilingualism and multilingual education (pp.  143–174). Clevedon: Multilingual Matters. Jahn, J. E. (1999). The political, ethnic and linguistic borders of the Upper Adriatic after the dissolution of Yugoslavia. Poznań Studies in Contemporary Linguistics, 35, 73–81. Larsen-Freeman, D. (2009). Prediction or retrodiction?: The coming together of research and teaching. In K.  Losey & C.  Pearson (Eds.), Spotlight on re-search: A new beginning. The selected proceedings of the 2008 MITESOL conference (pp. 5–16). Raleigh: LuLu Press. Letica Krevelj, S. (2014). Crosslinguistic interaction in acquiring English as L3: role of psychotypology and L2 status. Unpublished doctoral thesis. University of Zagreb, Faculty of Humanities and Social Sciences. Zagreb. Letica Krevelj, S. (2016a). Multilinguals’ perceptions of crosslinguistic similarity and relative ease of learning genealogically unrelated languages. SRAZ, 61, 175–206. Letica Krevelj, S. (2016b). The L2 status vs. psychotypology debate … and beyond. Paper presented at the 10th International Conference on Third Language Acquisition and Multilingualism, University of Vienna, September 1–3. Medved Krajnović, M., & Letica, S. (2009). Učenje stranih jezika u Hrvatskoj: politika, znanost i javnost. In J. Granić (Ed.), Jezična politika i jezična stvarnost (pp. 598–607). Zagreb: Croatian Association of Applied Linguistics. Milani Kruljac, N. (1990). La communità Italiana in Istria e a Fiume fra diglossia e bilinguismo. Rovigno: Centro di Ricerche Storiche. Orbanić, S. (1999). Rana hrvatsko-talijanska dvojezičnost u miješanim obiteljima u Istri. Doctoral dissertation: Humanities and Social Sciences, University of Zagreb. Pavličević-Franić, D. (2002). Analiza programa hrvatskoga jezika za niže razrede osnovne. In M.  Kovačević & D.  Pavličević-Franić (Eds.), Komunikacijska kompetencija u višejezičnoj sredini I: prikazi, problemi, putokazi (pp.  19–24). Zagreb: Naklada Slap and Sveučilište u Zagrebu. Poropat Jeletić, N. (2017). Italofona dijasistemska raslojenost u hrvatskoj Istri: jezični i komunikcijski status, korpus i prestiž. Annales: anali za istrske in mediteranske študije, 27(1), 191–204. Ringbom, H. (2001). Lexical transfer in L3 production. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 59–68). Clevedon: Multilingual Matters. Ringbom, H. (2011). Perceived redundancy or crosslinguistic influence? What L3 learners’ material can tell us about the causes of errors. In G. De Angelis & J.-M. Dewaele (Eds.), New trends in Crosslinguistic influence and multilingualism research (pp.  19–24). Bristol: Multilingual Matters. Scotti Jurić, R. (2003). Bilinguismo precoce: funzione e usi linguistici. Rijeka: Edit. Scotti-Jurić, R., & Poropat, N. (2012). Bilingual official policy and ideologies in formal interactions and development of plurilingualism and interculturalism in informal interactions among the young Croatian-Italian bilinguals in Istria. Journal of Teaching and Education, 1(6), 415–432. Williams, S., & Hammarberg, B. (1998). Language switches in L3 production: Implications for a polyglot speaking model. Applied Linguistics, 19(3), 295–333.

A Dominant Language Constellations Case Study on Language Use and the Affective Domain Richard Nightingale

Abstract  Globalisation, international mobility, and new technologies make current multilingualism qualitatively different to not only mono- and bi-lingualism but also to any of its historical incarnations. As a new linguistic dispensation (Aronin L, Singleton D: Int J Multiling 5(1):1–16, 2008; Aronin L, Singleton D: Multilingualism. John Benjamins, Amsterdam, 2012; Aronin L: Current multilingualism and new developments in multilingual research. In: Safont P, Portoles L (eds) Multilingual development in the classroom: current findings from research. Cambridge Scholars Publishing, Newcastle, 2015), current multilingualism is understood to be complex, suffusive, liminal, and super-diverse; four essential properties which necessitate alternative foci in multilingual research. From this view, multilinguals are the ‘glue’ that binds cultures and societies, it is therefore essential to focus on their socioculturally situated multilingual practices if we are to better understand the protagonists of this considerable social responsibility. In this regard, two concepts of profound interest are multilinguality (Aronin L, Ó Laoire M: Exploring multilingualism in cultural contexts: towards a notion of multilinguality. In: Hoffmann C, Ytsma J (eds) Trilingualism in family, school, and community, vol 43. Multilingual Matters, Clevedon, 2004) and, its expression/realisation in concrete time frames and sociocultural contexts, dominant language constellations (DLC – Aronin L, Dominant language constellations: an approach to multilingualism studies. In: Ó Laoire M (ed) Multilingualism in educational settings. Schneider, Hohengehren, 2006; Aronin L: Multicompetence and dominant language constellation. In: Cook V, Wei L (eds) The Cambridge handbook of linguistic multicompetence. Cambridge University Press, Cambridge, 2016). With this in mind, the current study examines the individual DLC of a Moroccan-born man living in the Valencian Community in Spain. In line with earlier research calling for more varied self-report data (Todeva E, Cenoz J: The multiple realities of multilingualism. De Gruyter Mouton, Berlin, 2009; Canagarajah AS, Wurr AJ: Read Matrix 11:1–15, 2011; Gorter D, Cenoz J: Mod Lang J 95:442–445. https://doi.org/10.1111/j.1540-4781.2011.01203.x, 2011), the study uses semi-structured interviews to investigate relationships between the R. Nightingale (*) Universitat Jaume I, Castellón de la Plana, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_12

231

232

R. Nightingale

DLC and the affective domain; specifically, attitudes, emotion, and identity. The resulting qualitative data explores the following questions: How does a multilingual speaker use their DLC to navigate specific sociolinguistic contexts? What influence does the DLC have on the expression of identity and emotions in concrete daily situations? What role does the DLC play in the formation of language attitudes? Moreover, these issues are framed within an acculturation context to articulate what Canagarajah and Wurr (Read Matrix 11:1–15, 2011) call ‘voices from the periphery’. While accepting that a case study limits any attempt at generalisation, it is hoped that this research focus may contribute by providing another small piece to the overall puzzle of multilingual practices realised in concrete social and cultural contexts. Keywords  Dominant language constellations · Multilinguality · Affective factors · Emotions · Attitudes · Identity

1  Introduction Aronin (2019) points out the importance of analysing language practices associated with the contemporary human condition. Globalisation has massively impacted our understanding of diversity and homogeneity, and how languages are used in super-­ diverse settings. This has led to the notion of multilingualism as a new linguistic dispensation (Aronin and Singleton 2008, 2012). Within this framework, it is now considered normal that individuals and communities essentially require sets of languages to cover communicative functions (Singleton et al. 2013); this is embodied in the concept of Dominant Language Constellations (Aronin 2006, 2016). However, while this concept has been well developed theoretically, to date there is relatively little work which applies insights deriving from this theory. For this reason, the current chapter presents a case study focusing on the dominant language constellation of a Moroccan-born man living in Spain, specifically on how his most expedient languages are reconfigured according to the multilingual environment and how they relate to his emotions, language attitudes, and identity construct.

2  A  New Linguistic Dispensation, Multilinguality, and Dominant Language Constellations Pioneering research in SLA and bilingualism (Cook 1992; Grosjean 1989) has lead multilingualism studies towards a more holistic, complex, and dynamic perspective (Jessner 2008, 2013; Todeva and Cenoz 2009; Cenoz and Gorter 2011). This focus has shown how multilingual systems interact from a psycholinguistic perspective (Herdina and Jessner 2002) and how they are eminently dependent on social contexts (Cenoz 2013). One of the most comprehensive theoretical frameworks for

A Dominant Language Constellations Case Study on Language Use and the Affective…

233

situating multilingualism in the postmodern social context is Aronin and Singleton’s (2008, 2012) concept of the new linguistic dispensation. These authors propound the idea that the scale and significance of multilingualism has reached a ‘critical point’ due to physical, economic, societal, and technological developments. Thus, in contrast to the rather supplementary nature of historical multilingualism, its modern counterpart is vitally interwoven with societal processes in all areas and at all levels (Singleton et al. 2013). As a new linguistic dispensation, multilingualism is characterised by four specific properties: complexity, suffusiveness, super-diversity, and liminality (Aronin and Singleton 2008, 2012; Aronin and Hufeisen 2009; Singleton et al. 2013; Aronin 2015). The way in which these properties are realised in concrete situations (see: Aronin 2015) means that research has moved away from a perceived monolingual/Anglocentric bias (Cook 1997) and new focal issues have emerged; dealing with diversity, identity, and emotions, among others. Two of the most pertinent focal issues to have arisen in this regard are multilinguality and dominant language constellations. While multilingualism refers to both the process of acquiring various non-native languages and the subsequent product at the level of society (Cenoz and Genesee 1998), multilinguality refers to the ‘inner constructs of a single speaker’ (Aronin and Ó Laoire 2004: 16). This means that while multilingualism is understood as the capacity for an individual to effectively and appropriately use several languages, multilinguality is understood as an inherent and intrinsic characteristic of that individual. Aronin and Singleton (2012: 80) point out that multilinguality relates to ‘individual aspects of multilingualism in their entirety’, and takes into account the influence of the social milieu as well as the cumulative effects that emotional, psychological, and linguistic aspects may have on a person. Thus, multilinguality goes beyond language per se to encapsulate aspects of identity. Moreover, multilinguality sees identity holistically, integrating sub-identities (cultural, national, gender, etc.) and language profile. In fact, multilinguality reflects ‘emotions, attitudes, preferences, anxieties, and personality type’ and has a reciprocal effect on ‘affective states and attitudinal orientations’ (Aronin and Singleton 2012: 81). Multilinguality is a unique possession of each multilingual individual. It is contingent on sets of languages and mastery of the languages in the sets, but also stretches further to include ‘cognitive and linguistic abilities, potential to gain knowledge, self-image as a language learner, preferences and the tangible impact of the cultural context’ (Aronin and Ó Laoire 2004: 19). The multilinguality that characterises a multilingual individual is bound to a specific time frame and cultural context and is expressed and realised from one context to another, from one moment to another, through that individual’s dominant language constellation. Dominant Language Constellation (henceforth: DLC  – Aronin 2006, 2016) refers to an individual’s most expedient vehicle languages, and the fact that they function as a whole. The DLC is complementary to, but quite distinct from, the language repertoire (Aronin and Singleton 2012). Language repertoire denotes the totality of language varieties shared by a group, or the linguistic skills and collective communication resources available to a group or individual (Gumperz 1964; Blommaert 2013). Conversely, DLC denotes the vehicle languages which are of ‘prime importance’ and ‘fulfil functions that are vital for a person or for a

234

R. Nightingale

community’ (Aronin and Singleton 2012: 62–63). The languages that make up the inner circle of the DLC tend to fulfil different functions and may have differing status within a given context. A DLC often comprises ‘an international language, a regional lingua franca, one or more state languages and one or more minority (local or immigrant) languages’ (Aronin and Singleton 2012: 60). In short, the DLC concept describes an individual or community’s most important languages in any given context, not necessarily all the languages present in their repertoire but those which allow them ‘to meet all needs in a multilingual environment’ (Aronin 2019: 15). A DLC is inherently dynamic and, therefore, has no reason to be constant over the same environment. The constituent languages may have different ‘weights’ depending on a user’s social-environmental needs and mastery of each language. Aronin and Singleton (2012) refer to this as the configuration of the DLC. Thus, DLCs are subject to dynamic reconfiguration as temporal change or sociocultural context makes one or another constituent language more pertinent to the user. DLCs are also indexical of certain states of affairs regarding the individual (origin, ethnicity, residence, social status, etc.) and regarding the constituent languages (usefulness, prestige, etc.) These qualities of dynamism and indexicality tend to reflect a person’s life trajectory, meaning the DLC may fluctuate or shift over the life span; that is, historical, personal, social, and environmental change may cause DLC reconfiguration. According to Aronin and Singleton (2012: 69), a DLC is ‘an evolving, emerging whole which transcends its parts’ and compels us to consider ‘whole sets of languages as units rather than […] the specific languages used by given individuals or groups’. Although DLC as a theoretical perspective is now well established, studies which apply this theory are as scarce as they are innovative and novel. Such work has hitherto only been presented as conference papers. The following state-of-the-art studies apply a DLC lens to issues as wide-ranging as: language diversity among teacher-training students (Sjoholm et  al. 2016); word guessing strategies (Otwinowska-Kasztelanic 2016); language interaction during production tasks (Letica Krevelj 2017); the relationship between external factors and individual DLC evolution (Kannangara 2017); teacher education (Björklund et  al. 2018); South African multilingual repertoires (Coetzee Van Rooy 2018); syntactic development (Fernández-Berkes and Flynn 2018); language system dynamics (Mayr-Keiler and Jessner 2018); language awareness in educational settings (Osterkorn 2018); children’s educational self-efficacy (Schwarzl 2018); family language and language education policies (Slavkov 2018; Vetter 2018); and, plurilingual competences in primary schools (Sugrañes 2018). Aronin (2019) has called for more descriptive studies on DLC, particularly which DLCs are essential for communities and individuals. She proposes various methods to research DLC as a unit and as a pattern, and invites work which explores how the constituent languages cooperate with or impede each other in various contexts, in different multilinguals, and under the cultural and historical influence that each language brings. In response, the current volume constitutes a first attempt to formalise and extend the application of the DLC framework to a range of specific situations of language use, and includes contributions from a number of the scholars mentioned above.

A Dominant Language Constellations Case Study on Language Use and the Affective…

235

In summary, the literature on multilingualism as a new linguistic dispensation, multilinguality, and DLC makes their intrinsic connection to the affective domain abundantly clear. However, it would appear that there is very little work which explicitly examines affective factors from this theoretical framework. Thus, in order to advance the current chapter, the next section will focus on existing work in the areas of sociolinguistics, SLA, and multilingualism which considers such affective factors as emotions, attitudes, and identity.

3  Affective Factors: Emotions, Attitudes, and Identity Earlier applied linguistics research covers many areas of the affective domain (see: Brown 1987). However, most of this research has been carried out in SLA contexts, essentially taking a monolingual perspective to the question of how the affective domain impacts language acquisition or learning. Sociolinguistics and multilingualism research provides deeper insights into affective factors such as emotions (Pavlenko 2005; Dewaele 2010) and identity (Dörnyei 2005; Norton 2013), and has vastly broadened our understanding of language attitudes (Pavlenko and Blackledge 2004; Dewaele 2005; Garrett 2010). Affective factors tend to overlap, and it is clear that emotions, attitudes, and identity are dynamically and inextricably intertwined. For instance, an attitude is the emotional expression of the external context in which the individual is situated and an emotion is an intuitive feeling which may influence attitudes towards external contexts and circumstances. Furthermore, as McCarthy (1994: 275) points out, ‘emotions have come to serve as one of the principal experiences of self-validation [...] from which to claim an identity and to build a self-­ conception’. To understand these factors and their relationship with language in bi−/multilingual contexts better, the balance of this section will summarise insights deriving from previous research regarding each one in turn. Emotions have a great deal to do with how and when we decide to use our languages. SLA research shows that the L2 has less emotional resonance than the L1 (Marcos 1976), meaning that L2 communication is often more cerebral. L2 use may also distance a speaker from their utterances (Bond and Lai 1986), meaning that language choice may act as an emotional shield. However, multilingualism research proposes that such language roles are not fixed because the affective impact of different languages is contingent on speakers and contexts. In fact, multilinguals ‘may use [their] languages to index a variety of affective stances, and they may also mix two or more languages to convey emotional meanings’ (Pavlenko 2005: 131). SLA research also relates language choice to emotions, specifically regarding compatibility with self/social image (Schumann 1997). However, multilingualism research considers this perspective to be reductive as it does little to recognise the relational nature of emotions, which may lead individuals towards certain language choices that are not easily rationalised academically (Pavlenko 2005). For instance, although a major factor in language choice for emotional expression is language dominance, very intense emotions or specific emotional language functions, such as

236

R. Nightingale

reprimanding children, also play a notable role (Pavlenko 2004). Moreover, languages influence how we express emotions, especially strong ones like love or anger, and the ‘language of the heart’ may not always be the L1. For example, perceptions of the emotional force of taboo words are mainly determined by linguistic history, and, in moments of extreme emotion, swearwords are often used in the L1 regardless of their comprehensibility to the interlocutor (Dewaele 2004). The highly charged phrase I love you usually has more weight in L1, however the weight can be equal in L1 and Lx, and occasionally even shift to the Lx (Dewaele 2008). Emotional speech acts are culture-specific, and bicultural multilinguals frequently navigate conflicting sociopragmatic norms. Codeswitching is often strategic when dealing with emotionally charged topics, but it can also be involuntary when particularly strong emotions are aroused (Dewaele 2010). Multilingual inner-speech is most frequently expressed in the L1, especially if it is of an emotional nature (Dewaele 2015). However, successfully acculturated migrants often come to assign this role to the host culture language (Panicacci and Dewaele 2017). In sum, language perceptions and choices regarding emotional expression in multilinguals is multifaceted, linked to past language experience, current language use, and sociobiographical and psychological factors (Dewaele 2013). Language attitudes constitute an essential affective variable in language learning (Dewaele 2005), and arise from complex, dynamic interactions of societal and individual factors (Cenoz 2009; Nightingale 2016). European studies highlight a number of variables significant in the formation of language attitudes. L1 is a major factor, as is regional, national, and politico-cultural identity, while other factors include media contact, self-perceived competence, and degree of multilingualism (Hoare 2001; Dewaele 2005; Lasagabaster and Huguet 2007). Language attitudes contribute to language choice (Gorter et al. 2001), are dependent on perceived symbolic capital and the economics of the linguistic marketplace (Gardner-Chloros et al. 2005), and are positively affected by intercultural contact (Fisher and Evans 2000; Dörnyei and Csizér 2005). Spanish studies show that the following variables have a significant impact: degree of linguistic competence, hometown size and dominant language, linguistic model of education, sociocultural, socioeconomic, and socio-professional status, and visiting or staying in a TL country, or participating in a study abroad programme (Lasagabaster and Huguet 2007; Lasagabaster and Sierra 2009; Nightingale 2012; Portolés 2014). Finally, while there is little research which takes a holistic approach to language attitudes, that which does exist shows positive attitudes to languages in contact (Lasagabaster and Safont 2008) and that attitudes are subject to socialisation processes and complex temporal and environmental interactions (Portolés 2015; Nightingale 2016). Identity and language are inseparable, and language mediates and expresses the link between individual and social identities (Tabouret-Keller 1998). Understood as a socially-situated practice, we use language to ‘accommodate, resist, subvert, and/ or transform the acts, stances, and activities that constitute particular social identities’ (Duff and Talmy 2011: 108), and languages themselves constitute symbolic resources through which ‘identities are forged, tried on, accommodated, imposed, resisted and changed’ (Harklau 2007: 649). Norton (2013: 2) establishes that linguistic identities are often ‘socially constructed in inequitable relations of power,

A Dominant Language Constellations Case Study on Language Use and the Affective…

237

changing across time and space’. Thus, poststructuralism allows us to see how language can organise society and construct subjectivities, moving beyond idealised and homogeneous signs to constitute a site of struggle as the signifying practices of heterogeneous linguistic communities clash over ‘conflicting claims to truth and power’ (Norton 2013: 54). Because of the fluid and multifarious nature of multilingualism, multilingual linguistic identity construction and performance processes are necessarily more complex and dynamic (Pascual and Rothman 2013). As multilingual identity is realised and negotiated through daily performance, feelings of authenticity is a prominent issue, especially in emotionally charged situations. Research shows that multilinguals tend to feel more authentic in their L1 and more fake in languages which were learned later in life (Pavlenko 2005; Dewaele 2010). Moreover, in online identity marking practices, multilinguals employ a wide range of linguistic and semiotic resources in order to express a unified identity through social networking platforms (Schreiber 2015). Finally, coexisting languages and cultures may induce a sense of hybrid identity in the minds of migrants; depending on how they are able to regulate their emotional responses, they may perceive their switching of languages as enriching rather than alienating (Panicacci and Dewaele 2017). In summary, the literature reviewed above highlights how emotions, attitudes, and identity have been explored in different areas of applied linguistics research. What we have not seen is any work connecting the affective domain to DLC. Therefore, we identify a research gap in that there is currently no work which (1) examines aspects of the affective domain using DLC as a framework, and (2) explores the impact of DLC on the specific affective factors of emotion, attitudes, and identity.

4  The Study 4.1  Aim of the Study To gain a more in-depth understanding of multilingual identity, Aronin and Singleton (2012) and Aronin (2019) have called for further descriptive work examining the relationships between DLC languages and their distinct functions in varying contexts. They also call for the exploration of attitudes towards the constituent languages and those languages within the DLC which have emotional force. In response, this chapter takes a DLC perspective to the use of languages in different contexts and the role of the DLC in terms of emotions, attitudes, and identity. The study explores the following questions: 1. How does a multilingual speaker reconfigure their DLC to navigate specific sociolinguistic contexts? 2. What influence does the DLC have on the expression of identity and emotions in concrete daily situations? 3. What role does the DLC play in the formation of attitudes towards languages and language use?

238

R. Nightingale

4.2  Participant As the current chapter is a case study, there is only one participant: Mehdi (henceforth referred to as M), a 37-year-old male, born in Dar Ould Zidouh in the Beni Mellal province of Morocco. M completed his obligatory education in Morocco and subsequently reached a postgraduate level in Spain. After briefly living in Switzerland, he settled in the Valencian Community in 2003. He currently works as an intercultural mediator for the Integration and Social Coexistence Mediation Agency, a local government-funded initiative set up and operated by the town hall of Castelló de la Plana. M represents a highly successful case of foreign-national integration. His long-term partner is Spanish and he has many Spanish friends. Furthermore, he enjoys permanent residency status, and has recently completed the process of gaining full citizenship. Alongside his L1s, Darija and Modern Standard Arabic, he has native-like command of French and Spanish, a good working knowledge of Catalan, basic competence in Italian, and some knowledge of English, albeit somewhat rudimentary. In this sense, M’s DLC can be represented by Fig. 1 below:

4.3  M  ultilingual Settings in the Valencian Community and Morocco Both the Valencian Community and Morocco constitute super-diverse sociolinguistic settings. The Valencian Community is characterised by use of Spanish and Catalan (Valencian). There is ubiquitous contact with English in both public and Chinese

Rumanian

Bulgarian

Russian

Ukrainian

Niger-Congo

French DLC

Italian

SPANISH DARIJA ARABIC

Catalan

Fig. 1  The participant’s DLC

English

A Dominant Language Constellations Case Study on Language Use and the Affective…

239

private education sectors, and, a notable immigrant population has brought with them a number of heritage languages (Rumanian, Arabic, Chinese, Russian, and numerous African languages). The unequal prestige that exists between these languages results in a palpable linguistic conflict: on one side, Spanish and English (more prestigious), and on the other, Catalan and specific community languages (less prestigious). This community provides a rich linguistic context which requires more investigation through multilingualism research (Safont 2015). Morocco is characterised by use of Modern Standard Arabic (MSA) and Darija alongside four main Berber languages and some western European languages. MSA is a high-­ prestige language used in official/formal contexts. Darija is widely used in every-­ day informal situations. As Darija is primarily a spoken language, for many Moroccans it is the original ‘mother tongue’, while MSA is learned later in formal education. Berber serves a vernacular function; it is a low-prestige language common in many rural areas. French is a high-prestige language used for diplomacy, government, and international commerce; it is also a vehicular language in education. Spanish is spoken in Northern Morocco and Western Sahara; it is also widely studied as a second language. Finally, English is rapidly becoming the foreign language of choice among young Moroccans, mainly due to its international prestige.

4.4  Approach to Data Collection As mentioned above, the current chapter presents a qualitative data analysis in the form of a case study. The relevance of this ‘emic’ approach has been expressly mentioned in earlier research (Todeva and Cenoz 2009; Canagarajah and Wurr 2011; Gorter and Cenoz 2011). Data were collected in two semi-structured oral interviews. The interviews were digitally recorded and subsequently transcribed. As the focus is on self-report data rather than conversation analysis, the interview extracts presented have been edited and cleaned up to focus on the essence of M’s contributions. The extracts are marked with line numbers to make it easier to highlight certain comments, and translations or interpretations are provided in English in the main text. The researcher’s comments are marked as R and the participant’s comments are marked as M. Finally, as the interviews took place in Spanish, a language common to both the researcher and the participant, the extracts are presented in the original language. Regarding linguistic competence in Spanish, although neither participant nor researcher have any official language qualifications, both have been living and working in Spain for over a decade (M = 16 years; R = 12 years) and are highly proficient users of this language.

240

R. Nightingale

5  Results and Discussion The first issue to explore is how the participant uses his DLC to navigate specific sociolinguistic contexts. Taking into consideration that M is a bicultural multilingual who has successfully integrated into the host culture, he constantly finds himself in complex sociolinguistic contexts, each of which supposes a reconfiguration of his DLC. From the interviews, we are able to identify four distinct yet overlapping sociolinguistic situations: his social relationships with Spanish-speaking friends, his multilingual encounters in his job as a cultural mediator, his relationships with friends and family in Morocco, and his online social practices. Extract 1: 01 02 03 04

R: M: R: M:

Qué lenguas utilizas para comunicarte con tu familia? Darija. Qué lenguas utilizas para ser sociable? Pues, aquí el castellano y allí, darija.

Extract 2: 01 02 03 04 05 06 07

R: Cuál es la lengua que utilizas más en tu día a día? M: El idioma? El castellano R: Qué opinas del hecho de que utilizas el castellano más que las otras lenguas? M: Opino que es necesario. Vivo en España, la lengua que se habla es el castellano. Mi pareja es castellanoparlante, con lo cual opino que, sí, que hay que utilizar el castellano.

Extract 3: 01 02 03 04 05 06 07 08 09

R: Cuando estás en España dame por ejemplo un porcentaje de español, árabe y darija. M: español ochenta por cien, darija quince por cien, árabe cinco por cien. R: Y cuando estás en Marruecos. M: Al revés, castellano un cinco por cien- también cuando estoy en Marruecos depende, si va mi pareja conmigo el español ya no es un cinco por cien, es un treinta por cien, y el árabe y darija, un setenta por cien.

In extract 1, we see that M uses Darija to communicate with his family (line 02). However, to be sociable, in general, the language he uses depends on the context; in Spain, Spanish, in Morocco, Darija (line 04). In extract 2, M mentions

A Dominant Language Constellations Case Study on Language Use and the Affective…

241

that the language he uses most in his every-day life is Spanish (line 02). The researcher asks what M thinks about the fact that he now uses his adopted language more than any other (lines 03–04), to which he replies that it is simply necessary because he lives in Spain and his peers and partner speak Spanish (lines 05–07). In extract 3, M talks about DLC reconfiguration when he travels between Spain and Morocco. In Spain, he uses around 80% Spanish, 15% Darija, and 5% MSA (lines 03–04). On the contrary, in Morocco, this reconfigures to around 5% Spanish (line 06) and the remainder in Darija/MSA. However, this reconfiguration is also contextual. M says that if he travels with his partner, his DLC is around 30% Spanish and 70% Darija/MSA (lines 08–09) (it must be stressed that these percentages are purely rough approximations used by M to give a convenient estimate of his language use). From his comments, it is curious to note that in Spain M distinguishes between Darija and MSA, giving prominence to the former, while in Morocco he considers the two languages together. Although, M is not explicit about why he makes this distinction, it would be reasonable to assume that, in the context of Morocco, not only would he have far more contact with MSA in its written form but it would also be essential for him to negotiate certain situations. Conversely, in Spain, MSA is only likely to be essential for the realisation of M’s professional activities. Extract 4: 01 02 03 04 05 06 07 08 09 10 11 12 13 14

R: Cuáles son las lenguas que utilizas en el trabajo? M: En el trabajo utilizo el francés, utilizo el castellano, utilizo el valenciano y utilizo el árabe. R: Vale. Sería posible trabajar con una sola lengua, en tu trabajo? M: En mi trabajo, no. No es posible. R: Por qué no? M: Porque mi trabajo consiste en generar espacios de encuentro entre población inmigrada y población autóctona, y muchas veces la primera parte suele ser gente recién llegada que no domina el castellano entonces, yo allí, tengo que hacer la labor de traducción e interpretación. R: Y en eso, usas muchas veces el árabe, no? M: El árabe.

Extract 5: 01 02 03 04

R: Me has dicho que usas el árabe en el trabajo pero usas darija o el árabe clásico? M: El árabe clásico, escrito, el darija, entrevista oral.

242

R. Nightingale

Extract 6: 01 02 03 04 05 06 07 08

R: Cuándo empezaste a entender o aprender el catalán, el valenciano? M: El valenciano, empecé un poquito a interesarme por él en el momento que mi carrera profesional dio un giro y me dedicaba al mundo social, que es a través de la administración pública, con lo cual sabiendo que para hacer la administración pública, pues, tienes que dominar la lengua- digamos, la segunda lengua autóctona.

As can be seen above the languages of M’s DLC necessarily change in order for him to navigate successfully the demands of his professional life. In extract 4, the researcher asks M which languages he uses at work (line 01). M replies that he uses French, Spanish, Catalan, and Arabic (lines 02–03), adding that it would be impossible to do his job without them (line 06). The researcher asks why this is the case and M explains that his work involves translating, interpreting, and generating points of contact between recently arrived immigrants and the autochthonous population (lines 08–12). The researcher asks if Arabic frequently plays a role in these activities (line 13), which M confirms. In extract 5, the researcher asks M if it is Darija or MSA he uses in his job (lines 01–02), M replies Darija for oral interviews but MSA for written work (lines 03–04). In extract 6, M indicates that Catalan became an important language when he began to dedicate himself professionally to social work organised by the local government (lines 03–06). As an official language of the Valencian Community, Catalan is essential in any professional activity under the control of public administrative bodies. What these extracts highlight is how easily other languages may enter the DLC as a necessary condition for the correct realisation of certain activities and practices. As well as more long-term and stable changes in an individual’s life trajectory (Aronin and Singleton 2012), this reconfiguration also appears to take place on short-term and temporal levels. Extract 7: 01 02 03 04 05 06 07 08 09 10

R: Cuáles son las lenguas que utilizas en internet? En Facebook, en WhatsAppM: Castellano. R: Alguna vez utilizas el árabe o el francés? M: El árabe lo utilizo en el trabajo para traducir, el francés tambiénR: -pero si hablas con tu familia a través de internet dices cosas en árabe? M: Sí. Cuando hablo con ellos a través de internet hablo en árabe.

243

A Dominant Language Constellations Case Study on Language Use and the Affective…

In extract 7, the researcher asks M which languages he uses on the Internet. M replies that he uses Spanish (line 03). The researcher asks if M ever uses French or Arabic (line 04), to which he replies that he sometimes uses these languages at work when he needs to translate texts (line 05). The researcher presses M further, asking if he uses French or Arabic when communicating with his family (lines 07–08), to which M replies that he uses Arabic (lines 09–10). However, this response seemed overly simplistic given the complex nature of multilingualism and the dynamism and fluidity that characterises the DLC. Therefore, with the aim of ascertaining more information, the researcher was granted access to M’s Facebook profile. Exploring a number of posts and subsequent comments, it became clear that although M almost exclusively uses Spanish with friends and work colleagues from Spain, when he speaks with his family in Morocco he makes frequent use of Spanish, French and, so-called, ‘Chat Arabic’. There was also one comment from a family member in English. The researcher even found examples of code meshing between French and Chat Arabic (mercie [sic] habiba diali = “thank you my dear” - here M was responding to a comment in French, courage mon cher), and more restrained intrasentential codeswitching between Spanish and MSA (recuerda mis consejos ‫ زعيرتة‬y cognac = “remember my advice thyme and cognac”). From these examples, we can see that M uses the languages of his DLC to navigate the specific sociolinguistic context of online social media and that, in this context, French momentarily enters the DLC through translanguaging practices which are necessary for the full expression of his multilingual identity. Furthermore, if we consider Chat Arabic not only as an innovative and technology-contingent translanguaging practice, but also as a separate linguistic code, we can observe the emergence of a new ‘language’ within the DLC which plays a vital and highly context-specific role in M’s communicative, identificatory, and cultural practice. These extracts give an insight into the way the DLC changes in order to navigate specific sociolinguistic contexts. Aronin and Singleton (2012) mention that DLC languages have different weights and are assigned different functions according to the context. However, their visualisation of the DLC in relation to the language repertoire tends to simply list these expedient languages. Taking Kannangara’s (2017, this volume) innovative visualisation of the ‘weight’ of the constituent DLC languages in different contexts and M’s comments into account, the current author proposes a more dynamic approach (Fig. 2).

S

S

A

A

D

i

D

S

A/D

ii

S

iii

A D

Fig. 2  visualisation of dynamic reconfiguration within M’s DLC

S

C

ChA A

F

iv

E

F

v

244

R. Nightingale

Configuration i represents all languages of M’s DLC equally balanced. However, his comments indicate that this is not representative of the way M uses his DLC in different contexts; in fact, such balance is not representative of any DLC because the constituent languages would be in a constant state of flux as they responded to long-term and short-term changes in the multilingual environment. Fig. 2 proposes that the DLC itself is an ever-changing nucleus in which there is constant reconfiguration among the constituent languages. This concept is visualised in configurations ii, iii, iv and v. Configuration ii represents M in every-day social contexts in Spain. Spanish (S) is by far the most prominent of his three DLC languages, Darija (D) is used more than MSA (A), and these two languages are considered separately, most likely separated by function (spoken/written). Configuration iii represents M when he visits his family in Morocco. Darija and MSA are considered holistically and are overwhelmingly prominent, yet Spanish still plays an essential role, albeit drastically reduced. However, when M travels to Morocco with his Spanish-speaking partner, the prominence of Darija/MSA reduces and Spanish use increases (indicated by the dotted lines). Configuration iv represents M at work in Spain. Spanish is the most prominent language in this context, Darija and MSA are again separated by function, the former being employed more than the latter, Catalan (C) now enters the DLC as it plays an essential administrative role, and French (F) also enters the DLC for commination with African immigrants. Finally, configuration v represents M socialising online. Spanish is still the most prominent language, but also Chat Arabic (ChA) emerges and overlaps with French and MSA as they merge in translanguaging practices. Moreover, English (E) also emerges, not necessarily as a productive language for M, but as a receptive one, vital for multimodal navigation of an overwhelmingly English-dominant Internet. Thus, it would appear that the DLC is constantly in flux, subject to social, cultural, and temporal changes, the latter of which may be long- or short-term. We have explored the dynamic nature of M’s DLC as he navigates the ever-­ changing multilingual environment, however, the examples we have seen are of a mostly practical nature. The remainder of the discussion will focus on the DLC in relation to the affective domain; specifically emotions, identity, and attitudes.

5.1  DLC and Emotions M’s comments reveal emotional issues relating to a new language entering the DLC, his responses to taboo topics, his emotional expression in intimate ‘heat of the moment’ interactions, and a certain conflict arising from the expression of emotionally charged terms. Extract 8: 01 R: Crees que la capacidad de expresarte en castellano te ha 02 proporcionado algún entendimiento nuevo sobre ti mismo como

A Dominant Language Constellations Case Study on Language Use and the Affective… 03 04 05 06 07 08 09 10 11 12

245

persona? M: Sí, porque todo está relacionado, al descubrir otro mundo a través de la lengua, otro mundo de emociones, pues descubres emociones que no las tenías, y descubres, a lo mejor, otra opción de pensamiento tuyo que no la tenías … la aprendes con la lengua, y yo personalmente lo valoro muy positivamente en el sentido de que te abre un abanico de posibilidades, o sea amplias el abanico de posibilidades, a nivel emocional, a nivel personal, a nivel de crecimiento, de capacidad de entender, de empatizar.

Extract 9: 01 02 03 04 05

R: El castellano te permite verte a ti mismo desde una perspectiva nueva? M: Es como ampliar el ángulo, … al incorporar otro idioma, otras emociones, otra manera de pensar otra manera de expresar.

In extract 8, the researcher asks if M’s competence in Spanish has led to a new understanding of himself (lines 01–03). M replies that the new language allows for the discovery of “another world of emotions” (line 04–05), and may also lead to the discovery of a new way of thinking (line 07). He values this very positively (lines 08–09) because it “opens a range of possibilities” in terms of personal and emotional growth, as well as understanding and empathy (lines 10–12). These comments are also indicative of language attitudes; M holds positive attitudes towards the ability to learn more about his emotional self through the languages of his DLC. In extract 9, the researcher asks if Spanish allows M to see himself from a new perspective (lines 01–02). M comments that incorporating another language, other emotions, and another way to think and express himself “broadens the angle” (lines 03–05). Thus, rather than change from one perspective to another, his comments indicate summation; all DLC languages play a vital role in advancing his understanding of emotional identity. Extract 10: 01 02 03 04 05 06 07 08 09

R: Y eso que has dicho de entender las emociones de otra manera, me puedes dar un ejemplo? M: Los temas tabúes. Como era un tema tabú el sentimiento a través de tu lengua que ya sabías y que te expresabas, la emoción que te genera este tema el tabú, no eres capaz de vivir el sentimiento, … cuando aprendes otro idioma y este tema no es tabú y con lo cual lo hablas abiertamente, pues igual llegas a conectar y empatizar con este tema y llegas a generar otra emoción totalmente distinta.

246

R. Nightingale

Extract 11: 01 02 03 04 05 06

R: Es posible que un tema que es tabú en árabe, te sientes más libre de explorarlo en castellano? M: Sí, claramente. R: Dirías que eres más abierto ahora? M: Más abierto, o también … con más herramientas para abarcar más sentimientos, más emociones.

In extract 10, the researcher asks M for an example of how his DLC languages give him another way to understand his emotions (lines 01–02). M gives the example of taboo topics (line 03), indicating that the emotions a taboo topic can generate in one language may inhibit the experience of the feelings deriving from this topic (lines 03–06). However, in another language the same topic may not be taboo, or less taboo, meaning that it can be discussed more openly (lines 06–07). This allows M to connect and empathise with the topic and means that a totally different emotional response may be generated (lines 08–09). In extract 11, the researcher asks if a topic were taboo in Arabic, would M feel greater freedom to explore it in Spanish (lines 01–02). M indicates that this is clearly the case (line 03). The researcher asks if M would now describe himself as more emotionally open (line 04), to which he replies that he would, and that now he has a wider range of tools to encompass a wider range of emotions and feelings (lines 05–06). These comments indicate that the constituent languages of M’s DLC cooperate with each other, and from this cooperation emerges an emotional intelligence which constitutes more than the sum of its linguistic parts (Aronin and Singleton 2012). Extract 12: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17

R: Cuando discutes con tu pareja, te sientes un poco limitado sólo poder discutir en castellano? M: No me siento limitado, pero me cabrea no poder hacerlo en árabe porque muchas cosas que diría vienen de un pensamiento profundo, ya que es una cosa sentimental, vienen del árabe y traducidas no valen lo mismo, no? No tienen el mismo valor. R: Entonces, tienes que buscar otras herramientas lingüísticas para poder discutirM: -para poder decir cómo te sientes, no? O qué es lo que piensas, no? Porque, claro aquel pensamiento florece o surge en otra lengua y tienes que buscar las castañas para decir lo que quieres decir y que se entienda y que tenga el mismo valor en otra lengua R: Para decir que estoy enfadadoM: -enfadado, sí. De hecho, muchas veces cuando discuto con mi pareja digo cosas en árabeR: -y no entiende-

A Dominant Language Constellations Case Study on Language Use and the Affective…

247

18 M: -a mí mismo y no entiende pero yo las digo a mí mismo 19 sabes? Es como que tengo la necesidad de decirlo, aunque sea 20 en árabe, sé que no vas a entenderlo, pero lo voy a decir.

In extract 12, the researcher asks M if he feels limited because he can only argue with his partner in Spanish (lines 01–02). M replies that he feels more frustrated than limited because many of the deeper and more sentimental things that he wants to express come from Arabic (line 05), and if the expression of these feelings is translated, their value is lost (line 06). M points out that, when an argument reaches a ‘heat of the moment’ point, he often says things in Arabic. However, he admits that the use of Arabic is no longer directed at his partner, it is self-directed, used as a conduit to get past the emotional block set up by the limitations of Spanish. He comments “it’s like I have a need to say it, even though it’s in Arabic, I know you’re not going to understand, but I’m going to say it” (lines 19–20). The block caused by Spanish is not lexical, M has no problem in expressing complex ideas in this language, rather it is emotional because the emotion terms have different values across the languages of his DLC. These comments reflect an affectively triggered language switch motivation (Pavlenko 2005). Despite the incomprehensibility to their interlocutor, when emotions are high, multilinguals may switch to their first, or earlier acquired, language because it feels right. Moreover, in the case of very strong emotions, this switch between languages may be involuntary (Dewaele 2010). Extract 13: 01 02 03 04 05 06 07 08 09 10

R: Te sientes más o menos auténtico si tienes que decir una cosa emocional como te quiero? M: Me siento más auténtico diciéndolo en árabe que diciéndolo en castellano, porque justamente con esta palabra yo tengo un dilema, de te quiero a la pareja en mi lengua por ejemplo, en el árabe, o inclusive en el francés, hay una palabra específica de querer a la pareja y hay palabras específicas de querer a un amigo y hay palabras específicas de querer una cerveza, aquí se utiliza quiero para todo … así, para mí, pierde valor.

Extract 14: 01 02 03 04 05 06 07 08

R: Justamente lo de querer a una persona, creo que tiene que ser un dilema, no? Porque si tu pareja habla castellano y le dices te quiero en vez de decirlo en árabe, para ella es más auténtico pero para ti no. M: Claro. Sí, sí. Es un dilema. R: Claro. Es que quieres decírselo de tal manera que para ella sea auténticoM: Claro. Algo especial. Pero para mí no es especial.

248

R. Nightingale

In extract 13, the researcher asks M about feelings of authenticity when using the emotionally charged phrase ‘I love you’ (lines 01–02). M indicates that he feels more authentic in Arabic (line 03). He faces a dilemma with the phrase in Spanish because Arabic, and even French, have separate expressions for romantic and platonic love (lines 04–08), while in Spanish the verb querer can be used for both (line 09). M says that, for him, the expression loses value (line 10). This leads to a greater dilemma expressed in extract 14. The researcher proposes that saying I love you in Spanish is more authentic for M’s partner but less so for him (lines 02–04). M agrees that this causes an internal conflict, he wants the expression to be something special for her, but for him it is not special (line 10). This illustrates that communicating love in a foreign language is ‘extra challenging if it has to be channelled through narrow and imperfect linguistic translations’ (Dewaele 2008: 1753), and also highlights the culturally specific nature of emotional speech acts (Dewaele 2010). In this sense, the languages which comprise M’s DLC do not cooperate, leaving him with an uneasy truce between the concept of love in his different languages and the sociolinguistic context as the only option available to express his feelings.

5.2  DLC and Attitudes M’s comments reveal connections between the DLC and attitudes towards self-­ appraisal and future possibilities mediated by linguistic knowledge, attitudes towards the majority and minority languages of the host culture, and attitudes towards intergenerational transmission of languages and heritage culture. Extract 15: 01 02 03 04 05 06 07 08 09 10

R: Quiero saber si tu conocimiento de lenguas te hace sentir orgulloso. M: Sí. Claro que sí. R: Te hace sentir satisfecho? M: Sí. R: De haber aprendido lenguasM: -aprendido lenguas, sí. R: Te hace sentir que cubres todas tus necesidades? M: También. Y también lo veo como una herramienta que me da muchas oportunidades y me abre muchas puertas.

In extract 15, the researcher asks M if his knowledge of languages makes him feel proud (lines 01–02), if it makes him feel satisfied (line 04), and if it makes him feel like he can cover all his needs (line 08). M replies affirmatively to all these questions and adds that he sees this knowledge as a tool that gives him many opportunities and opens many doors for him (lines 09–10). Above, we saw how languages

A Dominant Language Constellations Case Study on Language Use and the Affective…

249

from M’s repertoire enter and leave his DLC in response to fluctuations in his multilingual surroundings. Here, we see that M holds positive attitudes towards the languages of his DLC as a unified resource which may provide him with future opportunities. This type of attitude is also intrinsically connected to identity in the sense of the ‘ideal L2 self’ (Dörnyei 2005). M’s ‘future self’ orientation is realised through his DLC languages; they are how he positions himself socially (Garrett 2010). Extract 16: 01 02 03 04 05 06 07 08 09

R: Hay una posibilidad que el español puede convertirse en tu lengua materna? M: Yo creo que ya lo es. … Yo considero ya el español como una lengua materna. … El materno y el no materno, para mí, se traduce en el grado de uso, el grado de dominio. Si tú dominas una lengua a la perfección, si eres capaz de gastar bromas en esta lengua, si eres capaz de leer el trasfondo de una frase en una lengua, yo creo que allí es llegar a colgarte la medallita de materno.

Extract 17: 01 02 03 04 05

R: Crees que tienes algún vínculo emocional … al catalán? M: Al valenciano, al catalán, no. Ya no tanto. Al revés, lo veo como un palo en la rueda, me supone a mí- como algo que tengo que aprender obligado, sabes? más que un aprendizaje para mejorar mis oportunidades.

In extract 16, the researcher asks if Spanish could become a mother tongue for M (lines 01–02). M responds that he thinks it already is (line 03), stating that what makes a language a mother tongue for him is related to the extent of control and use (line 05). This implies that his attitude is positive towards Spanish entering his DLC over the long term. This is again highlighted when M comments that once a person can tell jokes in a new language or read into the deeper connotations of a text or phrase, that person can “pin a little maternal medal on themselves” (line 09). This medal metaphor indicates that M feels proud not only of achieving a high competence in a new language but also of adding that language cumulatively to his DLC. Contrarily, in extract 17, the researcher asks M if he feels any emotional connection to Catalan. M replies that he does not, calling the language “a thorn in my side” (line 03), and indicating that he has learnt it as an obligation rather than to improve his possibilities (lines 04–05). Further research into how attitudes shape or restrict the DLC could be very interesting. However, here, we propose that this negative attitude will restrain Catalan in becoming more than a very short-term and highly function-specific constituent of the DLC.

250

R. Nightingale

Extract 18: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17

R: Cuándo tengas hijos, cuáles son las lenguas que quieres enseñarles? M: Fundamentalmente, árabe. Eso sería el cometido por excelencia. R: Y por qué árabe más que otras lenguas? M: El árabe porque es la identidad, no? La identidad de uno. Yo creo que mi granito de arena identitario para mis hijos va a ser aprender árabe, no? Qué consigo con esto? … que ellos se relacionen con el mundo- la mitad del mundo de su padre, no? Que en realidad es la mitad pero ocupa una gran parte. Para mí, mi mundo, de Marruecos del árabe y todo esto, para ellos no va a suponer lo mismo, no van a tener el valor que yo le tengo. Obviamente, mis hijos si nacen y se desarrollan en España, en Castellón, no le van a tener el mismo valor. Qué pasa? Que si a eso le añadimos que no pueden comunicar con sus primos, con sus tíos, con sus abuelos y tal, lo van a perder para siempre.

Extract 19: 01 02 03 04 05 06 07 08 09 10 11 12 13

R: En qué lenguas castigarías a tus hijos? M: En árabe, en darija. R: Por qué? M: Porque lo voy a intentar transmitir en todos los aspectos, la lengua, en el castigo, en la alegría, en la fiesta, y en cantar nanas para dormir. R: En qué lenguas los elogiarías? M: En darija igual. … El idioma vínculo entre yo y mi hijo, que me identifique con éste. R: De esta manera crees que les darías a tus hijos un vínculo afectivo con la lengua? M: Claro. Me sería más fácil inclusive para transmitirles modos de afecto que no están en castellano.

In extract 18, the researcher asks M which languages he would like to teach his future offspring (lines 01–02). He replies that it would be fundamentally Arabic (line 03). When asked why, he replies that if his children were to learn Arabic he would have provided them with his identificatory “grain of sand” (line 07). He indicates that he wants his children to be able to relate to their father’s world (lines 09–10), but then goes on to state that “in reality it’s half [of my world] but it occupies a large part [of it]” (lines 10–11). M is also fully aware that Morocco and Arabic/Darija will not have the same value to his children as they do to him (lines 11–12), but that if this is compounded by an inability to communicate with their

A Dominant Language Constellations Case Study on Language Use and the Affective…

251

Moroccan family, they may completely lose their contact with the language and culture (lines 15–17). These comments indicate an attitude towards the DLC as a means of cultural expression; M’s cultural heritage cannot adequately be expressed solely through his most prominent language. In extract 19, the researcher asks M in which languages he would chastise his children (line 01). M replies that he would do so in Darija (line 02). When asked why, M explains that he would try to transmit this language in all aspects (line 04), going on to list such aspects as punishment, joy, festivity, and lullabies (lines 05–06). The researcher asks in which language M would praise his children (line 07). M replies that he would also do this in Darija because it would be the linking language between him and his children and that they should identify him through it (lines 08–09). The researcher asks if this would create an affective connection between M’s children and Darija (lines 10–11), to which M agrees, adding that it would facilitate the transmission of affection in ways not available in Spanish (lines 12–13). These comments show M’s attitudes towards the role of the DLC in the intergenerational transmission of emotional and identificatory values. M indicates that the expression of emotion is not the same between his languages; therefore, he wants to transmit a similar base DLC to his offspring so they may have the linguistic and emotional framework in place to deal with this contrast.

5.3  DLC and Identity M’s comments reveal that linguistic knowledge in general is a defining point in terms of his identity, the cultures attached to his DLC languages cohabit without conflict in his sense of self, negotiation of meaning or differing sociopragmatic norms has a cumulative effect on his personality, and the long-term addition of a new language to the DLC enriches his already strong self-perception. Extract 20: 01 02 03 04 05 06

R: Crees que el conocimiento de lenguas … determina tu identidad? M: Obviamente. Yo creo que sí. La identidad, al fin y al cabo, es un acumulo de aprendizajes, de cosas vividas, no? Cosas vitales … que al final constituyen la identidad de uno, sean idiomas, sean experiencias, sea educación

Extract 21: 01 02 03 04

R: Te identificarías como multilingüe o multicultural, o las dos cosas? M: Multilingüe sí, pero multicultural no. Me considero más intercultural. … La diferencia entre lo intercultural y lo

252 05 06 07 08 09 10 11

R. Nightingale multicultural es que intercultural es dos códigos culturales o estas diferentes culturas conviven en el espacio o en el ser. Yo no tengo una lucha interna entre lo que adopté de la cultura, por ejemplo, castellana o española o castellonense en este caso con lo mío original. R: O sea, no son culturas separadas. M: No, no. Qué va, qué va! No, no, no. Todo lo contrario.

In extract 20, the researcher asks M if he thinks his knowledge of languages determines his identity (lines 01–02). M replies that this is obviously the case (line 03), adding that identity is the accumulation of different types of learning, different things one has lived through (lines 03–04). He continues that these vital things, whether they are languages, experiences, or education, in the end, construct one’s identity (lines 05–06). In extract 21, the researcher asks M if he identifies as multilingual, multicultural, or both (lines 01–02). M replies that he considers himself multilingual but intercultural rather than multicultural (lines 03–04). He explains that his different cultures cohabit within his construction of the self (lines 06–07), and there is no internal struggle between what he has adopted from the host culture and what he originally had (lines 07–09). The researcher remarks that they are not considered separate cultures (line 11) and M confirms that they are totally the opposite (line 12); that is, they form a unified hybrid culture. As Norton (2013) points out, subjectivity is produced across different social sites and structured according to different power relations; under these conditions the individual may assume different subject positions. From the above comments and earlier comments relating to emotions and attitudes, we may argue that M takes different subject positions according to the languages of his DLC: son, brother, future father (Darija); mediator, translator (Arabic); friend, lover (Spanish). However, it takes all the languages of the DLC for him to realise his full identity, it is not possible to do so through any of the constituent languages alone. Furthermore, through language, the individual may reposition themselves from a marginalised to a powerful subject position. We may argue that the extent to which M has adopted Spanish into his DLC constitutes his resistance to being positioned by others (Harklau 2007; Duff and Talmy 2011). Through language, he identifies as an ‘intercultural’ citizen, in this way he takes control of his subject position and redefines himself as an expert professional (intercultural mediator), a role in which his heritage languages, far from being marginalised, perform an essential and highly valuable function. Extract 22: 01 02 03 04

R: Crees que cambia la personalidad cuando hablas lenguas distintas? M: Sí y no. … Yo no lo llamaría personalidad … la personalidad es la misma … todas la lenguas tienen un

A Dominant Language Constellations Case Study on Language Use and the Affective… 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23

253

patrón de comportamiento, un código de comportamiento que rige comunicarse en esta lengua no? Claro, cada lengua tiene el suyo, lo adoptas, lo incorporas, aprendiendo la lengua lo incorporas, puede que influya en tu personalidad de alguna manera R: Es sumativa. M: Claro, es sumativa, pero no es que cuando hablo en castellano tenga una personalidad y cuando hablo en árabe tenga otra personalidad, no. Cuando hablo castellano, hablo bajo un código de conducta … Yo no creo que sea cambio de personalidad, sino cambio de código de comunicación … Mi marco de referencia en árabe es éste, entonces hay cosas que si yo las quiero decir en árabe, las tengo que decir de esta manera. Cambio el repertorio al castellano, esto en castellano cómo se dice? se dice de esta manera. Sí yo quiero decirte te quiero, en castellano, te lo voy a decir de esta manera … en árabe es totalmente distinto, y yo no he cambiado de personalidad, yo te quiero igual, lo que cambio es el uso de la palabra con todo lo que conlleva.

In extract 22, the researcher asks M if he perceives a change in personality when he uses different languages (lines 01–02). M replies both yes and no, suggesting that for him it is not a case of personality change, rather that each language has a behavioural pattern or code (line 05) that governs communication, and that incorporating this code may have some influence on his personality (line 08). The researcher proposes that this influence could be cumulative (line 10), and M agrees but reiterates that his personality does not change as a result of switching from Arabic to Spanish. On the contrary, he points out that when he speaks Spanish he is under the influence of a specific code of conduct, he continues “I don’t think I change my personality, but I do change the communicative code” (lines 14–15). M points out that if his referential frame is Arabic, it will affect the way that he can express certain things, conversely switching to Spanish implies finding and understanding the new frame with which to regulate the expression of the same concepts (lines 16–19). He illustrates his point by returning to the phrase I love you, explaining that saying this in Spanish implies a specific manner which in Arabic would be totally different (lines 20–21). He says, “my personality hasn’t changed, I love you all the same, what I change is the word and everything that that brings with it” (lines 22–23). These comments are related to affective (re)socialisation (Pavlenko 2005). The incorporation of a new language into the DLC does not change the personality but rather adds to it cumulatively as the individual negotiates new communicative frameworks, necessarily sociocultural in nature, and compares them with those already present in the DLC.  The words may change but the basic feeling, the inner state, remains unchanged (Pavlenko 2005: 228).

254

R. Nightingale

Extract 23: 01 02 03 04 05 06 07 08 09 10 11 12 13

R: cómo te sientes utilizar otra lengua que no es tu lengua natural? M: De momento me siento muy cómodo utilizando el castellano, más que nada porque he tenido la suerte de dominar la lengua, con lo cual no tengo ninguna dificultad ni para escribir ni para hablarlo, con lo cual no me genera este sentimiento interno que muchas veces se genera. Es decir, necesito utilizar mi lengua materna, que es la original, que es el árabe o el darija en este caso, o sea, no tengo esta lucha interna, no? Más que nada porque tampoco me influye mucho a la hora de valorar mi identidad, no? Sé quién soy, sé dónde vengo, sé cuál es lo mío y cuál es lo que adquirí, y también lo cuento como mío.

In extract 23, the researcher asks M how he feels about the fact that he now uses Spanish more frequently than his mother tongue, Darija (lines 01–02). M replies that he feels very comfortable using Spanish (line 03), ascribing this mainly to his competence in the language. He goes on to say that he does not feel the need to use his mother tongue and this does not generate any feelings of internal conflict (lines 09–10). He points out that this switch of most prominent language does not have much influence when it comes to evaluating his identity (lines 10–11) because, on the one hand, he has a secure conceptualisation of it, and, on the other, he is fully able to distinguish between that which is originally his and that which has been subsequently added (lines 11–12). However, his final comment is perhaps the most revealing. Referring to what Spanish has subsequently added to his identity, M proposes that he also counts it as his (line 13). This comment indicates that M considers the effect of this long-term reconfiguration within the DLC as cumulative in relation to the construction of his identity. This finding can be related to Panicacci and Dewaele’s (2017) suggestion that certain personality factors (in M’s case, a secure identity conceptualisation) may push immigrants towards new cultural horizons, and a capacity to regulate emotions may convert long-term language switches from an alienating to an enriching experience.

6  Conclusion The interview extracts in this study suggest that, in practice, M’s DLC languages are assigned different functions to different degrees in different contexts. Other languages from his repertoire may enter the DLC on a moment-to-moment basis in order to carry out context-specific activities and practices which are vital to his daily life. In this sense, the reconfiguration of the DLC appears to be dynamic and constantly in flux as an individual perceives new conditions in the multilingual

A Dominant Language Constellations Case Study on Language Use and the Affective…

255

environment. However, as the interview data indicate, the functions assigned to DLC languages are not only practical but also affective. We have seen that on some occasions M’s languages appear to work in conjunction, especially in terms of his identity, while on other occasions has to negotiate notable conflict between them, especially in terms of his emotions. Regarding emotions, M is able to learn more about his emotional self through the languages of his DLC. Although emotional terms and topics do not have a uniform value across the languages of his DLC, rather than switching perspectives between languages he uses them all together in order to come to new understandings about his emotional identity. However, in intense emotional situations such as arguing or expressing love, the languages of M’s DLC often impede each other, leaving him in paradoxical situations. Thus, we see that DLC languages may not always cooperate when it comes to expressing emotions. Where there is cooperation it may result in the emergence of an emotional intelligence that surpasses what can be realised by the constituent languages individually. Regarding attitudes, M is very positive towards the future opportunities his DLC may offer as a unified resource. Moreover, he is proud of his high competence in Spanish and the new role this language plays in his DLC.  Conversely, he shows quite negative attitudes towards Catalan, which he sees as an obligation. We posit that this attitude will restrain the role of Catalan and its durability within the DLC. M indicates acute awareness of the interrelated nature of language and culture and displays a positive attitude towards his heritage languages as means of cultural expression. Furthermore, he is highly positive towards transmitting emotional and identificatory values to his offspring through the languages of his DLC. He is also interested in providing them with a suitable framework to be able to negotiate emotional expression in all their languages. Thus, we see a range of attitudinal evaluations of the languages that comprise the DLC. Regarding identity, M uses his DLC languages to take different subject positions. He also uses the host language to resist being socially positioned by others, and may use all the languages of his DLC to take control of and redefine his subject position from that of an immigrant to that of an intercultural citizen. However, although M negotiates different aspects of his identity through different DLC languages, it can only be fully expressed by all the languages working together as a set. Switching the most prominent language from Darija to Spanish has not influenced M’s evaluation of his identity. The fact that M counts the host language as his own shows a cumulative effect stemming from this long-term DLC reconfiguration. Thus, we see that DLC languages may work in conjunction when it comes to understanding and negotiating identity. The current study is not without its limitations. On the one hand, it is a case study reliant on purely qualitative data. In this sense, although very rich data have been acquired, it is not possible to generalise the results. On the other hand, the conclusions reached reflect the author’s interpretive judgement on applying the DLC model to the data after the fact. While care has been taken to make the most faithful interpretation of the participant’s comments, it is not possible to know if his views have been fully represented as he intended them. Nevertheless, the author believes

256

R. Nightingale

that this chapter contributes to our ongoing understanding of multilingual practices realised in concrete sociocultural contexts and, more specifically, fills a significant gap in DLC research. That is, providing an avenue to advance the model by linking it to the affective domain and suggesting that it may be more dynamic and flexible than initially theorised. Acknowledgements  As a member of the LAELA (Lingüística Aplicada a l’Ensenyament de la Llengua Anglesa) research group at Universitat Jaume I (Castellón, Spain), I would like to acknowledge that this study is part of a research project funded by (a) the Spanish Ministerio de Economía y Competitividad (FFI2016-78584-P), (b) the Universitat Jaume I (UJI-B2019-23), and (c) Projectes d’Innovació Educativa de la Unitat de Suport Educatiu 3821/20.

References Aronin, L. (2006). Dominant language constellations: An approach to multilingualism studies. In M. Ó Laoire (Ed.), Multilingualism in educational settings. Hohengehren: Schneider. Aronin, L. (2015). Current multilingualism and new developments in multilingual research. In P. Safont & L. Portoles (Eds.), Multilingual development in the classroom: Current findings from research. Newcastle: Cambridge Scholars Publishing. Aronin, L. (2016). Multicompetence and dominant language constellation. In V. Cook & L. Wei (Eds.), The Cambridge handbook of linguistic multicompetence. Cambridge: Cambridge University Press. Aronin, L. (2019). Dominant language constellation as a method of research. In E.  Vetter & U.  Jessner (Eds.), International research on multilingualism breaking with the monolingual perspective. Basel: Springer. Aronin, L., & Hufeisen, B. (Eds.). (2009). The exploration of multilingualism: Development of research on L3, multilingualism and multiple language acquisition (Vol. 6). Amsterdam: John Benjamins. Aronin, L., & Ó Laoire, M. (2004). Exploring multilingualism in cultural contexts: Towards a notion of multilinguality. In C. Hoffmann & J. Ytsma (Eds.), Trilingualism in family, school, and community (Vol. 43). Clevedon: Multilingual Matters. Aronin, L., & Singleton, D. (2008). Multilingualism as a new linguistic dispensation. International Journal of Multilingualism, 5(1), 1–16. https://doi.org/10.2167/ijm072.0. Aronin, L., & Singleton, D. (2012). Multilingualism. Amsterdam: John Benjamins. Björklund, S., Björklund, M. & Sjöholm, K. (2018). Embracing multilingualism in teacher education in Finland? DLC as a tool for analyzing the state of multilingualism in policy and practice. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Blommaert, J. (2013). Language and the study of diversity. Tilburg Papers in Culture Studies, 74, 1–14. Bond, M. H., & Lai, T. M. (1986). Embarrassment and code-switching into a second language. Journal of Social Psychology, 126(2), 179–186. Brown, H. D. (1987). Principles of language learning and teaching (2nd ed.). Englewood Cliffs: Prentice Hall Regents. Canagarajah, A. S., & Wurr, A. J. (2011). Multilingual communication and language acquisition: New research directions. The Reading Matrix, 11, 1–15. Cenoz, J. (2009). Towards multilingual education: Basque educational research from an international perspective. Clevedon: Multilingual Matters.

A Dominant Language Constellations Case Study on Language Use and the Affective…

257

Cenoz, J. (2013). The influence of bilingualism on third language acquisition: Focus on multilingualism. Language Teaching, 46(1), 71–86. https://doi.org/10.1017/S0261444811000218. Cenoz, J., & Genesee, F. (Eds.). (1998). Beyond bilingualism: Multilingualism and multilingual education (Vol. 110). Clevedon: Multilingual Matters. Cenoz, J., & Gorter, D. (2011). A holistic approach to multilingual education: Introduction. The Modern Language Journal, 95(3), 339–343. https://doi.org/10.1111/j.1540-4781.2011.01204.x. Coetzee Van Rooy, S. (2018). Dominant language constellations in the repertoires of multilingual South Africans. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Cook, V. J. (1992). Evidence for Multicompetence. Language Learning, 42(4), 557–591. https:// doi.org/10.1111/j.1467-1770.1992.tb01044.x. Cook, V. J. (1997). Monolingual bias in second language acquisition research. Revista Canaria de Estudios Ingleses, 34, 35–50. Dewaele, J. M. (2004). The emotional force of swearwords and taboo words in the speech of multilinguals. Journal of Multilingual and Multicultural Development, 25(2/3), 204–222. https:// doi.org/10.1080/01434630408666529. Dewaele, J. M. (2005). Sociodemographic, psychological and politicocultural correlates in Flemish students’ attitudes towards French and English. Journal of Multilingual and Multicultural Development, 26(2), 118–137. https://doi.org/10.1080/01434630508668400. Dewaele, J. M. (2008). The emotional weight of ‘I love you’ in multilinguals’ languages. Journal of Pragmatics, 40(10), 1753–1780. https://doi.org/10.1016/j.pragma.2008.03.002. Dewaele, J. M. (2010). Emotions in multiple languages. Basingstoke: Palgrave Macmillan. Dewaele, J. M. (2013). Multilingualism and emotions. In C. A. Chapelle (Ed.), The encyclopedia of applied linguistics. Wiley-Blackwell. Dewaele, J. M. (2015). From obscure echo to language of the heart: Multilinguals’ language choices for (emotional) inner speech. Journal of Pragmatics, 87, 1–17. https://doi.org/10.1016/j. pragma.2015.06.014. Dörnyei, Z. (2005). The psychology of the language learner: Individual differences in second language acquisition. Mahwah: Erlbaum. Dörnyei, Z., & Csizér, K. (2005). The effects of intercultural contact and tourism on language attitudes and language learning motivation. Journal of Language and Social Psychology, 24(4), 327–357. https://doi.org/10.1177/0261927X05281424. Duff, P. A., & Talmy, S. (2011). Language socialization approaches to second language acquisition. In D. Atkinson (Ed.), Alternative approaches to second language acquisition. New York: Routledge. Fernández-Berkes, E. & Flynn, S. (2018). Where syntactic development meets DLC. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Fisher, L., & Evans, M. (2000). The school exchange visit: Effects on attitudes and proficiency in language learning. Language Learning Journal, 22, 11–16. https://doi. org/10.1080/09571730085200191. Gardner-Chloros, P., McEntee-Atalianis, L., & Finnis, K. (2005). Language attitudes and use in a transplanted setting: Greek Cypriots in London. International Journal of Multilingualism, 2(1), 52–80. https://doi.org/10.1080/17501220508668376. Garrett, P. (2010). Attitudes to language. Cambridge: Cambridge University Press. Gorter, D., & Cenoz, J. (2011). A multilingual approach: Conclusions and future perspectives: Afterword. The Modern Language Journal, 95(3), 442–445. https://doi. org/10.1111/j.1540-4781.2011.01203.x. Gorter, D., Riemersma, A., & Ytsma, J. (2001). The Frisian language in the Netherlands. In G. Extra & D. Gorter (Eds.), The other languages of Europe: Demographic, sociolinguistic and educational perspectives. Clevedon: Multilingual Matters. Grosjean, F. (1989). Neurolinguists, beware! The bilingual is not two monolinguals in one person. Brain and Language, 36, 3–15. https://doi.org/10.1016/0093-934X(89)90048-5.

258

R. Nightingale

Gumperz, J.  J. (1964). Linguistic and social interaction in two communities. American Anthropologist, 66(6), 137–153. Harklau, L. (2007). The adolescent English language learner: Identities lost and found. In J.  Cummins & C.  Davidson (Eds.), International handbook of English language teaching. Basel: Springer. Herdina, P., & Jessner, U. (2002). A dynamic model of multilingualism. Perspectives of change in psycholinguistics. Clevedon: Multilingual Matters. Hoare, R. (2001). An integrative approach to language attitudes and identity in Brittany. Journal of Sociolinguistics, 5(1), 73–84. https://doi.org/10.1111/1467-9481.00138. Jessner, U. (2008). A DST model of multilingualism and the role of metalinguistic awareness. The Modern Language Journal, 92, 270–283. https://doi.org/10.1111/j.1540-4781.2008.00718.x. Jessner, U. (2013). Complexity in multilingual systems. In C. A. Chapelle (Ed.), The encyclopedia of applied linguistics. New York: Wiley-Blackwell. Kannangara, S. (2017). The evolution of personal dominant language constellation. Paper presented at the 11th International Symposium on Bilingualism. Limerick, June 11–15. Lasagabaster, D., & Huguet, A. (Eds.). (2007). Multilingualism in European bilingual contexts: Language use and attitudes. Clevedon: Multilingual Matters. Lasagabaster, D., & Safont, P. (2008). Un análisis de las actitudes lingüísticas en dos comunidades bilingües. XXVI Congreso AESLA. Lasagabaster, D., & Sierra, J. M. (2009). Immersion and CLIL in English: More differences than similarities. ELT Journal, 64(4), 367–375. https://doi.org/10.1093/elt/ccp082. Letica Krevelj, S. (2017). Interaction of languages within a DLC. Paper presented at the 11th International Symposium on Bilingualism. Limerick, June 11–15. Marcos, L.  R. (1976). Linguistic dimensions in the bilingual patient. American Journal of Psychoanalysis, 36(4), 347. https://doi.org/10.1007/BF01250860. Mayr-Keiler, K. & Jessner, U. (2018). On the dynamics of the dominant language constellation. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. McCarthy, E. D. (1994). The social construction of emotions: New directions from culture theory. Sociology Faculty Publications, 4. Nightingale, R. (2012). Bridging the gap between the internal and the external: The effect of sociocultural factors in adolescent learners’ attitudes towards English. Saarbruken: Lambert Academic Publishing. Nightingale, R. (2016). The effect of out-of-school media contact on language attitudes in multilingual adolescents: a complex psychosociolinguistic system. Unpublished Ph.D dissertation. Castelló: Universitat Jaume I. Norton, B. (2013). Identity and language learning: Extending the conversation (2nd ed.). Clevedon: Multilingual Matters. Osterkorn, P. (2018). The awareness of dominant language constellations in educational settings. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Otwinowska-Kasztelanic, A. (2016). Dominant language constellations affect multilingual guessing of words. Paper presented at the 10th International Conference on Multilingualism and Third Language Acquisition. Vienna University, September 1–3. Panicacci, A., & Dewaele, J.  M. (2017). ‘A voice from elsewhere’: Acculturation, personality and migrants’ self-perceptions across languages and cultures. International Journal of Multilingualism, 14(4), 1–18. https://doi.org/10.1080/14790718.2016.1273937. Pascual, D., & Rothman, J. (2013). Multilingualism and identity. In C.  A. Chapelle (Ed.), The encyclopedia of applied linguistics. New York: Wiley-Blackwell. Pavlenko, A. (2004). ‘Stop doing that, Ia Komu Skazala!’: Language choice and emotions in parent-­child communication. Journal of Multilingual and Multicultural Development, 25(2–3), 179–203. https://doi.org/10.1080/01434630408666528. Pavlenko, A. (2005). Emotions and multilingualism. Cambridge: Cambridge University Press.

A Dominant Language Constellations Case Study on Language Use and the Affective…

259

Pavlenko, A., & Blackledge, A. J. (Eds.). (2004). Negotiation of Identities in Multilingual Settings. Clevedon. Multilingual Matters. Portolés, L. (2014). Analysing prospective teachers’ attitudes towards three languages in two different sociolinguistic and educational settings. In A. Otwinowska-Kasztelanic & G. De Angelis (Eds.), Teaching and learning in multilingual contexts: Sociolinguistic and educational perspectives. Clevedon: Multilingual Matters. Portolés, L. (2015). Multilingualism and very young learners: An analysis of pragmatic awareness and language attitudes. Boston: Mouton de Gruyter. Safont, P. (2015). The promotion of multilingualism in a Catalan-speaking area. Familial challenges in the Valencian community. In U. Jessner & C. Kramsch (Eds.), Multilingualism. The challenges. Berlin: De Gruyter Mouton. Schreiber, B.  R. (2015). “I am what I am”: Multilingual identity and digital translanguaging. Language Learning and Technology, 19(3), 69–87. https://doi.org/10125/44434. Schumann, J. (1997). The neurobiology of affect in language. Maiden: Blackwell. Schwarzl, L. (2018). The impact of dominant language constellations on children’s self-efficacy in educational settings. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Singleton, D., Fishman, J. A., Aronin, L., & Ó Laoire, M. (Eds.). (2013). Current multilingualism: A new linguistic dispensation. Berlin: De Gruyter Mouton. Sjoholm, K., Björklund, M. & Björklund, S. (2016). Dominant language constellations in multilingual Finland  - past and present perspectives at the societal and individual level. Paper presented at the 10th International Conference on Multilingualism and Third Language Acquisition. Vienna University, September 1–3. Slavkov, N. (2018). Family language policy and dominant language constellations: A Canadian perspective. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Sugrañes, C. (2018). Promoting plurilingual competences in primary schools in Barcelona: A DLC approach to teaching and learning languages. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15. Tabouret-Keller, A. (1998). Language and identity. In F. Coulmas (Ed.), Handbook of sociolinguistics. Oxford: Blackwell. Todeva, E., & Cenoz, J. (2009). The multiple realities of multilingualism. Berlin: De Gruyter Mouton. Vetter, E. (2018). Language education policy through a DLC lens: The case of urban multilingualism. Paper presented at the 11th International Conference on Multilingualism and Third Language Acquisition. Lisbon, September 13–15.

Quo Vadis, DLC? Joseph Lo Bianco

Abstract  What are the likely future directions of the concept of the Dominant Language Constellation? What is the role of the DLC within multilingualism studies? This chapter uses the simile of the heavenly constellation to develop a line of inquiry about the uses and usefulness of the DLC concept. It reviews the contributions, data and arguments made in the chapters that comprise this book and tracks possible future directions in areas such as public policy, education, grammatical exploration, sociology of multilingualism, intercultural relations and personal and group identity. The chapter argues that the DLC concept makes a vital contribution to understanding language questions today as an important addition to the analytical concepts of the field, by offering a new perspective. Keywords  Multilingualism · Public policy · Language teaching and learning · Heuristics · Dominant Language Constellations

1  Introduction One of the striking features of the night-sky all humans share is the arrangement of its celestial objects into constellations. In 1930 the International Astronomical Union (IAU n.d.), determined that star constellations could be named and ‘standardized’ into 88 stable formations. Individual stars only ‘belong’ to a specific constellation by virtue of how they appear in the two-dimensional perspective available to us from our earth-based vantage point. Given their different distances from earth no star has an inevitable and physical connection with any other star in its constellation. If they could be viewed from another planetary location, or from a three-­ dimensional perspective, they would form different relationships with each other. Just like a stellar constellation, earthly Dominant Language Constellations vary according to vantage point. When the role an individual language performs in a J. Lo Bianco (*) Language and Literacy Education, University of Melbourne, Melbourne, VIC, Australia e-mail: [email protected] © Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7_13

261

262

J. Lo Bianco

DLC is transformed in its prevailing socio-linguistic situation, or when an individual’s life is remade through changes in circumstance or through personal upheaval, a new DLC emerges. The patterns we assign to humanly created concepts, such as languages and their significant groupings, also differ from patterns we assign to phenomena in the physical/material world. This is because humancreated phenomena rely on our subjectivity and interests, as well as reflecting objective conditions. In this volume we have subjected the Dominant Language Constellation to scrutiny from multiple angles. Aronin began by describing the scope of the concept of DLC, a concept she has pioneered and which today is an important theoretical innovation in the study of multilingual phenomena and practices. She distinguishes DLC from the similar but distinct concept of linguistic repertoire. The latter aims to capture all the skills and registers of an individual’s languages and other communicative resources. By contrast, the DLC restricts its focus to the languages that serve most immediately and fully the communication needs of that individual. As such the DLC involves a selection based on saliency and prominence of a group of languages for an individual and its study involves gathering data about concurrent use of these expedient languages across settings and contexts, time and interlocutor. As well as conceptual innovation, DLC represents a heuristic innovation in one of the senses originally proposed by the cognitive scientists, economists and political scientists who coined this term. Among these scholars US decision making expert Herbert Simon (Nobel Laureate in Economics in 1978), focused on problem-­ solving processes in complex systems (Simon 2008). In his conception, heuristics are systems that permit relatively straightforward judgement. This is because they allow individuals to form a rapid sense and apprehension of an entity or an environment. This facility to form rapid and practical judgement is a repeated and daily necessity which helps people to make decisions that correspond with their intentions and predilections as they are engaged in activity of various kinds, and to organize appropriate action to follow the judgement they make. In effect, a heuristic function operates through a kind of withdrawal and re-focus, in which we remove our attention from the entire complexity of a phenomenon to then re-focus attention only on its most salient characteristics. We can apply this notion of a heuristic to the contrasts I want to make here between multilingualism and the DLC. Multilingualism is clearly a heterogenous phenomenon. It is complex because its very nature involves multiple languages co-­ existing in particular environments, but it is also complex due to historical structuring. This means, for example, that what is described as multilingualism in some environments can be radically different from the meaning of multilingualism in a different environment. For example we could contrast multilingualism as a map and descriptive account of all the languages of Spain, understood as separable, named, and bounded entities, with another setting where languages are treated as overlapping, fluid and unbounded processes of communication. As a concept the DLC is a depiction of key elements, selected in some heuristic process of judgement an individual makes, or a collection of individuals in an institution make, of the utility or relevance of a grouping among the multiple languages

Quo Vadis, DLC?

263

and semiotic codes in a single environment. This applies to broader groupings, such as a public institution or a nation, as it does to the communicative behaviours of individuals. In its clear-eyed approach to the dominant codes used by individuals and groups a DLC therefore responds to and functions heuristically, with clear face validity, as demonstrated in the chapters in this volume. In my own discussion of the utility of DLC in policy making settings (Lo Bianco, this volume) I noted how wider use of the notion of DLC would connect scholars more closely to how we see power-wielding non-experts operate in concrete policy contexts. These categories of people (politicians, decision makers and public officials) are entrusted with the management of public institutions, the disbursal of resources, or the running of schools and museums. While the existence of multiple languages as understood in multilingualism research might occasionally impinge on their work, their main focus is on effective functioning of organizations and personnel management, or efficient delivery of a service, rather than on conceptual and critical analysis, hence a DLC is conceptually more aligned with their professional function. Heuristic reasoning is based on the idea that instead of aiming for full comprehension of a phenomenon we typically apprehend only its essential characteristics, and that we do this because we need to take action or respond to the demands the particular phenomenon places on us in some way. By rapidly isolating salient or dominant features of a phenomenon a functional (yet always provisional) sense of the entity or phenomenon is formed. This ‘sense’ is provisional because there is still the possibility of later modification or refinement since heuristics are dynamic processes of judgments being formed and reformed, allowing deviation from initial apprehension of a situation as new information is obtained, and newer characterizations become more convincing or useful (Bazerman 2017). Heuristics therefore can be seen as a form of probabilistic reasoning, bounded by the need to align with action or policy or practical decisions of some kind. Hence DLC gains functional relevance to policy and decision-making environments, especially when contrasted with currently popular and more wide-ranging depictions of multilingualism which seem to insist on limitless inclusion of new forms of communication diversity, with no apparent rules for determining boundaries. Intuitive judgments based on incomplete provisional knowledge are typical of policy environments, and also in teaching and other practical activity, because acting in these environments draws on praxis, understood here as a form of practice informed by reflective knowledge. Praxis, as rapid expert action is an essential quality and feature of policy related domains, of teaching, or interpreting and translating, as of other language related professional activity. Time constraints, competing pressures, limited or perhaps ‘interested’ stocks of information (meaning information gathered specifically with the purpose of supporting decision making rather than open-ended exploration), are all contained in what Simon originally proposed as the ‘bounded rationality’ of heuristic reasoning. However, despite these compelling similarities, DLC remains a scholarly concept more than a heuristic, but a scholarly concept which contains a valuable heuristic-­ like quality. In making action-oriented judgments in real-world settings, the

264

J. Lo Bianco

conditions of available information are radically different from those that scholars typically associate with the processes of exploration and discovery. The risk of error, bias and uncertainty inherent in heuristic reasoning is corrected in scholarly DLC. Aronin’s opening chapter in this volume locates DLC in a temporal dimension, showing how individuals and groups utilize their stock of salient vehicle languages holistically to allow them to achieve all their needs in a particular environment, and how this differs from linguistic repertoire because it includes only ‘the most expedient languages’ for that individual. The model of the DLC therefore ‘delimits, specifies, and systematizes’ information about the characteristics of the environment to critical elements, those utilized and taken up by the individual or group because of their saliency. This empirical orientation highlights the theoretical rigor of DLC, attuned to the best traditions of multilingual sociological research. Aronin extends and bolsters this through her genealogical account of the genesis of DLCs in longue durée (Braudel 1958). A DLC in a specific here-and-now is a language formation distilled from historical time under specific conditions. These conditions might be the education policies, political events, power relations, population mobility, economic circumstances etc., that prevail in a given setting and which determine the current order of language arrangements. In effect, this diachronicity also provides an account of ‘valency’ of a DLC in the present as well as its emergence over time. This is because each distillation is a here-and-now settlement of dynamic past possibilities, what Aronin and Singleton (2008, 2012) have called a Linguistic Dispensation, and any new linguistic dispensation is the qualitative and quantitative modifications or, in more radical environments, transformations, that are made to existing communicative arrangements. Linguistic dispensations have no natural form or qualities, they are only the outcome of the work of institutions and social forces on the communicative raw material of given communities. A linguistic dispensation can therefore be monolingual, bilingual, or multilingual, determined by the play of social and economic forces competing in particular real-world environments. Given that most states historically and even today support only official or national languages, usually the codes of dominant groups and privileged traditions, and they do this by limiting, marginalizing or even exterminating domestic multilingualism, the resultant linguistic dispensations typically represent the outcome of these processes. As public authorities and dominant cultural forces shape the sociolinguistics of a given polity in this way their efforts are limited only to the extent that speakers of threatened or repressed languages can achieve success in resisting the linguistic assimilation desired by public authorities. The DLC allows a shift in focus for both scholars and practitioners from the investigation of individual languages, or the totality of meaning making forms, i.e., of the entire communicative repertoire, or other lenses directed at sociolinguistics and communication, towards an investigation of identifiable clusters of languages, and how these clusters coalesce into relatively stable constellations. As such these languages relate to each other as they share the load of overall communication needs of individuals and groups, and in this way justify being analyzed together. DLCs vary over time, according to need and circumstance but also as a result of conscious

Quo Vadis, DLC?

265

or planned intervention. It is in this sense, as I argue in my chapter on language policy and the DLC (Lo Bianco, this volume), that the DLC concept is productive in various ways, firstly as a tool of research that overlaps with both a popular heuristic device of apprehending the existence of a DLC in the lives of a community and its members, and secondly, a DLC is productive in its connection with decision making directed at language change, such as we see in language policy and planning. The dual nature of DLC as an emergent but still autonomous whole and at the same time as a subsystem of the entirety of multilingualism is extended in language policy and planning discussions into an overlapping conceptual tool shared between scholars and policy makers. This shared apprehension has the potential for facilitating more understanding between experts and officials, with the promise of more productive interactions between them. Language policy can be seen as the formal and deliberate component of how a new linguistic dispensation might emerge in a given setting. This is despite the fact that most language change arises from less formal processes of social transformations, or arises from unplanned developments and innovations in technology, power relations, political events or economic conditions. Nevertheless, more and more societies are making language decisions in planned, deliberate and explicit ways. This is because multilingualism has emerged as a ubiquitous problem of contemporary society, in response to migration and population mobility, foreign language teaching choices and the demands of international diplomacy and trade in more tightly interconnected modern commercial relations. This new salience for multilingualism can be noted in the efforts of more and more countries to produce explicit and occasionally comprehensive language policies. The increase in claims for political asylum across the world in recent decades has also led to investments in research and documentation of multilingual environments and especially of language tests. This too poses a problem for university based experts who analyse language phenomena as they are called in to advise on or critique public decision making. Their involvement fulfils the criterion that policy should be ‘evidence based’ but the evidence that academic researchers provide is ‘purchased’ in consultancy arrangements, and doesn’t strictly belong to the academic, or even to academia, but to commissioning agencies (Lo Bianco 2019). These interactions push more and more scholars of multilingualism into discussions with policy officials, combining heuristic and research oriented approaches to knowledge. In Science in Democracy Mark Brown (2009) discusses similar dilemmas for scientists who want their academic work to stand above politics but are located in institutions and societies and need to deal with political agendas and bureaucracies that influence and attempt to manage the production, reception and use of scientific findings. Given the central role that language plays in national identity, economic opportunities and education participation, dilemmas of this kind confront the language sciences also, especially related to the findings from sociolinguistic research, language test development, linguistic description and classification studies. Specifically, for language testing for high stakes decision making such as determine asylum seeker claims, questions of fairness and justice in assessment (McNamara and Ryan 2011) are ever present.

266

J. Lo Bianco

Fairness is related to the technical quality of tests while justice is linked to values in the constructs on which language tests are written, or how tests are used in different social settings. Increasingly across the world policy makers use language tests as a way to solve problems of citizenship (do applicants have sufficient knowledge of the official/national language?), of asylum claims (is the applicant a member of the oppressed community that he or she is claiming?), etc. Because the offer of ‘sanctuary’ to persecuted individuals is a moral obligation on states (enshrined in the 1948 United Nations Universal Declaration of Human Rights and the 1951 Convention Relating to the Status of Refugees and a 1967 Protocol developing the 1951 Convention) individuals with a credible case of persecution are expected to be offered asylum in most cases, yet many recipient nations are reluctant to concede this to all applicants. The Dominant Language Constellation is an important notion that should be applied to such settings to build a more accurate picture of communicative life in contemporary societies. It can replace obsolete ideas that nations and individuals are monolingual or that ordinary communication is typified by use of rigidly discrete and separately bounded languages. The DLC can build a shared stock of concepts about language in society between public officials and scholars whereas other analytical approaches to multilingualism have less traction of this sort. The seventy-two indigenous languages spoken alongside English, the main language of education in Zambia, form one powerful demonstration of the sociolinguistic utility of the DLC notion. Banda, this volume, shows persuastively how Zambians typically speak more than three languages. In his analysis the DLC concept is developed into a multi-layered form as it is discussed across boundaries of individual/household, community, regional and national settings. These tiers of DLC distill the current linguistic dispensation from the conditions of past movement of peoples, trans-regional and transnational migration, and the interethnic contact this movement has produced within the nation state of Zambia. In Banda’s analysis we also see that in ten provinces language policy and planning has allocated seven indigenous languages the status of “regional language” for fulfilling local government administration and early primary school education, both of these are subsidiary functions to the colonial legacy of English as the code of national government, and of post-primary education, and the formal economy. Banda’s analysis emphasizes the dynamic or fluid quality of multilingualism. His analysis reveals how across different domains of life in Zambia, individuals/ households, communities and regions, the languages that make up the DLCs vary in predictable and patterned ways. In effect, vastly heterogenous configurations and multiple strata of languages constitute the DLCs. This dynamic multilingual character is not simply the result of recent post-colonial arrangements but a reflection of traditional African communicative patterns of ethno-cultural co-existence, of worlds “without borders before the onset of colonialism in 1880” (Banda, this volume), arising in pre-colonial movements of and relations between agriculturalists, pastoralists and hunter-gatherers across spaces that are now delimited by colonially inspired political borders.

Quo Vadis, DLC?

267

Nation making in the post-colonial environments of traditional multilingualism extended by introduction of and domination by colonial Arabic and several European languages add the element of national administration and national economic life to social structuring derived from traditional arrangements of communication in a complex overlapping system of official and semi-official forms of language power, cultural power and prestige and various other kinds of language spread or restriction. What is represented here is not just fluidity of language use across the boundaries between languages, nor even just use of many languages, but dense and intense communication in general, frequent and regular patterned use of multiple languages, but still arranged according to DLCs that are specific to domains or individuals. The more than 2000 languages of the African continent and their confinement into national spaces inevitably generate a proliferation of solutions, practical, ideological, administrative and educational to the challenges and diverse purposes for communication. From Banda’s analysis therefore the Dominant Language Constellation is shown to be a plurality of DLCs for individuals and communities. One of the key reasons for this, as has been stated, is policy which favors some languages over others for exclusive or semi-exclusive use in literacy and elaborated functions, making the evolution of DLCs inevitable. The vehicle languages of individuals therefore are selections from what is required in specific settings and adaptation to what is required in others. These are accommodations to the practical demands of living in a multilingual environment with layers of overlapping literacy and orality, and their different functionality in the diverse speech communities that comprise a large nation state structure. Here the qualifier “dominant” is important, not in an evaluative way to suggest an order of intrinsic worth or value, but in terms of an empirical depiction of power and utility differentials in sociolinguistic arrangements and the reverberating effect of these power asymmetries on what an individual or group must construct as their linguistic DLC to interact within that environment. If there were one region, province, town or valley of Zambia that was fully monolingual individuals residing in that place might still need to construct a DLC to interact beyond that space with predictable others in predictable activities, but that region, province, town or valley itself would not have a DLC, but instead a (possibly dialectally variegated) monolingualism. The DLC might arise to allow individuals to traverse bounded multilingual settings, but many of these co-exist in precisely the same domains and are differentiated, and therefore maintained, but the stably existing speech practices, some of which may be monolingual, within that domain. The overlaps, to complete the references to Zambia, also arise because of asymmetrical power: Bemba and Nyanja are spoken outside official zones and these languages are incorporated in the DLCs of Zambians from diverse regions, as common forms of speech with different sets of additional languages in overlapping, but individually unique DLCs. As has been pointed out above, multilingualism’s emergence in any specific setting is particular to the historical conditions that have pertained there. Fernández-­ Berkes and Flynn (this volume) argue that this inherited linguistic dispensation can be seen as a natural extension of first to second language acquisition. The authors

268

J. Lo Bianco

offer the initial state (S0) concept to depict this temporal emergence of a DLC. How does this occur? Fernández-Berkes and Flynn argue that first to second language acquisition process can be discerned in the syntactic knowledge multilingual learners possess on commencement of subsequent language acquisition, and then what function or role a newly and additionally acquired language performs in subsequent language over time. They attempt therefore to define S0 for multilingual acquisition from the extant language capability, the specific languages of a learner, and the impact of these on the success of a learner’s cumulative acquisition of new languages. This attention to the temporal sequencing of language grammars in learners’ minds prior to acquisition of Ln is pursued through use of the Dominant Language Constellation concept to evaluate syntactic knowledge available to learners as a resource to identify how and what language connections exist in the minds of learners. In this way Fernández-Berkes and Flynn shift discussion of DLC towards the multiple grammars that people have in their minds, that must multiply further due to the exigencies of globalization so that larger numbers of people use multiple languages on a daily basis, thereby making some kind of multilingualism more normatively typical than it has been previously. This focus on multiple grammars links to the “implied competences in multilingual development” (Fernández-Berkes and Flynn, this volume), and asks what these precisely are, and at what level a learner needs to have attained these to be considered a multilingual. The radical difference from the sociolingistic analysis of Zambia by Banda suggested by these questions underscores the fecundity of the DLC notion. The analysis by Fernández-Berkes and Flynn ranges deep and far beyond sociolinguistic principles and realities into the challenge of building on proficiency indicators in mental and cognitive understandings of multiple language knowledge, across all skills. The task is then extended even further into attempting to ascertain what are the stocks of syntactic knowledge of learners when they commence study of additional languages. This prodigious intellectual task aims to understand the already-multilingual language learner. What does this multilingual learner (the trilingual or quadrilingual learner) know, and what language knowledge do they draw on to acquire additional languages? To address these questions Fernández-Berkes and Flynn report studies of learners of English who use the same languages on a daily basis. These learners are grouped together and subcategorized according to English proficiency as measured by a standardized test instrument. This procedure generates what we might call a skill platform or base from which they to explore the “initial state for multilingual acquisition”. In essence this approach contributes a mode of exploration of the role and contribution of accumulated linguistic knowledge as it sustains, or informs, subsequent language development. This ambitious aim of detecting initial state development from a multilingual perspective, essentially the mental connections of all languages known by subjects, maps out one promising conceptual line of future research for DLC informed research. Shifting focus from mind and learning to interaction between language systems in a multilingual subject and the effects on language acquisition is the work by

Quo Vadis, DLC?

269

Krevelj (this volume) on cross-linguistic interaction in multilingual production. The DLC concept is deployed in this quantitative research on simultaneous interaction of the languages of given multilinguals during a production task in L3 English. The research is based in Croatia, in settings using Croatian and Italian as official languages of the target research community, with the addition of English as a shared international language. Krevelj’s approach adds new insights into how supple and productive DLC proves to be as a conceptualization and research tool characterised additionally by a dynamic view of language as embodied by speakers and embedded in situated, highly specific settings. Critical to this study is an emergentist usage-based approach and the combination of qualitative and quantitative methodology for L3 acquisition studies. Cross-linguistic influence and effects studies are only now turning their attention to the contribution that DLC can offer to accounts of multilingual learning of additional languages. This research reveals DLC as a useful concept in multilingualism quantitative research on background languages, understood here in terms of how background proficiency is treated as a research variable. This shifts attention away from the perceived need for controlled proficiencies of existing languages, towards interactional effect of both high and low proficiency, high and low usage, existing languages. The DLC concept allows the researcher to describe and analyze data on the relationship between benefits and constraints of existing languages in new learning towards seeing the two languages as holistically and mutually relating, a coherent single unit tied together by the functional and regular presence of the languages in the communication behavior of the individual. As Krevelj notes “DLC allows for detailed analysis of interconnections and interactions between languages, language users, and the environment” and allows her to generate research that portrays the communicative social reality in which the participants reside and to depict precisely defined patterns in the configurations of the languages based on their interdependence. In this way DLC allows for a non-reductionist account of the role of languages in the life of individual and communities, and invites and facilitates deeper research investigations conceptualization. Through the work by Björklund, Björklund and Sjöholm in the context of Finland (this volume) our expansion of DLC turns towards the practical realm of teacher preparation. The writers’ analysis method was to apply the DLC as a tool to ascertain how multilingualism is embraced in policy and practice. Teachers in Finland are, according to Finnish curriculum guidelines, expected to promote the development of language awareness and multilingual competence among their students. In order to do so, teachers need to have well-developed multilingual awareness themselves. It is here that DLC shows its further usefulness in serving as a tool to describe domains and functions of different languages as unitary wholes. Björklund, Björklund and Sjöholm pursue their qualitative research agenda through data derived from language objectives and content as stipulated in national curricula in the years 2004 and 2014, supplemented by focus group interviews with pre-service student teachers in Swedish-medium teacher education, alongside immersion student teachers.

270

J. Lo Bianco

Finland is an officially bilingual nation and both of its official languages also serve as identity markers for its constituent national communities, the Finnish and Swedish speaking populations of the nation. Therefore the Finnish context supplies an already existing authoritative DLC for the entire nation, which is in turn variegated by how deeply individuals and sub-national groups are attached to either Finnish or Swedish, and other languages, such as personal, familial or ancestral ones which might differ from the official two, or still other languages they might have acquired for professional, recreational of occupational purposes. Interestingly, despite its official and national bilingualism, Finland is the least multilingual and multicultural Nordic polity. Only 6–7% of Finns speak languages other than Finnish (88.7%) and Swedish (5.3%). Finland also hosts a much smaller proportion of immigrants than its neighbor Sweden, where, the authors show, the proportion of immigrants reaches 20% of the national population. The research question however, and its relevance for future lines of development of DLC, concerns the robustness of DLC in research, and its ability to characterise a subset of existing national multilingualism, however great or small, and to inform national policy directions that aim to extend, in this case, ‘language awareness’, to all students. In this bilingual mix the official DLC however must be further widened to include English, whose global presence already made it locally important prior to Finland’s entry into the European Union in 1995, but significantly more so since. Inevitably most linguistic dispensations, arising as they do from historical struggles, are contested, as is the official bilingualism of Finland, where the minority Swedish presence is asymmetrically ordered compared to the nation-naming Finnish. Hence the official DLC of the nation, despite being entrenched in law and practice and of longstanding presence in the institutions of Finnish life, is contested and perhaps even unstable. Björklund, Björklund and Sjöholm pursue the questions of mandated language awareness as its connection to DLC in societal and individual domains and reveal the patterns of DLCs that emerge, of a “distinct development towards a more explicit recognition of diversity among students and in society”. This growth in inter-­ linguistic awareness is strongly linked to assessment descriptors used for Swedish and Finnish as L1s and SNLs and FL English, but absent for Sámi, a set of indigenous languages in the north of Finland and ranging across its national boundaries, as L1. In this mix need to be included the ‘languages of newcomers’, immigrant minority languages such as Arabic and Somali, geographically concentrated in cities, and exogamous but not global languages, of neighbor countries, Russian and Estonian in particular. The ultimate form that the Finnish DLC takes is therefore of a dominant endogenous Finnish, extended to include a dominant exogamous English, a non-dominant endogenous Swedish and a marginalized and separated Sámi. Extending the study of the DLC to the views of student teachers and others the authors reveal an even more patterned DLC that can be characterized along a temporal continuum of past, current and future orders of communication. The past pattern comprises mainly Swedish and Finnish and regional languages. The current communication order is of an eroding past pattern (through a weakened Swedish), with introduced

Quo Vadis, DLC?

271

immigrant languages (in subaltern position), and an expanded order of important global languages. The future pattern Is what is planned and imagined in curriculum documents and in emerging social arrangements and revealed in the present-day teacher attitudes and abilities, projected into the future. There is no guarantee that the future pattern will be realized but it is broadly visible. This work by Björklund, Björklund and Sjöholm enriches understanding of DLC by attaching a perspective of becoming, so that the national idealized DLC and multiple ‘in reality’ DLCs, in the dynamic space of education where future patterns of language competence are being fashioned, are clearly amenable to DLC-informed analysis that enrich the concept and help illuminate the operations of policy. Susan Coetzee-Van Rooy (this volume) continues the attention in other chapters on the nature of multilingualism at societal and individual levels but shifts the geographic focus far to the south in South Africa, using survey data to expose the functioning and nature of DLCs in “the multilingual minds of South African students”. The linguistic focus of the research by Susan Coetzee-Van Rooy is on Southern Sotho and Zulu as the home languages of participants in urban multilingual environments of Gauteng. Like previous authors, Coetzee-Van Rooy also links the DLC to language repertoire concepts to reveal the complementary relationship that exists between the two depictions of multilingualism. The size of a typical DLC is three languages tied together, as DLC stipulates, but also understandable through the specifiable repertoire skills individuals take from their DLC component languages, which is a wider multilingualism comprising of diverse kinds of language skill and usage. Using this reasoning Coetzee-Van Rooy proposes that “multilingualism scholars should approach their work from a language repertoire and a DLC view” because her language repertoire survey instruments and the complementary approach she deploys produces “empirical and comparable baseline data” that can in turn illuminate a great deal of how urban multilingual communication actually operates. Svaitlana Karpava’s study also addresses and compares a view of multilingualism from a language repertoire perspective with the view of multilingualism from the DLC view in a study on identity, language use and maintenance among Russian speakers in Cyprus. Using questionnaires and oral interviews the research studied 30 international Russian-speaking students living in Cyprus and 50 adult females, native speakers of Russian, members of Russian monolingual or mixed-marriage families, of an older age cohort. While all participants speak Russian not all identify as ethnic Russians originating as they do from Russia, Ukraine and Moldova. Hence, it is Russian language ability and use that is an integral component of their DLC, prominent in their individual and collective lives. English and Greek are the other main named languages that fill out the DLC, as Cypriot immigration, student status and trajectory, and local relations, are the situational and contextual envelope in which the DLC and repertories take shape. As the author explains it, the communication environment on the island of Cyprus is “diglossic, bidialectal” involving two varieties used by Greek Cypriots (Standard Modern Greek and Cypriot Greek) and also multilingual given various Cypriot minorities (e.g., Latin, Maronites), British colonial descended groups and

272

J. Lo Bianco

more recent immigrants from various countries. The study’s conclusions add a significant gender and family status dimension, so that female adults, mixed-marriage family members, have “either Russian or mixed (Russian and Cypriot Greek) cultural and linguistic identity, while the representatives of Russian-speaking families in Cyprus have mainly Russian or mixed (Russian and English) identity.” The DLC of Russian speakers is trilingual, Russian, English and Greek, the three critical functioning languages for “cognitive, social and emotional needs” but predictably there is variation according to the social status, marriage status and occupational position of individuals. By contrast the DLC of “Cyprus society” is complex and likely affects the DLC of the Russian-speaking minority community. The linguistic repertoires of the Russian L1 community are broader and more varied than their DLCs. As in Susan Coetzee-Van Rooy’s South African analysis we can see in this study the enrichment that emerges from fusing perspectives and data from repertoire analysis with DLC. In Nikolay Slavkov’s contribution on profiling of language background in Canadian elementary schools the DLC is extended in a different direction by testing the underlying understanding of language in procedures of language profiling against what such procedures might be if DLCs were more widely appreciated. The author notes that while language background profiling is under researched there are grounds to believe that it is widespread and used for important purposes, often involving large populations and with significant policy and life chance repercussions. How profiling of language backgrounds occurs is highly variable, often taking the form of simple questions of forms in which individuals make application for registration to participate in a service provided by public authorities, such as admission to formal education. In this process in Canada Slavkov finds that there is a “lack of explicit recognition of simultaneous bilingualism, a surprising finding in a country with two official languages” and that this omission is compounded by widespread use of categorical concepts that may suggest underlying monolingual assumptions and norms. The study notes the prospect of opening up these practices to more sophisticated procedures through use of insights and concepts developed by multilingualism studies (plurilingualism, translanguaging, multicompetence etc) and in this specific case the DLC. The overall result of this analysis is that DLC could assist administrators, but also parents of school enrolling children, to think more deeply and widely about how multiple languages can operate together as a single unified system in the lives of individual and groups, and that by doing so fairer, more nuanced and useful information would be generated. The research on DLC covered in this volume futher expands to include individual cases of different kinds. The potential that the DLC can offer as a lens or perspective to understand language use and affective domain is demonstrated in Richard Nightingale’s (this volume) examination of the communicative life of a Moroccan immigrant living in the ‘Valencian Community in Spain’. Using

Quo Vadis, DLC?

273

self-­report data Nightingale investigates relationships between the DLC and attitudes, emotion, and identity of the individual concerned. Generating qualitative data he asks how does a multilingual speaker use his or her DLC to navigate specific sociolinguistic contexts? What influence does the DLC have on the expression of identity and emotions in concrete daily situations? What role does the DLC play in the formation of language attitudes? Located in an acculturation context the case study offers a “small piece of the overall puzzle of multilingual practices realised in concrete social and cultural contexts”. The DLC revealed for the research participant, M, is one where the differentiation of languages is highly variegated, according to axes of function and degree of intensity. There appears to be a generic pattern of distribution that is situationally disrupted according to specific circumstances, events or feelings. Some languages from M’s repertoire enter the DLC on a “moment-to-moment basis in order to carry out context-­specific activities and practices which are vital to his daily life”. This reveals that the DLC is a stable undergirding of languages, “but is dynamic and constantly in flux” because new needs, or conditions in the communication environment, demand a response or change. Nightingale’s interview data highlights how the functions assigned to M’s DLC languages are both practical and affective as seen on occasions when M’s languages flow together in his processes of identity and yet at other times they sharply conflict, imposing on M the need to reorder or negotiate between them, and the emotional consequences of some new demand or change. Emotions and topics vary in their prominence and value in M’s DLC, sometimes producing separated use and at other times simultaneous use. M’s languages include his ‘heritage pair’ of Darija and Modern Standard Arabic, French and Castilian Spanish, Catalan, Italian and some English. Across this remarkable range of languages affective and practical life tasks are negotiated, distributed and evaluated. This DLC has been accumulated in a life biography of movement, study, professional life, friendships and sociality, all marked by both intellectual, practical and emotional/affective components. Sarasi Kannangara continues this evaluation of DLC from the personal lived experience in the context of Sri Lankan data based on the amount or quantity of total usage of the languages available to speakers in their DLC. Kannangara stresses that a central feature of a DLC is particularity or specificity, regardless of whether this refers to individuals or groups. The writer’s own individual DLC is revealed and discussed through illuminating graphs of a language repertoire charted according to proportion of usage time, revealing a “non-traditional childhood and adult multilingualism”. This all takes place within the context of the wider multilingualism of Sri Lanka, and constitutes therefore an auto-ethnographic socially situated communicative biography. The graphic representation of a dynamically lingual life reveals dramatic change involving English, Sinhalese, German and a distanced or marginalised, yet societally official, Tamil, varying in proportion and time from the wider societies in which the language choices are based.

274

J. Lo Bianco

2  Striking out in New Directions Like any new infant entering into the world many practical struggles have faced Larissa Aronin’s Dominant Language Constellation. Yet, like most infants, with nourishment and encouragement from intimates, DLC is now a fully ambulant, upright and well fed concept. DLC has taken many steps towards full independence, and having navigated scrutiny and responded to critique, it is now fortified with the antibodies of vitality and resilience. The locus within which the DLC concept resides has also expanded. It now robustly includes individuals but also groups, institutions within societies as well as whole societies, and sub-national parts or regions, of states. These overlap in the organic way that all human organization functions, but not in predictable or uniform ways, since any specific DLC is the outcome of historical processes of particular societies, and biographical ones for individuals. As the outcome of new linguistic dispensations, forged by pressure between past processes and current pressures, a DLC is always unique and at the same time intimately connected to the diachronic flow from which it emerges and the synchronic context in which it currently functions. A DLC has valency to the extent and for as long as its user, the individual or group, finds practical effect in its whole structure. I have claimed that the DLC is also a heuristic allowing rapid assessments to facilitate decision making, and also a scholarly concept buttressing sociolinguistic investigation. Multilingualism is a bigger and all-encompassing whole, a condition of humanity, and a DLC is a selection made from this condition of possibility for the exigencies of users. The many insights and extensive empirical evidence presented in our volume show that the concept of a Dominant Language Constellation is productive and generative. The direction of research, where the infant is now headed, is open ended, but the pathways are not limitless. They have been plotted in the chapters of this book and in recent thinking that will be followed by contributors and of course by Professor Aronin. What we can be certain of is that the DLC is likely to be a permanent and substantive citizen of the community of multilingual scholars because it makes a valuable contribution to the self-knowledge and comprehension we make of our communicative selves.

References Bazerman, M. H. (2017). Judgment and decision-making. In R. Biswas-Diener & E. Diener (Eds.), Psychology. Champaign: DEF publishers. Brown, M. (2009). Science in democracy: Expertise, institutions, and representation. Cambridge, MA: MIT Press. IAU (n.d.), The Constellations. International Astronomical Union https://www.iau.org/public/ themes/constellations/. Retrieved 20 April, 2019.

Quo Vadis, DLC?

275

Lo Bianco, J. (2019). Talking to the Pollies: Academic researchers and public officials. In C. Roever & G. Wigglesworth (Eds.), Social perspectives on applied linguistics and language testing: Studies in honour of Tim McNamara (pp. 89–105). Berlin: Peter Lang. McNamara, T., & Ryan, K. (2011). Fairness versus justice in language testing: The place of English literacy in the Australian citizenship test language assessment quarterly. An International Journal, 8(2), 161–178. https://doi.org/10.1080/15434303.2011.565438. Simon, H. A. (2008). Economics, bounded rationality and the cognitive revolution. Cheltenham: Edward Elgar Publishing.

Language Index

A African languages, 76, 77, 143, 239 Afrikaans, 85, 144–146, 148 Arabic, 13, 52, 78, 85, 98, 114, 190, 205, 239, 242–244, 246–248, 250, 252, 253, 267, 270 Arabic Chat Alphabet (ChA), 243, 244 B Bemba, 11, 79–92, 267 Berber languages, 239 C Čakavian dialect of the Croatian language, 218, 219 Caribbean, 122 Catalan (Valencian), 13, 238, 239, 242, 244, 249, 255, 273 Chinese, 48–53, 67, 191, 239 Chu Han, 48, 49, 51, 52 Chu nho/Chu Han, 49 Chu Nom/Nom, 49 Chu Quoc Ngu/Quoc Ngu, 49 Cognition, 8, 24, 25, 58, 69, 159, 160, 189, 213 Cypriot Greek, 188, 205, 206, 271, 272

D Darija, 13, 238–242, 244, 250–252, 254, 255, 273 E English, 5, 7, 9, 11–13, 51, 58, 59, 66–69, 76, 78–80, 82–92, 98–105, 107, 108, 110–113, 118, 119, 122, 124–127, 129, 144–146, 148, 151–160, 171–180, 182–184, 188, 190, 191, 193–206, 214, 217–219, 221–223, 225, 227, 238, 239, 243, 244, 266, 268–273 Estonians, 98, 114, 270 European languages, 76, 188, 239, 267 F Finnish, 11, 97–114, 269, 270 French, 13, 22, 49–53, 83, 108–110, 118, 119, 127, 148, 179, 181, 182, 190, 191, 205, 238, 239, 242–244, 248, 273 G Georgian, 191, 194, 198, 205 German, 67–69, 108, 109, 145, 179–183, 190, 205, 216, 273 Greek, 12, 188, 190, 191, 193–199, 201, 203–206, 271, 272

© Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7

277

Language Index

278 H Hebrew, 7, 9, 52, 77, 78 Hungarian, 60, 68, 69

Rumanian, 239 Russian, 7, 9, 12, 67, 68, 77, 78, 98, 109, 114, 187–207, 239, 270–272

I Indian, 122, 180 Istroventian dialect of the Italian language Italian, 217, 218, 221–225

S Sámi, 99, 101–106, 112, 270 Sanskrit, 179 Shona, 145, 148 Sign language, 100 Sinhala, 125, 171–174, 176–180, 182, 183, 185 Sinhalese, 170, 171, 174, 273 Slovenian, 216 Somali, 270 Southern Sotho, 12, 143–148, 151–161, 271 Spanish, 13, 67, 68, 78, 108–110, 145, 190, 191, 205, 236, 238–249, 251–255, 273 Swati, 144–146, 148 Swedish, 11, 98–113, 190, 270

J Japanese, 51, 60, 66–68, 179, 181, 182 K Kaonde, 11, 80, 82, 83, 85 L Latins, 78, 101, 188, 216, 271 Lozi, 11, 79, 80, 82, 83, 85, 90 Lunda, 11, 79, 80, 82, 83, 85 Luvale, 11, 80, 82, 83, 85 M Modern Standard Arabic (MSA), 238, 239, 241–244, 273 N Ndebele, 144–146, 148 Ndonga, 145 Northern Sotho, 144–146, 148 Nyanja, 11, 80–92, 267

T Tamil, 125, 170–174, 177–179, 181, 182, 184, 185, 273 Tonga, 11, 80–83, 85, 90, 91 Tsonga / Shangaan, 145, 148 Tswana, 144–148 Turkish, 37, 52, 53, 188, 190, 191, 205 U Ukrainian, 190, 191, 194, 198, 205 V Venda, 144, 145, 148 Vietnamese, 37, 48–53

P Portuguese, 49, 78, 145, 148, 171, 174 X Xhosa, 85, 144–148 R Roman alphabetic orthographic conventions, 50 Romance languages, 60 Romani, 100

Z Zambian languages, 76, 80, 81, 85, 91 Zulu, 12, 79, 88, 89, 143–148, 151–161, 271

Subject Index

A Additional languages, 87, 120, 122, 133, 212, 226, 267–269 Affective domain, 235, 237, 244, 256, 272 African languages, 76, 77, 143, 239 African youth language, 76 Age of onset, 65, 190, 191, 215, 217, 219, 221 Awareness, 4, 8, 87, 100, 102, 105, 106, 112, 114, 133, 176, 179, 189, 190, 199, 200, 234, 255, 269, 270 B Background languages, 58, 212, 214, 215, 223, 226, 269 Bilingual education, 44, 124, 175 Bilingualism, 1, 22, 98, 113, 118, 126, 133, 136, 172, 176, 177, 216, 223, 232, 270, 272 Bilinguals, 2, 4, 5, 11, 21, 67, 76, 98, 100, 101, 110, 111, 113, 121, 124, 127, 132, 133, 135, 146, 171, 173, 175–178, 180, 188, 192, 194, 203, 212, 217, 223, 227, 264, 270 C Code-mixing, 133 Cognition, 8, 24, 25, 58, 69, 159, 160, 189, 213 Colonial language, 48, 76, 77 Common (“the same”) DLC, 12 Common European Framework of Reference for Languages (CEFR), 100, 103, 104, 123

Communities, 3, 5, 7, 10, 13, 24, 27, 28, 30, 38–41, 44, 47, 54, 58, 77–79, 81, 83, 84, 90–92, 106, 112, 121, 124, 140, 141, 158, 160, 162, 170–172, 174, 187–190, 192, 193, 205–207, 215–219, 223, 227, 232, 234, 237, 239, 264–267, 269, 270, 272, 274 Community DLC, 5, 7, 10, 28, 39, 77, 81, 84, 90–92, 140, 162, 170, 234, 265, 266, 269, 272, 274 Competences, 11, 45, 58–61, 66, 69, 100, 102–105, 108, 109, 112, 113, 121, 123, 180, 183, 218, 219, 221, 225, 234, 236, 238, 239, 245, 249, 254, 255, 268, 269, 271 Complex, 3, 5, 9, 11, 12, 21, 22, 24, 26, 42, 50, 53, 68, 76, 79, 81, 84, 90–92, 121, 129–131, 134, 135, 141, 143, 159, 160, 170, 188, 189, 193, 205, 206, 213, 224–225, 232, 236, 237, 240, 243, 247, 262, 267, 272 Complexities, 5, 7, 10, 22–26, 29, 30, 59, 63, 121, 126, 133, 135, 143, 161, 176, 177, 184, 190, 195, 197, 212, 233, 262 Composition of the DLCs, 12 Configurations, 6, 11, 12, 29, 39, 43, 50, 77, 84, 85, 92, 158, 159, 189, 190, 192, 194, 199, 201, 202, 204–207, 213, 217–227, 234, 244, 266, 269 Constellation maps, 125 Culture identity, 190, 192–194 Cumulative Enhancement Model for Language Acquisition (CEM) Curriculum/ curricula, 10, 64, 65

© Springer Nature Switzerland AG 2020 J. Lo Bianco, L. Aronin (eds.), Dominant Language Constellations, Educational Linguistics 47, https://doi.org/10.1007/978-3-030-52336-7

279

Subject Index

280 Curriculum/curricula, 2, 11, 40, 98–108, 111–114, 119, 120, 122, 201, 202, 217, 220, 221, 269, 271 Croatia, 12, 13, 214–221, 223, 225, 269 Croatophones, 217–221 Croatophone speakers, 217–221 Cross-linguistic influence (CLI), 10, 13, 59, 212, 269 Cross-linguistic interactions (CLI), 29, 59, 64, 66, 69, 70, 212–218, 222–228 Culture identities, 190, 192–194 Cumulative enhancement model for language acquisition (CEM), 10, 64–67 Curriculum/curricula, 2, 11, 40, 98–108, 111, 113, 114, 217, 221, 269, 271 Cyprus, 12, 187–207, 271, 272 D Diaspora, 40, 41 Diversity ethnic, 118, 216 linguistics, 2, 6, 11, 26, 42, 43, 91, 98, 100, 107, 118, 120, 136 DLC maps, 12, 30, 135 DLC units, 11, 29 DLC valency, 264, 274 Domains of use, 219 Dominant Language Constellation (DLC) common (“the same”) DLC, 12 communal, 77 groups, 170 individuals, 170 institutional, 11, 184 multi-layered, 10, 76–92 personal, 170, 173–184 societal, 78, 102, 112, 184 E Education, 1, 4, 5, 7–11, 13, 21, 22, 26, 30, 37, 39, 40, 42–46, 48–50, 52, 53, 76, 78–80, 84, 85, 91, 92, 98–102, 104, 106, 111, 113, 119, 120, 122, 127, 133, 136, 140, 143, 144, 146, 152–156, 159–163, 171, 172, 174–177, 179–182, 189, 191, 196, 198, 201, 202, 204–206, 217, 218, 220, 222, 223, 225, 228, 234, 236, 238, 239, 252, 264–266, 271, 272 Educational attainment, 99 Educational contexts, 11, 97–114, 118 Educational systems, 11, 23, 118, 119, 132, 133, 201, 217, 227 Education of language immersion teachers, 99

Elementary schools, 11, 117–136, 272 Emergences, 5, 24, 48, 159, 243, 255, 264, 267, 268 Emergent, 5, 20, 24, 29, 81, 101, 158, 172, 190, 192, 205, 265 Emotions, 13, 25, 29, 140, 199, 232, 233, 235–237, 244–247, 251, 252, 254, 255, 273 English language learner (ELL), 120 European language, 76, 188, 214, 239, 267 Evaluations, 10, 99, 255, 273 F Family language policies, 12, 133, 187, 189, 190, 192, 201, 205, 207 Finland, 11, 98–101, 105, 107, 109–114, 269, 270 First language, 60, 100, 102–104, 112, 114, 120–122, 125–134, 177, 212 First-language (L1) acquisition, 61 Foreign languages, 40, 101–103, 105, 108, 112, 176, 188, 193, 194, 205, 217–219, 239, 248, 265 Francophone, 119, 126, 127 Functions, 3, 11, 12, 22, 26–28, 38, 40, 44, 50–52, 62, 78, 90, 101, 104–113, 140–142, 146, 150, 152, 160, 161, 170, 172, 173, 183, 184, 189, 193, 195, 197, 206, 207, 213, 218, 222–225, 232–234, 237, 239, 243, 244, 252, 254, 255, 262, 263, 266–269, 273, 274 G Generative framework, 61 Germany, 12, 110, 181–183 Globalization, 13, 58, 268 H Heritage culture, 248 Heritage languages, 8, 239, 252, 255 Holistic, 13, 30, 77, 114, 124, 212, 224–227, 232, 236 Home languages, 12, 84, 92, 120, 121, 126, 127, 129–134, 143–149, 151–159, 161, 192, 271 I Identification with (a language), 223 Identities, 5, 6, 12, 13, 40, 41, 43, 46, 47, 77, 100, 101, 103, 105–113, 122, 140, 142,

Subject Index 154, 156, 160, 170, 188, 189, 192, 194, 206, 213, 223, 225, 232, 233, 235–237, 243–245, 249, 251, 252, 254, 255, 265, 270–273 Immigrants, 4, 12, 39, 41, 47, 52, 98, 99, 118, 133, 187, 188, 191, 197, 202, 206, 207, 232, 234, 239, 242, 244, 254, 255, 270–272 Indexicalities, 224, 234 Indigenous languages, 11, 76, 79, 83, 84, 91, 92, 98, 118, 266, 270 Initial state of language development/initial state (S0), 58 Institutional DLC, 11, 37 Institutional policymaking, 13 Interlingual awareness, 106 Istria County, 13, 216, 217 Italophones, 216, 217, 221–222 L Language attitudes, 232, 235, 236, 245, 273 Language background profiling, 11, 120–125, 127, 129, 132–135, 272 Language behaviors, 125 Language choices, 53, 90, 201, 235, 236, 273 Language contact, 13, 76, 77, 81 Language domain, 11 Language faculty, 3, 61 Language functions, 189, 235 Language identities, 102, 103, 109, 111, 184, 187, 193 Language maintenance, 41, 188, 189, 207 Language policies, 1, 2, 10, 11, 13, 22, 23, 30, 35–54, 88, 160, 170–173, 184, 189, 228, 265, 266 Language practices, 2, 3, 20–22, 24–26, 30, 91, 196, 201, 232 Language repertoires, 3, 5, 9, 11, 12, 27, 70, 101, 109, 113, 121, 123, 124, 139–162, 175, 176, 178–180, 205, 233, 243, 271, 273 Language skills, 3–5, 8, 21, 48, 58, 141, 142, 149, 271 Languages other than English (LOTE), 122, 127 Language status, 85, 91, 100, 215 Language syllabi, 100 Language use, 6, 12, 21, 23, 24, 26, 30, 53, 76, 77, 102, 103, 114, 121, 127, 133, 143, 176, 182, 187, 189, 190, 192, 198, 205, 207, 227, 234, 236, 237, 241, 267, 271, 272 Language vitality, 76, 188

281 Lingua franca, 5, 84, 92, 99, 184, 188, 205, 206, 216, 234 Literacy/literacies, 42, 50–53, 77, 84, 267 Local languages, 80, 83, 85, 87, 206 Longue durée, 10, 22, 25–26, 30, 264 M Marginalization, 41, 122, 217 Migrations, 2, 40, 77–80, 91, 124, 195, 265, 266 Minority languages, 41, 76, 81, 100, 104, 112–114, 121, 133, 189, 201, 207, 248, 270 Models, 1, 3, 6–9, 12, 29, 30, 46, 62–67, 70, 140, 213, 224, 227, 236, 255, 256, 264 Monolingual standards, 122 Morocco, 12, 238–241, 243, 244, 250 Multiculturalism, 118 Multilingual awareness, 11, 98, 101, 269 Multilingual contexts, 10, 21, 42, 75–92, 140, 159, 235 Multilingualism, 172, 174, 267 current multilingualism, 2, 3, 22–26, 123, 161, 170 South African multilingualism, 161 Sri Lankan multilingualism, 171 Zambian multilingualism, 11 Multilinguality, 45, 46, 190, 192, 203, 205, 212, 233, 235 Multilingual learners, 58, 59, 63–68, 70, 268 Multilingual portraits, 140 Multilingual production, 211–228, 269 Multilinguals, 2–7, 10–13, 19–30, 37, 39, 40, 42–46, 50, 51, 57–70, 76–78, 80, 81, 85, 91, 92, 98, 100, 101, 106, 110, 111, 113, 114, 119, 121–126, 132–136, 139–161, 170–172, 182, 188–190, 192, 193, 203, 205, 207, 212, 213, 215, 222–228, 232–237, 240, 243, 244, 247, 249, 252, 254, 256, 262, 264–271, 273, 274 Multiple language use, 44, 101, 102, 113 N National languages, 4, 11, 37, 42, 48, 51, 52, 87, 98–100, 102, 103, 105, 106, 109, 110, 112, 171–174, 219, 221, 264, 266 Native speakers, 38, 113, 121, 122, 124, 176, 180, 190, 201, 202, 218, 271 Non-official languages, 91, 92, 118 Northern Morocco, 239

Subject Index

282 O Order of acquisition, 121, 215, 218, 219 Orthography/orthographies, 10, 48, 49, 51, 52 Other languages, 5, 50, 59, 63, 78–80, 82–84, 91, 98–100, 106, 107, 109–111, 113, 122, 123, 125, 127, 129–132, 134, 144, 190, 191, 195, 196, 198, 200, 205, 213, 224, 242, 254, 263, 270 Ownership (of a language), 106, 222 P Patterns, 4, 6, 9, 11, 20, 23, 24, 26, 29, 30, 36, 38, 45, 47–50, 58, 59, 84, 97–114, 125, 129–133, 175, 183, 191, 192, 196, 212, 215, 224–227, 234, 253, 262, 266, 269–271, 273 Plurilingual, 123, 234 Plurilingualism, 2, 41, 45, 123, 124, 136, 272 Praxis, 35–54, 263 Pre-service programs, 99 Prestige, 40–42, 122, 173, 184, 188, 216, 217, 234, 239, 267 Primary education, 76, 80, 83, 174, 179 Primary schools, 174, 180, 217, 218, 221, 234, 266 Productive abilities, 121, 134 Productive skills, 134, 136 Proficiency, 8, 9, 12, 13, 28, 46, 58, 65, 123, 142, 149, 170, 172, 176, 206, 212, 214, 215, 218, 220, 221, 224–227, 268, 269 Profiling, 11, 117–136, 272 Provincial languages of wider communing, 81–83 Public schools, 11, 120, 203 Q Questionnaires, 118, 120, 121, 123, 127, 132, 149, 150, 191, 192, 271 R Receptive abilities, 121, 134 skills, 134, 136 Russia, 190, 191, 193–195, 204, 271 Russian, 7, 9, 12, 67, 68, 77, 78, 98, 109, 114, 187–207, 239, 270–272 S Script reform, 37, 53 Scripts, 10, 37, 48–54

Self-image, 47, 193, 233 South Africa, 12, 42, 45, 143, 144, 146, 147, 152, 161, 162, 271 South African urban multilingualism, 161 Southern Sotho, 12, 143–148, 151–161, 271 Spain, 12, 232, 238–241, 243, 244, 262, 272 Specific Language Repertoire Survey Instrument (SLRS), 11, 12 Sri-Lanka, 12, 170–177, 180, 182–184, 273 Standardization (of profiling questions) Super-diversity, 5, 6, 21, 23, 30, 43, 44, 233 Surveys, 11, 12, 140–143, 145, 149–151, 157–161, 271 Swedish immersion, 99, 100, 108, 109, 113 Swedish-medium schools in Finland, 99 Syllabus/syllabi, 100, 103, 105, 112 Syntactic development, 57–70, 234 Syntax, 10, 23, 25, 60, 69 T Target language, 58, 60, 63–70, 106, 107, 122, 214 Teacher education, 11, 99–101, 112–114, 234, 269 Teacher education programmes, 99 Third language acquisition (TLA), 212 Transfers, 39, 59, 62–66, 177, 214, 215 Translanguaged discourses, 78 Translanguaged unit, 78 Translanguaging, 1–4, 12, 41, 46, 78, 123, 124, 136, 243, 244, 272 Trigraphia, 51, 52 Trilingual, 268, 272 U Universal Grammar (UG), 61–64, 66 V Valencian Community, 238, 242, 272 Vernaculars, 49, 173–175, 239 Visualisation, 30, 125, 243 W Western Sahara, 239 Writing systems, 37, 49, 50 Z Zulu, 12, 79, 88, 89, 143–148, 151–161, 271