Maintaining Control Autonomy and Language Learning Edited by Richard Pemberton, Sarah Toogood and Andy Barfield © RASH RE KonG UNIVERSITHong Kong University Press Contents Contributors vn jon of ti pubtnton Introduction 1 7 3 Theories and discourses of autonomy and language learning 11 2. 1g sense of autonomy in language 13 Benson 3. Crash or clash? Autonomy 10 years on 27 Edith Escl 4, Discursive dissonance in approaches to autonomy 45 Philip Riley Practices of learner autonomy 65 5. Controlling learning: Learners’ voices and relationships 7 between motivation and leaner autonomy Terry Lamb | 6. Learner autonomy in a mainstream writing course: 87 Arti Sara Ci arning gainsji Contents 7. Reflective lesson planning: Promoting learner autonomy. in the classroom Lindsay Miller 8, The use of logbooks — a tool for developing learner Practices of teacher autonomy 9, Learner autonomy, the Europe and teacher development David Little inguage Portfolio 10. The teacher as learner: Developing autonomy in an interactive learning environment Barbara Sinclair Il, Defending stories and sharing one: Towards ‘understanding of teacher autonomy Naoko Aol with Hiroaki Kobayash 12, Autonomy and control in curriculum development: “Are you teaching what we all agreed?’ ‘Mike Nix and Andy Barfield Commentary 13. Autonomy: Under whose control? Richard Smith and Ema Ushioda Notes References dex 109 125 145, 7 175 199 239 2a Contributors Naoko Aoki is Professor in the Graduate School of Letters at Osaka University, Japan, and works with pre-service and in-service teachers of Japanese as a Second Language (SL). She is a co-founder of JALT’s Learner Development SIG and has published on the topics of learner autonomy and teacher autonomy in the context of JSL. Her current pedagogical and research interests are narrative-based teacher education, particularly applications of narrative inquiry, and second language user (rather than learner) stories. Her latest pet projects are the Japanese Language Portfolio and bringing a plurilingual Japan into Andy Barfield is Professor in the Faculty of Law, Chuo University, Japan. His publications include Autonomy You Ask! (2003; co-edited with Mike Nix), Reconstructing Autonom edited with Steve Brown), Lexieal Proces working nother Language: Multiple interpretations (2009, co-edited with Henrik Gyllstad). Andy has taught in the UK, France, Spain and Yugoslavia, and he is co-editor of the IATEFL Learner Autonomy SIG newsletter, Indep Phil Benson is Professor in the Department of English at the Hong, Kong Institute of Education, He has worked in Hong Kong since 1991, and formerly taught English and Applied Linguistics in the English Contre at the University of Hong Kong. His main research interests are rrative-based research. He is the author nomy in Language Learning (2001), and nce and diversity in language learning in autonomy in of Teaching and co-editor of Learners’ Stores: Dif Sara Cotterall taught in the School of Linguistics and Applied Language Studies at Victoria University of Wellington, New Zealand, from 1986vill Controutors to 2005, and from 2005 to 2008 was Associate Professor at Akita International University in Japan, where she taught courses in Independent Language Leaming and Communication, Between 1996 and 2002 Sara was co-convenor of the AILA Scientific Commission on Learner Autonomy in Language Learning and in 2007 she was co- convenor of the Independent Learning Association Japan Conference. Sara is currently a PRD student at the University of Macquatie, Austra Her publications include Learner Autonomy in Language Learning: Def the field and effecting change (1999, co-edited with David Crabbe). Leni Dam took her first steps towards developing leamer autonomy in language teaching and learning in 1973 at Karlslunde school near Copenhagen, Denmark, where she taught for 40 years. From 1979 she also worked as an in-service teacher trainer and pedagogical adviser at University College, Copenhagen, a post which provided excellent opportunities for implementing the principles of learner autonomy in teacher education. Her publications include Learner Au 3: From theory to classroom practice (1995). From 1993-1999, Leni was co-convenor of the AILA Learner Autonomy in Language Learning Scientific Commission, In 2004 she was awarded an honorary doctorate from Karlstad University, Sweden. In 2007 Leni retired and is now working freelance, applying her experiences with school children to the language education of adult refugees, Edith Esch is Senior Research Fellow in the Faculty of Education, University of Cambridge, UK, and a Fellow of Lucy Cavendish College. Her interest in autonomy dates to the beginning of her career when she worked with the CRAPEL (Centre de Recherches et d’Applications Pédagogiques en Langues) in Nancy, France. She developed this interest further in Cambridge as Director of the University Language Centre Her main current interest is in second language education, especially the British and French pedagogical cultures in post-colonial contexts ‘These sociolinguistic and sociocultural themes are the extension of her life-long, interest in bilingualism and cross-linguistic communication and language change. Her publications include The Bilingual Family: A handbook for parents (1985 & 2003, with Philip Riley) and Self-Access and the Adult Language Learner (1994), Terry Lamb taught languages in secondary schools for 16 years, and ow works at the University of Sheffield, UK, where he is Director of Contoutors Initial Teacher Education and of the MA in Applied Professional Studies in Education, as well as supervising doctoral students in his research fields of learner autonomy, teacher education, linguistic diversity and language policy. Terry has published widely in these fields, and is also convenor of the AILA Research Network on Learner Autonomy in Language Learning and founder editor of the journal Innovation Language Learning and Teaching. Terry's work with a number of key national and international bodies involves him in policy development in the UK and beyond. Since January 2007, he has been President of the ationale des Professeurs de Langues Vivantes. Fédération Inter David Little retired in 2008 as Head of the School of Linguistic, Speech and Communication Sciences and Associate Professor of Applied Linguistics at Trinity College Dublin. His principal research interest is the theory and practice of learner autonomy in second language education. From 2001 to 2008 he was Director of Integrate Ireland Language and Training, a government-funded unit that provided English language courses for adult newcomers with refugee status and supported the learning of English as a second language in Irish schools. He is currently chair of the Council of Europe's European Language Portfolio Validation Committee. His numerous publications on learner autonomy include Learner Autonomy 1: Definitions, issues and problems (1991), Lindsay Miller is an Associate Professor in the Department of English at the City University of Hong Kong, where he teaches courses in self- access learning, materials development and critical pedagogy on BA and MA TESOL programmes. He also trains secondary school teachers. He researches and publishes in the areas of self-access learning and listening, He has published Second Language Listening (2005, with John Flowerdew) and Establishing Self-Access: From theory to practice (1999, with David Gardner). He has worked in primary, secondary and tertiary level educational establishments in the UK, the Middle East, Thailand and Hong Kong. Mike Nix is Professor in the Law Faculty at Chuo University, Japan His research and teaching focus on helping students to use English to engage critically through research, discussion and writing, with legal, political and global issues of concern to them. He is also interested in questions of identity in language learning, and use, especially in selationx Contibutrs to the glocalisation of English, as well as in exploring possibilities for collaborative curriculum and teacher development with full-time and part-time colleagues. His publications include Autonomy You Ask! (2003, co-edited with Andy Barfield) Richard Pemberton is Associate Professor in TESOL in the School of Education at the University of Nottingham, UK, where he teaches and supervises MA and PhD students, He taught for nearly 15 years at Hong Kong University of Science and Technology, where he co-ordinated the Self-Access Centre team, and before that taught ESL at secondary and tertiary level in the UK, Zimbabwe and Papua New Guinea. He co- edited Taking Control: Autonomy in language learning (1996, with Edward Li, Winnie Or and Herbert Pierson) and his interests include learner autonomy, L2 listening and vocabulary acquisition, and technology- enhanced language learning, Philip Riley is Emeritus Professor of Ethnolinguistics at the University of Nancy, France, and a former Director of the CRAPEL (Centre de Recherches et d’ Applications Pédagogiques en Langues). He has taught English, Linguistics and Language Didactics in Finland, Malta and France and has made extended visits to Italy, Hong Kong and New Zealand. His main areas of interest include autonomous language learning, identity studies and intercultural communication. His latest n-specific English: Textual practices across ms (2002, edited with Giuseppina Cortese) and ty: An ethnolinguistic perspective (2007). Barbara Sinclair is Associate Professor of Education (TESOL) in the ‘School of Education at the University of Nottingham, UK, where she has worked since 1992, She is currently Director of the MA TESOL and supervises research students. For over 25 years, her research has focused on issues relating to the development of autonomy in language learners, including metacognition, ‘learner training’, the assessment of learner autonomy and, more recently, autonomy in on-line learning. Her publications include Learning to Learn English (1989, with Gail Ellis) and Learner Autonomy, Teacher Autonomy: Future directions (2000, co-edited with lan McGrath and Terry Lamb). She has worked in Spain, Germany and Singapore, and has provided consultancies and courses on learner autonomy in Mexico, Turkey, UAE, Austria, Malaysia, Brunei and Cyprus Conrbuters 6 Richard Smith is an Associate Professor in the Centre for Applied Linguistics, University of Warwick, UK. His research interests and professional activities are mainly in the areas of learner autonomy, teacher development, cultural issues in ELT and the history of language teaching, He co-edited Learner Autonomy across Cultures: Language education perspectives (2008, with David Palfreyman) and has a particular interest in innovative approaches to language teaching ‘in difficult circumstances’. He runs the DAHLA project (‘Developing an Archive and Histories of Learner Autonomy’) as a means of recording practice in the field of pedagogy for autonomy, oversees the online journal English Language Teacher Education and Development and is co-editor of Independence, published by the [ATEFL Learner Autonomy SIG. Sarah Toogood is an Instructor and Self-Access Language Learning ‘Adviser at the Hong Kong University of Science and Technology. She ‘was President of the Hong Kong Association for Self-Access Learning, and Development from 2001 to 2005 and continues to be an active member. Her main interests are concerned with devising ways of integrating self-access language learning into the curriculum and looking in particular at how to advise and assess for'selh-access language earning projects. She has published widely in these areas and has conducted teacher-training workshops on self-access language learning, and advising for tertiary and secondary level contexts, Her current research is developing a Virtual English Language Adviser (VELA): http:/ /vela.ust.hk/. Ema Ushioda is Associate Professor in the Centre for Applied Linguistics, University of Warwick, UK, where she teaches MA courses and coordinates the Education Doctorate programme. Her main research interests include language learning motivation and its theoretical interface with learner autonomy, and relevant implications for classroom practice and teacher education (Learner Autonomy 5: The role of mot 1996). In recent years, she has been exploring language motivation from the perspective of Vygotskian sociocultural theory as well as theories of social and cultural identity, and is currently developing an approach to examining how language motivation is socially constructed in discourseIntroduction1 Maintaining Control: An introduction Richard Pemberton, Sarah Toogood and Andy Barfield The origins of this book lie in a major conference entitled ‘Autonomy and Language Learning: Maintaining contro!’ held in Hong Kong and Hangzhou (mainland China) in June 2004, That conference was the younger sibling of another important conference held 10 years earlier, also in Hong Kong and mainland China, which formed the basis of the book Taking Control: Autonomy in language learning (Pemberton et al 1996). Back in June 1994, atthe time of the first of these two conferences, the concept of autonomy in language learning — together with related. practices of self-directed and self-access language learning (SALL) — had been around for some 20 years, starting out from the Centre de Recherches et d’Applications Pédagogiques en Langues (CRAPEL) at the University of Nancy in France in the early 1970s (cf. Harding-Esch 1977a; Holec 1979, 1981; Riley 1985) and spreading to the UK, Denmark, Ireland and elsewhere in Europe. At the wider international level, the concept of autonomy in language learning was starting to become more popular: a Learner Autonomy Scientific Commission had been formed as part of the Assocation Internationale de Linguistique Appliquée (AILA) and had just held its first symposium. However, autonomy had not yet become part of mainstream theory and practice in second language education. East Asia, the region that the three of us are most familiar with, is a case in point. In Japan, interest was just beginning to develop, as evidenced and aided by the formation of the Japan Association of Language Teachers (JALT) Learner Development Special Interest Group in 1993, and the publication of its first newsletter Learning4 Fichard Pombortn, Sarah Toogood and Andy Batoks Learning in Spring 1994 (see Andy's ‘story’ below and Richard Smith’s ‘autobiography’ in Chapter 13). In Hong Kong, the concept of learner autonomy was familiar toa relatively small group of teachers involved in supporting SALL at university level (and in some companies) (e.g, Gardner & Miller 1994), but was not familiar to local language teachers in general. In mainland China, where self-access had yet to ‘take off the concept was even less well known, and to our knowledge there had been no publications or presentations on the subject, ‘Ten years later, however, presentations. at the 2004 conference in Hangzhou by both local and overseas participants were filled to overflowing, and to date more than 35 papers written by participants, from mainland China have resulted from this landmark event (see Sarah's ‘story’ below). This growth of interest in China is part of a global trend, which, as Phil Benson (this volume) details, has seen autonomy take up a central position in second language education literature since the turn of the millennium, becoming, in Benson's words, “an idea that researchers and teachers ignore at their peril’ Similarly, Richard Smith and Ema Ushioda argue in their concluding chapter that the expansion in the ‘autonomy movement’ from small university-based circles scattered here and there to a much larger and more diverse grouping means that new voices need to be listened to and conflicting interpretations engaged with. But just as itis vital to keep opening out to different narratives and understandings, it is also important to understand how we got to where we are today. this spirit, then, that we would like to share with you our stories of how the two conferences and this book came to be. Taking control in 1994: Richard's story In the early 1990s, when I arrived in Hong Kong, the University Grants ‘Committee had decided to provide each of the seven UGC universities / polytechnics with a language enhancement grant, in order to improve the language proficiency of Hong Kong undergraduate students (which was perceived to be in decline) and to maintain Hong Kong's position as a regional international financial centre in the face of competition from Singapore and Shanghai. These funds, involving, very large sums of money, allowed universities to employ more language teachers, so that the number of language classes could be increased and class sizes reduced. They were also a vital ingredient in the mushrooming of Maintaining Contot 5 university self-access centres (SACs) across Hong Kong in the early 1990s. Within a few years, each institution had set up its own SAC so that learners could develop their language skills outside regular class time; there was an active Association for Self-Access Learning and Development (HASALD) drawn largely from university teachers; and Hong Kong quickly developed a ‘cutting edge’ reputation for the design of SACs and support of SALL at university level. In the early days, as we planned and started running our SACs, we were very much focused on practical issues such as lay-out, shelving, ataloguing and copyright. However, extended consultancy visits to Hong Kong in 1992 and 1998 by Philip Riley helped me make the connection between the ‘what’ of self-access and thé ‘why’ of autonomy. —to sce why we were supporting self-access in the first place. To borrow Phil Benson’s (2002b: 4) description of his own growing awareness under Philip Riley's guidance: “Ever so gently, Philip made me aware that there was not much point to self-access without autonomy” However, in 1993, as an SAC coordinator, with our own centre about to open, SACs and SALL were still very much to the forefront of my mind, So when, in the same year, the director of our Language Centre at the Hong Kong University of Sciencé and Technology (HKUST) asked for volunteers to convene the next ini our series of annual joint-venue conferences, my first suggestion for a theme was not ‘autonomy’ but ‘self-access’. Luckily, our director suggested a broader theme, and so ‘autonomy’ it was. (Interestingly, Phil Benson [2002b] reports a similar intervention by his director, David Nunan, which resulted in the Benson and Voller [1997] book having an ‘autonomy’ rather than a ‘self-access’ theme.) With an appropriate conference focus in mind, I then asked Herbert Pierson if the Chinese University of Hong Kong (CUHK) would be our institutional partner for the conference. At the time, CUHK had just created a very impressive open-plan Independent Learning Centre, and Herb, as ILC director at the university, had been responsible for bringing Philip Riley to Hong Kong to advise SAC teams at all the tertiary institutions about both the theory and practice of SALL. Herb agreed, and my colleague Austin Conway joined as co-organiser from HKUST. ‘Together we set about planning the conference. ur first decisions concerned who to invite as keynote speakers. With two sponsoring institutions and support from the British Council, ‘we were able to fund the attendance of the following five pioneering figures in the fields of learner autonomy, self-access and learning to16 Richard Pemberton, Sarah Toogood anc Andy Batol learn: Edith Esch (University of Cambridge, UK), David Little (Trinity College, Dublin, Ireland), Philip Riley (University of Nancy, France), Barbara Sinclair (University of Nottingham, UK) and Ken Willing (Macquarie University, Sydney, Australia). We were also lucky that David Nunan had recently joined the University of Hong Kong as director ofits English Centre, and had accepted our invitation to give a plenary presentation — and so there were six keynote speakers. The schedule we decided on for the conference was to have two days at HKUST, followed by a morning at CUHK, and then an afternoon trip across the border into mainland China to visit Shenzhen University, where the final keynote presentation, by David Little, was to be given Cramming three locations and two countries into three days seemed like a good idea at the time, but the final afternoon trip across the border with no air conditioning in a packed coach (Austin Conway had to perch on David Little's knee) was — in hindsight — overdoing, it, memorable though the trip was. the conference was very successful, attracting some 150 enthusiastic participants from Asia, Oceania and Europe. It was one of those occasions where something occurs in the right place at the right time. Hong Kong University Press had just published Directions in Self Access Language Learning (Gardner and Miller 1994), which had been well received, and offered to publish selected papers from the conference as soon as they heard about it. My SAC team colleagues Edward Li and Winnie Or agreed to join Herbert Pierson and myself as editors, and the four of us set about editing the volume that came to be called ‘Taking Control ittle did 1 think that 10 years later 1 — along with many of the participants at the 1994 conference — would be helping to bring Taking Control's younger sister into the world. Maintaining control in 2004: Sarah's story The idea behind holding the 2004 conference was not, as some may have initially thought, ‘a good excuse for a reunion party’, although that may have been one of its positive outcomes. The opportunity to hold a 10-years-on conference was, in fact, quite serendipitous. One afternoon in early 2003, I was called into our director's office. He wanted me to convene the next Language Centre conference and had some ideas as to we and potential collaborative partners. As listened, ‘Maintaining Contr’ 7 I decided that I would be happy to convene a conference but would prefer the focus to be on autonomy and language learning. My proposal ‘was met with approval and I was given full control to organise the conference as I wished. {As soon as I left the director's office, I sought out Richard and told him the news. He thought it was a wonderful idea and, being good at maths, saw an immediate marketing angle if we connected the 2004 conference with the 1994 conference that he had convened. I saw an immediate opportunity for a bit of word play. As Taking Control was the name of the book that resulted from Richard's 1994 conference, ing Control was the name I proposed for our 2004 conference the prospect that in another 10 years, we night have fully lost control and would have a hat-trick to mark the end of our careers Richard, weary of my puns, wasn't too keen on the idea. There was more to it, however. Having spent almost 10 years at HKUST researching ‘ways in which to improve the support of SALL for our learners through the SAC and through course integration, my feeling, was that the issues we faced as frontline promoters of language learner autonomy were now more to do with progress and maintenance rather than beginnings and taking, As Richard mentioned above, many tertiary institutions in Hong Kong had been given rich resources to set up SACs, providing. learners with the opportunities required to ‘take’ control. Yet there were instances where some of these institutions had lost their space, people had moved on and materials were merged with the main library collections. At HKUST we had-experienced a situation where teachers, had begun to refer to our SAC as a white elephant. This galvanised a small group of us to propose changes to our provisions which aimed to ‘maintain understanding, interest and collaboration among colleagues and learners. It seemed to me that the idea of maintaining control as a progression from taking control could be an inspiring concept for practitioners and researchers dealing with the need to ensure progress and continuation not only of ‘set-ups’ in the form of SACs but also in provisions that scaffold autonomous language learning inside and outside a SAC. Richard and I then agreed on Maintaining Control as a working title. Our next step was to set up a team of commitice members and, in keeping with the idea of ‘one conference, more than one venue’, we proposed a collaboration with Pang Jixian, vice-dean of the School of International Studies at Zhejiang University in Hangzhou. Pang readily agreed to 2 joint venture whereby participants would be at HKUST for18 Fichard Pemberton, Sarah Toogood ard Andy Bartel the first two days and then move on to Zhejiang University for the following two days, with a day for travelling in between, The committee then sat down and discussed who to invite. We ended up with a very long list of people, not just to maintain a connection with the 1994 conference but also to represent progress in research and practice 10 years on. Luckily our Language Centre was extremely supportive and allocated enough funding to invite 12 key speakers (six plenary and six invited). This, however, gave us quite a challenge in working out our programme, given that we only had a total of four days for plenary and parallel sessions. (Andy explains how we got round that conundrum in his ‘story’ below.) The conference proved a great success. Phil Benson (2007c: 1) calls it “the largest to be held on the subject of autonomy and language learning to date... attracting participants from all over the world”. I also possibly the only conference to generate five publications (so far), including this one. On the first day of the conference it struck me that it might be a good idea if we could publish a variety of volumes instead of one proceedings. David Little (editor of the Authentik Learner ‘Autonomy series) was one of our plenary speakers and the way in which we (with Phil Benson’s great help) had organised the abstracts into clear ‘themes’ headed by invited speakers seemed to cry out for separate Authentik Learner Autonomy volumes with invited speakers as the editors. On our way back from Hangzhou, David agreed to the idea. Learner Autonomy volumes 8, 9 and 10 (Benson 2007; Gardner 2007; Miller 2007) are currently on the shelves in the form of one set of proceedings from the conference. Another volume of Chinese papers produced by the School of International Studies at Zhejiang University, tled Selected Papers from the International Conference on Autonomy ard Language Learning: Maintaining control (Fan & Pang 2005), came out in 2003. Its taken a litte longer for us to produce this special volume, also called Maintaining Control, containing the work of our plenary and invited speakers. ing momentum 2004-07: Andy's story I joined the JALT Learner Development SIG in late 1994 and started reading in the SIG’s newsletter, Learning Learning, fascinating accounts of a conference in Hong Kong that had recently taken place. Although Vhad missed the conference itself, {soon met Richard Smith in Japan; Maintaining Contra! 9 and through a shared interest in teacher education and learner development, we became involved in bringing David Little and Leni Dam as main speakers to the international JALT Conference in Tokyo in November 1998. Their joint plenary lecture, delivered to a packed conference hall with people standing in the aisles and everyone listening, with rapt attention, was perhaps one of those moments when the waves of interest in autonomy in language education from Hong Kong in 1994 started to gather speed, at least in Japan, and scattered groups began to surf together. ‘A few years later at the AILA Conference in Singapore in 2002, T ‘met up with Richard (Pemberton) and Sarah and many others, swapping, stories and different ideas for conferences and collaborative participation by speakers. I mentioned the 1998 joint plenary in Tokyo, and we brainstormed ideas for creating a different kind of format for plenary partners for the upcoming autonomy conference in Hong Kong. From these different conversations came the idea of twinning David Little and Leni Dam, Philip Riley and Edith Esch, and Phil Benson and Naoko Aoki for the plenaries. Later, together wi Terry Lamb, Lindsay Miller and Barbara Sinclair, Mike Nix and myself ended up as fed speakers. The conference was a huge success, but what about the proceedings? Having submitted our original paper tw6 years earlier, Mike and I felt, when I rang Richard in late 2006, that a decent enough interval had now passed to ask anew about the intended publication date. “Isit...2", “Well, not quite ...” —and then [ said the fateful words: “Thave some free time coming up in the next few months ..." Within a short while, Richard, Sarah and myself had worked out a tentative schedule and draft plan of action. Now, several drafts — and ‘many draft action plans later — with free time itself fading in memory more quickly than an editorial deadline ever did, we have almost completed our editorial work. It is good to get here, and it just remains for us to say a word or two about the organisation of this volume. Reading on We have organised the chapters in this book into three main sections. ‘As you read on, you will find that the next three chapters (Chapters 2 to 4, by Benson, Esch, Riley) provide the theoretical foundation for the rest of the book, looking at current conceptualisations of autonomy from a critical and sociocultural perspective, Benson and Esch highlight10 Richard Pemberton, Sarah Teogood and Andy Bartld problems that occur in a globalised world in which ‘autonomy’ has come to be seen either as a‘ for members of a flexible workforce or as the ‘freedom’ to make ill-informed and self-constraining ocuses on problems that occur when autonomy jople and the discourses of ‘As you will see, we have grouped the rema ‘main sections: one that is concerned ig mainly with developing teacher autonomy ( r, Aoki, and Nix and Barfield). As i: sions are somewhat rough and ready — for examy ve come under the “Teacher ind Aoki could also ha serve as usefal starting poin Another way of navigi concluding chapter by Smith and Ushioda (Chapter 13) — or last. If you read it before you read the other chapters, it will help you identify the major themes of the book, and serve as a useful yomy as it has changed in the ike its predecessor Taking C momentum on your own voyages into we next 10 years and beyond, Sarak Toogood, The Hoi Hong Kong Anily Barfield, Chu Note: For further detail f the 2004 co Ihupe/ /lesastthk / contre /con®2004/ e, see the conference website: Theories and discourses of autonomy and language learning2 Making sense of autonomy learning language Phil Benson Introduction In December 1976, a group of language educators gathered at the University of Cambridge to discuss an idea that was, atthe time, largely unheard of in the field of language teaching and learning, The idea was ‘autonomy’ and the discussion that took place was preserved for posterity in a mimeographed collection of papers that has recently been made available once again on the Web (Harding-Esch 1977a). Reading, the collection for the first time, almost 30 years later, I was struck by the fact that the issues addressed by the contributors were very similar to those that we continue to discuss today. In particular, the editor noted in her preface that “there were heated arguments about the definition of autonomy on the one hand and its intrinsic value on the other” (Harding-Esch 1977b: iii). Although the context for discussion of autonomy in language learning has changed considerably, the questions, of what exactly we mean by autonomy and how we see its value to the individual and society remain with us and are likely to do so for some time to come, It also occurred to me, however, that the contributors to the Cambridge collection could hardly have anticipated the explosion of interest in the idea of autonomy in language teaching and learning, that we are now experiencing. The aim of this chapter, therefore, is to revisit the issue of what we understand by autonomy from the perspective of the difficulty that we seem to experience in agreeing upon a single definition of the term, What strategies do we have to ‘make sense of the concept of autonomy in the light of this difficulty,14 Phi Bencon and how are they related to the broader context of growing worldwide interest in autonomy in language learning? The problem of definition Since the mid-1970s, and in particular since the tur of the 21st century, there has been a remarkable growth of interest in autonomy in language teaching and learning across the globe. In Benson (2007a), for example, I noted that the number of studies on autonomy published since the turn of the century far exceeds the number published over the previous 25 years. This literature includes more than 20 book-length publications. Chapters on autonomy have also begun to appear regularly in books in overviews on language teaching methodology (Hedge 2000; Harmer 2001; Kumaravadivelu 2003) and in more specialised work, such as Nation (2001) on vocabulary, Thornbury (2005) on speaking, Dirnyei (2001a) on motivation, White (2003) on distance learning, Littlemore & Low (2006) on figurative thinking and Cameron (2001) on young, earners. Chapters on autonomy have also begun to appear with increasing frequency in edited collections on a variety of topics (Aoki 1999; Blin 1999; Healy 1999; Hoven 1999; Wachman 1999; Lamb 2000; Little 2001a; Littlemore 2001; Lynch 2001; Schalkwijk et al. 2002; Holliday. 2005; Lamb & Reinders 2005). And at the AILA 2005 Congress in Madison, Wisconsin, there were no less than 36 presentations from 18, countries listed under the heading of autonomy in language learning. Put simply, autonomy seems to have become part of the current orthodoxy of language teaching and learning research and practice: an idea that researchers and teachers ignore at their peril. In addition to taking those who have advocated autonomy for many years somewhat by surprise, the current popularity of the idea also presents us with two major problems. The first of these concerns the definition of autonomy, or perhaps more accurately the meanings that are currently being attached to it. As interest in autonomy has spread, so has the need to provide, and at times defend, an adequate definition the concept. A degree of consensus has evolved around the idea that autonomy best refers to a capacity to take charge of (Holec 1981) or control (Benson 2001) one’s own learning, but it has proved very difficult to specify exactly what this capacity consists of. In an appendix to his, compendious work on adult self-directed learning, Candy (1991) lists more than 100 competencies associated with autonomy in the Making sense of autonomy in langusge earring 15 educational literature. Little (1991: 4) argues that autonomy in learners can “lake numerous different forms, depending on their age, how far they have progressed with their learning, what they perceive their immediate learning needs to be, and so on”, and I have taken a similar position, suggesting that it may be neither necessary nor desirable to define autonomy too precisely, because it is “a multidimensional capacity that will take different forms for different individuals, and even for the same individual in different contexts or at different times” (Benson 2001: 47). In the face of statements of this kind, a newcomer to the field could be forgiven for having the impression that autonomy can be almost anything that we want it to bel ‘The second problem is that those of us who have been working with the idea of autonomy for a number of years are yet to factor the reasons behind the current worldwide interest in autonomy into our understanding of the concept. While this interest is clearly welcome, there is an underlying feeling that it is not entirely, or even primarily, a consequence of our arguments and examples. Language teachers and researchers are not, it seems, the only ones who have an interest in autonomy and it is the interest of these mysterious others that often concerns us most. There is also a concern that as the idea of autonomy ‘popularised’, it may also be ‘misunderstood’, ‘watered down’, or even put to purposes for which it was not originally intended (Little 1991; Kenny 1993; Benson 1997; Pennycook 1997; Holliday 2003; Schmenk 2005). As Pennycook (1997: 41) has put it ‘The idea of autonomy has therefore moved rapidly from a more ‘marginal and politically engaged concept to one in which questions are less and less commonly asked about the larger social or educational ‘aims of autonomy. Broader political concerns about autonomy are increasingly replaced by concerns about how to develop strategies for learner autonomy. The political has become the psychological. In the light of our reluctance to be pinned down on the definition of autonomy, however, we find ourselves in something of a dilemma. ‘While we worry that the concept of autonomy might be ‘watered down’, we also appear to be unwilling to say exactly what it means. In this chapter, I want to explore this dilemma further in three ways first, by looking at how the idea of autonomy entered and has gained ground within language teaching, and learning; second, by exploring some of the strategies we have used to make sense of this idea; and,18 Phil Benson third, by looking more closely at the wider context of current interest in autonomy in education and social life. Where did autonomy come from? The history of autonomy in language learning has been outlined elsewhere (Gremmo & Riley 1995; Benson 2001) and here I want to limit myself to two points that are particularly related to the argument of this chapter. The first point is that the idea of autonomy is not indigenous to language teaching and learning, Rather, itis an imported, essentially non-linguistic, concept that has been brought into language teaching, via psychology and educational theory, from the field of ‘moral and political philosophy. One reason for beginning this chapter with a reference to the Cambridge collection on autonomy (Harding- Esch 1977a) is that it allows us to place an approximate date on its arrival. Contributors to the collection make a few references to earlier papers on autonomy in Mélanges Pédagogiques (the journal of the CRAPEL at the University of Nancy, France, which began publication the early 1970s — see Riley 2000), but beyond this point the sources appear to dry up. Up to the late 1970s, therefore, we can assume that, if language educators were familiar with the idea of autonomy in other contexts, they did not necessarily see its relevance to language education. One reason for this is the fact that the idea of autonomy lies a focus on learners and learning, For much of the 20th century, however, language teaching theories and methodologies were largely grounded in theoretical and applied linguistics. It was only in the 19605 that theories concerned with learners and learning came into the field and, in this sense, we can see autonomy as one of a number of non-linguistie concepts that have been borrowed from psychology and social theory in order to make sense of the learner's role in the language learning process. Other examples include the idea of ‘motivation’, which first came into the field at the end of the 1950s (Spolsky 2000) and, most recently, ‘agency’ and ‘identity’ (Norton 2000; Lantolf & Pavlenko 2001). The second point is that interest in the idea of autonomy has grown largely through its association with various forms of practice, including \dividualised learning, self-instruction, self-access, computer-assisted language learning, distance learning, the use of authentic materials, Jangunge advising, leamer training and strategy training, collaborative Making sense of autonomy in language learning 17 learning, project work and the process or negotiated syllabus. ‘Autonomy has often been posited as a goal that lends meaning and direction to practices of these kinds. In tum, these practices have been viewed as being supportive of the goal of autonomy to various degrees. In this sense, the ‘rise’ of autonomy is closely connected to the deconstruction of the traditional language classroom in the 1970s and 1980s and to the emergence of innovative forms of learner-focused practice. More recently, however, autonomy has been presented as a more general goal equally applicable to more conventional classroom situations (see, for example, Dam 1995; Gardner & Miller 1996; Nunan_ 1997; Scharle & Szabé 2000; Harmer 2001; Nation 2001; Benson 2003). It is largely in the form of ‘autonomy in the classroom’ that the idea of autonomy has become part of the fabric of present-day thinking on. language education. These two aspects of the history of autonomy in language learning ‘mean that autonomy is necessarily a complex multifaceted concept. Because our ideas about autonomy have drawn on a range of sources beyond the field of education and evolved in association with a wide range of practices, they are necessarily diverse. In an earlier paper, L attempted to make sense of this diversity using the idea of ‘versions’ of, autonomy (Benson 1997). Although we may all have out own opinions about the relative merits of the different versions of autonomy available to us, we also have to acknowledge that it is not easy to set these opinions on firm theoretical grounds. The alternative, however, seems to be a relativist view that accepts the validity of almost any definition of autonomy. The remainder of this chapter is, therefore, devoted to an exploration of the possibility of going beyond the idea that autonomy can be whatever we want it to be, either through definitional and argumentative strategies, or through connections between the idea of, autonomy in language learning and conceptions of autonomy in wider educational and social contexts Strategies for making sense of autonomy ‘The first definition of autonomy in language learning to gain widespread acceptance was formulated by Holec (1981: 3), who described it as “the ability to take charge of one’s own learning”, Elaborating on this definition, Holec stated that18 Phi Benson is to have, and te the ity for all the decisions concerning all aspects of this learning, ie + determining the objectives; + defining the contents and progres + selecting methods and techniques to be used; ing the procedure of acquisition properly speaking (rhythm, ‘The autonomous learner is himself capable of making all these decisions concerning the learning with which he is or wishes to be This definition is so widely quoted that itis perhaps superfluous to repeat it here. The point that | want to highlight, however, isthat although there isa good deal of consensus on the first part of this definition, there isfarlessconsensuson Holec’s more detailed description of what “taking charge of” one's learning involves. I have suggested, for example, that autonomy involves control over three major levels of the teaching and learning process: learning management, cognitive processing and the content of learning (Benson 2001: 50). Holec’s definition clearly focuses more on learning management than it does on cognitive processes or learning content and, as we will see below, alternatives to his definition have emerged over the years. At the same time, ithas become clear that, if autonomy can be defined at all, it must be defined as a composite of abilities, attitudes or dispositions, Within this basic framework, to describe what I see as three major sense-making strategies in the literature on autonomy in language learning, which I will call the kaleidoscopic, exegetical and quintessential strategies. These are, of course, xno more than metaphors representing a certain way of making sense of a complex body of work, and although Ifind them useful, [also feel that they do not stand up to too close a scrutiny. Anyone who more deeply into the theoretical work on autonomy wi discover other strategies and other metaphors to describe them. want The kaleidoscopic strategy ‘This metaphor involves the idea of shaking up a number of objects, in this case components of a capacity for autonomy, until they fall into Making sense of autonomy in language learsing 19 some meaningfully ordered pattern, This was essentially the approach used in identifying technical, psychological and political versions of autonomy (Benson 1997), a model that was critiqued by Oxford (2003), who pointed out that it “privileged” the more political versions. While itis certainly true that I favour a more political approach, and this was explicit in the paper, my intention was actually to suggest that cach of the three versions was valid in its own way. Oxford (2003) also used the kaleidoscopic strategy to considerable effect in order to construct a in which technical, psychological, sociocultural and political-citical perspectives were placed along one axis and themes of context, agency, motivation and learning strategies were placed along the other. The essential feature of the kaleidoscopic strategy is that it accepts all definitions of autonomy as equally legitimate and attempts to amalgamate them into a kind of ‘macro-definition’. As Oxford (2003: 90) puts it, in her model, “no single perspective should be considered antithetical to any other perspective". The exegetical strategy ‘This metaphor is drawn from theology, in which an exegesis is a critical interpretation of an ancient sacred text. In the context of autonomy, this strategy involves going back to an earlier source, interpreting it and arguing that this interpretation represents the core meaning of autonomy. In spite of the fact that many variations upon his definition cof autonomy have emerged, Holec (1981) remains the most authoritative source of this kind. His definition is the most frequently cited in the literature and, in defining autonomy as “the capacity to control one’s ‘own learning” (Benson 2001: 47), I made reference to Holec’s definition as one around which consensus could be built. The significance of the exegetical strategy is, of course, that it privileges work within the tradition of the authoritative source and defines the boundaries of the field in its terms. The implication of my own definition was, for example, that whatever autonomy migitt mean, it has by and large come to be associated with the idea of control over learning within the body of work that has emerged around the idea of autonomy in language earning.20° Pht Benson The quintessential strategy ‘The third metaphor comes from medieval philosophy and its belief that the universe was composed of five elements: earth, air, fire, water and a fifth unknown element, the quintessence. Although philosophers did not know what this fifth element was, they believed that it could be extracted from matter, if only they knew how. In the context of writing on autonomy, the quintessential strategy involves an attempt to try to discover, or isolate, what is most essential to autonomy. In contrast to the kaleidoscopic strategy, the quintessential strategy privileges one perspective over others, usually the psychological or political perspective. Little (2004: 69), for example, defines autonomy as follows: Autonomy in language learning depends on the development and exercise of a capacity for detachment, critical reflection, decision ng and independent action (see Little, 1991: 4); autonomous the purpose, con loring its progress and rhythm and method of their learning, m iatng its outcomes (see Holec, 1981: 3) Little suggests here that an ability to control learning at the level of learning management does not in itself constitute autonomy. Rather, this ability depends upon the presence of certain psychological capacities, which define autonomy. Wenden (1995) makes a similar argument concerning the fandamental importance of metacognitive knowledge to autonomy. I have argued, on the other hand, that “there is good reason to believe that control over content is fundamental to autonomy” and that “if learners are self-managing methodological aspects of the learning process, but not learning what they want to learn, their learning may not be authentically self-directed” (Benson 2001: 99). In this case, 1 was suggesting that the ‘quintessence’ of autonomy lay not so much in learner psychology as it did in their capacity and freedom to control learning content (see also Kenny 1993; Pennycook 1997; Holliday 2003). ‘As a means of going beyond the idea that autonomy can mean whatever we want it to mean, however, al three of these strategies are relatively weak. The kaleidoscopic strategy explicitly accepts the legitimacy of any definition. The exegetical strategy succeeds in narrowing down the scope of autonomy; it has been successfully used, for example, to dissaciate the idea of autonomy fram the practice of Making sence of autonomy in language learning 21 individualised ‘programmed’ learning (Riley 1986). Its broad emphasis on the idea of learner control nevertheless allows for a wide variety of different defi my based on different ideas of what the most important features of ‘control’ are. The quintessential strategy is clearly the strongest of the three, but it depends on an ability to show either that the non-preferred dimensions of autonomy are trivial or that they are dependent upon the preferred dimension, So fat, this strategy has shown how definitions of autonomy based on learning management skills alone are problematic, but the priority of the psychological or political perspectives remains unresolved. jons of autor Why autonomy? Why now? Itseems, therefore, that itis difficult to establish or defend any particular definition of autonomy against any other definition through logical or reasoned argument alone. I now want to explore what a consideration of the broader contexts of our current interest in autonomy can add to the picture, In particular, I want to make reference to the fact that the last several decades have been marked both by the wotldwide expansion of education systems and institutionalised second language learning and by a worldwide tendency to harness educational practices to broader economic and social goals connected to the idea of globalisation. One consequence of this process is the expectation that ever-growing, educational sectors will produce individuals who fit the ‘needs’ of the new global economy. In the following discussion, then, T want to touch upon three aspects of the process of globalisation that may help to answer the questions: ‘Why autonomy?’ and ‘Why now? These are the global expansion of second language learning, the emergence of the self as a reflexive project and the technologisation of the self. 1 also want to explore the implications of these aspects of globalisation for the ways in which language education researchers represent the idea of autonomy. The global expansion of second language learning Since the early 1960s, there have been several developments in language teaching and learning that can be traced back to the fact that the number learning and teaching languages around the world has22. Phi Benson increased exponentially. These include: a greater biographical diversity and diversity of purposes among teachers and learners; the growth of igration for purposes of language learning or teaching; learner diversity within classrooms; and growing diversity in the settings and. modes of practice involved in language teaching and learning. All of this is widely recognised, but I would argue that we have tended to underestimate the impact of the current scale and diversity of language teaching and learning on the ways in which we think about theory and. practice. The rise of the principle of learner-centredness in the late 20th, century, for example, and the entry of non-linguistic concepts such as motivation, autonomy and identity into the field can be seen as part of an attempt to make sense of this scale and diversity. And from this, perspective, we may view current interest in autonomy as a sense- making strategy that works in either of two ways: first as a means of encouraging sensitivity to diversity and promoting the recognition that learners are individuals with unique histories and purposes for second language learning; or, second, as a practical solution to the problems posed by mass second language education, in which the responsibility for learning is pushed onto the student. The self as a reflexive project ‘The idea that the self has become a reflexive project comes from Giddens (1991) and is based on a contrast between individual-culture relationships in traditional and late modern societies. In traditional societies, Giddens argues, individual identities were largely determined by strong cultural frameworks held together by the expectation that children would for the most part follow in their parents’ footsteps. In late modern society, these frameworks are breaking down and individuals are increasingly obliged to construct their own identities. It is in this sense that the self has become a reflexive project, or something, that individuals have to work on for much of their lives. Again, the fact that the cultural frameworks in which our parents and grandparents lived are breaking down and that we are increasingly obliged to find our own way in the world is well known, but [ would argue that its implications for language teaching and learning have been underestimated. Itis, perhaps, the identities of second language learners that are most at risk in the processes of globalisation, because language learning so often involves a struggle to achieve a balance between first Making senae of autonomy in language learning 23, and second language identities. From this perspective we may, perhaps, find a deeper significance to the idea of autonomy in the context of second language learning — one that is connected to the emergence of new and relatively stable multilingual ‘selves’ out of poten disorienting processes of second language acquisition The technologisation of the self Late modernity also involves, however, what Cameron (2002: 75) calls, a “self-improvement culture” comprising “a range of practices and text- types focusing on the individual and her or his relationships with others, and particularly on the problems of modern personal life”. According to Cameron, the most accessible expressions of this culture include self- help and popular psychology books, and TV shows of the ‘confessional’ type where people talk out their experiences, problems and feelings in public. In this context, we might also refer to discourses on individual responsibility for one’s own health and safety and gendered discourses ‘on physical fitness, beauty and bodily improvement, Once again, these are familiar facts of our lives, but I would argue that their imp for language teaching and learning have not been fully taken into account, Cameron’s examples focus on the importance of ‘communication skills’ within this self-improvement ci emphasis which reflects the importance of such skills as a recognised qualification for entry into the new globalised world of work and their current importance as a focus of much second language teaching. Collectively, however, these discourses of self-improvement may add up to what we might call a ‘technologisation of the sel’, based on the problematic assumption that our bodies, minds and personalities should be worked upon as if they were components of a complex ‘machine that is in constant need of maintenance and upgrading, This technologisation of the self is, I would argue, the other side of the coin to Gidden’s (1991) self as a reflexive project. Italso points to a potentially disturbing role for the idea of autonomy in language learning — one that emphasises the learner’ responsibility to create and maintain a self that is adequate for the ever-changing demands of the new global economy.24 Phi Benson The ambiguity of autonomy Once the idea of autonomy is placed globalisation, the reasons behind our currer in this concept in language education become somewhat clearer. It also becomes clear that autonomy is a highly ambiguous concept. We have tended to represent it as an idea that serves the interests of learners, but we can also see how the idea of autonomy can serve the interests of a global economy in which individuals are increasingly being held responsible for the processes involved in their own development. It has become clear in recent years, for example, that employers increasingly expect their workers to be able to train and re-train themselves and that there is a corresponding expectation that education will produce individuals who are capable of doi i sense, is no doubt in the interests of learners, but only inasmuch as it helps them accommodate to the needs of the new economy (Schmenk 2005; Auerbach 2007; Toohey 2007). Placing the idea of autonomy in this wider context, however, also gives usa different way of looking at the problem of defining autonomy, Which is less concerned with the issue of what it means and more wit want to look fhout direct reference to earlier definitions. First, Nation (2001: 394) includes a long and learning. He defines autonomy by stating that “autonomous learners ibility for their own learning”. Here “control” sides of the coin of autonomy. mn offered by Scharle and Szab6 their teacher's handbook on learner autonomy: (2000: 4) interrelat We may conclu: autonomy, we clearly need to devi also encourage learners to take an active part in making decisions about their learning. Making sense of autonomy in language learning 25 ‘The word that concerns me here is, of course, ‘responsibility’ — a word that appears in the earlier literature, but has not been emphasised in definitions of autonomy. My concern arises from the possible implication that autonomy is a matter of instilling into learners a sense of responsibility for their own actions, which they would otherwise lack, and without which they are likely to be left behind. Indeed, Scharle ‘and Szab6 (2000: 1) make this implication explicit, when they rationalise autonomy as follows: Most language teachers have experienced the frustration of investing ‘were reluctant to use the target language in pair or group work, did not listen to each other, who did not use opportunities to learn ‘outside the classroom, and so on, Such behaviour very often stems from one common cause: the learners’ over-eliance on the teacher. Even otherwise motivated learners may assume a passive role if they feel the teacher should be in charge of everything in the classroom. In terms of the discussion earlicr in this chapter, this will perhaps be recognised as a line of argument based on the idea-that autonomy is a matter of control over learning management. But what is missing from this argument? And what support can we marshal for an argument that this may not, in fact, be an argument for autonomy. What is missing, I would argue, is a sense that many learners are already autonomous in certain respects and that itis not primarily their ‘Tack of autonomy’ but the suppression of their autonomy by educational systems that is the problem (Smith 2003). And in such a context, an ‘argument for autonomy as ‘responsibility’ seems to imply that we are icular kind of autonomy — one that will be displayed argument improvement and self-improvement. The line of argument that I would prefer to pursue would be one based more on Lantolf and Pavenko's (2001: 145) idea that researchers should treat language learners as ‘people’:26 Phil Benson Although Lantolf and Pavlenko do not use the term ‘autonomy’, their ‘use of the term ‘agency’ is, I think, very close to the sense of autonomy for which I would like to argue. This sense is based on the idea second language learners are already autonomous in important ways and that it is part of our role as teachers to support their autonomy as far as we are able by creating the conditions in which it can flourish. The argument is based, not so much on logical analysis of what the concept of autonomy actually means, but more on a strong belief that it is our responsibilty as teachers to help students learn to lead the kinds of lives that they wish to lead, rather than to fit them out with the skills and attributes that society demands of them. Conclusion To sum up, this chapter has been concerned with two problems: the problem of defining autonomy in language learning and the problem of accounting for the current interest in this concept within our field. 1 have tried to argue that these two problems are closely related. From a purely theoretical or intellectual perspective, it is difficult fo establish grounds for the legitimacy of any definition of autonomy over any other. Autonomy has many sources and contexts of application and we may be inclined, along with Oxford (2003), to regard them all as equally legitimate. But when we place our own interest in autonomy wider context of a world in which powerful economic and social forces increasingly require education systems to provide them with ‘autonomous learners’, the question of what kind of autonomy we want to aim at becomes an important ground of legitimacy in its own right. To dichotomise, and perhaps oversimplify, the issue, we might put it this way. Do we view autonomy as a matter of the production of responsible, active, flexible and adaptable worker-learners who are capable of fitting into and matching the demands of the new economic order? Or do we view autonomy as a matter of leamer agency — the production of critically aware learners who are capable of controlling their own learning and lives and of participating in the authoring of the worlds in which they live? The more interest in the idea of autonomy within the field of language learning grows, the more important our answers to these questions will become. 3 Crash or clash? Autonomy 10 years on Edith Esch Introduction Inthe first part of this chapter, I show how the notion of autonomy has spread into language pedagogy in the past 10 years and how this mainstreaming has been accompanied by conceptual distortions and discursive dissonances. Such dissonances can be located in the contradictions between the discourses of individual personal autonomy and of critical socially situated autonomy. | argue that we are at a ‘crossroads and that if we take the notion of social learning seriously, opting for the road of individual personal autonomy is not sufficient. We need to take a whole-community approach to autonomy and reassert the critical dimension originally associated with autonomy and foreign language learning, This redirection will help us engage in a new research agenda in the years to come. The notion of autonomy over the last 10 years A period of dissemination Over the past 10 years, interest in the concept of autonomy in relation to language learning and language teaching has developed in many directions and at many levels, At the conceptual level, one could argue that the publication of the Council of Europe's Framework in 2001 reflects the impact of Henri Holec’s Autonomy and Foreign Language28 Esch Learning published by the Council in 1979 (Holec 1979, 1981). As David Little rightly reminds us, Holec explicitly linked the concept of autonomy to the Council's work on adult education Janne 1977) which “emphasised the importance of equipping adult learners with the knowledge and confidence to participate in the democratic process” (Little 2004: 70). It is difficult to assess whether the educational aims of the Council have been achieved, but when it comes to language learning, the size of the impact can be observed at the level of practice. The last decade can be characterised as a period of dissemination of the idea that autonomy is more than a commodity and that it needs fostering and developing as advocated by Trim (1977). While the pedagogical principle per se can be pursued in many language learning, and teaching environments, efforts to implement it have been particularly notable in university Language Centres, not only throughout Europe and in former Commonwealth countries but also in mainland China, Hong Kong, Japan and South America — in short, wherever the need to hamess new technologies has imposed a complete reassessment of language learning and teaching practices. Innovations associated with particular individuals or institutions 20 years ago — such as the Language Learning Advisory Service of the Cambridge Language Centre (Harding-Esch & Tealby 1981; Harding, Esch 1982; Esch 1994) — have now become mainstream throughout the educational sector. Moreover, in Europe, dissemination has frequently been based on government-funded evidence-based research or action research so that, in parallel, a process of validation has taken place at the policy level Let me illustrate briefly the way in which the process of mainstreaming has interacted with trends in research and influenced the way we conceptualise the pedagogical relation. In using the term ‘pedagogical relation’, I refer to Yves Chalon’s (1970) characterisation of the relationship between teacher and learner (‘enseignant’ and ‘apprenant’ in French), where both words denote active participants, in opposition to ‘teacher’ and ‘taught’ (‘enseignant’ and né’), where the word enseigné refers to a passive recipient. Let us start with ‘the learner’. This construct from the late 1960s became popular when developments in second language acquisition made applied linguists switch focus from the process of laiguage teaching to that of language learning, The 1973 collection of edited papers Focus on the Learner Pragmatic perspectives for the language teacher by John Oller and Jack Richards is a typical example of that turning point (Oller & Richards Crash or clash? Autonomy 10 years on 29 1973). The disembodied character of the learner became the object of much research activity in the innatist and/or cognitivist tradit as the term became increasingly popular, use of the terms ‘pupils’ and ‘students’, which both reflect a socially dependent and responsible role in the pedagogical relation, declined, In the past 10 years, ‘the learner’ has undergone a process of resocialisation as sociocuitural theories have regained currency (Lantolf 1994, 2000). During this time, the abstract and idealised construct has been literally fleshed out as researchers have increasingly acknowledged the role of the social and cultural context in the learning process. The motivations (Ushioda 1996; Ridley & Ushioda 1997), decision-making, processes (Simmons & Wheeler 1995), beliefs (Benson & Lor 1998), learning styles (Duda & Riley 1990; Griffiths & Sheen 1992), discourses (Crabbe, Hoffmann & Cotterall 2001) and earning cultures (Pierson 1996) of actual learners in particular learning environments have all been analysed. During the same period, the spread of pedagogical practices involving the development of learners’ self-awareness and their ability to engage in self- and peer-assessment throughout the educational sector from primary (Dam 1995) to adult education (Esch 1994) has made apparent how slow the process of changing individual learners’ conceptions of learning is. Such innovations have at the same time made it clear how much progress is, dependent on a social-interactive view of development (Little, Ridley & Ushioda 2002; Little 2003). In effect, the process of dissemination of the idea that pursuing autonomy is a valuable educational aim has coincided with a shift in focus from abstract theoretical models of the learner to constructs embracing the importance of language learning as an artefact that is locally and collectively realised in situations subject to specific political, social and cultural constraints (Little, Dam & Timmer 1998). Similarly, if we look at the teacher pole of the pedagogical relation, the past decade in Europe has witnessed a repositioning of teachers as ‘advisers’ (Mozzon-McPherson & Vismans 2001; Pemberton et al. 2001) co ‘conseillers & Vapprentissage’ (Gremmo & Abé 1985; Gremmo 1995) in pedagogical discourse. What is notable here is the passage from experiments by influential groups such as the CRAPEL in Nancy, or individuals (Leslie Dickinson in Edinburgh, David Little in Dublin) to government-funded networks dedicated to collaborative training in best practices fora whole sector. To take one example from higher education, the SMILE (Strategies for Managing an Independent Learning Environment) project coordinated by the University of Hull (1997-2000)30° th Econ has led to the creation of online courses for the professional development of advisers leading to European qualifications. So here, the process of ‘mainstreaming has been associated with a shift from experiments conducted by a small network of similarly minded educationalists to the professionatisation of new pedagogical roles and their integration into local structures and institutions (Mozzon-McPherson & Dantec 2006; Mozzon-McPherson 2007). As to resources, challenging developments have taken place as new technologies have demanded a reassessment of the relationship between teacher, learner and tools for learning (Trim 1977). This has prompted new thinking about the role of technology in supporting autonomy (Little 1996). However, technological advances giving direct access to language learning materials have come much faster than imaginative uses of the technology for the development of learners’ independence and self-awareness. In the UK, the government funded a nationwide initiative to “make teaching and learning more productive and efficient by harnessing new technology” in the higher education sector. This programme (ook place from 1992 to 2000 and had three phases, with over 76 projects being funded overall, many by consortia. While the Teaching and Learning Technology Programme (TLTP) (1992 to date) concerned all disciplines, its impact on the teaching and learning of languages at university level was considerable. The production of computer-assisted language-learning materials was made possible and new technical and pedagogical expertise had to be brought in for the development work required to make content-support materials as well as a whole range of learning-process-support materials accessible online. However, the balance between content support and learning support varies widely depending on the pedagogical context, the funding (Reinders & Lazaro 2007) and user-friendliness. Tools for learning which require that learners organise their own learning programme such as concordancers for data-driven learning have now become increasingly easy to use — although their actual efficiency is being currently questioned (Boulton 2007). Crucially, networking, and broadband technology have made possible the distribution and exchange of multimedia materials (Esch & Cleary 1999; Esch & Zihner 2000; Zalner, Wong, & Fauverge 2000). Increasingly, the integration of e-mail, chatrooms, MOOs etc. intd language-learning programmes (Mollering 2000; Shield, Davies & Weininger 2000) has led to the formation of virtual language learning communities (Mozzon- McPherson 1996) and the widespread use of virtual learning Crash or clash? Autonomy 10 years ont environments to improve learning in schools (National College for School Leadership 2005). In recent years, however, the possibi opened to language learners for the creation of their own materials and for communication with people around the globe have also made it apparent that the integration of technologies in education is in fact, problematic. The educational world reproduces other social divisions so that many of these opportunities can be accessed only by a privileged minority (Education for Alt 2004) and, as pointed out by Hawisher and Selfe (2000: 9), “the global-village myth is far from culturally neutral” The increasingly ambiguous discourse of autonomy 1e very brief characterisation I have offered is sufficient to make the point that the process of mainstreaming of pedagogical practices to support autonomy in language learning appears to have been successful but leads us to ask whether the radical aspects of the concept of autonomy as proposed by Holec (1981: 3) have been distorted in the process. Bearing in mind the nature of autonomy as “the ability to take charge of one’s own learning’, has the aim become to sell wall-to-wall autonomy, to have autonomy validated, assessed and made available in easy self-assembly packs at university supermarkets? A number of issues associated with the process of mainstreaming need to be addressed as the ‘discourse of autonomy’ becomes increasingly ambiguous and leads to discursive dissonances. Inevitably, expansion and recognition go along with issues of integration into existing structures and changes: when diversity appears, interpretations differ. Many of the misconceptions identified by Little (1991: 3) are being recycled in the course of these mainstreaming processes. The main ones remain that ‘autonomy is synonymous with self-instruction’ and that ‘any intervention on the part of the teacher’ is, detrimental to autonomy. Indeed, at a recent conference in Spain, many young researchers seemed to assume that designing computer programs to allow students to learn on their own would lead to learner autonomy. There is also much variation in the way particular cultures and educational systems conceptualise learner autonomy as evidenced by their discursive practices (Alexander 2000: 522). Moreover, priorities vary depending on which educational sector and age range one is dealing with. While adults need to analyse the learning routines they have been using automatically for years, children have yet to discover32 Euth Esch self-regulation. Similarly, even though the pedagogical principles might be the same, integrating leaming-to-learn practices into the primary and secondary curriculum raises different implementation problems. Crucially, educational institutions are not islands and there can be considerable outside resistance towards pedagogies which can be perceived as jettisoning the notion of authority and in so doing may constitute a danger to society (Turner 1999: 92). For example, many parents are concerned by the lack of control exercised by teachers on the out-of-school context of language learning, particularly if the children are expected to work online. Finally, there are frequent pressures relating to cost-effectiveness in an educational culture — at least within the UK — that are increasingly based on accounting and national targets. Elements of distortion in the very notion of autonomy as an educational goal seem to be the common outcome of conflicting interpretations reflected in language use. Indeed, the more the word autonomy (sometimes replaced by independence) is used, the more it seems to refer to contexts where the word means the opposite of what it used to refer to, even in colloquial usage. Another such word is choice. Ten years ago, I proposed the provision of choice or genuine alternatives as ‘one of the evaluation criteria for assessing systems supporting autonomy (Esch 1996). In current usage, though, ‘choice’ often relates to the provision of a range of options that people do not want or cannot afford. The following quote from the Times Higher Education Supplement in 2003 is a clear example of the way discourse “constitutes the social” (Fairclough 1992) and illustrates the double talk in relation to the students’ so-called ‘choice’ while having to pay increased tuition fees in UK higher education, War on Mickey Mouse degrees will kill autonomy However unpopul the hands of customers rather than bureaucrats. Unive drop Mickey Mouse courses if they do not attract students, What higher education really needs isa way of mitigating the effects of top-up fees con those from poor backgrounds. Labour is indeed vulnerable to the arge of allowing such students to be priced out of the most igious universities, But destrayi ‘out of the hands of students is not the answer. (Editorial, Times Higher Education Supplement, 13 June 2003: 14, my erophosis} (Crash or clash? Autonomy 10 years on 33 Not surprisingly, distortions are reproduced in pupils’ discourse ‘The second example comes from a TV documentary on current obesity in children. In the US, apart from improving, the diet offered in the canteens in schools, some governors are trying to find ways of making the schools abandon the habit of providing vending machines in schools. In the programme, the reporter mentioned the problem of “the ‘dollar factor’ [which] prevents schools from pursuing their educational agenda’, Immediately afterwards, a schoolgirl (12 years old at most) was interviewed and declared: “There's no way they [ie. the school authorities} are going to control us. We'll still eat junk food at home "(BBC TV, Thursday 15 April 2004). This example shows that the conflict between children and authority is “no longer between what the child would like to do and what should not be done but between what the child wants to do and what he cannot do” (Marcelli 2003, quoted in Truong 2003, my translation). Here the autonomy of the child means the authority of the infantile since she does not understand that the ‘dollar factor’ is a much more pernicious constraint than the school governors’ aims. Confronted with these discursive dissonances (Riley, this volume), language teachers and researchers need to make a choice between two roads to guide their future practice and research: the road giving prominence to individual personal autonomy or the road giving prominence to aufonomy as the capacity fo exercise critical thinking about learning as a participant in a social miliew. This critical dimension highlights the socially transformative potential of learning when it arises from the constructive questioning of knowledge. The process whereby we identify conflicts in discourse between, on one hand, ‘school knowledge’ or “the knowledge which someone else presents to us” and, on the other hand, ‘action knowledge’ or “that view of the world on which our actions are based” (Barnes 1976: 81, quoted in Little, Ridley & Ushioda 2002: 11) is central to this kind of learning. It engages our responsibility as social participants in the production and interpretation of discourse and requires collective action. Such a view is far from original, but needs to be reasserted at a moment when the social control orientation of educational systems tends to marginalise more and more leamers into adopting individualistic attitudes to autonomy.