Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reframing research into 'self-direction' in adult education : A constructivist perspective 1987

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1987_A2 C36.pdf [ 21.61MB ]
JSON: 1.0055826.json
JSON-LD: 1.0055826+ld.json
RDF/XML (Pretty): 1.0055826.xml
RDF/JSON: 1.0055826+rdf.json
Turtle: 1.0055826+rdf-turtle.txt
N-Triples: 1.0055826+rdf-ntriples.txt

Full Text

R E F R A M I N G R E S E A R C H I N T O ' S E L F - D I R E C T I O N ' I N A D U L T E D U C A T I O N : A C O N S T R U C T I V 1 S T P E R S P E C T I V E by PHILIP C. C A N D Y B.Com., The University of Melbourne, 1972 B.A. , The University of Melbourne, 1977 Dip.Ed., The University of Adelaide, 1977 M.Ed. , The Victoria University of Manchester, 1979 Dip.Cont.Ed., The University of New England, 1981 A DISSERTATION SUBMITTED IN P A R T I A L F U L F I L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E O F DOCTOR OF E D U C A T I O N i n T H E F A C U L T Y O F G R A D U A T E STUDIES Administrative, Adult and Higher Education We accept this dissertation as conforming to the required standard T H E UNIVERSITY OF BRITISH C O L U M B I A September 1987 © Philip C. Candy, 1987 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Administrative, Adult and Higher Education The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date )E-6 n/811 A B S T R A C T Research into self-direction has been hampered by the absence of a consistent theoretical framework, and the indiscriminate application of the term 'self-direction' to different phenomena. The purposes of this study were: (a) to critically analyse the use of the term 'self-direction' in adult education and to ascertain whether there are differences among the phenomena subsumed under that label; (b) to critically survey the literature, and synthesise research findings; (c) to compare the significance of 'self-direction' in adult education with other sectors of education; (d) to identify and evaluate assumptions underlying past and present research traditions in 'self-direction'; and (e) to reconceptualise 'self-direction' from a constructivist perspective and to formulate themes for future research. It was shown that 'self-direction' has been used to refer to three different phenomena: (i) as a personal quality or attribute (personal autonomy); (ii) as the independent pursuit of learning outside formal instructional settings (autodidaxy); and (iii) as a way of organising instruction (learner-control). Two distinct approaches were used in undertaking the study. The first involved a critical analysis and review of literature in each of the three domains, the second was based on a form of conceptual analysis. Major paradigms in educational research were surveyed. It was asserted that assumptions underlying the interpretive paradigm were congruent with the phenomenon of self-direction and that, despite its limitations, there are advantages to adopting a constructivist perspective. Major findings were: (1) lack of internal consistency in the literature precludes the development of a coherent 'theory of self-direction' from ii within the literature; (2) autodidaxy can be usefully distinguished from learner-control; (3) autonomy in learning does not necessarily lead to personal autonomy, nor does personal autonomy always manifest itself in the learning situation; (4) autonomy has both personal and situational dimensions; (5) understanding the perspective of learners is vital to understanding strategies used and outcomes attained; (6) personal autonomy in learning comprises both cross-situational and situation-specific dimensions; (7) research into learning outcomes should stress qualitative rather than quantitative dimensions of knowledge acquisition; and (8) constructivism sanctions action-research and other naturalistic inquiry modes. The study incluuded an agenda for reaseach into autodidaxy and learner-control from a constructivist perspective. iii TABLE OF CONTENTS A B S T R A C T » Table of Contents v List of Figures ix List of Tables x Acknowledgements x I. Background to the study 1 A. Introduction 1 B. Statement of the problem 3 C. Context of the study 5 D. Nature of the study 7 E . Purpose of the study 8 F . Methodology 9 G. Significance of the study 10 H . Outline of the dissertation 10 II. The place of 'Self-direction' in adult education 13 A. Adult education's search for an identity 13 1. Definitions of adulthood 14 B. A question of terminology 17 1. Objections to the term 'self-direction' 19 2. Suggested alternative terms 21 3. 'Self-direction' as a personal attribute 22 4. 'Self-direction' in natural societal settings 22 5. 'Self-direction' in formal instructional settings 24 C. Summary 31 III. 'Self-direction' as a personal attribute 33 A. Introduction 33 B. The concept of personal autonomy 36 1. Towards a definition of autonomy 41 2. Threats to autonomy 43 3. Personal autonomy as situationally variable 47 C. The. development of autonomy 48 1. Autononry as an innate disposition 48 2. Autonomy as an acquired quality 50 3. Autononry as a learned characteristic 52 D. Summary 58 IV. Autodidactic activity - a critical analysis 62 A. Introduction 62 B. Descriptive and verification studies 65 C. The autodidactic 'method' 71 1. The autodidactic process 73 iv 2. Sources of information 82 3. Assistance with the learning project . 84 D. Theoretical, conceptual and policy studies 103 1. Autodidaxy and lifelong education 106 2. Implications for policy 110 3. Reservations concerning the concept of autodidaxy 116 E . Summary 126 V. The autodidact 130 A. Introduction 130 B. Skills and competencies of the autodidact 131 1. Towards a profile of the autodidact 132 2. Criticisms of this approach 133 3. Self-confidence versus learned helplessness 140 4. Scales purporting to assess 'self-directedness' 143 5. The development of competence as an autodidact 149 C. The learner's purposes and intentions 158 D. Autonomy with respect to knowledge 163 1. Autononry and public knowledge 164 2. Autonomy and private knowledge 174 E . Summary 176 VI. Learner control in adult education 180 A. Autonomy and the adult student 180 B. Learner-control versus teacher-direction 185 1. Differing levels of learner-control 185 2. Dimensions of learner-control 187 C. Arguments for increasing learner-control 193 1. Increased motivation 193 2. Greater meaningfulness 195 3. Enhanced learning outcomes 202 4. Individual differences among learners 211 5. The moral preferability of learner-control 219 6. The primacy of learning 228 D. Summary 229 VII. The transition from teacher-direction to learner-control 232 A. Introduction 232 B. Teachers and increased learner-control 234 1. Goodbye teacher? 234 2. Towards a new role for the adult educator 239 3. Difficulties in the transition 240 4. Teacher beliefs and the promotion of learner-control 244 5. Capitulating to pressure: Pseudo-autonomy 252 C. Learners and increased learner-control 257 1. Individual differences in the acceptance of learner-control 258 2. A preference for dependent learning 261 v 3. Learned helplessness and jarring loose the 'passive set' 262 4. Adapting to the situation: A deliberate strategy 265 D. Summary 269 VIII. Approaches to educational research 273 A. Introduction 273 B. Educational research: Three different paradigms 274 1. Positivism 277 2. Interpretive approaches 279 3. Critical approaches 282 4. Relationships among paradigms 286 C. Self-direction - a psychological or sociological issue? 290 D. Self-direction and the interpretive paradigm 291 E . Summary 295 IX. Constructivism 297 A. Introduction 297 B. A constructivist view of people 303 C. The constructivist epistemology 308 D. Constructivism, learners and learning 317 E . Summary 321 X. Constructivism and 'self-direction': A review 327 A. Introduction 327 B. The learner's sense of personal control 329 1. Self-management skills 331 2. Academic skills 333 C. Comparing constructivism with other paradigms 334 D. Summary 343 XI. Reframing research into 'self-direction' 345 A. Introduction 345 B. Autonomous learning from the learner's perspective 347 1. The learner's view of learning in general 348 2. The learner's view of this specific learning endeavour .. 353 3. The learner's view of assistance or direction received .. 359 4. The learner's view of autonomous learning and the development of autonomy 363 C. Autonomous learning from the facilitator's perspective 364 D. Summary 366 XII. Conclusions and implications 369 A. Introduction 369 B. Distinguishing autodidaxy from learner-control 370 1. Implications of the difference for theory building 374 2. Implications of the difference for the learner 377 C. Main findings of the study 379 D. Conclusion 385 v i XIII. Notes 388 Appendix A A profile of the autonomous learner 393 Appendix B A note on research methodologies 399 XIV. References 401 vii List of Figures Figure 1: A hypothetical learner-control continuum 25 Figure 2: A hypothetical learner-control continuum showing different instructional strategies 26 Figure 3: Developmental model of the mentoring relationship 89 Figure 4: Framework of the development of self-directed learning (mathetics) capacity 150 Figure 5: The development of autonomy in adult second language learning 167 Figure 6: Typology of programs classified by the degree of learner autonomy 191 Figure 7: Image used for test of perception 197 Figure 8: Learning task involving a series of numbers 199 Figure 9: Approaches to inquiry 287 Figure 10: Research into 'self-direction within three paradigms 292 Figure 11: Three-dimensional portrayal of research into 'self-direction 293 Figure 12: A conceptual model of the development and consequences of the student's sense of personal control 331 Figure 13: Changing balance of teacher-direction and learner-control in the instructional domain 371 Figure 14: A hypothetical continuum of autodidaxy and assisted autodidaxy 372 Figure 15: The relationship between autodidaxy and learner-control of instruction 372 Figure 16: Learner-control and autodidaxy as laminated domains 373 Figure 17: Schematic representation of the variables influencing autonomy in learning 383 viii List of Tables Table 1: Comparison of assumptions underlying research into 'self-direction' 337 IX A C K N O W L E D G E M E N T S In my case, and I suspect for many others, the writing of a doctoral dissertation has been a difficult, demanding and, above all, a lonely activity. In fact, it is easy to lose sight of the collaborative nature of an enterprise such as this; to overlook or diminish the critical role played by many people in getting me to this point. However, when I look back over my time here at U B C , I realise just how numerous and diverse are my debts to various people. First of all, I must thank my committee members, each of whom has contributed to my progress in his unique way. I am grateful to Roger Boshier, who first enticed me to U B C , for a multitude of small kindnesses and, in conformity with his nautical leanings, for steering me through the perilous waters of preparation for the final defence; to Gaalen Erickson for his ceaseless encouragement and for his encyclopaedic knowledge of constructivism in educational research; to Jean Hills for his patient and exhaustive comments on successive drafts of each chapter; and to Dan Pratt for his help during the many informal discussions which we had at the Faculty Club and elsewhere (as well as for the loan of a dishwasher and other furniture!). Like all students, I am enormously indebted to the authors of many books, articles and research reports, and to the faculty members and students who have wakened me to the richness of ideas to be explored. As so often occurs, especially in adult education, the greatest obligation is to other students, fellow travellers along the avenues of academe. Countless times have they pointed out references or perspectives x which might otherwise have escaped me, as well as providing the sort of camaraderie so often associated with desperate situations. Although it is invidious to single out some for special mention, I want to acknowledge, with thanks, the encouragement of Paula Brook, Shauna Butterwick, Carmel Chambers, Maurice Gibbons, Nand Kishor, Mikaella Latieff, Michael Law and Jane Munro. I especially value the support, advice and friendship of Joyce Costin, with whom I have trudged many weary miles on this particular 'Pilgrim's Progess', enjoying and enduring, by turns, the vicissitudes of graduate stud3r. Aside from these spiritual and intellectual debts, there has been much practical support as well. I have to thank the Council of the South Australian College of Advanced Education for granting (and extending) study leave to allow me to undertake this work; to my colleagues in the Centre for the Study of Adult and Further Education for taking over my teaching duties so that I could come; to the University of British Columbia for providing me with a Graduate Fellowship and with the opportunity to teach - both of which enabled me to 'make ends meet' financially. I cannot fail to acknowledge the staff of the interlibrary loan section of the Main Library, who patiently procured many obscure and often esoteric items from libraries in various parts of the world; the Centre for the Study of Teacher Education for kindly providing me with relatively luxurious office space; and also the mainly unseen people who run what must be one of the finest academic computing centres in North America. However all this pales beside the adjustments which my family have made to accommodate me. A doctoral student must be among the most difficult of people to live with: constantly pre-occupied, wracked by self-doubt, inattentive, quarrelsome and obsessive. Somehow, my family have survived me. My children xi have continued to grow, in spite of my neglect, and have certainly benefited from the experience of living in Canada. M y wife, Mary-Anne, instructed me not to mention her in the acknowledgements. But I simply must, for of all the people to whom I am indebted for their patience, support and forebearance, she is certainly the foremost, and accordingly it is to her that this dissertation is lovingly dedicated. xii I . B A C K G R O U N D T O T H E S T U D Y A . I N T R O D U C T I O N In recent years, the literature of adult education has reverberated with the call for adult educators to surrender to learners some measure of control over the teaching situation. This demand is not recent. As early as 1816, Pole wrote of the need for a specialized method for teaching adults in which they were to be treated as the "sincere friends" (p. 35) and equals of the teacher in many respects. Neither is the call for a democratisation of teaching limited to adult education, but has been a recurring preoccupation in the literature .of education at all levels from kindergarten to university. In adult education, the term which embraces this form of 'self-direction' is andragogy. In other sectors of education, a similar concern can be found, to a greater or lesser extent, in a variety of guises including, amongst others, open education, individualised instruction, discovery learning, student-centred instruction, independent study, and collaborative learning. At first sight, perhaps, there seems little to unify such diverse themes which, to use Griffin's (1977) phrase, tend to look more like a 'mish-mash' than a 'movement.' Some advocates of open education, for instance, would shudder to be mistaken for supporters of individualised instruction, which they might view as extremely narrow and, in its competency-based form at least, the complete antithesis of 'open' education. Similarly, those with an interest in collaborative learning might regard independent study as altogether too solitary and lacking in what they see as essential interpersonal contacts, either with teachers or other learners. To some extent, such mutual suspicions are well-founded, and these 1 2 various terms are by no means synonymous. However, they do seem to constitute a constellation of ideas: collectively they represent an ideology "in which many . . . initiatives have passed over to the [learners], who are now expected to be much more independent, self-directed or, in a word, autonomous" (Dearden, 1972, p. 449). The move toward increasing autonomy for adult learners is buttressed by many supporting arguments. Its proponents invoke such varied considerations as: rapid social and technological change, and the consequent need for constant new learning; the concept of democracy, with its vision of equality; changes in psychology and in the view of how individuals learn; cherished ideals such as liberty and individualism; and, especially in adult education, certain ideas about what it means to be adult in this culture. It is also based on a large and growing body of literature concerning the fact that adults can, and do, learn many things for themselves outside formal institutional structures. This phenomenon of people teaching themselves represents the second of several usages of the term 'self-directed' learning, although in this dissertation, it will be referred to as 'autodidaxy.' The interest in both learner-control and autodidaxy is part of a much larger preoccupation with 'autonomy,' both as a social ideal, and an educational phenomenon, and many people refer to this broader goal, as well as its more specific manifestation within education, as 'self-direction.' The focus of the dissertation is 'self-direction' in adult education. However, it can be seen that the term 'self-direction' is used to refer to at least three different phenomena: 'self-direction' as a generalised personal attribute (personal autonomy); 'self-direction' as the independent pursuit of learning in 3 non-institutionalised settings (autodidaxy); and 'self-direction' as a way of organising instruction (learner-control). This has accordingly necessitated an analysis of several distinct bodies of literature. It would have been possible to limit the study to any one of the three components which appear here, but as it is part of the present purpose to establish that these domains are separate, a study of each was called for. B. STATEMENT OF T H E PROBLEM In the past two decades, a thin trickle of interest in the area has swollen into a veritable torrent of books, journal articles, dissertations, research reports and conference presentations. This abundance of material presents a problem, threatening, as it does, to engulf and overwhelm the researcher. But the lack of precision and clarity has even more undesirable consequences. Although it is rarely made explicit, it is commonly assumed that there is some sort of connection between autonomy in learning, and personal autonomy in a wider sense. Some theorists, particularly in adult education, link the incidence of 'self-directed learning' outside formal instructional settings—'autodidaxy'—with personal autonomy. Others claim that the enhancement of personal autonomy is an outcome of increasing learners' control over certain features of the instructional setting. Others again seem to assume that an increase in learner-control will lead to an increase in autodidactic activity which, in turn, will result in enhanced personal autonomy. The situation is complicated by the fact that many theorists also hold the existence of personal autonomy to be a necessary prerequisite to the exercise of autonomy in learning. Accordingly, personal autonomy is viewed simultaneously as a means and as an end of 4 education. This state of confusion is a stumbling block for practitioners and theorists alike. While professional differences of opinion are only to be expected, it is hard to take seriously a concept such as 'self-direction' which is used by different writers to mean so many different things. Many practical and theoretical problems ensue from this tendency of authors to confuse ends and means, and to lump together phenomena as diverse as independent study and autodidaxy. In particular, it is not unusual to find authors who begin by writing about autodidaxy, and end up making recommendations for the conduct of instruction in adult education, or vice versa. It will be argued that this tendency to view autodidaxy as simply one end of a continuum of instructional techniques ignores its unique features and has stifled research and thinking. A second problem is that research into autodidaxy has effectively been stale-mated for several years. Despite one or two interesting findings, there has not been any major breakthrough or dramatic new line of inquiry opened up by researchers. A potentially fertile area of educational inquiry seems to have 'dried up,' and it will be argued in this dissertation that this is largely because investigators have not had a sturdy and defensible theoretical framework in which to ground research. A third problem area concerns learner-control which, like autodidaxy, has yielded confusing and contradictory research findings and where, in recent years, there have been no significant new insights into the dynamics of the phenomenon. It will be argued in this dissertation that, for the most part, adult educators have not familiarised themselves with earlier research into learner-control, nor with related phenomena in other domains of education. 5 In summary, while the area of 'self-direction' is held to be central to the field of adult education, it is plagued by terminological imprecision, and by a lack of progress in research and, in the case of learner-control, in practice as well. C. C O N T E X T O F T H E S T U D Y It is widely acknowledged that the way in which any research problem is formulated or 'framed' (Schon, 1983) will influence the actual conduct of inquiry (Cohen & Manion, 1985; Garfinkel, 1981; Koetting, 1984; Pepper, 1942; Popkewitz, 1984; Sarbin, 1977). One of the most basic distinction concerns whether an issue or question is fundamentally one of psychology or sociology, and the choice of perspective is a matter of no small concern, for the adoption of one approach necessarily directs attention to certain aspects, processes or qualities of the learning situation, while at the same time obscuring or suppressing others. Accordingly, it is important to ascertain whether the issue of individualism in learning is basically a sociological or a psychological phenomenon (Garfinkel, 1981, p. 13). At one level, it is clearly a psychological entity, and researchers have been justified in considering 'self-direction' as essentially a matter of individual preference or personal inclination. Within an environment in which individualism is widely, albeit tacitly, approved as a societal ideal (Lukes, 1973; Spence, 1985), it is not surprising to find that much research has been directed at exploring ways in which people's individuality may be recognised and enhanced in the learning situation. Alternatively, it is also clear that relatively little learning occurs in complete isolation, and that people are significantly influenced by the 6 expectations and perspectives of others - notably those who have had, or continue to have, a major impact on their attitudes, habits, values and beliefs. As Sullivan (1984) expresses it: The embedded in real historical relations. She or he comes into a world that is already a momentum and where there is a solid, weighty and dense social structure in which the person is influenced and which he or she operates on. The personal embedded in larger structured totalities that are impersonal in nature but nevertheless affect the viability of the personal world, (p. 53) Acknowledging the importance of social and historical influences on individuals in society, there are strong grounds for examining individualism in learning as a sociological phenomenon, and some authors have actually adopted this perspective. Brookfield (1984a), for instance, argues for a consideration of independent learners within their socio-cultural milieux; Hargreaves (1980) claims that the educational system is excessively concerned with individualism, and that this threatens society's "organic solidarity," Borgstrom (1985) points to the role of 'self-directed learning' in reproducing, and even exacerbating, social inequalities; and Shapiro (1984) argues that individualisation of instruction is part of the hegemony-creating and sustaining aspects of education generally. These writers are in the minority, however, by far the majority of researchers and theorists have chosen to consider self-direction from a psychological point-of-view. Like many others, the present study is concerned mainly with psychological questions. However, a more interpretive approach is advocated and it is hoped that, as a result, these questions will be more broadly based than some previous research in this domain. Even so, it is recognised that the adoption of this approach will still leave untouched important and provocative considerations of a sociological nature, and this, while constituting a distinct 7 limitation of the study, might serve to stimulate interest in some of the broader questions which are only treated superficially here. D. NATURE OF THE STUDY All fields of study develop and progress through the cumulative efforts of many researchers and theorists (T. S. Kuhn, 1970; Lakatos, 1970; Laudan, 1977). Individual contributions sometimes confirm and consolidate existing knowledge in a field, sometimes they make a novel contribution which results in the pursuit of new directions of inquiry. From time to time, an attempt is made to stop and take stock of the existing state of knowledge, and current directions in research. Such stock-taking is called meta-research, which is defined as "systematic study of the processes and products of inquiry which characterize a discipline or field of study" (Sork, 1982, p. 1). Sometimes, research in a particular domain gets 'bogged down' or 'stale-mated', and little progress is made until some new perspective—often a revised epistemological formulation or a new research methodology—is proposed and accepted. Very often, the new approach is imported or borrowed from some other field of inquiry, but it may have the effect of restarting research, which subsequently makes quite rapid progress in new directions. In adult education, there is, as Sork (1982) and others point out, an emergent tradition of meta-research. There are also instances of research traditions (for example that concerning motivational orientations of adult learners) which undergo something resembling a paradigm shift, and proceed with renewed vigour. The present study concerns the field of 'self-direction' in adult education which, it is argued, seems to have made relatively little progress in recent 8 years. The study is in the nature of meta-research, in that it draws on the work of other theorists and writers. It (iii) to compare the significance of 'self-direction' in adult education with its place in other sectors of education; follows several previous studies which have attempted to review and summarise parts of the same literature base (Moore, 1973; Coolican, 1974; Geis, 1976; Tough, 1978; Skager, 1979; Jankovic et al., 1979; Cross, 1981; Brookfield, 1982; Mocker & Spear, 1982; Caffarella & O'Donnell, 1985, 1986), but it differs from these in two major respects. First, .an attempt has been made to review and synthesise major themes in three distinct areas of research (personal autonomy, autodidaxy and learner control), and to search for underlying similarities and differences between the domains. The second major difference is that it considers the possibility of reframing research from a particular epistemological position. Thus, it contains not only what Sork (1982) has called a "critical or interpretive review of research on specific topics," but also a "taxonomy of needed research" from a particular perspective. E . P U R P O S E O F T H E S T U D Y The study was conceptual in nature, and had the following purposes: (i) to critically analyse the use of the term 'self-direction' in adult education and to ascertain whether there are differences among the various phenomena presently subsumed under that label; (ii) to critically review the literature, and to synthesise research findings on 'self-direction'; 9 (iii) to compare the significance of 'self-direction' in adult education with its place in other sectors of education; (iv) to identify and evaluate major assumptions underlying past and present research traditions in 'self-direction'; and (v) to reconceptualise 'self-direction' from a constructivist perspective, with a view to formulating themes for future research. F. METHODOLOGY Two distinct approaches were used in undertaking this study. The First comprised a critical analysis and review of the literature in three domains - personal autonomy, autodidaxy and learner-control. In the case of autodidaxy, the bulk of the literature is in adult education, but in the case of personal autonomy and learner-control, material from elementary, secondary and higher education is also included. As a result of this survey of the literature, a number of dilemmas, paradoxes or impasses in research were identified. These became the centrepiece for a second stage, based on a form of conceptual analysis. In this second stage, a particular world view or metaphysical commitment—constructivism—was examined. It is argued that many of the difficulties presently manifest in the literature could be resolved, or would not have arisen, if a constructivist perspective were adopted. 10 G. SIGNIFICANCE OF THE STUDY As will be discussed in the body of the dissertation, adult education has long sought a star to which to hitch its wagon. As a field of study and practice, it has looked for a unique theoretical framework to distinguish it from other sectors of the educational domain. In choosing 'self-direction,' however, it seems to have chosen a 'falling star'; a field characterised by confusion, conflicting claims and, in the case of learner-control, a disappointing lack of success in enhancing learning outcomes. This dissertation will clarify some of the imprecise thinking about the subject, and provide researchers and practitioners with a way of framing their thinking and practice that, it is hoped, will generate new insights and hypotheses for study. H. OUTLINE OF THE DISSERTATION Chapter one comprises an introduction, and background to the study. Chapter two begins with an overview of 'self-direction,' and an attempt to explicate why it has become such a central theme in the discourse and recent practice of adult education. It is shown that the term 'self-direction' is used in the literature to refer to at least three different phenomena, various objections to this situation are noted, and it is argued that, to avoid confusion, the term 'self-direction' should be abandoned and replaced in each of the three usages. Chapter three analyses literature pertaining to personal autonomy, derives a working definition of personal autonomy, and examines the extent to which its attainment is influenced by educational (particularly adult educational) interventions. 11 Chapter four presents a critical analysis and review of the literature on 'self-directed learning' in what Jensen (1960) referred to as 'natural societal settings,' or autodidaxy. Chapter five presents an analysis of research on the skills and competencies of the autodidact, and on the development of such competence in self-teaching. Chapter six contains an overview of learner-control, its various degrees and dimensions, derived from a study of the literature in elementary, secondary and higher, as well as adult, education. A number of the arguments commonly raised in favour of increasing learner-control within adult education are critically analysed. Chapter seven deals with the transition, for both teachers and learners, Chapter nine acts as a bridge between the literature surveys which constitute the first part of the study, and the subsequent reconceptualisation of self-direction from a constructivist perspective. It introduces the notion of the learner's sense of personal control, and then compares and contrasts the assumptions implicit in much previous research into self-direction with those underlying constructivism, from situations of teacher-direction to those of learner-control. It is demonstrated that frequently programs which ostensibly lead to increased It demonstrates how, in the context of adult education, a constructivist perspective might lead to productive new directions in research, theory-building and practice in 'self-direction' (autodidaxy and learner-control). learner-control do so in fairly minor or inconsequential ways, and that there are significant conceptual as well as practical difficulties in trying to promote autonomy within formal instructional settings. Chapter eight acts as a bridge between the first and second parts of the 12 dissertation. It reviews the assumptions underlying varying paradigms in educational research, and demonstrates the inadequacy of the positivistic approach. It argues that the interpretive paradigm (in the form of constructivism) is more congruent with, and appropriate to the study of, the phenomenon of 'self-direction.' Chapter nine contains an explication of constructivism as a way of viewing educational phenomena. It considers the constructivist view of human nature, the constructivist understanding of knowledge, and the constructivist view of learning. Chapter ten introduces the notion of the learner's sense of personal control, and then compares and contrasts the assumptions implicit in much previous research into self-direction with those underlying constructivism. Chapter eleven attempts to reframe research into 'self-direction' using a constructivist perspective. It demonstrates how, in the context of adult education, a constructivist perspective might lead to productive new directions in research, theorj'-building and practice in 'self-direction' (autodidaxy and learner-control). It is shown that constructivism allows for a new way of looking at enduring problems. However, because it is a different paradigm from that which underpins most research in this field, it actually calls for a whole new approach to research in this domain. The twelfth and final chapter is a synthesis of the preceding ones. It explains the distinction between learner-control and autodidaxy from a constructivist perspective, and summarises the main findings of the study. II. THE PLACE OF 'SELF-DIRECTION' IN ADULT EDUCATION A. ADULT EDUCATION'S SEARCH FOR AN IDENTITY Although the primary focus of this dissertation is 'self-direction in learning,' it seems appropriate to begin by exploring why self-direction should be a valued feature of adult education at all. Throughout its history, but more particularly in the past decade or so, attempts have been made by scholars and theorists to identify, analyse, define, redefine, map or otherwise delineate the essential characteristics or boundaries of the field of adult education (e.g., Lindeman, 1926; Bryson, 1936; Jensen, Liveright & Hallenbeck, 1964; Schroeder, 1970; Champion, 1975; Campbell, 1977; Little, 1979; Boyd, Apps & Associates, 1980; Rubenson, 1982; Tight, 1983; Sinnett, 1985). A number of these attempts are the result of "professors of adult education, nervously trying to stake out a territory separable from other territories, both within educational studies in particular, and the social sciences and humanities in general" (Welton, 1986, p. 8). Several features are commonly mentioned which ostensibly differentiate adult education from other sectors of education: it has an extremely diffuse and nebulous mandate; it is distinguished by an ethos of voluntarism among both teachers and learners (including what Ranger (1985) describes as its 'nocturnal ritual'); it claims to place a higher emphasis on meeting the needs of learners than other sectors of education do; and much of its activity is characterised by what Bernstein has called weak classification and weak framing (Bernard & Papagiannis, 1983; Stalker-Costin, 1986). As Keddie (1980) points out, many of these claims are ideological, rather than empirical, deriving from what Welton 13 14 (1986) has characterised as the attempt by adult education to develop from within itself its epistemological foundations. According to Welton (1986), attempts to provide a theoretical framework for adult education have been based on a "shaky and porous foundation," namely an 'adult characteristics episteme.' He goes on to examine critically the three 'modalities' within this episteme: firstly, the claim that adult education is distinctive because there are forms of knowledge which are distinctively adult (the 'adult knowledge' modality); secondly, because adult educators seek to meet the needs of their clients through flexible and responsive provision and open access (the 'needs, access and provision' modality); and thirdly because of something special and unique about teaching methods 1 employed with adults (the 'methodological' modality). This last has, in turn, been dominated by two concepts and their attendant bodies of literature: andragogy and self-directed learning. Whilst recognising the close relationship, and strong interconnections, between these two notions, it is the latter which claims the attention of this present work. 1. Definitions of adulthood Central to each of these three modalities, and indeed to the 'adult characteristics episteme' itself, is the notion of adulthood, and accordingly, over the years, the construct of adulthood has received a good deal of attention. In 1964, in a paper on "The Definition of Terms," Verner wrote: . . . the precise meaning of the term adult is actually quite vague - particularly when it is used to identify the clientele of adult education. The notions of who is an adult vary from "those past school age" through "grownups" to "mature individuals" - perceptions so indefinite as to be all but meaningless. Attempts to arrive at a 15 precise identification of an adult tend to fall into the categories of age, psychological maturity and social role. (p. 28) Subsequent research has tended to emphasise one or other of these three categories; age, psychological maturity or social role. However, notwithstanding nearly two decades of further research and enquiry (Bova & Phillips, 1985), there are still few, if any, satisfactory and comprehensive conceptualisations of adulthood. Perhaps this is because adulthood is a residual concept, what is left after defining other stages in the human life cycle (Jordan, 1978). Perhaps it is simply because adulthood is such a broad, amorphous and diffuse phenomenon. Despite Paterson's (1979) assertion that; "Adults are adults, in the last analysis, because they are older than children" (p. 10), age has proven to be an unsatisfactory criterion for determining the threshold of adulthood. Studies have variously cited 16, 18 or 21 as the 'magic age,' based on laws which permit one to vote, drink, drive or be drafted into the armed services (Bova & Phillips, 1985, p. 38). Yet, it is not difficult to think of instances in which age alone is a poor indicator of adult status: the eldest child who, orphaned at age 15 becomes responsible for her or his younger brothers and sisters or, at the other extreme, the 25 year old student who, still living at home, is protected from life's vicissitudes by his or her doting parents. According to H . M . Kallen (1962), "adulthood, even if determinate biologically, is culturally a variable . . . Images of it are collective ideals which the societies committed to those ideals strive to have their young embody. The common name for those strivings is education" (p. 38). In our society, it is true that our ability to recognise ourselves and others as adult is based, at least in part, on developing independence, along with the adoption of responsibilities (such 16 as worker, spouse, parent, citizen etc.). However, definitions of adulthood based on social roles have a disconcerting tendency towards circularity: "The adult . . . can be distinguished from a child or adolescent by his or her acceptance of the social roles and functions that define adulthood." (Darkenwald & Merriam, 1982, p. 77) a. The place of autonomy in defining adulthood The third class of definitions, namely those concerned with psychological maturity, are potentially the most promising for the present purpose. Some of these definitions portray adulthood as the development or acquisition of an interrelated set of psychological characteristics, usually including independence or autonomy or freedom from the influence of others. Other definitions, such as those of Maslow or Rogers, "stress the idea that adulthood is a process rather than a condition, a process in which men and women continually strive toward self-actualization and self-fulfilment." (Darkenwald & Merriam, 1982, p. 40) Whether viewed as a process or a condition, however, the common element is the achievement of autonomy (Birren & Hedlund, 1984). This fact has profound significance for adult education. Darkenwald and Merriam (1982) claim that the mission of adult education "is not preparatory, so much as it is one of assistance - helping adults to realize their potential, make good decisions and in general, better carry out the duties and responsibilities inherent in the adult role." (p. 77) Thus, it would appear that one of the primary tasks of adult education is to develop and to permit the exercise of individuality (Hostler, 1981, p. 37) and autonomy; . . . while the fostering of mental autonomy is an important objective in the education of children, it is of special importance in the 17 education of adults. In deeming someone to be an 'adult,' we are ascribing to him various rights and responsibilities in virtue of certain distinctive moral and personal qualities which we presume him to have . . . the qualities of impartiality, objectivity and balance, at least in some minimum degree, and the ability to draw on his experience with some measure of sense and skill . . . The project of fostering mental autonomy is the project of helping adults to be adult . . . (Paterson, 1979, pp. 120-1) The argument thus far might be summarised as follows: adult educators have sought a unique and distinctive foundation for their work. Of all the criteria which are alleged to provide such a distinctive foundation, the most compelling and unique, is probably the nature of the client group itself (i.e., adults), and central to this construct of adulthood is the notion of autonomy. This then raises the question: What is meant by the term autonom3' in this context? B . A Q U E S T I O N O F T E R M I N O L O G Y In adult education, the term most commonly used as a synonym for autonomy is 'self-direction.' However, it is not necessary to venture far into the literature to discover that 'self-directed' and 'self-direction' have a number of meanings. For many authors, self-direction is seen simply as a method of organising instruction. Thus, in 1967, MacNeil undertook " A comparative study of two instructional methods . . . Lecture-discussion and self-directed study". Three years later, in 1970, Himmel presented a dissertation entitled; "A critical review and analysis of self-directed learning methods utilized in the teaching of undergraduate psychology courses". Redditt's (1973) doctoral dissertation comprised; "A quasi-experimental comparison of a group lecture method and a self-directed 18 method in teaching basic electricity at the college level," and in 1978 Harrison wrote an article which counselled on "How to design and conduct self-directed learning experiences"! For others, self-direction is not so much a method of teaching as a characteristic of learners. Cheren (1983) for instance writes of; "Helping learners achieve greater self-direction," Kasworm (1983b) presents a model of increasing self-directedness in her article on "Self-directed learning and lifespan development" and, since the appearance of Guglielmino's Self-directed Learning Readiness Scale (1977), there has been a succession of studies based on the notion that self-direction is a measurable attribute, distributed throughout the adult population (Bayha, 1983; Box, 1982; Brockett, 1983b; Curry, 1983; Mourad, 1979; Sabbaghian, 1979; Savoie, 1979; Skaggs, 1981; Torrance & Mourad, 1978). Moreover, lurking beneath these different interpretations is an even more basic distinction. Although 'self-directed learning' inside formal instructional settings has captured the imagination, and consumed the energies, of many adult educators, the popularity of the term is due to research findings about the extent of 'self-directed learning' outside formal instructional settings. In the past twenty years, dating from the original work of Tough (1966), there has been a growing awareness that most, and perhaps all, adults engage in self-initiated, self-planned and self-executed learning projects largely independent of any institutional affiliation or formal support. It is the rapidly burgeoning body of literature about this phenomenon which has thrust the term 'self-directed' into such prominence, and in many respects transformed 'self-direction' into a rallying point for adult educators. 19 1. Objections to the term 'self-direction' These are all important concepts. However, they are by no means synonymous, and it is confusing when one term—self-direction—is applied to describe such varied phenomena. Accordingly, and although it might be considered somewhat quixotic to attempt to dislodge a term which has become so firmly embedded in the discourse of adult education, there are many sound reasons for advocating the abandonment of the terms 'self-direction' and 'self-directed' in favour of something else. The first objection is that 'self-directed learning' has been contaminated through overuse. In particular, indiscriminate use has blurred the distinction between the sort of 'self-directed learning' which is possible in formally constituted adult education programs, and that which takes place in situations not formally designated as 'educational' or 'natural societal settings' (Jensen, 1960). It will be argued in this dissertation that there are material differences between these activities, differences which should be reflected in the adult education lexicon. Secondly, there is confusion as to whether it is a process or a product: 'self-directed learning' can be an activity in which people engage, or the outcome of such an activity. This derives from the fact that the word learning is, as Brookfield (1984a) points out, "a gerund; that is a word which functions colloquially as both a noun and a verb" (p. 61). A further disadvantage of the term 'self-directed learning' is that in may respects the notion of self-direction is redundant. Since it is impossible for anyone to learn on behalf of another, one could argue that all learning is in effect self-directed. Sometimes, self-directed learning is contrasted with other-directed 20 learning, but this usually refers to control of the external conditions, rather than control of the act of learning itself. This leads to the final point. The final charge against 'self-directed learning' is that—even disregarding the above criticisms—it is too narrow. It does not adequately represent the phenomenon of managing one's own education, which is usually implied. In 1983, at the Annual Conference of the American Education Research Association, Boshier argued that the term 'education' "should be reserved to describe the process of managing external conditions which would facilitate . . . internal change [i.e., learning]. Hence an adult . . . who assigned his or her own learning goals, who located appropriate resources, and who evaluated the progress made in attaining those goals would be engaged in self-education, rather than self-directed learning" (Brookfield, 1984a, p. 61). However, even the term 'self-education' is unacceptable. As Hamm (1982) points out, it may imply anj' one of several things: education of the self, education about the self, or education by the self. Of these three, it is only the third which is of interest for the present purpose, and many authors, from Plato onward, have written at length of the inconceivability of anyone actually educating himself or herself in the fullest sense of the term. As Hamm states: Proponents of self-education make much of the notion of self-teaching. But is this logically possible? One cannot teach if one is not able in some way to display the subject matter to be learnt. If one does not know that subject matter, then it is not only pointless to teach oneself, but it becomes impossible to do so, because one cannot learn what one already knows. If one does not know the subject matter, it is still possible to learn it, say from experience or trial and error, but that is not teaching unless one mistakenly equates teaching with learning, (p. 95) All in all, despite its widespread use in the adult education literature, and even its adoption by authors who had earlier advocated alternative terms (e.g. Brookfield, 1980a, 1980b, 1981a, 1982a, 1982b, 1984a, 1984b, 1985c, 1985d; Penland, 1977, 1979, 1981), the term 'self-directed learning' has been found to be too vague for the present purpose. Accordingly, it was decided to make use of alternatives to cover the various phenomena presently subsumed by the one term 'self-directed learning.' 2. Suggested alternative terms One potential solution would have been to invent some new words— a process which Glaser and Strauss (1967) refer to derisively as 'Intellectual capitalism' (p. 11). The alternative was to scan the literature for a term, or terms, considered to be more precise, or at least with discrete ranges of meaning. This procedure yielded no less than thirty words and phrases2 which are either partially or wholly interchangeable with 'self-directed learning' in one or other of its meanings. Unfortunately, many of these alternative terms also suffer from some of the logical or linguistic deficiencies expressed above, or else they, too, have been used rather indiscriminately. Nonetheless, it is hoped that their use in this present context will avert some of the ambiguity which attends their appearance in the wider literature. Basically, there are three uses for which alternative terms were selected. These are: 'self-direction' as a personal attribute (which embraces self-directedness in learning); 'self-direction' as the individual, non-institutional pursuit of learning opportunities in the 'natural societal setting'; and 'self-direction' as a mode of organizing instruction in formal settings. 22 3. 'Self-direction' as a personal attribute When adult education authors apply the term 'self-direction' or 'self-directed' to people, they usually mean one of two things. Either they refer to a general disposition towards taking control of, and giving direction to, one's life or else they mean capable of undertaking learning without outside direction. These are assumed to be linked. It has been decided to refer to 'self-direction' in the former sense as 'personal autonomy.' This is dealt with in chapter three. In the second case, self direction or self-directedness in learning also has two meanings. It may refer either to the propensity to accept and exercise control over valued instructional functions within an instructional setting, or else the ability and willingness to learn things for oneself, without institutional support or affiliation. The term 'independent study' or 'learner control, will be used to apply to the first situation, 'autodidaxy' or 'autodidactic learning' to the second. There is a relationship between these various characteristics (personal autonomy, autodidactic learning and independent study) but, as will be discussed later, they are not necessarily synonymous, nor does the existence of one necessarily imply the existence of the others. 4. 'Self-direction' in natural societal settings To identify the situation where the learner voluntarily initiates, plans, conducts and evaluates his or her own learning project or enquiry without any institutional framework (Jankovic et al., 1979, p. 5), the term 'autodidaxy' will be used. This is equivalent to Brookfield's (1981, 1982) formulation of 'independent learning,' Gibbons and Phillips' (1978, 1979, 1982) notion of 'self-education,' Strong's (1977) concept of 'autonomous learning,' and the 23 individual, self-planned component of Tough's (1967, 1978, 1979a) 'learning projects' or 'major learning efforts.' The term autodidaxy is favoured for several reasons. For a start, it describes a unique educative situation, which should not be confused, in the reader's mind, with any other. Secondly, it embodies the dual notions of teaching and learning, and moreover, the prefix 'auto' means more than just 'self,' but implies both: 1. "independence, freedom of action, and 2. a process carried out with one's own resources, without outside authority" (Jankovic et al., 1979, p. 15). Thirdly, after an extensive review of current usage and literature in both English and French, it was endorsed by the 1979 'European expert meeting on the forms of autodidactic learning' (Jankovic et al., 1979). As Tough and others have pointed out, there will be occasions when an autodidact will turn to someone else for help. It may be a relatively minor interchange, or it may turn into a more extensive and protracted relationship. However, so long as the initiative rests with the learner, who still has undisputed control over the twin dimensions of objectives and evaluation, this phenomenon is still treated as autodidactic learning, albeit in a modified form, and accordingly it will be referred to as 'guided' or 'assisted' autodidaxy. In practice, it may prove extremely difficult to distinguish assisted autodidaxy from independent study. In view of the propensity of adult education agencies to conduct their activities outside 'formal' settings, there may be virtually no detectable difference in the external appearance as judged by an outside observer, and the only wa}- to ascertain which it is, would be by referring to the 24 intentions and understandings (and perhaps misunderstandings) of the people involved. This issue is dealt with in greater depth later, but it raises the fact that autonomy may best be viewed as a variable which has both personal and situational characteristics and that, as Dearden and others have commented, one of the defining characteristics is whether or not, in any particular situation, the actor considers himself or herself to be acting autonomously. 5. 'Self-direction' in formal instructional settings Many adult education activities occur in contexts recognisably similar to other sectors of education. The physical setting may be different (sometimes more luxurious, but frequently more frugal and ascetic); the pupils may be referred to as 'learners' or 'participants'; and the teacher may have some other title such as trainer, helper, tutor, guide, facilitator, coach or instructor. Nevertheless, there is a recognized, pre-existing relationship, based on a complex network of mutual expectations between the teacher and the taught, and a number of (mostly tacit) rules governing their conduct towards each other. Among these is the expectation that the teacher will accept responsibility for certain functions in the instructional situation, and that the learner will be responsible for other, different functions. Although some authors write as if the teacher's responsibilities and those of the learner are simple, mutually exclusive domains which can be distinguished from one another on the basis of objective criteria, others acknowledge that control over the teaching/learning situation is more like a continuum than a dichotomy. It is perhaps useful to think of these actors as occupying positions on a continuum extending from teacher-control at one extreme to learner-control at the other, and the deliberate surrendering of certain prerogatives by the teacher is accompanied by the concomitant acceptance of responsibility by the learner or learners. In the sense that there can be a dynamically changing equilibrium in this arrangement, it is reminiscent of Mark Hopkins' famous image of the teacher on one end of a log, with the learner on the other end. Diagrammatically, the situation may be portrayed as a Tannenbaum and Schmidt type continuum, where each diminution in the teacher's control may be compensated for by a corresponding increase in the learner's, so that it resembles a sliding scale from -complete teacher-direction at one extreme to total learner-control at the other. Both Gibbons and Phillips (19S2, p. 76) and Millar et al. (1986, p. 43 71 include such diagrams to express the gradual shift in control from one party to the other. To describe this continuum, the term 'learner-controlled instruction' will be used. Stance of Tells Sells Tests Consults Joins the teacher: Exercise of authority by the teacher Exercise of responsibility by the learners Stance of Submits the learner: Buys Approves Chooses Shares Figure 1: A Hypothetical learner-control continuum (adapted from Millar et al., 1986, p. 437) Although the term 'learner-control' was reasonably commonplace in the 26 1960s (e.g., Campbell & Chapman, 1967; Mager & McCann, 1961), it suffered a decline in usage, presumably because of its unfashionable 'behaviouristic' connotations. However, it was rehabilitated by Snow in 1980 and, for the present purposes, has the advantage that it is logically possible to speak of learner-control as "both a dimension along which instructional treatments differ, and a dimension characteristic of individual differences among learners." Thus, as Snow points out, "it is perhaps the first instance of an aptitude and treatment variable being potentially definable in common terms" (1980, pp. 157-158). If learner-control is conceived of as a range or continuum (or more likely a series of continua, for it is possible to exert differing degrees of control over various dimensions), then one end of the range will involve a great degree of learner control over valued instructional functions. Various instructional strategies could be placed at intervals along this continuum, to imply the differing balance of teacher-direction and learner-control: Teacher direction Learner control a b o d e f g h i Figure 2: A hypothetical learner-control continuum showing different instructional strategies At the far left might come indoctrination (a), with almost total teacher-direction and little room for learner-control at all. Then might come, in sequence, lectures (b), lessons (c), programmed instruction (d), individualised instruction (e), personalised instruction (f), computer managed learning (g), discovery learning (h) 27 etc., until finally the point is reached where learners have accepted almost all control over valued instructional functions. This point, at the far right-hand edge of the continuum (i), has been called here 'independent study.' In discussing the relationship between individualised instruction, and independent study, Percy and Ramsden (1980) write: Individualization of learning tasks, observe Dressel and Thompson (1973), can be the 'first step toward independence.' . . [it] may foster the motivation for independent work and, if properly conducted, will merge into independent study as responsibilitj' for direction is transferred from teacher to student. But independent and individualized study are not equivalent, (p. 6) As will be discussed later, independent study is the form of learner-control which most closely resembles autodidactic learning, but this resemblance is misleading, and the concepts are not the same. Just as individualised instruction is not synonymous with independent study, it will be argued here that independent study is not synonymous with autodidaxy. a. Independent stud3? as a form of learner-control Like 'self-directed learning,' 'independent study' is a catch-all for all manner of educational practices having some bearing on the notion of learner-control. In reporting attempts to survey usages of the term, and to distill out of them common themes, Moore (1973b) wrote: in our first explorations of the literature, among references to 'independent study.' . . we found the following: "Why S U N Y students fail to complete Independent Study courses" (the term 'Independent Study' here referring to correspondence courses); "a system of Instruments for the management of independent study" (here it meant individualized, programmed instruction in a school setting); "Independent Study in secondary schools" and "Final report on an Independent Study program for the academically able" (which described supervised reading programs in schools); and "A rationale and a role for 28 Independent Study" (which focuses on out-of-school, part-time degree programs for adults) . . . (p. 663) In addition to all these other applications, the term 'independent study' has been used since the 1920s in American higher education to refer to; "teaching and learning which focuses on the individual instead of the group, which emphasizes the person-to-person relationship between teacher and student" and "the pursuit of special topics by individual students under the guidance of faculty advisers apart from organized courses" (Bonthius et al., 1957, pp. 3-8) For a long time, such practices were restricted to 'superior' or 'honors' students (Felder, 1963, 1964; Stein, 1954; Umstaddt, 1935), yet by 1960, Hatch and Bennet were able to state that: Of late, there has been much experimentation with independent study quite outside of 'honors' programs. In addition, there are other programs or practices that advance the purpose of independent study, but are not always identified with it. Included are Socratic, problem and case methods of instruction, student research and administrative and curricular practices that introduce greater flexibility into academic programs and so provide an opportunity for independent study, (p. 1) Thirteen years later, in 1973, Dressel and Thomson published another major survey in which they defined 'independent study' as: "the student's self-directed pursuit of academic competence in as autonomous a manner as he is able to exercise at any particular time" (p. 1). Aside from importing yet another term into the field ('autonomous'), this definition also introduced the notion that independent study might be a situationally variable construct, depending on the individual student's capability to act 'independently' in a particular situation. The part of the term which has caused much of the confusion is 'independent,' and it was this which prompted Perc}' and Ramsden (1980) to observe: 29 Talk of student 'independence' needs to begin from the question: of what is the student to be independent? In theory, at least, he might be independent of teachers, of fellow students, of prescribed course content or methods of learning, of specialisms and publicly acknowledged categorizations of knowledge, of limitations on sequence or pace of learning, of assessment, even of academic conventions in the use of evidence and sources. When a student simply works on his own on individually set tasks, when he has some control over the pace or mode of learning, or some choice of options, it may be more realistic to talk of 'individualized' study, (pp. 5-6) This quote clearly acknowledges the notion of varying levels of independence, a point also recognised by Passmore et al. (1963) in their survey of teaching in Australian universities: We need . . . to begin by distinguishing a number of different levels of independence. A t one extreme, the student is thought of as, in general, doing no work whatever beyond attending lectures, taking part in practical classes, reading his textbook, preparing for examinations. At this level, he is regarded as doing 'independent work' if he so much as opens his mouth in a discussion class, works at a set assignment, or reads any book or periodical except a set text. Thus in some departments the introduction of a 'tutorial' of any sort—even a 'tutorial' which is basically a class for the working out of exercises—the setting of any sort of assignment, the recommendation of any reading whatsoever, is thought of as the encouragement of independent work. At the opposite end of the scale, 'independent' work is defined as consisting in the tackling by the student of problems he has thought up by himself, by methods he chooses to employ, with the teacher acting only as a supervisor. So when departments describe themselves as making provision for independent work or say that they cannot possibly do so, they may be thinking of independent work at different levels, (p. 216) For many, the term 'independent study' has special connotations relating to a physical separation of teachers and learners (e.g., Wedemeyer, 1975, p. 57). In the present context, no such significance is attached to it. Admittedly, it is hard to imagine an instance of independent study which would be carried out in a contiguous or face-to-face mode, but the separation of the learner from the 30 teacher is not necessary. Instead, it is more a function of independence in the form of "freedom in the self-determination of goals and activities" and of "learning programs which are carried on to the greatest extent possible at the convenience of the learner" (Wedemeyer, 1971, p. 550). Overall, the definition which has been found most appropriate for the present dissertation is that proposed by Forster (1972) in her philosophical and historical analysis of independent study: Independent study is a process, a method and a philosophy of education: (1) in which a student acquires knowledge by his own efforts and develops the ability for inquiry and critical evaluation; (2) it includes freedom of choice in determining those objectives, within the limits of a given project or program and with the aid of a faculty adviser; (3) it requires freedom of process to carry out the objectives; (4) it places increased educational responsibility on the student for the achieving of objectives and for the value of the goals. The primary goals of independent study are to contribute: (1) to the development of an intellectually free and responsible person by placing greater responsibility on the student for his own education; (2) to the development of an independent person by creating instances in the educational process which require students to choose objectives and carry out decisions; (3) to the development of a person capable of continuing his education . . . by developing the skills of inquiry and critical thinking and increasing the student's active participation in his own education, (p. ii) In this dissertation, the convention to be observed is that when the teacher or trainer determines objectives, sets tasks, or gives only some limited control over dimensions such as pace, mode or sequence of learning, such a situation would be classed as 'individualised instruction.' But the more the learner exercises independence from situational constraints or from 'publicly acknowledged categorizations of knowledge' the more they engage in independent study. For the present purposes, independent study does not necessarily imply solitary activity, it might well embrace 'self-directed groups' (Beach, 1965) as well as individual 31 learners. 'Independent study' is characterised by a high degree of learner-control over many instructional elements, including the setting of objectives, choices about pacing, content and methodology, and assessment of learning outcomes. As a result, it is frequently confused with the learning undertaken outside institutional structures, or what is referred to here as autodidaxy. C . S U M M A R Y In this chapter, it has been argued that adult education as a field of study and practice has sought unique characteristics to distinguish it from other sectors of the educational domain. Central to the enterprise of adult education is the construct of adulthood, and central to adulthood is the notion of 'self-direction.' Self-direction, however, is applied to a number of different phenomena. Some see it as a personal characteristic or quality - either a general disposition or one which manifests itself in learning situations, others as the independent pursuit of learning beyond institutional frameworks, and others as a way of conducting instruction. For a number of reasons, this chapter advocates the abandonment of the term 'self-direction' in favour of several alternatives. Thus self-direction as a personal attribute is referred to as personal autonomy, self-direction in natural societal settings is called autodidaxy, and self-direction in instructional situations is referred to as learner control. It was argued that learner-control might be thought of as a continuum, and that the end of the continuum which maximises learner-control could be referred to as 'independent study.' In many respects, independent study appears to resemble aotodidactic learning. However, it is argued here that the concepts 32 are not synonymous, and that educational agents such as teacher cannot make learners into autodidacts simply by diminishing their exercise of authority. The next chapter will consider what it means to be personally autonomous, and whether or not personal autonomy can be influenced by educational interventions or educational experiences, in particular in adult education. III. 'SELF-DIRECTION' AS A PERSONAL ATTRIBUTE A. INTRODUCTION As mentioned in chapter two, the term 'self-direction' is applied to people—as a personal attribute or characteristic—and to learning situations. In this dissertation, 'self-direction' in the former sense will be referred to as 'personal autonomy.' It is the purpose of this chapter to analyse literature pertaining to personal autonomy, to derive a working definition of it, and to examine the extent to which its attainment may be attributed to educational (particularly adult educational) interventions. The term autonomy literally means 'self-rule'. Originally, the term applied to the property or characteristic of cities in ancient Greece; "The city had autonomia when its citizens were free to live according to their own laws, as opposed to being under the rule of some conquering or imperial neighbour" (Dearden, 1972, p. 448). In due course, the term was extended to smaller social groupings and the parallel between a self-governing city-state and an individual person eventually led to the adjective 'autonomous' being applied to persons as well as cities: "By analogy, the autonomous person is an independent agent, one who is in command of himself, the author of his own work, deeds and way of life, not subject to the authority of other persons or things" (B. Gibbs, 1979, p. 119). In his paper on 'Autonomy and authority in education', B. Gibbs defines the essential characteristics of autonomy as intellectual self-determination, fortitude and temperance. These personal qualities, he argues, are precisely the cardinal virtues which Plato delineated in The Republic. Gibbs (1979) goes on to 33 34 demonstrate the parallels which Plato drew between the city and the individual soul, where the proper task of education is to establish within the soul something analogous to the constitutional government of the city. He deals with each of the three aspects of autonomy in turn, and notes that; "Just as the several elements of a self-governing nation must be unified and at peace with one another, so in a temperate soul there is 'friendship and accord' between the elements" (p. 123). In everyday discourse, the term 'autonomy' is used to denote a state of freedom, of independence, and perhaps of self-sufficiency. Many people have attempted to define what may be meant by autonomy, and its multidimensional nature is reflected in its many possible definitions. Thus, an autonomous person is one: * whose life has a consistency that derives from a coherent set of beliefs, values and principles (Benn, 1976); * who engages in a "still-continuing process of criticism and re-evaluation" (Benn, 1976); * who "is obedient to a law that he prescribes to himself (Rousseau); * whose thoughts and actions, being determined by himself, "cannot be explained without referring to his own activity of mind" (Dearden, 1972); * who demonstrates a responsiveness to his or her environment, and the ability to make creative and unique responses to situations as they arise, rather than patterned responses from his or her past (Jackins, 1965); * who is "capable of formulating and following a rule, pattern or policy of acting and working" (Gibbs, 1979); * who has independence from external authority, free from the dictates and interference of other people (Gibbs, 1979); 35 * who has mastery of himself or herself, free from disabling conflicts or lack of co-ordination between the elements of his or her personality; and * who, instead of taking over unquestioningly the judgements and opinions of others, scans evidence, examines assumptions and traces implications - in short, uses his or her reason (Paterson, 1979). Two comments with respect to the above list are called for. First, many of these definitions emphasise the highly individualistic, situationally variable and psychologically complex nature of personal autonomy; features which should be, but often are not, reflected in the research paradigm used to study 'selfdirection' in learning (see chapter eight). Second, definitions such as the foregoing accord with common-sense understandings of the concept of autonomy. However, if the development, enhancement, or recognition of personal autonomy is to be acknowledged as a central feature of the enterprise of adult education, it is essential to have a more specific idea of what an autonomous person is like. Such an idea is expressed by Krimerman (1972): . . . if we cannot distinguish persons from objects, self-actualized individuals from those who lack any distinctive human excellences, then we can have no hope of finding reliable ways of liberating the former from within the latter, nor of developing (Rogerian) men "from whom creative products and creative living emerge" . . . We will be unable to assist ourselves or others in enlarging the scope of choice and autonomy or in moving towards what we think of as the most distinctive and desirable forms of human activity. (Krimerman, 1972, pp. 333-4) Accordingly, the next part of this chapter will review literature on personal autonomy and will include an operational definition of what might usefully be meant by the term. B. T H E CONCEPT OF PERSONAL AUTONOMY In discussing the sort of evidence one might search for to ascertain whether or not a person was autonomous, B. Gibbs (1979) distinguishes two conceptions of autonomy; those which concentrate on its intellectual dimensions, and those which regard it as more of a moral quality: 1. In one view, autonomy is equated with critical intelligence, independence of thought and judgement, discernment, involving not necessarily a high degree of intellectual originality and enterprise, but at any rate a readiness to think things out for oneself free from bias and prejudice . . . This conception of autonomy is probably the most familiar, for it is part of an individualistic, anti-authoritarian ideology which is very deep-rooted in Western capitalist democracies . . . (p. 121, emphasis added) 2. The other conception "envisages autonomy as fundamentally a moral virtue or a disposition of character rather than intellect: self-mastery or self-discipline, having command of one's own feelings and inclinations." (p. 121), where "self-mastery is conceived as something like what used to be called fortitude, or . . . temperance, or a combination of the two." (p. 122) Crittenden (1978), for his part, writes of three overlapping components or dimensions to autonomy: 1. "Intellectual autonomy would require in the first place that a person not accept any of his important beliefs primarily on the authority of others, but on his own experience, his own reflection on evidence and argument, his own sense of what is true and right . . . 2. "Moral autonomy . . . in addition to independence of thought in determining and applying criteria of moral judgment . . . also includes the 37 executive capacities for carrying into practice what one decides should be done. The possession of these capacities is commonly described by such terms as tenacity, resoluteness, strength of will, self-mastery3. 3. ["Emotional autonomy] implies not simply that a person would exercise self-mastery in the face of strong emotional involvement, but that he would remain emotionally detached in his relationships to other persons and things" (p. 108, emphasis added). Partridge (1979) argues that the development of personal autonomy is a sufficiently important goal that it justifies the compulsory imposition of liberal education on children, and the violation of the prima facie right to non-interference. She claims that three conditions distinguish the autonomous person: freedom of choice, rational reflection and strength-of-will. The first of Partridge's criteria, freedom of choice, concerns "freely chosen acts . . . for which the agent has causally operative reasons as opposed to rationalizations" (p. 65) for acting in a particular way. "We do not attribute the exercise of autonomy to anyone whose freedom is constrained either outwardly or inwardly" (p. 65). Outward freedom implies the absence of physical constraints (such as violence or threat of violence), and psychological constraints (including, but not limited to, hypnosis and other forms of psychological manipulation). Inward freedom has been dealt with at length by R. S. Peters (1973), and includes the absence of acute deprivation, hysteria, paranoia, obsessions and delusions, psychopathy and various forms of compulsions such as kleptomania (pp. 123-124). Partridge's second criterion for identifying autonomous people is rational reflection, and she goes on to define what is meant by this term; "in saying 38 that rational reflection is a necessary condition of autonomy, we are saying two things: (1) that one must have reasons for one's behaviour; and (2) that one's reasons must be good ones" (p. 69). Reasons are defined as "considerations which the actor takes into account in holding certain beliefs, proving certain points etc." (p. 69). Partridge states that this criterion does not mean that one has to consciouslj' review one's reasons for doing everything, "but [that] very likely one could supply the reasons if asked to do so . . ." (Partridge, 1979, p. 69). Being able to supply reasons is one thing, but being able to show that they are good is more difficult, and Partridge lists four criteria for deciding if reasons are good or not: (1) they must be deliberated on; (2) using non-arbitrary criteria; (3) in as objective a way as possible (which implies the ability and willingness to change one's mind or alter one's belief in the light of new evidence or changed circumstances); (4) using relevant and adequate evidence. She goes on; "We do not require that one's beliefs be true or one's reasons wholly accurate before we attribute the possession or exercise of autonomy . . . neither false belief nor errors in judgement necessarily constitute a threat to autonomy . . . " (pp. 73-74). The third of Partridge's criteria for judging autonomy sounds somewhat quaint; it is strength-of-will. Where there is no strength-of-will to carry through with the choices one has made, according to Partridge, there can be no autonomy. This does not mean that a strong-willed person will not experience conflicts or indecision (indeed the more one engages in rational reflection, the greater may be the conflict one experiences). What it does imply, however, is that the strong-willed person, having systematically organised his or her priorities into some sort of hierarchical structure, is more likely to resolve conflicts and 39 dilemmas and arrive at a new state of equilibrium. The weak-willed person, on the other hand, is more likely to be swayed by whims and impulses, to be immobilised by indecision, and to act anomically rather than autonomously (pp. 74-75). According to R. S. Peters (1973); "The strong-willed [person] . . . sticks to his [or her] principles in the face of ridicule, ostracism, punishment and bribes" (p. 125). It will be noted that these quotes are increasingly specific as to the type of behaviour which an autonomous person might be expected to exhibit. Krimerman (1972) goes even further than this. He concerns himself with autonomously selected beliefs and desires, which may be thought of as approximately equivalent to Partridge's criteria of rational reflection and strength-of-will respectively. He discusses them as follows: 1. Autonomous belief. The following lend support to the claim that a person's belief in a certain proposition (P) is autonomous: a. has the ability to explain P to others using words and in circumstances substantially unlike those in which P was first encountered; b. has tested and evaluated P against alternatives, even when there are no extraneous rewards (social, psychological or physiological) for doing so; c. is willing to relinquish or decrease belief in P when relevant counterevidence is presented; and d. is not angered, threatened or incapacitated when objections or alternatives to P are presented. 2. Autonomous desire. The following are among the considerations which might 40 be used to test whether a person's desire for a goal (G) is autonomous: a. has the ability to explain (in terms and circumstances different from those in which the goal was first encountered) what G consists of, how it differs from other goals, and how it might be achieved; b. has personally experimented with alternative goals without the threat of sanctions, or hope of rewards, for such experimentation; c. is willing to curtail or eliminate the pursuit of G when autonomously held beliefs concerning G alter, or when it becomes apparent that attainment of G is incompatible with other, more highly valued goals; and d. is not angered, incapacitated or threatened when exposed to criticisms of the value of G , or when temporarily prevented from pursuing or attaining G . (Krimerman, 1972, pp. 334-336 passim) Dearden (1975) has also attempted to operationalise the concept of autonomy, and suggests that an autonomous person would characteristically: 1. wonder and ask, with a sense of the right to ask, what the justification is for various things which it would be quite natural to take for granted; 2. refuse agreement or compliance with what others put to him, when this seems critically unacceptable; 3. define what he really wants, or what is really in his interests, as distinct from what may be convenient!}' so regarded; 4. conceive of goals, policies and plans of his own, and form purposes and intentions of his own, independently of any pressure to do so from others; 5. choose amongst alternatives in ways which could exhibit that choice as the deliberate outcome of his own ideas or purposes; 6. form his own opinion on a variety of topics that interest him; and 7. govern his actions and attitudes in the light of the previous sorts of activity (p. 7). A l l of these actions imply an intentional control of one's life, and accordingly; " A 41 person could not be to any marked degree autonomous, without this being an important part of his self-concept" (Dearden, 1972, p. 460). With respect to all the criteria for judging an autonomous person, Dearden (1975) concludes: To be autonomous therefore, is very much a matter of degree . . . Unlike being six feet tall, married or a British citizen, whether a man is quite simply autonomous or not is something we will quite often rightly refuse to say. And our hesitation will be related to at least three dimensions of variability: the extent to which he shows initiative in forming judgements of his own, the firmness with which he then adheres to those judgements, and finally the depth of ramifying reflection which lies behind the criteria which he employs in making those judgements, (p. 9) Krimerman (1972) concurs with this judgement that autonomy is a matter of degree, and even goes so far as to suggest that any given act might be assessed using the various criteria and rated along a scale according to how many of the criteria were satisfied, from "unequivocally or paradigmatically autonomous" to "an ideal case of non-autonomous behavior" (p. 336). Even though one might blanch at the practical, not to mention ethical, difficulties of attempting to classify people along a continuum of personal autonomy, nonetheless the notion that autonomy is a 'matter of degree' is a useful one. To the extent that autonomy is regarded as a developable capacity, the notion may have important educational implications because it would allow practitioners to identify and, with learners, work on areas of perceived weakness. 1. Towards a definition of autonomy As a result of the foregoing discussion, it is now possible to conclude that an individual is autonomous to the extent that he or she: 1. conceives of goals, policies and plans, and forms purposes and intentions of 42 his or her own, independently of any pressure to do so from others; 2. exercises freedom of choice in thought or actions, without inward or outward constraints or restrictions on his or her capacities to act or to reason; 3. uses the capacity for rational reflection, judging among alternatives; a. on the basis of morally defensible, non-arbitrary beliefs as to what is true or right, derived from personal experience and/or reflection; b. as objectively as possible; c. using relevant and adequate evidence; 4. has the will and the capacity fearlessly and resolutely to carry into practice, and through to completion, plans of actions arrived at through (1), (2) and (3) above, without having to depend on others for encouragement and reassurance, and regardless of opposition; 5. exercises self-mastery in the face of strong emotional involvements, reversals, challenges and setbacks, and remains emotionally detached as far as possible; and 6. has a concept of himself or herself as autonomous. A list such as this may provide a profile of an 'ideal' or prototypic autonomous person, and might even serve as a checklist for designing learning activities to encourage personal autonomy. However, it is still not easy to be clear whether or not a person is autonomous. There are two reasons for this: first, there are many threats to autonomy not evident to an observer (and sometimes not even known to the subject himself or herself), and second, autonomy is situationally variable. These considerations will be dealt with in turn. 43 2. Threats to autonomy In the original political context, autonomy was contrasted on the one hand with 'heteronomy' (meaning domination and rule by others) and, on the other hand, with 'anarchy' (meaning chaos and disorder occasioned by absence of government). These two situations have their parallel in the case of the individual people who may lack autonomy either because they are under the jurisdiction or influence of another, or alternatively because of discord and disharmony within themselves. The situation of being under the control or influence of others is by no means easy to identify, and it is this which perhaps causes many educators to mistake the absence of overt or apparent constraint, for autonom}'. At one level, it is often assumed that as long as people are not physically and psychologically threatened, they will behave autonomously. Clearly, if someone hands over his or her possessions at the point of a gun or, in less dramatic circumstances, enrols in a course because of threatened retrenchment, he or she is not acting autonomously. However, these are by no means the only types of pressure to which people are subject, and it is seriously questioned whether they can ever be entirely free from external pressures and considerations in living their lives. The fact of being part of a social community implies the acceptance of certain standards of behaviour, and rules (e.g., language). While freedom may be necessary to the exercise of autonomy, and even to its development4, the absence of external constraints in a particular context is not, in itself, a sufficient condition for autonomy. Certain other factors must be 44 present, yet this is frequently overlooked by adult educators who advocate certain freedoms in the instructional setting in the belief that this will inevitably lead to the exercise of autonomy. As Dearden (1972) says: A long term prisoner might gain his freedom, but have been so incapacitated for ord ina l life by the institutional life of the prison that he exhibits only anxiety and withdrawal in this state of freedom, rather than the capacities of self-direction and choice which are characteristic of autonomy, (p. 451) This issue will be dealt with in chapter seven, in considering whether or not it is logical^ possible for adult educators to give learners autonomy, to assist them to develop autonomy, or merely to create circumstances within which they might exercise autonomy. But even supposing that people were ostensibly free to think and act as they like, this would still not guarantee that they would behave autonomously. A person would not be regarded as autonomous if, for instance, he or she were merely following some anomic whim or falling under the influence of some propaganda, advertising claim, opportunity or point-of-view which had been encountered. What is required is some stable set of personal beliefs (or 'rules') which guide and give consistency (but not rigidity) to their actions. It will be recalled that, in the political domain, autonomia referred to 'self-rule' and the notion of rules or laws is one of the entailments which was carried over when autonomy was transplanted from the political to the personal domain. Accordingly, the existence, development and status of personal rules, laws or norms of behaviour or judgement are central to discussions of personal autonomy. One position holds that the fully autonomous person is, in every respect, the author of his or her own destiny, and that the criteria used to 45 make personal decisions are in themselves the product of his or her own enquiries, analysis and reflection. A n alternative point-of-view acknowledges that it is impossible for a person to achieve the maturity of adulthood without innumerable encounters with the environment and with other people; encounters which inevitably shape values or conceptions of right and wrong, good and bad, worthwhile and worthless. Few, perhaps, would agree with Skinner's (1971) observation; "as we learn more about the effects of environment, we have less reason to attribute any part of human behavior to an autonomous controlling agent" (p. 96). However, since there is a limit to the absolute number of such guiding principles in existence, and in view of the pervasive and profound influence of early conditioning, it is unlikely, if not impossible, for anyone to escape entirely the influence of others in forming personally relevant rules. This is not to say, however, that people must always be passive victims of their biographies, condemned forever to an acceptance of values and rules uncritically internalized at an early age. A n autonomous person is able to assent to rules, or modify or reject them, if they are found wanting. Irrespective of whether rules are autonomously derived, as in the first case, or critically assented to, as in the second, there still arises the question of what criteria people bring to bear in determining the value, legitimacy, or appropriateness of their 'first-order' rules. What is called for is a superordinate or 'second-order' value system by which to judge the 'first-order' rules. However, these 'second-order' criteria also have to be subjected to some sort of scrutiny, or else to be derived autonomously and in either case, they too need to be critically evaluated according to yet another higher order set of criteria. This line of reasoning is followed backwards in an infinite regress, until the point is reached 46 of the autonomous person making some 'criterionless choices.' At this point, as D. C. Phillips (1975) indicates, for all practical purposes, it becomes impossible to distinguish the Autonomous Person (AP) making 'criterionless choices' from the Person Lacking in Autonomy (PLIA), whose behaviour is based on following rules which have been internalized without being subjected to critical reflection5. Such a picture runs counter to the usual notion of the autonomous person as one who makes decisions on the basis of carefully considered values and beliefs. From the foregoing, it can be seen that autonomy is a difficult concept to operationalise, and that it is not easy for an outside observer to be clear whether any given pattern of behaviour is autonomous or not. This is because autonomy cannot be detected solely from behaviour, but must also be understood in terms of the actor's (for instance a learner's) own intentions and understandings. Behaviour which may, on the surface of it, seem to be autonomous, may simply be independent, but in fact determined by some 'script' or 'program' implanted in the learner at an earlier time. In an attempt to get round this problem, Benn (1976) acknowledges the pervasive effect of socialising influences, but goes on to write: Within this conception of a socialized individual, there is room to distinguish one who simply accepts the roles society thrusts on him, uncritically internalizing the received mores, from someone committed to a critical and creative conscious search for coherence . . . (p. 126) However, even this generalised definition is of limited use to a researcher or observer, unless he or she is somehow able to have access to the perspective of the research subject. This is because coherence, like autonomy in behaviour, is not inherent in the act itself, but is attributed to the act by the individual actor. This issue will be considered again in chapters eight to eleven. 47 A second reservation about the definition of autonomy presented earlier in this chapter is the failure to account for situational variability, and accordingly this will be considered next. 3 . Personal autonomy as situationally variable Much of the research into personal autonomy has been based on the notion that it is a context-free disposition; that once people 'become' autonomous, they will behave autonomously in whatever situation they find themselves. In terms of the argument put forward in this dissertation, there are two flaws with this line of reasoning. The first is that autonomy is more akin to a process than a product. That is, one does not 'become' autonomous in any final or absolute sense. The second flaw is that, although some people manifest more self-assurance, or clarity of purpose across a range of situations, it is impossible to judge whether or not a person is autonomous without specifying the context within which this autonomy will , or might, manifest itself. In other words, autonomy is not simply a personal quality or characteristic, but is a relation involving the interplay of personal and situational variables. According^, any person could vary in the degree of autonomy he or she exhibits from situation to situation. For instance, a person who may be autonomous with respect to career or family, may lack autonomy (i.e., be dependent) when it comes to learning; or alternatively, a person who is autonomous with respect to learning how to sail a boat, might prove to be dependent when it comes to learning calculus or Spanish. This reservation about research into personal autonomy will also be discussed later in this dissertation. 48 C. T H E D E V E L O P M E N T O F A U T O N O M Y As mentioned at the start of the dissertation, there is a widespread belief that permitting or encouraging 'self-direction' in learning will lead to the development of personal autonomy as an educational outcome. It is the purpose of this section of the chapter to examine evidence concerning whether or not personal autonomy is attained as part of a natural developmental process, or whether it is susceptible to educational interventions. This is important because of the reciprocal relationship which is assumed to exist between personal autonomy and autonomy in learning. In the discussion which follows, three alternative points of view are put forward. These are: (1) that autonomy is a trait, or innate disposition; (2) that it is acquired through non-educational processes of socialisation and maturation; and (3) that it is learned through educational experiences. Each of these will be dealt with in turn, and its educational implications considered. 1 . Autonomy as an innate disposition In the first case, which is sometimes called 'nativism,' autonomy is seen as a condition of freedom from the dictates and interference of other people, and is thus the situation into which children are born. One of the best known contemporary advocates of this position is Carl Rogers, who writes of the child; "Unlike many of us, he knows what he likes and dislikes, and the origin of those value choices lies strictly within himself (C. R. Rogers, 1969, p. 243). Rousseau is another famous advocate of the essential goodness and autonomy of the 'natural child,' as typified by the spontaneous, uninhibited and free behaviour of Emile. So, too, is A. S. Neill, father of the famous experiment 49 in free education at Summerhill. One corollary of viewing autonomy in this way is that education must interfere as little as possible with the learner's natural inclinations and interests, as these represent the outworking of the person's autonomous preferences. For those who regard autonomy as an innate quality of childhood, one which is diminished or even extinguished by the processes of socialisation, education should consist of liberating the real self from within the social self (Strike, 1982, p. 151). Teachers should limit themselves to supplying resources required by the student's natural inclinations and, since liberty is the removal of restraint, autonomy is facilitated by having minimal intervention by the teacher. According to Dearden (1975), it is difficult to know what method of education is appropriate to such innately autonomous people: the only appropriate method would seem to be that of personal discovery. But if this method were taken quite literally, its effectiveness for most individuals would be limited, and it would make impossible the cumulative achievement of knowledge and skill from one generation to another. Nor would it be possible to apply any public criteria to the quality of what an individual discovered for himself. It could not be said, for example, that a conclusion he had reached was false, or insignificant or biased. It is difficult to see how we can speak seriously at all of the education of human beings, if they are interpreted as asocial and ahistorical atoms, (p. 115) A variant of this 'nativistic' model is the view that there is one autonomy of childhood based on impulse, and another of adulthood based on reason, and that over time there is a progressive shift in emphasis from one to the other. Some see this 'handover of power' or transition as a purely developmental process which might potentially be deflected or otherwise adversely affected by education which, in this view, is seen as a socializing agency, capable of extinguishing the tender shoots of moral autonomy, and of perverting 50 the natural development of intellectual autonomy. Others, such as Dewey (1963), clearly see a mandate for education when he writes; "The crucial educational problem is that of procuring the postponement of immediate action upon desire until observation and judgement have intervened. Overemphasis upon activity as an end, instead of upon intelligent activity, leads to identification of freedom [autonomy] with immediate execution of impulses and desires" (p. 69). In this view, autonomy does not consist of following one's whims, but rather in the selective application of "observation and judgement," both of which skills can be sharpened through education. 2 . Autonomy as an acquired quality Although it may be plausible to attribute some dimensions of autonomy to childhood, few would seriously maintain that children are capable of the mature rational reflection, objectivity or emotional detachment implied by the term autonomy in the present context. As people develop and "become fully human" (Strike, 1982, p. 153), they "internalize available social and cultural resources." Earlier in this chapter, it was argued that autonomous adults will not simply accept what is thrust on them, but "children are in no position to judge the value of these cultural and social resources" (p. 153), and thus "all of us, in the first stage of our education, are Persons Lacking in Autonomy" (Phillips, 1975, p. 9). As early as 1932, Piaget, discussed the socialization process of the child, describing the evolution of sequential stages which he identified as moving from 'heteronomy' to 'autonomy.' The following summary provides a delineation of behavioral and attitudinal components characteristic of heteronomy and autonomy: Heteronomy egocentrism Autonomy cooperation mutual respect individual creativity flexibility rational criticism inner directed independence. unilateral respect conformity rigidity blind faith in authority other directed dependence According to Piaget, the normal process of development in the healthy person involves progressing from heteronomj' in the direction of autonomy Another way of expressing this process of development is to talk of maturity. As people grow older, they mature and it is this which distinguishes children from adults; "most human beings acquire [maturity] gradually and informally over a period of years" (Strike, 1982, p. 129). There is no criticism implied in saying that a child is immature, although such a comment has quite a different connotation when applied to adults, because, "with the exception of the mentally ill or senile, all adults are in the maturity of their faculties" (p. 130). Maturity, in turn, is linked to autonomy, in that "passive and uncritical acceptance of one's situation is characteristic of an essentially immature mind" (Overstreet, 1950, p. 250). As early as 1941, for instance, Angyal, in a foundational study on personality, argued that increasing maturity in the healthy adult is accompanied by an increase in independence. "According to Angyal, the psychological aspects of the individual . . . move toward greater autonomy, becoming less and less bound by the immediate situation, and the individual is more and more able to weigh possible outcomes and select that which is most advantageous, advantageous meaning that which leads to increased autonomy" 52 (Birren & Hedlund, 1984, p. 63). If the linkages between age, maturity and autonomy were invariant, all older people could be expected to exhibit the hallmarks of autonomy. However, as has already been mentioned, it is evident that, in any given context, some people will behave more autonomously than others; that is, they will generate alternative goals, select decisively from among them, persevere with their intentions, exert disciplined self-control and so on. R. S. Peters (1973) notes; "that since people are not autonomous when they are born, and since many people reach old age without attaining very high levels of autonomy, some learning process which is not purely maturational is involved in becoming autonomous" (p. 176). To the extent that characteristics of autonomy may be demonstrated by people who have had limited exposure to formal education, one can dismiss the claim that education is the sole (or for that matter, even the main) contributor to the development of autonomy. However, this does not preclude the possibility that education, including adult education, can contribute to the enhancement of autonomy. The question which arises is: what is the place of education in this "learning process which is not purely maturational"? 3. Autonomy as a learned characteristic The development of personal autonomy is almost universally proclaimed as a goal of education6. However, if one refers to the definition proposed earlier, it is apparent that some aspects of autonomj' are more amenable to educational intervention than others. For instance, some qualities or characteristics, such as those of emotional autonomy or perseverance, are partly personal attributes or in 53 any case attributes whose origins in the person are rooted deeply in their very earliest experiences at home and school. Other components of autonomy (for instance the ability to rational^ reflect) may be taught, at least in part, as curricular content (Telfer, 1975, p. 28; Wang, 1983). Assertiveness training, life-planning skills, values clarification, synectics and creativity training, learning-to-learn, moral philosophy and critical thinking have all been advocated as content areas because of their alleged contribution to the development of personal autonomy. Other dimensions (such as a self-concept of being autonomous) are not only more difficult to operationalise, but are best enhanced ancillary to, or concomitant with, some other content-oriented instruction. Thus, certain techniques of instruction, or ways of conducting teaching, have also been linked with autonomy. These include collaborative learning, guided learning, discovery learning, contract-based learning, individualised instruction, learner-controlled instruction, open learning, and independent study (or self-directed learning). This notion is hardly new; it is enshrined in Snyder's (1971) formulation of the 'hidden curriculum,' and in McLuhan's oft-quoted dictum that "the medium is the message." Experience clearly shows, however, that even when people are exposed to such educational approaches, not all become more autonomous. In fact, none of these interventions—content or process—leads invariably to increased personal autonomy, and none has a monopoly on the development of personal autonomy because of the multidimensional nature of the latter (Dittman, 1976, p. 467). Moreover, there is a high degree of interdependence amongst the various components of autonomy. Any attempt to develop one or two parts in isolation is likely to fail in the end. Similarly, if a particular strategy (such as a program 54 on learning-to-learn) is successful, it will predictably lead to increased critical awareness, and demands for more freedom of choice, and other manifestations of autonom}'. a. Personal autonomy and adult education Before leaving the question of educational interventions, it is important to discuss, with respect to the development of autonomy, the proper and legitimate role of adult education. Five points will be made. The first is that all forms of education should have aims which "cohere with the aims we set for life itself. Our philosophy of education must connect with, and in the end be justified by, our ethical, political and even our religious beliefs" (Hostler, 1981, p. 14). Adult education, too, derives its mandate from the society in which it occurs. This point is made by Botkin, Elmandjra and Malitza (1979) in their report to the Club of Rome, No limits to learning: Bridging the human gap; There is a near-universal demand for increased participation at all levels. More people are aware of, and are using, their capacity to obstruct rather than to support decisions reached without their concurrence, regardless of the merits of such decisions... The term participation is not new. Few words convey so powerfully the idea of the individual's aspiration to be a partner in decision-making, of the unwillingness to accept unduly limited roles, and of the desire to live life more fully. Few terms suggest so forcefully people's claim to influence both local and global decisions that shape their environment and lives, coupled with people's aspirations for equality as well as their refusal to accept marginal positions or subordinated status, (pp. 29-30, emphasis added) This trend toward participatory democracy is clearly one of the "wider social changes" (Dearden, 1972, p. 449) contributing to increased interest in autonomy. 55 It affects education in two, related ways. The first is that for people to be able to participate fully, they need to be capable of autonomous thought and action and; "since [personal autonomy] is a fundamental value of our society, a responsibility is obviously placed on education to enable such autonomy" (Strong, 1977, p. i). The second is through the inevitable demands for increased participation within education itself; as people become more accustomed to, and skilled at, informed participation and 'choosing,' in other aspects of their lives, they are likely to make increasing demands for similar power-sharing in relation to their education. Thus, to the extent that personal autonomy is a valued societal goal, it is reasonable to expect to find various forms of adult education attempting to foster personal autonomy as a goal. It is, however, improper to speak of adult education as if it were a single, unified system—either conceptually or administratively. Adult education encompasses a greater diversity of locations, goals, clients, techniques and contents than other sectors of education. Again, as Hostler (1981) notes, "It comprises a host of courses, usually part-time and mostly conducted in the evenings, . . . [It has] a multitude of different goals, some of which are broad and long term, while others are immediate and more restricted, which overlap, diverge and even conflict in subtle and complicated ways" (p. 1). Thus, it is not reasonable to expect that all forms of adult education will have, either implicitly or explicitly, the goal of promoting personal autonomy. The second point relates to the first. It is not uncommon to find, in the literature of adult education, the assumption that all adults are autonomous or 'self-directing,' and that instruction