i The copyright of this thesis vests in the University of the Witwatersrand, Johannesburg, South Africa, in accordance with the University’s Intellectual Property Policy. No portion of the text may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, including analogue and digital media, without prior written permission from the University. Extracts of or quotations from this thesis may, however, be made in terms of Sections 12 and 13 of the South African Copyright Act No. 98 of 1978 (as amended), for non-commercial or educational purposes. Full acknowledgement must be made to the author and the University. An electronic version of this thesis is available on the Library webpage (www.wits.ac.za/library) under ‘Research Resources’. For permission requests, please contact the University Legal Officer or the University Research Office (www.wits.ac.za). ii THE EXPERIENCE OF BECOMING A PHD Jennifer Ann Hadingham A thesis submitted to the Wits School of Education, Faculty of Humanities, University of the Witwatersrand in fulfillment of the requirements for the degree of Doctor of Philosophy Johannesburg, 2011 iii ABSTRACT AND KEY WORDS The development of the next generation of researchers is a priority if South Africa is to make a significant contribution to the international knowledge economy and establish itself as a force to be reckoned with in international research circles. In the context of this knowledge economy, researchers are increasingly being recognised as agents of economic growth. In order to be competitive, therefore, an extensive pool of active researchers needs to be cultivated. One way of doing this is to promote and develop doctoral capacity at the country’s universities. This entails, among other things, a move away from the traditional focus on what the supervisor does, to a more student-centred understanding of how the doctoral candidate experiences the process, and by implication, how this impacts on their research contribution. In this qualitative study, thirty doctoral candidates from the Faculties of Science and Humanities at the University of the Witwatersrand, Johannesburg, were interviewed in order to establish firstly, how they had experienced their supervision at this level, and secondly, whether or not these experiences had influenced the successful completion of their doctorates. One of the principal findings of the research was that the role of the supervisor was not central to the achievement of their degree; rather, many of the doctoral candidates asserted a significant level of agency in both identifying and making contact with experts beyond their university-appointed supervisors in order to supplement their access to relevant knowledge. In the majority of cases, this was encouraged by the supervisors. Such enterprises represent a marked departure from the traditional models of supervision in the Science and Humanities faculties. In the case of the former, the customary co-supervision arrangement is increasingly being augmented by student- initiated collaboration with authorities located outside the formal boundaries of the institution. The traditional Humanities model of supervision is also transforming from a one-on-one relationship to a style characterised by multiple supervisors, each from separate but cognate disciplines. The research suggested that these emergent models are eclipsing their predecessors as doctoral candidates increasingly recognise the value of extending the breadth of their disciplinary exposure beyond the confines of the university. iv Keywords: PhD; doctorate; supervision; postgraduate studies; narrative inquiry; communities of practice; international knowledge economy; agency; power v DECLARATION I declare that this thesis is my own unaided work. It is being submitted for the degree of Doctor of Philosophy at the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination at any other University. Jennifer Ann Hadingham __________ day of _________________________ in the year _________________ vi DEDICATION This thesis is dedicated to family, friends and colleagues who have shared my journey with me: Mike and Rhoda Hadingham – my ever-patient parents Tim and Karen Hadingham – my brother and sister-in-law Ziningi, Andy and Cailin Rose – my nieces and nephew who have added smiles and laughter along the way Jacqui, Charles, David and Timmy de Matos Ala – travelling friends and Friday night movie companions Margaret Orr – boss, role-model, and revolutionary Cheryl Chamberlain, Alison Button, Hilary Geber, Gill Mooney and Judy Backhouse – friends and sounding boards par excellence Del King, Candice Michael, Shereez Peffer, Yasmin Dadabhay and Mpumi Mazibuko – colleagues, friends and takers of no nonsense Jane Castle and Ruksana Osman – doctoral supervisors second to none Thank you all. vii ACKNOWLEDGEMENTS Professors Jane Castle and Ruksana Osman have been extraordinary supervisors – they picked me up, dusted me off and helped make a confident doctoral candidate out of me. Their feedback, advice and insights have added significant value to this thesis. Without them, it would never have been completed. My first PhD supervisor, who guided my selection of a research topic and saw me through the proposal and ethics clearance stages. Professor Margaret Orr provided the workplace support: time, photocopiers, strong coffee and amused tolerance when my mind wandered. viii TABLE OF CONTENTS List of Tables and Figures Page xiv GLOSSARY xv PROLOGUE – MY EXPERIENCE OF POSTGRADUATE SUPERVISION 1 CHAPTER 1 – INTRODUCTION TO THE WORK 6 Chapter overview 6 1.1 Introduction 6 1.2 Research aims and objectives 9 1.3 Research questions 9 1.4 Contextualising the research 10 1.4.1 The Institution – Wits 11 1.4.2 The National Context – the CHE and the HEQC 15 1.4.3 Going Global – The International Knowledge Economy 22 1.5 Some final directions 27 CHAPTER 2 – LITERATURE REVIEW (PART 1) Plotting the parlance of the PhD: from pedagogy to practice 29 Chapter overview 29 2.1 Introduction 29 2.2 Concept clarification 31 2.2.1 The doctorate/Doctor of Philosophy 31 2.2.2 Supervision 35 2.2.3 The pedagogy of supervision 40 2.2.4 The discourses of supervision 48 2.2.5 Communities of practice 50 ix 2.2.5.1 Mutual engagement 53 2.2.5.2 Joint enterprise 56 2.2.5.3 Shared repertoire 59 2.2.6 Beyond communities of practice 62 2.2.6.1 Compartmentalisation of practice 63 2.2.6.2 The contested nature of ‘community’ 65 2.2.6.3 The disputed nature of participation 69 2.2.6.4 A critique of the master-apprentice model 74 3 Conclusion 76 CHAPTER 3 – LITERATURE REVIEW (PART 2) The ‘how’ and the ‘who’: mapping the route and destination of supervision 78 Chapter overview 78 3.1 Introduction 78 3.2 Supervision guidelines 79 3.3 Institutional and national guidelines 90 3.4 Identity and the doctoral candidate 97 3.5 Conclusion 101 CHAPTER 4 – METHODOLOGY 104 Chapter overview 104 4.1 Introduction 104 4.2 Qualitative research 104 4.3 The narrative method of inquiry 107 4.4 The interview technique 116 4.5 Interview questions 124 4.6 Sample selection 126 x 4.7 Validity, reliability and generalisability 127 4.8 Ethics 131 4.8.1 Informed consent 131 4.8.2 Deception 135 4.8.3 Privacy and confidentiality 136 4.8.4 Accuracy 137 4.9 Conclusion 138 CHAPTER 5 – TALES FROM THE HUMANITIES 139 Chapter overview 139 5.1 Introduction 139 5.2 The narratives 139 5.2.1 Hannah 139 5.2.2 Thabang 142 5.2.3 Nkosinathi 145 5.2.4 Marcus 147 5.2.5 Faith 149 5.2.6 Patricia 151 5.2.7 Zodwa 155 5.2.8 Xolani 158 5.2.9 Sophia 160 5.2.10 Bridget 163 5.2.11 Rebecca 166 5.2.12 Mojalefa 168 5.2.13 Ntombi 171 5.2.14 George 174 xi 5.2.15 Alphons 176 5.3 Conclusion 179 CHAPTER 6 – THE SCIENTISTS’ STORIES 180 Chapter overview 180 6.1 Introduction 180 6.2 The narratives 180 6.2.1 Ace 180 6.2.2 Emily 182 6.2.3 Indira 184 6.2.4 Seamus 186 6.2.5 Nomsa 189 6.2.6 Kenneth 191 6.2.7 Karabo 195 6.2.8 Leigh 197 6.2.9 Günther 200 6.2.10 Bill 203 6.2.11 Dumisani 205 6.2.12 Megan 209 6.2.13 Bruno 212 6.2.14 Shanti 215 6.2.15 Kristelle 218 6.3 Conclusion 220 CHAPTER 7 – ANALYSIS Through the looking glass: students reflecting on their supervision experience 222 Chapter overview 222 xii 7.1 Introduction 222 7.2 Setting 228 7.2.1 Context 228 7.2.2 Master’s experience 232 7.3 Conclusion 247 CHAPTER 8 Through the looking glass: the journey, fellow travelers and milestones en route 249 Chapter overview 249 8.1 Introduction 249 8.2 The process 249 8.2.1 Feedback 250 8.3 Activities and events 255 8.3.1 Conferences, co-publications and communities of practice 255 8.3.2 Social interaction 266 8.4 Relationships and social structure 269 8.4.1 Power 269 8.4.2 Agency 277 8.5 What was missing from the candidates’ narratives 282 8.6 Conclusion 283 CHAPTER 9 – DISCUSSION AND CONCLUSION 286 Chapter overview 286 9.1 Introduction 286 9.2 Formula-ting supervision 287 9.3 The presence of pedagogy 287 9.4 Discourses of supervision 292 xiii 9.5 Communities of practice and beyond 297 9.6 Power and the doctoral relationship 299 9.7 Reflections on the design of the study 300 9.8 Further research 300 9.9 Final thoughts 302 REFERENCE LIST 304 xiv LIST OF TABLES AND FIGURES Page Table 1: Pile-up effects of Master’s postgraduate students 95 Figure 1: Evaluative questions for measuring the quality of institutional postgraduate supervision and research training 97 Figure 2: Levels of representation in the research process 110 Figure 3: Letter of consent: Interview 128 Figure 4: Letter of consent: Transcription of the recorded interview 129 Figure 5: The picture window analysis structure 220 Figure 6: Time-to-completion continuum 240 Figure 7: A typology of the x-centred approach to supervision 241 Figure 8: Feedback response time continuum 248 Figure 9: Feedback quality continuum 250 Figure 10: Levels of supervisor engagement with the doctoral candidate 258 Figure 11: The power continuum 266 Figure 12: The different aspects of personality that may impact on the supervisory relationship 273 Figure 13: The agency continuum: from student-led research to student detachment 277 xv GLOSSARY ANC African National Congress CHE Council on Higher Education CLTD Centre for Learning, Teaching and Development CoP Community of Practice DoBE Department of Basic Education DoHET Department of Higher Education and Training GET General Education and Training IKS Indigenous Knowledge Systems HAU Historically Advantaged University HDU Historically Disadvantaged University HEI Higher Education Institution HEQC Higher Education Quality Committee HWU Historically White University MA Master of Arts degree MRC Medical Research Council MSc Master of Science degree NGO Non-Governmental Organisation NRF National Research Foundation NSFAS National Student Financial Aid Scheme PhD Doctor of Philosophy degree PPO Postgraduate Projects Office SARChI South African Research Chairs Initiative SEnC Student Enrollment Centre UGCH University of the Cape of Good Hope xvi Wits University of the Witwatersrand 1 PROLOGUE MY EXPERIENCE OF SUPERVISION In 1997, I registered to do my Master’s degree (by dissertation) in Department X. I was looking forward to the challenge of writing an extensive analysis of my chosen topic because it was something that had intrigued me since my early days in this Department. The lecturer who had taught the section in my undergraduate days had been so passionate about the subject that I could not help but feel passionate about it too. I mistakenly made the assumption that, since he had been such an amazing undergraduate lecturer, he would be just as amazing as a postgraduate supervisor. I was wrong. I should have realized this when I was left entirely to my own devices to develop a topic. I ended up with a vague, cover-all kind of topic, which successfully camouflaged my lack of direction and focus. My supervisor did not appear to notice. When I embarked on the degree, I had the realistic expectation of a highly supportive and feedback-driven experience from the Department. I was doomed to disappointment. The proposal-writing stage was a disaster. Despite the Faculty’s ‘how- to’ guide to proposal writing, the process seemed impossible. I could not articulate the difference between the ‘aims’ and ‘rationale’ sections, and received conflicting advice - where any was forthcoming - about what information ought to appear in the literature review, and how this ought to be presented. Ultimately, pressurized by the Faculty to submit a proposal, I scavenged the proposal of a peer of mine who had nearly completed his degree; I then copied it verbatim, adapting it only slightly to suit my topic. What I couldn’t adapt, I simply omitted. Thus, my theoretical section was non- existent, my rationale weak and my literature review a joke. Surprisingly, the proposal passed through both my supervisor and the Higher Degrees Committee with no changes suggested. On this shaky basis began my Master’s degree. It was clear that I was bumbling around in the dark, and yet my supervisor (the ‘expert’) did not provide me with the guidance I needed. He unquestioningly accepted everything I wrote as being true, and failed to give me the kind of formative feedback that I needed in order to develop an intelligently argued piece of research. He had access to the wider community of 2 practice of scholars in my area of research, both in South Africa and abroad, and yet he never encouraged me to become a part of it. At the time, I assumed it was because I was too young, and not ‘good enough’ to join this community. Perhaps it was because I was a woman in a man’s realm of study (there is only one woman of significance in my research area). I had no contact with my supervisor’s other research students, either, so we were not able to compare notes about our research experience – were they experiencing the same issues as I was? If so, how did they cope? In the absence of such a community, I assumed that I was the ‘problem’ and that I needed to work harder. At the same time, I was engaged in an informal, but fierce, ‘competition’ with a friend of mine in the Department about who would be the first one to graduate. I envied her supervision experiences - her supervisor always seemed to give constructive and timely feedback (although he often forgot to arrive for scheduled supervision meetings). It did nothing for my self-confidence when I saw how he treated her as an equal in the Department, how excited she was at the feedback he had given her work and how she relished the challenge of using this feedback to improve her dissertation. I remember thinking at the time: “What am I doing wrong?” I began to internalize the notion that I was a weak, incompetent student whose work would forever remain ‘mediocre’. As a result, I began to feel incredibly demotivated, and wished that I had someone to talk to – somebody who could support and guide me. I was frequently frustrated at my supervisor’s lack of urgency when it came to reading and commenting on the chapters that I had submitted – it took him up to two months to mark each chapter; during this period, I had managed to write another one. His marking also left a great deal to be desired – little feedback and even less insight into what was good and what could be improved upon. At one point, I felt that this non-responsiveness was a deliberate attempt to derail my dissertation. But to whom could I complain? Methodologically, the dissertation was a mess. For instance, I had wanted to undertake interviews with local and international scholars so that I could reflect the various views on the topic; however, with no access to this community of practice, not the slightest idea about how one went about undertaking an interview and no input from my supervisor, these interviews never materialised. Instead, I relied on the opinions of journalists writing in local and overseas newspapers, and authors of books and journal 3 articles that I found in the library. At the end of the two years that I spent undertaking the ‘research’, all I had to show was a uni-dimensional, descriptive piece of writing, essentially a shopping list recording which author said what about whom and why. My voice is glaringly absent, as is any valid argument. In 2000, as I made my first attempt at a PhD, I was painfully reminded of my research and writing inadequacies, especially with regard to theory. There remained a gap in my academic knowledge that should have been filled in 1997, but wasn’t. I wondered whether it was too late to rectify this state of affairs. Fast forward a decade and two (aborted) attempts at a PhD proposal in Department X. I was beginning to despair of ever achieving the elusive doctorate. In the meantime – 2001, to be precise – I left Department X to take up a position at the university’s Staff Training Unit (STU). The need for a PhD remained in the back of my mind, but I was quite content to throw myself whole-heartedly into the business of developing academic staff into passable teachers while I waited for inspiration to strike. Naturally, that never happened. However, one afternoon, a colleague that I was ferrying to a function on the Education Campus told me about some research that was being undertaken by a senior professor in her department in an area of staff development that I knew to be under-researched: postgraduate supervision. Was I interested in developing a PhD topic in this area, asked my colleague? The answer was a resounding ‘I don’t know’ – the topic sounded interesting and fitted neatly into my work, but at the same time, I also knew that the senior professor who would be my supervisor was someone quite intimidating. She had a formidable reputation at the university – both for her vast knowledge and her abrasive personality when the occasion called for it. The first few meetings that I had with her were generally nerve-wracking, adrenalin-rich roller coaster rides, in which she would talk at me, and I would listen and nod mutely. With hindsight, I realize that her vision of the research and mine were very different – mainly because I never actually knew what her vision was, and was far to intimidated by her intellect to ask. Over the course of the three years that I was supervised by her, I never quite ‘got’ what it was that she had in mind for the research. I adopted a ‘hit-and- miss’ approach to my reading and writing, hoping that something would be good enough to earn praise from her. 4 Our meetings never got any easier. In the hours (and sometime even days) before a supervisory meeting, my anxiety levels would spike – Would my work be ‘good enough? Or would I be wasting her time? I constantly berated myself for not working harder, smarter, longer and better. I shone a harsh spotlight on myself and time after time, found myself to be lacking. Again and again, I found myself questioning my ability to undertake research at this level – after all, I reasoned, my interest is in teaching and not research; maybe good teachers could not be good researchers as well. Matters finally came to a head, when, for the umpteenth time, my supervisor insisted that I present ‘something’ at the School’s quarterly research weekend. These weekends were ostensibly designed to bring postgraduates together to share their research, which in theory I did not have a problem with. In practice, however, my research had little in common with that of my fellow postgraduates, and I had come to resent being forced to attend these sessions. I was uninterested in their research, and held no expectation that they would be interested in mine. In my mind, they were certainly not the ‘community of practice’ that my supervisor opined them to be. Presenting work at these weekends became a resented chore. I reached my breaking point in March 2009 when my supervisor instructed me to prepare a presentation for an upcoming research weekend; when I resisted, she looked me in the eyes and said bluntly ‘Either you present something, or I won’t supervise you any more!’ I went back to my office, shocked at the ultimatum. I remember sitting zombie-like at my desk, unable to think, unable to speak, unable to even begin to comprehend the enormity of what had just transpired. I started mentally playing around with the idea of breaking away from my supervisor, and finding someone else to work with. Perhaps not surprisingly, this possibility took root very quickly. At a meta-level, I marveled at how such a (misleadingly) simple solution could make me feel so empowered. My session with a psychologist served to confirm that I needed to get out of what had effectively become a toxic relationship. I subsequently set up an appointment with the assistant Dean of the faculty and my supervisor to discuss a parting of ways. Both were dismayed when I informed them of the purpose of the meeting, and both tried to sway my decision through a raft of backwards and forwards emails. I stuck to my guns – probably the first time I had done this in my entire 5 postgraduate career. Ultimately, the assistant Dean suggested that I take a few days to think about things before I took a final decision – she committed herself to honouring whatever choice I subsequently made. She was as good as her word, and a week later my supervisory relationship with the professor was officially terminated. There are times in one’s life when one knows that a decision is the right one, no matter how hard it was to make. This was one such occasion. Two new professors came forward, expressing their interest in co-supervising me. Both were (and remain!) high profile, highly regarded and most of all, welcoming and accessible. They immediately disabused me of any lingering notions of unworthiness, and set about getting my doctorate and me back on track. That was just over a year ago; now, as I sit at my desk, typing furiously, I marvel again – this time at something positive: how confidently I am writing, how much I have written, how much I look forward to supervision sessions, and how highly I value the feedback that I receive. For the first time, I feel like a ‘real’ research student, not just a cardboard cutout of one. One final reflection (the irony of which is not lost on me): a few months after I started working with my first PhD supervisor, she asked me to write a narrative of my Master’s supervision experience for her. Once she had read it, she labeled me ‘damaged goods’ and seemed to take her supervision cues from there. The narrative that I wrote for her constitutes the first half of this prologue. Who would have thought that I would not only use it, but add to it as I have in the prologue to the degree I thought I might never get? 6 CHAPTER 1 INTRODUCTION TO THE WORK Chapter overview An introduction to is essential to both the contextualization of the work and the presentation of a roadmap for the following chapters. This introductory chapter fulfils this task: it begins with a brief and general contextualization of the current state of postgraduate supervision studies. From there, it explores the institutional, national and international perspectives in which the research is located with a view to highlighting the fact that postgraduate research and supervision does not occur in a vacuum, but rather within a broader system of competing rules and obligations. The chapter concludes with an itinerary of the remainder of the thesis. 1.1 Introduction Undertaking doctoral study represents an exercise in focus and commitment that many candidates have never experienced before. It is also a time of uncertainty during which candidates may question their ability to undertake research at this level. For instance, we are told that the thesis is not our magnum opus, and yet it often feels like it is because it seems to be colossal when we are in the midst of working on it. We are bombarded with demands for originality of research, but live in fear that our work won’t be original enough. We read the theses of those who went before us, and think ‘Argh! My thesis will never be this good!’ And yet, every year, several hundred doctoral candidates are capped at university graduation ceremonies around the country, implying that there must come a time in the lives of all PhD students when their original, non-magnum opii are deemed to be both ‘good’ and ‘enough’ for the award of a doctorate. The paths that they have traveled to reach this destination may have differed from those of their fellow doctoral travellers, but the destination itself – the doctoral degree – remains constant. From a purely technical point of view, the path to success (if that is how one perceives the award of the degree) has been well-documented in the literature of postgraduate studies (Booth, Colomb & William 1995; Craswell 1996; Delamont, Atkinson & Parry 1997; Evans and Green 1995; Grant 1999, 2000 and 2005; Gullan 1996; Hart 2001; James & Baldwin 1999; Knowles 2003; Leder 1995; Phillips & Pugh 1994; Shannon 1995; Singh & Martijntje 2003; Thomson & Walker 2010; Vartuli (ed) 1982; Walker & Thomson 2010; Wisker 2001 and 2005; Yeatman 1995; Zhao 2001). These texts offer 7 thorough – although often quite technicist - guidelines for supervisors and students alike, setting out what processes should be followed, from conceptualization through to the writing up stages, as well as how the relationship between supervisor and postgraduate student should be managed. They set out a linear rendering of what both the supervisor and the postgraduate student have to bring to the process in order to ensure that it is a successful one. By ‘linear’, I am referring to the understanding that the process of knowledge creation and development always takes place in a sequential fashion, and never loops back on itself when obstacles or digressions occur. In my experience, postgraduate research is inherently recursive in nature; that is, each step – from the development of the proposal, to the collection of the data, to its analysis and recording in a thesis format – does not necessarily occur in a particular, unchanging order. Rather, it tends to twist and circle back on itself in often-unanticipated ways, depending on the nature of the data and the direction in which its analysis leads the researcher. So, for instance, the data collection stage may not adhere strictly to plan – the candidate may need to go back and re-collect information if the initial data set was tainted, or re-interview research participants because their responses were found to be incomplete. Candidates may also feel the need to collect follow-up data when the initial analysis throws out more questions whose answers will lead to a more thorough piece of research (or re-search – the process of rooting around for more data, information, answers). These loops are an unavoidable – and arguably invaluable - part of the process. Despite the current abundance of such guides – and many more are published every year - the literature that describes the pedagogies involved in postgraduate supervision is less than prolific. If anything, there is a conspicuous gap in the literature regarding the theorization of supervision pedagogy, possibly because the focus of the international knowledge economy (discussed in more depth later in this chapter) has only relatively recently turned to the financial cache that derives from original research (often in the form of doctoral studies). While I do not propose to close the gap in the theory by developing one of my own based on the research that I will present in this thesis, I will demonstrate that there is much work to be done in the field of postgraduate pedagogy if we are to understand the variety of philosophies, assumptions 8 and practices that underpin it. Put another way, it is important to establish what theories exist in the field and which ones can be harnessed effectively in order to understand the process better. If in the process, these theories are developed beyond their current understandings, so much the better for the practice of postgraduate supervision. Doctoral candidates experience the supervision of their PhDs in different ways, some positive and others not. Although much has been written on postgraduate supervision practices in general, little exists about the specific perceptions that doctoral candidates have of the process, and of the supervision practices employed by their supervisors. The link between the intention of supervisory practices on the one hand, and the experience of receiving supervision (‘being supervised’) on the other, is vital if we are to understand what motivates candidates to undertake doctoral research in the first place, what keeps them stimulated (and hence registered), and finally, what guides them through to the final stage of thesis submission. In a global context dominated by the knowledge economy (see section 1.4.3 below for more), higher education institutions are feeling increasingly pressured to contribute to and advance new knowledge through research and development – work that is usually carried out by doctoral candidates. As a result, these institutions need to recognize the factors that promote the retention of doctoral researchers so that they can remain competitive. However, to date, insufficient work has been undertaken to achieve such an understanding, and it is this knowledge gap that I hope to address (albeit only partially) in this thesis. Until recently, the role of the supervisor and the role of the doctoral candidate have been positioned in separate academic boxes and accordingly discussed as two issues that run parallel, rather than in concert with one another. At best, when the interaction between the two was recognized, it was usually in the context of uneven power relations, misaligned expectations and general academic disharmony. Certainly, during the course of my experience as a research student – in both my Master’s degree and with my first PhD supervisor – all three of these interactions were the norm. Thus, the purpose of this research is to explore the supervision experiences of doctoral students as they navigate their way through the research process. It intends to research 9 how PhD students experience their supervision and how the practices of their supervisors impacted on the success or failure of their candidature. Its contribution to current research in the field of postgraduate supervision therefore lies in the ‘voice’ that it gives to doctoral candidate. In doing this, I hope to foreground the kinds of practices that motivate and support doctoral candidates, and those that undermine them. What remains in this chapter is for me to set the scene – institutional, national and international – in which my research is located. Even though most postgraduate students are probably unaware of it, all three contexts inevitably impact on their research experiences in some way or the other. The final section of the chapter will present a preview of the thesis’ organization. 1.2 Research aims and objectives This research sought to investigate how doctoral candidates in the Faculties of Humanities and Science at the University of the Witwatersrand experienced their research supervision. In order to do this, I had to determine what kinds of supervisory practices were being undertaken in these two Faculties, and subsequently, how the doctoral candidates understood and responded to them. A secondary objective was to elicit the practices which are more likely to lead to positive supervisory relationships, and by implication, to graduation rates that are consistent with national imperatives. 1.3 Research questions The questions for which I sought answers were: • What was the student experience of the doctoral supervision process? How did it unfold for different students in different faculties? How did it unfold for different students who had the same supervisor? • What were the factors that both (or either) inhibited and promoted progress? How did the students respond to these? • What kinds of supervision practices are being undertaken at the University? How are these experienced by the candidates? Included in this question are the following sub-questions: Are supervision meetings formal or informal 10 affairs? Are they scheduled regularly, or are they only called when the student feels there is something to discuss? What kinds of feedback characterise the supervisor’s practice • How did the students experience power relations between themselves and their supervisors? • Was the supervision focused only on the short-term goal of graduation, or was it geared towards developing the students to continue to undertake research beyond the attainment of the degree? From these questions, I wanted to determine how doctoral candidates experienced their supervision, and how this, in turn, impacted on their development as researchers. This data is significant if institutions are to produce knowledge workers/researchers who can play a significant role in South Africa’s contribution to the international knowledge economy (see section 2.3). 1.4 Contextualising the research In this introductory chapter, I will provide a ‘road map’ of the route along which my thesis will travel. In the Prologue, I contextualized how I came to undertake this particular area of research. What follows is a broad overview of the institutional, national and international frameworks in which doctoral research is located. I will present each of these in separate sections below, although I would caution the reader against trying to understand each in isolation from the other – although they exist at separate ‘levels’, there are clear overlaps between them which need to be acknowledged. The first framework that I will examine will be that of the institution – the University of the Witwatersrand (Wits). This is both my alma mater and my current employer, and the doctoral candidates that I interviewed for this research are registered here, or are alumni. The university does not exist in a vacuum, though. While it does possess a certain degree of freedom from government intervention, it is nonetheless bound by government statutes designed and implemented by the national Department of Education (DoE) and, since 2009, the Department of Higher Education and Training 11 (DoHET). It is through the statutory bodies established by DoE – namely the Council on Higher Education (CHE) and its only permanent committee, the Higher Education Quality Committee (HEQC) - that Wits is held accountable to government. I will provide a brief description of these bodies, with specific emphasis on the CHE, since it is from here that much of the direction for postgraduate supervision comes from at the national level. Finally, I will take a further step back to look at the international framework in which higher education as a sector is situated. I will place particular emphasis on the role of the international knowledge economy in defining where higher education is located at the global level. I will argue that it is this knowledge economy that seems to be the driver behind what Neuman (2002: 168) describes as the ‘commodification of knowledge’, with a knock-on effect for universities around the world. Essentially, what I aim to demonstrate in this section is how each level has a profound impact on the other two, and why each one needs to be taken into account when researching postgraduate supervision practices. 1.4.1 The Institution – Wits The University of the Witwatersrand is located in Johannesburg, in the heart of South Africa’s economic hub, Gauteng. It is an urban university, spanning five campuses across Braamfontein and Parktown. However, its historical origins are somewhat more humble than this: The origins of Wits lie in the South African School of Mines, which was established in Kimberley in 1896 and transferred to Johannesburg as the Transvaal Technical Institute in 1904, becoming the Transvaal University College in 1906 and renamed the South African School of Mines and Technology four years later. Full university status was granted in 1922, incorporating the College as the University of the Witwatersrand, with effect from 1st March (History of Wits, http://web.wits.ac.za/AboutWits/IntroducingWits/HistoryofWits.htm). From these relatively modest beginnings, Wits has developed into one of the country’s leading tertiary education institutions, making significant contributions to South Africa’s research output. For instance, at the time of writing (in mid 2010), the university was celebrating the discovery of a new species of hominid – thought to be 12 over two million years old – by a team of Wits scientists. The contribution that this find will make to understanding human evolution is immeasurable. In terms of numbers, Wits’ student population is relatively representative of South African society. According to statistics released by its Strategic Planning Unit (SPU) in early 2010, the university has 28, 442 students registered across all five of its Faculties1. Of this figure, 28% are postgraduates – a number that Wits hopes to grow in the next few years in response to ever-increasing national and international demands for greater research production (see 1.4.3 of this chapter). This will suppo rt it in its efforts to increase the quantity of its research output, but not necessarily its quality. In terms of two other vital indicators of Wits’ representivity – race and gender - the university is making significant strides: just over half (54%) of its student body is made up of women, while approximately 56% are black2. These figures tend to mask the real significance of the metamorphosis of this historically white university (HWU) into one that is far more representative of the country’s demographics than ever before. However, it is not similarly mirrored in Wits’ staff complement: in 2009, only 20.8% of its permanent academic staff was black3 Over and above its responsiveness to national needs, Wits is also the South African university of choice for students from numerous other countries on the African continent and beyond: approximately 7% of its current student body is made up of international students (University of the Witwatersrand, Facts and Figures 2010: 5). What these figures fail to illustrate are the diverse elements of language, culture, religion and sexual orientation that pervade every aspect of the university’s day-to-day existence. Seen in the context of an undergraduate lecture theatre (of 200 – 300 students), this diversity is not always immediately obvious to the observer; by contrast, and 47% female (University of the Witwatersrand, Facts and Figures 2010: 9). 1 The five Faculties are: Humanities; Science; Engineering and the Built Environment; Commerce, Law and Management; and Health Sciences. 2 For the purpose of this work, the classification ‘black’ excludes Coloured and Asian students, who constitute approximately 3.5% and 14% of the total student population respectively (Facts and Figures 2010: 5). 3 As compared to whites who make up 67.8% of the staff. Coloured staff make up 2.6%, and Indians 8.5% (Facts and Figures 2010: 9) 13 when the focus shifts to the more individual postgraduate student-supervisor relationship, this diversity tends to be more noticeable. The diverse nature of the university’s student numbers is not a recent phenomenon – although constructed as an historically white university by successive apartheid governments, Wits has historically had a liberal ethos when it comes to the composition of the student body, which its website describes as follows: From the outset, Wits was founded as an open university with a policy of non- discrimination – on racial or any other grounds. This commitment faced its ultimate test when the apartheid government passed the Extension of the University Education Act in 1959, thereby enforcing university apartheid. The Wits community protested strongly and continued to maintain a firm, consistent and vigourous stand against apartheid, not only in education, but in all it’s manifestations. These protests were sustained as more and more civil liberties were withdrawn and peaceful opposition to apartheid was suppressed. The consequences for the University were severe – banning, deportation and detention of staff and student, as well as invasions of the campus by riot police to disrupt peaceful protest meetings (History of Wits, retrieved 10 June 2010, from http://web.wits.ac.za/PlacesOfInterest/Archives/UniversityArchives/History.ht m) Since the dismantling of institutionalized apartheid in South Africa, and the subsequent restructuring of the higher education sector4 Since the stated focus of this thesis is postgraduate – specifically doctoral – education, I now turn my attention to this aspect of the university’s function. In the latest issue of Wits’ postgraduate prospectus, the Vice-Chancellor and Principal, Professor Loyiso , new challenges have emerged for Wits, such as a rapid growth in student numbers as well as increased government demands for transformation (see section 1.4.2 below) of both the student and employee profile. This does not necessarily have any impact on the supervision process – although it could change as increased undergraduate numbers develop into a larger postgraduate cohort – but it certainly foregrounds them more powerfully. 4 The 2001 National Plan for Higher Education called for the restructuring of the number of HEIs in South Africa from 36 to 21.This restructuring, observe Hall, Symes and Luescher (2004), was politically-motivated: ‘Mergers and incorporations have been prescribed by the state as part of an explicit agenda of transformation, redress and equity in the sector. Restructuring has as its goal the dismantling of the apartheid landscape of higher education and the configuration of a new landscape which will allow higher education to achieve the goals set for it in national policy’ (2004: 11). 14 Nongxa, offers potential students the following insight into the learning context in which postgraduate study at the university takes place: Currently, Wits is home to 135 What Professor Nongxa is arguing is that Wits offers its postgraduate cohorts the benefit of extensive expertise and infrastructure in the pursuit of research excellence. This assertion is backed up by the fact that Wits is one of only two African universities ranked as ‘… a leading institution in the world and is the only university in the country that features in the top 1% in the world in seven defined areas of research according to the 2007 ISI National Research Foundation (NRF) A-rated researchers, 13 Chairs funded by the South African Research Chairs Initiative (SARChI), eight Medical Research Council (MRC) Units, three Centres of Excellence, three new state –of-the-art electron microscopes and two new research thrusts, bringing the total number of institutional research thrusts to nine.…Another measure of research excellence is that we have 172 NRF-rated researchers and have maintained that level for three consecutive years (University of the Witwatersrand, Postgraduate Prospectus 2010: 6). 6 international rankings’ (University of the Witwatersrand, Postgraduate Prospectus 2010: 4). It is against this backdrop that the students from my study undertook their doctoral research. As the discussion chapters will illustrate, each one of them experienced the aforementioned features quite differently, suggesting that there are patches of distinction, rather than wholesale excellence when it comes to research expertise and infrastructure, as well as postgraduate supervision. 5 Since the publication of this Prospectus, one more staff member has been given an A rating, thereby increasing the final number to fourteen. 6 The Institute for Scientific Information (ISI) is an international citation indexing system. According to Wits librarian Maryna van den Heever, ‘Journals indexed on ISI are evaluated against strict criteria and are also known as the ISI Master list which forms part of the accredited journals. These journals are also on ISI journal citation reports with impact factors to rate and rank the journals’ (personal communication, 7 September 2010). Furthermore, the ISI is used by bodies such as ARWU (Academic Ranking of World Universities) and THE (The Times Higher Education) to rank the top 100 universities. 15 1.4.2 The National Context – the CHE and the HEQC The national landscape in which the University of the Witwatersrand is located is a complex one. South Africa has 23 public universities and universities of technology (formerly known as technikons) which fall under the auspices of the Council on Higher Education (CHE). The CHE is a statutory body, established in terms of the Higher Education Act of 1997, and is …responsible for advising the Minister of Higher Education and Training on all higher education policy issues, and for quality assurance in higher education and training (www.che.ac.za, accessed on 26 September 2009). The policies alluded to deal not only with pedagogic and governance issues in the South African higher education sector, but also with the twin goals of improved access and transformation – both of which are of historical and political importance to the country, given the skewed provision of education at all levels that characterized the previous regime. In South Africa, the terms ‘access’ and ‘transformation’ have deep historical and political connotations stemming from its Apartheid education system past (Christie 1991; Cloete et al 2002; Jansen 2009; Soudien 2008). At a recent Higher Education Summit on Transformation held in Cape Town, the country’s deputy- president, Kgalema Motlanthe, defined the term ‘transformation’ as follows: ‘Broadly speaking, … we mean a process that addresses the accumulated disabilities in our society ranging from cultural, economic, racial and gender exclusions as they manifest themselves in numerous institutions and organizations’ (2010: 4). At the same Summit, Minister of Higher Education and Training, Dr Blade Nzimande, reframed this concept as a process of contestation rather than of negotiation: The term ‘transformation’ is sometimes used rather narrowly almost as if it were synonymous with black economic empowerment. More broadly, it is also used to refer to the process of overcoming racial divisions. This is more accurate and acceptable. However, I believe that transformation should be understood to be about more than eradicating the purely racial aspects of apartheid, as important as this is. It is essentially about radically changing our society, including our education and training system and all other areas of life to ensure that they can serve the interests of all South Africans in a democratic, equitable and prosperous society. Put differently it is about confronting the 16 deeply interrelated challenges of class, race and gender inequalities, including confronting the HIV and AIDS pandemic and being an inclusive society for the disabled. This means ensuring that the working class and the poor, women, youth and the disabled, become significant beneficiaries not only economically and politically, but also in terms of cultural and educational development. All of us here have a duty to ensure that the higher education system serves this purpose (2010: 1). By the same token, ‘access’ has taken on a political meaning in the South African context: essentially, it refers to the number of students who are able to gain entry (that is, access) into tertiary institutions on financial (not academic) grounds. Put another way: while students may achieve the required university entrance results, many are not able to take up their places at these institutions because they do not have the financial means to do so. Students who find themselves in this situation are more likely to come from black African backgrounds than from white – an Apartheid legacy that has not yet been corrected. Akoojee and Nkomo (2007) suggest that this access should be the ‘… key mechanism by which to forge a new order’ (2007: 392), although this has yet to happen7 7 The South African government has sought to deal with this obstacle by setting up a national bursary fund aimed at opening access to tertiary study to students from disadvantaged backgrounds. In a CHE publication Higher Education Monitor: Access and Throughput in South African Higher Education – Three case studies (2010), the authors describe the role of this national bursary fund (NSFAS) as follows: ‘The NSFAS was set up in 1999 with the aim of increasing access to education and training. It provides students who qualify with loans of which up to 40% can be converted to a bursary, depending on the student's academic results. Students repay loans once they are employed and earning above a threshold amount. In 2003 the NSFAS disbursed R304 million to 27,783 students at HAUs [Historically Advantaged Universities] and R219 million to 32,069 students at HDUs [Historically Disadvantaged Universities]. The NSFAS has facilitated access for a significant number of students who would otherwise not attend higher education and has become a significant source of income for HDUs, improving cash flow and reducing student debt. And the NSFAS awards to students at HAUs have been increasing, indicating that more students from lower income groups are enrolling at these institutions, in line with the goal of increasing access’ (2010: 5) . In addition, for those who are able to overcome this financial barrier, there is another obstruction to their success: epistemological access. Morrow (2009) offers the following explanation of the term: 17 Epistemological access is not a product that could be bought or sold, given to someone or stolen; nor is it some kind of natural growth, such as the growth of plants or bodies. Epistemological access cannot be supplied or ‘delivered’ or ‘done’ to the learner, nor can it be ‘automatically’ transmitted to those who pay their fees, or even to those who also collect the handouts and attend classes regularly. The reason for this is that epistemological access is learning how to become a successful participant in an academic practice. In the same way in which no one else can do my running for me, no one else can do my learning for me (Morrow 2009: 78. Italics added) Given the fact that South African universities were largely developed along the lines of their Western counterparts, it seems logical (although still unacceptable) that only those with ‘Western’ or colonial backgrounds would be able to access the ‘academic practices’ to which Morrow refers. Discriminatory apartheid policies that denied black students access to quality education served to compound this problem (Jansen 2009). As a result, African students have been effectively excluded from the kinds of learning that are valued by the academy. The inevitable result is a form of intellectual elitism that obstructs the learning of many students of African descent. The dilemma that this poses for South African higher education is profound: the massification of higher education, on the one hand, in which HEIs are being pressurized to register ever- increasing numbers of students; and the lack of epistemological access to the academy that a large number of these students possess, on the other. Thus, when faced with the double difficulties of the denial of financial and epistemological access, it is perhaps no wonder that black student numbers (as a percentage of enrolments and graduates, and relative to their share of the national demographic) at HEIs are not as high as the government would like them to be8 8 According to Statistics South Africa, black Africans constituted 79.3% of the population of this country in 2009 ( . This has an obvious knock-on effect for postgraduate enrolments among black students, particularly at the Masters and Doctoral level. According to figures released by the CHE, for instance, in 2005 only 29% of doctoral candidates who graduated were black, compared to 59% white, 6% coloured and 7% Indian (2009: xxi). What these statistics point to is an education system that www.statssa.gov.za/publications/P0302/P03022009.pdf, accessed 28 June 2010). This mismatch between the demographic representation of black people and their participation rates in higher education is understandably a concern for the South African government because it seems to imply that the education system has not yet ‘caught up’ with the changes in the political system – sixteen years after the advent of democracy. 18 remains top heavy in terms of white enrolment and graduation – at least partially a consequence of the aforementioned lack of financial and epistemological access that black students continue to experience. How this situation can be altered is the subject of another thesis altogether. The connection between a strong higher education sector and a functioning democracy is fundamental to understanding the work of the CHE. As Saci Macozoma, the chairperson of the CHE remarked in 2004: High quality higher education is crucial for social equity, economic and social development and a vibrant democracy and civil society. If higher education does not produce knowledgeable, competent and skilled graduates, generating research and knowledge, and undertaking responsive community service, then equity, development and democracy will all be constrained. The challenges of reconstruction, social transformation and development are tremendous. Higher education must not fail to establish the new priorities and satisfy the new needs of a democratic South Africa. Hence, one of the principle roles of the CHE is to actively promote quality and participation in the sector as a means of consolidating democracy and the values that underpin it. However, it is important to recognize that the Council on Higher Education and its activities are merely a government intervention to provide guidelines for good practice – it is ultimately the institutions that enact these. While none of the CHE policies focus specifically on postgraduate education (yet), they nonetheless seem to implicitly recognize the centrality of a healthy postgraduate research community in higher education institutions for the purpose of advancing both democracy and South Africa’s standing in the international knowledge economy (discussed in more depth in the following section)9 9 For instance, Habib and Morrow have argued that ‘… South Africa’s share of global research output has been declining for over a decade, from 0.8% in 1990 to 0.5% in 2001’ (2006: 9). In addition, they point out that this country’s scientific output – measured in terms of publications from its public research sector – has been at a virtual standstill for at least fifteen years (2006: 9). The logical conclusion to be drawn from these statistics is that South Africa is becoming less competitive in terms of its contribution to international research and development innovations. This, in turn, means that the country will attract less funding for any potential research avenues that its HEIs may wish to pursue, causing it to fall even further behind. Unless there is a concerted effort on the part of all stakeholders to arrest this decline, the downward spiral has the potential to undermine all political, economic and intellectual gains that South Africa has achieved since the advent of democracy. . 19 The CHE has one permanent committee, the Higher Education Quality Committee (HEQC), whose principle mandate is to promote quality assurance at all HEIs in the country. Two tools –quality audits and accreditation – are used to these ends. Both take a global view of the various programmes and courses offered by HEIs, but at the same time, recognize that postgraduate education and research are vital contributors to the missions of HEIs. This is evident from the content of the various documents that the CHE has published on quality audits and accreditation in South African higher education. For instance, Criterion 15 of the HEQC’s Criteria for Institutional Audits document, states: Effective arrangements are in place for the quality assurance, development and monitoring of research functions and postgraduate education. In order to meet this criterion, the following are examples of what would be expected: … (iv) Appropriate strategies for the support and development of postgraduate education, including effective postgraduate supervision, which are implemented and monitored. (v) Regular review of the effectiveness of arrangements for the quality, development and monitoring of research functions and postgraduate education (2004: 16) In the Institutional Audit Manual, published in 2007, two additional criteria are added to this. In essence, these deal with research participation, productivity and resources (Criterion 16), and efficiency in the linked processes of quality assurance, development and monitoring of postgraduate education (Criterion 17). Point (iv) of the latter criterion explicitly sets out the institutional responsibilities in respect of their postgraduate students as follows: (iv) Clear and effective policies and strategies which facilitate the development, support and improvement of postgraduate education; these include the availability of – • training and development opportunities for new supervisors • research design and methods courses for postgraduate students • access to support services for postgraduate students 20 • facilitation of regular access to supervisors and other researchers in the field • special funds to support postgraduate research • additional support and development programmes for previously disadvantaged students (2007: 24) These audit criteria point to a recognition on the part of the CHE (and by implication, of the South African government) that the development of quality postgraduate education through multiple, specific interventions such as those listed above is a priority. The criteria are complemented by Criterion 9 of another CHE document, the Criteria for Programme Accreditation, the aim of which is to provide universities with descriptions of the minimum standards that are required for academic programmes to be recognized by this body. This criterion reads: Postgraduate programmes have appropriate policies, procedures and regulations for the admission and selection of students, the selection and appointment of supervisors, and the definition of the roles and responsibilities of supervisors and students etc. (2004: 14). At minimum, universities need to meet the following requirements in order to have their postgraduate programmes accredited by the CHE: (i) Appropriate policies, procedures and regulations are in place for student admission, selection and assessment. These are communicated to all postgraduate students, and academic and administrative staff, and implemented consistently across the institution and programme. (ii) The selection and appointment criteria in place for postgraduate supervisors are acceptable to the research community in the area of study. These include the following: • The supervisor has a qualification in a relevant field of study higher than, or at least at the same level as, the exit level of the postgraduate programme he/she is supervising. • The supervisor has an appropriate research track record, as well as experience, expertise and peer recognition in the field of study. • In the case of inexperienced or new supervisors, there is ongoing staff development and support, and joint supervision is explored as an option. 21 (iii) Explicit guidelines exist on the roles and responsibilities of supervisors and students and other matters relevant to the performance of research. These include the following: • The nature, format and expected turnaround time for work submitted to the supervisor. • Forms of assessment, and the communication of feedback to the student, which includes:  The periodicity of contact between student and supervisor, and the schedule for the submission of progress reports and written work.  Research ethics, code of conduct, regulations on plagiarism and intellectual property rights.  Examination and qualification requirements. (2004:14) The point of including such detailed information in the introduction to my thesis is to foreground the initiatives that South Africa’s national higher education quality assurance body has taken to improve the postgraduate education capacity at HEIs. Although the listed criteria do not exist outside the general tertiary education sector publications and documents of the CHE – that is, in a stand-alone volume of their own – the fact that they exist at all, and in the detail that they do, is significant. It is evidence of a national commitment to growing a postgraduate community capable of contributing to the country’s long-term development. In addition, it represents recognition that postgraduate education makes significantly different demands on both students and academics than undergraduate education, and hence requires specialized support from national government structures such as the CHE. At this point, it would be useful to describe how Wits measures up to the CHE- mandated responsibilities. At best, the university’s response to Criterion 15 and 16 is perfunctory and superficial: while it’s Centre for Learning, Teaching and Development (CLTD) does offer workshops for both novice and experienced supervisors, there has been no systematic follow-up by the facilitators to ensure that what has been learned has been implemented. However, a questionnaire designed to elicit this information was recently administered; its results are currently being processed. From the perspective of the postgraduate student, numerous research design and methods courses 22 are offered by various faculties and the Postgraduate Project Office (PPO)10 The knowledge economy consists of a mix of commercial production and knowledge goods that are freely created, disseminated and exchanged. In their form as ideas and know-how and as first creations of works of art; that is, as original goods; knowledge goods have little mass and their production is annually. However, these tend to cater for full-time students who are able to attend the lunchtime and early afternoon slots; part-time students who are employed elsewhere and who are unable to get time off are disadvantaged by the time slots. Furthermore, access to support services and special funds (Criterion 17) is patchy, at best. For instance, one of the science students that I interviewed described how the university’s Research Office – to whom he had applied for conference funding – kept him waiting for months before telling him that no funds were available for his request. He felt that the office could have informed him earlier so that he could have sourced funds from other entities. He missed the conference. At the end of 2009, a Postgraduate Office was set up within the university’s Student Enrolment Centre (SEnC) as a ‘one-stop shop’ to deal with postgraduate registrations. It is too early to judge whether this has improved the registration process for staff and students. Numerous resources have also been made available to postgraduate students, including access to accommodation, health services, sport and recreation and career counseling (all of which are also available to undergraduate students). A Research Commons has been established in one of the University’s main libraries, combining library services and computer technology to support Masters and PhD students, as well as academic staff and researchers, in their research. While these services and responses appear to be beneficial at face value, they are patchy. In other words, they are not uniformly applied across Faculties, Schools and departments. Therefore, they are arguably less effective than was intended. 1.4.3 Going global – the international knowledge economy 10 The PPO was established in 2006 as a ‘…a cross-faculty, cross-discipline graduate centre that provides a home for the intellectual and social life of postgraduate researchers. It also acts as a facilitator to increase awareness of postgraduate needs on campus’ (retrieved 10 June 2010, from http://web.wits.ac.za/Prospective/Postgraduate/StudentResources.htm ). Its future is uncertain, though, as the two senior professors who managed it on secondment have returned to their full-time academic positions, and their replacement was employed on a contractual basis. 23 sustainable, requiring little or no industrial energy. It rests on donated human energy and time. Subsequently these knowledge goods can be copied, mostly with minimal resources, energy and time (Marginson 2008: 1) The emergence of the international knowledge economy has fundamentally altered the global higher education landscape, re-directing its focus away from ‘knowledge for knowledge’s sake’ and towards market demand for tangible (not to mention economically viable) outputs, as is evident from the opening quote. Hence, Usher (2002) has described the phenomenon as one that ‘…replaces an epistemological with an economic definition of knowledge’ (p. 144). He continues: Knowledge becomes a factor of production, more critical in the production process as economic performance comes to rely more and more heavily on knowledge inputs. At the same time, the knowledge economy is one where knowledge is not only a key input but also an increasingly significant output that can be grown in an unlimited way… Economic growth is now seen to be vitally dependent on the development of an infrastructure that facilitates and enables sustainable knowledge development. With universities therefore now increasingly seen as significant knowledge producers, they take on a hitherto unrecognized role as agents of economic growth (Usher 2002: 144). What Usher seems to be arguing is that since universities are one of the key producers of knowledge, it follows that these institutions will be (and indeed, are) increasingly called upon to produce the kinds of knowledge and research that have a measurable market value. He concludes that these ‘corporate conceptions’ of the role of the university conflict with the academic conception of the university. This is most evident in the context of doctoral study: The university is conceived as a community of scholars, discipline-based and autonomous. A doctorate … doesn’t figure in this [corporate] conception because such an activity would be considered neither educational nor doctoral. Equally, the notion that universities existed to service the knowledge economy would be rejected on the grounds that the university’s main function is cultural and social (Usher 2002: 152). This contradiction points to a tug-of-war, in which the two opposing communities – academics in one corner and corporations/markets in the other – are battling to assert dominance over the knowledge issuing from the higher education sector. As Usher 24 indicates, directing doctoral research towards the needs of the market has little educational value for students (unless they have a financial stake in the results of the research output). I tend to agree with this view because in scenarios where research topics are defined by markets rather than academics, there may not be any inherent interest on the part of candidates to discover ‘the answer’. Put another way: doctoral candidates may not feel an innate desire to make discoveries because the research agenda has been imposed on them – they only undertake it because they can get funding for it. In such a scenario, the doctoral candidate becomes a drone, undertaking other people’s research – for their financial gain - and receiving little satisfaction (educational or otherwise) in return. While this may be a crude generalization, but it captures the crux of the tension between research for the sake of learning and self- development, on the one hand, and research for the financial gain, on the other. However, the role of HEIs in the international knowledge economy does not end at the production of new knowledge through research, argues Neumann (2002). On the contrary, universities are also being tasked with the development of ‘intellectual capital’ for this economy (p. 168); that is, graduates who are able to produce and advance such knowledge. She points out that More and more there is a need for people who can understand, interpret and undertake research outside the university context. Their knowledge and skill is recognized as forming a vital, if intangible, economic asset (Neumann 2002: 168). This has arguably been one of the principle drivers behind the vast massification of higher education that has occurred in the last decade or so11 11 According to Guri-Rosenblit et al (2007), ‘By 2000, the total enrolment in higher education institutions was about 100 million students, representing about 20% of the age cohort worldwide, whereas at the start of the twentieth century, only about 500,000 students were enrolled in higher education institutions over the globe’ (p. 374). Admittedly, these figures have probably not taken into account the emergence of universities in the former colonies in the post-colonial era. Nonetheless, they do offer a useful insight into the extent of the international tertiary education sector and by implication, into how market demand for skilled workers has impacted on universities worldwide. – the market needs workers with high-end tertiary qualifications to advance its economic agenda. The more advanced the level of the qualification (undergraduate, Honours, Masters or PhD), the more in demand the student/knowledge worker. Since the international knowledge economy appears to be organic - in that it is in a constant state of flux and change - I 25 anticipate that its demand for well educated human capital will continue indefinitely, which in turn will have long-term implications for the way in which universities function to meet this demand. Wright (2004) has highlighted the role of capitalist competition for the advancement of the knowledge economy in her historical analysis of the genesis of the term: It is perhaps in international arenas that the fiercest debates over the purpose and policies for higher education are being contested … Reich (1992), in particular, influenced [then US President Bill] Clinton and then [Prime Minister Tony] Blair with arguments that capitalism was reorganizing around an ‘international knowledge economy’ and that to compete in this economy, nations would need a workforce with high levels of tertiary education. This in turn has made higher education a potentially very lucrative facet of the knowledge economy (Wright 2004: 73). Even international finance organizations such as the World Bank have, in recent years, come to recognize the changing role of higher education in world markets. A 2002 Bank policy document stated that: Tertiary education institutions support knowledge-driven economic growth strategies and poverty reduction by (a) training a qualified and adaptable labour force… ; (b) generating new knowledge; and (c) building the capacity to access existing stores of global knowledge and to adapt that knowledge to local use. Tertiary education institutions are unique in their ability to integrate and create synergy among these three dimensions (World Bank 2002: xx, Italics added). Further, St George (2007), drawing on the work of educational theorist Riddell, points out that the international knowledge economy places heavy demands on the higher education sector because it virtually compels it to become a slave to international trade patterns (2007: 591). Put another way, markets seem to have become the ultimate deciders of what kinds of knowledge possess value (measured in financial terms), and hence what kinds of curricula universities ought to follow in order to meet these market needs. She also adds a caveat – relevant to the South African context – that ‘… developing countries [may] get left behind in a global market place, if adequate attention is not paid to developing countries’ ‘knowledge economy’’ (2007: 591). In other words, if the intellectual capital located in the developing world does not evolve at the same pace as (or in conjunction with) that of the developed world, there is the 26 very real danger that existing economic imbalances between developed North and the less-developed South will widen significantly, leading to even more inequality between the two. In this never-ending decline, the socio-economic and political realities of the latter will inevitably suffer. Marginson (2008) concurs with this sentiment: In the knowledge economy the patterns of knowledge production and the flows of knowledge are conditioned by and reproduce global relations of power in other domains such as the economic, technological and political (2008: 2) When one puts together these elements of the international knowledge economy, what emerges is a global higher education system is under siege by market forces to produce both knowledge and knowledge workers, the quality of which is invariably measured in financial terms. Gone are the days, it seems, of ‘blue sky’, curiosity-driven research in which knowledge was celebrated in and of itself. However, this state of affairs has not gone either unnoticed or unchallenged. In February 2009, for instance: Calling for a "modest revolt," 20 United Kingdom scientists, including one Nobel laureate and eight Royal Society fellows, have launched a scathing attack on the U.K.'s seven research councils for now requiring grant applications to include a 2-page statement on the economic impact of the proposed work. In a letter to the Times Higher Education (THE), they call for peer-reviewers to ignore those summaries, arguing that they have no business predicting what research may produce an economic windfall. They also blame this financial mindset and a lack of private industry investment for causing a decline in the U.K.'s leadership in science, as reflected in a decreasing frequency of Nobel Prizes (Travis 2009: 1). Even though this ‘modest revolt’ has been fomented by a small group of researchers – whose views can by no means be extrapolated to all academics everywhere – the backlash against knowledge-for-profit that it represents is nonetheless significant. Whether it spreads beyond the borders of Great Britain will ultimately depend on how far global markets are willing to push academics to produce ‘an economic windfall’ mentioned above, on the one hand, and whether or not (and how) academics will conform or resist this commodification of knowledge, on the other. Clearly, the impact of the international knowledge economy on the global higher education sector – particularly on the contribution of postgraduate research – has been universally felt: no state or university is (or indeed, can be) immune to ever-increasing 27 market demands for more knowledge and more knowledge workers. South Africa, a developing country, is no exception; in turn, the University of the Witwatersrand, one of this country’s leading research institutions, is also feeling the inexorable pull towards applied research that represents financial gain for it12 1.5 Some final directions . However, this is arguably not as mercenary as it sounds – universities are expensive entities to maintain, and to a large extent rely on the inflow of capital from research projects to ‘top up’ what they receive from national government in terms of subsidy. Striking a balance between undertaking research for research’s sake (or educational enquiry), on the one hand, and research for financial gain, on the other, is clearly a challenge that will determine the shape of higher education for decades to come. The chapter immediately following this one – the literature review - takes the reader through the existing scholarly work on postgraduate supervision, and includes a survey of the literature on pedagogy, power, identity and communities of practice. The review comes with a caveat, though: it is limited by the impossibility of a single researcher reading every piece written on each of the aforementioned topics and including them in the thesis. However, it offers the reader a ‘big picture’ of the field of postgraduate supervision, as well as some of its ‘feeder fields’ – that is, research areas whose paths may lead them into the study of supervision. Subsequently, the methodology chapter will unpack the qualitative data collection system that I chose to use for my research, namely the semi-structured interview. I felt that this method would offer me an opportunity to obtain rich narrative descriptions of the supervision experiences of PhD candidates. Ultimately, this technique proved to be 12 For instance, one of the 6 goals listed in its Strategic Research Plan: 2007 – 2011 is entitled ‘Challenging, relevant and innovative research’. Included in this goal is the following statement: ‘IP management tools will be enhanced and commercialization of marketable intellectual property will be encouraged’ (p.1). This is developed later in the document, when the need for a ‘…highly developed and university-wide systems to manage commercial exploitation including patents, diversity and quantum of funding grants, based in Wits Enterprise’ is listed, alongside a ‘…a well-developed culture of entrepreneurship amongst students in the most applied disciplines’ (p. 5) 28 very effective, not only in eliciting the narratives from the candidates, but also in allowing them the space to reflect on their experiences. Many of my interviewees provided additional reflections on their experiences after I had sent them copies of their transcribed interviews for confirmation. They were intrigued at how their experiences ‘looked’ in writing. Some were even shocked at how harshly they had described their supervisors. The common thread that ran through their narratives, though, was how difficult it was to articulate their experience of a form of pedagogy which for them was largely a ‘silent’, undiscussed one. The discussion and analysis chapters constitute the next section of the thesis. Here, the journey through the literature and methodology culminates in a critical review of the lived experiences of the thirty doctoral candidates who allowed me access to their stories of supervision – the good, the bad and the ugly. In these chapters, I attempt to foreground their voices because what they had to say – about their experiences, their feelings and their opinions of the process – was too valuable to subsume in abstract theoretical discussions. It is also my hope that anyone who reads this thesis will spend time reflecting on the voices and conversations captured in these two chapters, and pondering how they could best use them in their own practice as supervisors (or, indeed, as doctoral candidates). If this work triggers any (constructive) changes in the way that its readers approach their supervision activities, then I believe that it has done what it set out to do. What remains after this discussion is the conclusion, in which I will draw together the threads that have been interwoven throughout the thesis. It is my hope that the tapestry that emerges will not only illustrate the key arguments of the thesis, but also offer the kinds of conclusions that can be taken forward by future researchers in the field. 29 CHAPTER 2 LITERATURE REVIEW (Part 1): Plotting the parlance of the PhD: from pedagogy to practice Chapter overview No exploration of any research topic can begin before its key concepts have been both identified and clarified. This allows the reader to follow the author’s train of thought and to understand her point of departure for the remainder of the work. It is therefore the purpose of this chapter to unpack the thesis’ key concepts – the PhD, pedagogy, discourse, and communities of practice - and to examine how each is (re)presented in the literature in order to lay the foundation for the remainder of the research. 2.1 Introduction In this section of the thesis, I will present a detailed overview of the scholarship in the area of postgraduate supervision. The aim is to contextualize my study within the broader scope of the field of postgraduate studies because the experiences of the doctoral students that I interviewed for this research do not exist in a vacuum – they are a part of a wider institutional, national and global picture. There is a caveat, however, to this contextualisation: the literature on postgraduate supervision is a vast one, and it will not be possible to list and/or discuss all of it without losing sight of the research questions that I posed in the first chapter. Therefore, I have been selective about what I have included in this chapter, and in so doing, I acknowledge that I may have omitted certain texts and authors. I begin the review with a clarification of the terminology that will be used throughout this research. As an educator, I have learned over the years never to assume that everyone shares an understanding of commonly used terms, even when there seems to be a shared discourse. If anything, these assumptions are often at the root of confusion and thus I have decided to make explicit what I mean when I use these terms. Following on from this, I will survey the scholarship that has been undertaken on research supervision – specifically the introductory guides (for both supervisors and students) which have become prominent in the last decade or so (see, for instance, Bartlett & Mercer 2001; Cryer 1996; Delamont, Atkinson & Parry 1997; Dunleavy 2003; Graves & Verma 1997; Leonard 2001; Mouton 2006; Murray 2002; Phillips and 30 Pugh 1989 Remenyi & Money 2004; Walker & Thomson 2010; Wellington 2010; Wisker 2001 and 2005). One of the main aims of this section will be to foreground what these guides consider to be good supervisory practice. This segues neatly into a section dealing with both national and international guidelines for postgraduate supervision – that is, what national education bodies, both in South Africa and abroad, have identified as being core to successful supervision. It is interesting to note where the ‘how to guides’ and the national guidelines converge and where they diverge because it demonstrates an important tension between good teaching and the development of proficient researchers, on the one hand, and the imperatives of governments for the financial rewards associated with research output in an international knowledge economy, on the other. The review will also examine the assertion that research supervision is a particular form of pedagogy, or teaching. This assertion has implications for the process of supervision – and for supervisors - because it requires some reflection about how supervisors perceive their role vis-à-vis postgraduate students: are supervisors there to simply ‘get the student through’ regardless of whether or not key skills are imparted; or is it the role of the supervisor to teach, mentor, guide and/or develop the student into a skilled researcher? In the latter case, ideally, the student could continue to undertake research (and hence contribute to the knowledge economy) even when she has completed her doctoral studies. Finally, I will review a range of literature on power and identity, and connect these largely invisible concepts to the process of research supervision. I have deliberately used the term ‘invisible’ to describe them because it captures the idea that while one cannot always physically see them, their effects (on the student, the thesis and the supervision relationship) are unmistakable. Further, by virtue of the fact that they are intangible, they are often dismissed as either not existing, or of not having any significant impact on supervision. As the literature will show, both of these assumptions are problematic. 31 2.2 Concept clarification Many of the key concepts that will be used in this research have meanings that are contested – by this, I mean that there is not necessarily a commonly agreed-upon understanding of what each means. There is some value in having disparate understandings of the same term, since each one tends to foreground what individuals, institutions of higher learning, governments and other stakeholders hold to be critical to them. However, this kind of selective interpretation by the various stakeholders cultivates ambiguity, which is where misunderstandings occur. Therefore, I have provided a range of definitions from the literature in order to both highlight these contestations, as well as to demonstrate why many of the practices of doctoral supervision are fraught with unmet expectations, unspoken assumptions and uninterrogated understandings. 2.2.1 The doctorate/Doctor of Philosophy It is no longer just about producing an original piece of excellent research; producing a trained researcher is an equally important output – (Purpose of the doctorate 2005: 2). This quotation aptly captures the crux of what a doctorate means at its most fundamental level in the twenty-first century: not simply a piece of written work that it considered to be ‘worthy’ by one’s peers, but more importantly, an individual who has the capacity to undertake excellent and often ground-breaking research – one who has the skills and capacity to add to the existing knowledge base in her discipline (or in more than one, if the researcher has multi- or cross-disciplinary interests). In other words, it suggests a move away from an understanding of the doctorate as simply a text, to one that recognizes it as a more holistic, researcher-based concept. It stands in stark contrast to how the original doctorate was understood: the first doctorates were offered in medieval Europe as a ‘license to teach in universities’ (Park 2007: 4). According to Powell and Green (2007): Doctorates have been awarded in the UK since the thirteenth century – though titles and purposes have clearly changed much over his period (2007: 89). The doctorate as a research degree emerged in Germany at the start of the nineteenth century, where after it ‘spread’ throughout the rest of the European continent and 32 beyond, to countries such as the United States, Canada and Australia. The first doctorate in the United States was awarded in 1861 and by the turn of the century, approximately 3 500 more had been conferred. According to Walker et al (2008: 19), Now, as the twenty-first century begins, more than 40 000 doctoral degrees are awarded each year, adding to the more than 1.36 million doctorates granted by US universities during the twentieth century. In Australia, the PhD is a relatively new degree offering – ‘No Australian university had regulations for the award of the PhD until after 1945, and in the first postwar years the numbers of doctoral students were small’ (Nelson 1996: 2). This picture changed significantly in the last decade of the twentieth century, when, according to Ward and West (2008: 61): …[PhD] candidate numbers more than doubled between 1990 and 2000, when a total of 37,374 enrolled candidates were reported (Department of Education, Science and Training 2003). Although no detailed current figures are publicly available, the number of candidates can be reasonably estimated at around 40, 000. The outcome of the doctoral process in this country is a written thesis which has been ‘designed to test their [doctoral students’] capacity to carry out independent research’ (Nelson 1996: 3). In addition, doctoral students in Australia are expected to make a ‘significant contribution to research’ (Singh and Kulski 2003: 1), although it is not always clear how this significance is determined. In South Africa, the doctorate is also a relatively new feature of the higher education landscape. According to Herman, ‘The first PhDs were awarded by the UCGH (University of the Cape of Good Hope)13 …the degree in which you achieve depth in scholarship, and specialize in a certain area so that you are able to make a contribution to the existing body of knowledge. A Ph.D. or D.Litt. et Phil. signifies that you have produced new knowledge. You have progressed beyond the level of reproducing and at the beginning of the 20th century’ (Chaya Herman, personal correspondence, 10 June 2010). Despite its relative newness, however, the PhD shares many of the characteristics mentioned above. For instance, Mouton (2006: 5) describes it as: 13 The University of the Cape of Good Hope changed its name to the University of South Africa (UNISA) in 1916. 33 mastering existing knowledge (master’s) to the point where you have made a unique contribution to the scholarship in a particular domain. Participation in doctoral study in South Africa is worryingly low. A recent article in The Sunday Times pointed out that only 1 300 doctorates were awarded by South African universities in 2007, dropping to 1 182 in 2008 (Philp 2010:3). Of this figure ‘a third … were going to foreign students, mostly from elsewhere in Africa’ (Philp 2010: 8). This falls well short of the government target of producing 6 000 (predominantly South African) PhD graduates per year by 2018. Furthermore, latest statistical projections by Statistics South Africa (StatsSA) project that the country has an estimated current population of 49, 99 million people (retrieved from http://www.statssa.gov.za/publications/P0302/P03022010.pdf on 10 August 2010). Thus, as a percentage of the population, the number of doctorates being awarded annually in South Africa is a meager 0.002%. This figure highlights the ‘downward spiral’ of the country’s contribution to international research and development innovations referred to earlier in the chapter (cf. pages 18 and 27). But what is a doctorate? Park (2007: 3) articulates his understanding of the concept as … the highest academic degree that a university can award to a student who has successfully completed a defined programme of work in a particular field of study. However, even this is a vague definition, and one that is open to multiple interpretations. Matos (2006: 2) suggests that the PhD has ‘historically lacked a precise definition’ because there is no common understanding of what the core aims of this degree are (or should be). He cites numerous authors to support his assertion: Bent (1962) reflects over the meaning of the PhD and argues the case for the need of clear departmental and institutional definitions of the doctoral degree. This lack of definition may be due, he suggests, to the fact that no one really knows what a PhD is. He thus foresees the need for a new degree, a post-PhD, with clearly defined aims and that will symbolize ‘the highest values in scholarship and research (1962:17). Walters (1962) also alludes to the lack of a clear definition of the doctoral degree that institutions offer and administer. He defines this silence as an ‘amiable anarchy’. Bowen and Rudenstine (1992) confirm this view and go even further by stating that the PhD is the least controlled ‘activity’ within higher education institutions (Matos 2006:2). 34 What is evident from this is that the doctoral degree is an inexact and nebulous qualification, and has been for decades14 While laudable, neither deals with substantive issues such as how quality is understood by the institution (in real, not abstract, terms), what is considered to be an ‘original contribution’ to knowledge, or what the assessment criteria for a postgraduate dissertation or thesis are. Consequently – and in the absence of outstanding supervision – students often proceed blindly through their studies, knowing what their goal is (the . While this may allow doctoral candidates an unconstrained opportunity in which to investigate their chosen phenomena, with a positive knock-on for knowledge production, it also poses a threat to the integrity of the degree itself. This is because it opens it up to claims of unevenness of quality and rigour: what one institution may consider a scrupulously researched thesis may be considered to be the opposite by another, and vice versa. Thus, it is difficult to develop a standard set of benchmarks for what constitutes a PhD thesis when there is no consensus on what a PhD actually is. It is surprising that this glaring gap has existed for so many decades, with no documented attempts being made by government to develop a national understanding of the term – for the purpose of quality assuring doctoral candidates’ work, if nothing else. In spite of this, there have been some limited efforts by institutions (often at the level of individual faculties) to standardize expectations for both supervisors and doctoral candidates. For instance, at Wits, the Faculty of Humanities has produced an online document of guidelines which attempt to provide some direction to postgraduate students, and presumably to supervisors as well. Subtitled ‘All you need to know’, this extensive document covers a range of topics, including the study programmes that are available in the Faculty, the application and enrolment processes, fees tables, a list of workshops for support postgraduate students, as well as supervision guidelines. It also provides a list of services that postgraduate students can access during their registration, such as IT support, library access and health and wellness assistance (including psychological and legal counseling). The Faculty of Science has a similar online document. 14 This inexactness is arguably the result of disciplinary differences – that is, the requirements for a PhD in, for instance, Philosophy are distinct from those in the medical sciences. From this perspective, it could be asserted that the ‘amiable anarchy’ described by Walters, in Matos (above) is in fact the most appropriate way of approaching any description of the PhD. 35 degree) but uncertain about the route to it. As mentioned earlier, disciplinary differences might prevent the development of a single, universal understanding of the term; however, it would arguably be an exemplar of good practice to encourage individual disciplines to develop (and disseminate) their specific requirements for a doctorate in that discipline. By doing this, they could eliminate much of the uncertainty that has historically plagued the process. 2.2.2 Supervision In the literature that I surveyed, there seemed to be an unspoken assumption that the term supervision did not require explanation because its meaning was self-evident (especially to those engaged in the practice). Mirroring this, I found no clear-cut definition for the term in the Wits staff guidelines for the supervision of higher degree students, Strategies for Postgraduate Supervision. In its place was a ‘shopping list’ of the ‘desirable qualities in a supervisor’15 1. Supervision is about excellence in research: its scientific principles, meticulous practice and strict adherence to ethical standards. This must be conveyed to the student both through teaching and by example. . This was followed by a set of ‘principles of good supervision’, which I have reproduced below in order to illustrate the point that while the University is able to identify such principles, it falls short of a precise definition of the concept – in other words, what supervision is, and what it is not. However, perhaps an authoritative definition of supervision is not necessary, given that each supervisor-student relationship is unique. Each has its own particular dynamic which may defy any attempt at a singular and universal representation. What this means for the practice of supervision is that, literally, anything goes; that is, any interaction between a supervisor and a student that is related to the research being undertaken could be considered to be supervision. This would include both the intention as well as the practice itself. 2. Good supervision involves the basics of good teaching. Among these are concern for students, interest in their progress and the provision of thoughtful and timely feedback – while keeping in mind that the supervisory relationship 15 The qualities – described as ‘intangibles’ – are: availability; the ability to hear both what is said and what is meant; circumspection; good interpersonal skills; respect; transparency, openness and a sense of fairness; the ability to be obliging and accommodating; broadmindedness; the capacity for clear goal- setting; innovation; the ability to accept failure with equanimity; and a healthy sense of the absurd (2007: p. 2). 36 has a personal dimension, especially when students face crises of confidence or personal problems. 3. Good supervision is more than just information transfer. The sustained complexity of supervision involves much time and energy. Good supervisors are aware of this, and of the professional commitment they must make to every student. 4. Research students are highly individual, with different preferences, expectations of the relationship and approaches to study – some of which may be related to their cultural background. Good supervisors recognize and value this diversity, and adjust their practices accordingly. 5. Good supervisors extend their students well beyond what those students thought possible, by setting high but realistic standards. They encourage independence by building students’ confidence in their personal research capabilities. 6. Finally, good supervisors are conscious of their mentoring role. They aim to be models for first-rate scholarship, while at the same time introducing students to a scholarly community in which they can then establish their own networks and scholarly identity (Strategies for Postgraduate Supervision 2007: 2) While all of these prescriptions of what constitutes ‘good’ supervision and the ‘good’ supervisor are useful, they nonetheless fail to adequately explain the concept itself – what postgraduate supervision is, and what differentiates it from other forms of pedagogy (to be discussed in more depth in section 2.2.3). It seems that there is an assumption being made that since all supervisors have themselves been supervised at some stage in their academic careers, they simply know what supervision is. This is a problematic assumption, especially given the range of contexts, environments and styles of supervision to which they may have been exposed. Indeed, this assumption is borne out in some of the early literature on supervision. For instance, Rudd (1985: 79) states that The supervisor of postgraduate research is appointed for his knowledge of the student’s field… It is generally assumed that anyone capable of holding a lectureship is capable of supervising a research student – after all, many universities’ contracts specify that their staff shall engage in research, and if one can do research, one can, presumably, supervise it. (Italics added). The danger of this kind of assumption in the high-stakes domain of postgraduate research is clear: undertaking research and being an effective supervisor are not synonymous practices, nor should they be understood to be. Attempting to equate one with the other could create the false expectation that the ‘brilliance’ of the researcher 37 will be transferred to the student without much effort on the part of either – what could be termed the ‘osmosis effect’. Left uninterrogated, this assumption has the potential to undermine the perception that postgraduate supervision is (or at least, should be) a conscious pedagogic practice in which the supervisor engages with the student in such a way as to advance learning (see section 2.2.3 for more on the pedagogy of supervision). In other words, learning (both skills and content) does not happen automatically, but rather is promoted by the active participation of the individuals involved in the supervisory relationship. This perspective is supported by South Africa’s Council on Higher Education (CHE) - a statutory body whose primary responsibility is to advise the National Minister of Higher Education on all policy issues relating to Higher Education, including those of quality assurance and promotion. This body has offered the following description of postgraduate supervision: … the supervision or promotion of students’ research activities leading in whole or in part to the awarding of a Master’s or Doctoral degree. The goals of postgraduate supervision are both the production of a good thesis and the transformation of the student into a competent and independent researcher. The supervision process is essentially a complex teaching and mentoring activity that includes a range of activities, such as • Assisting students to refine a research topic and design an acceptable research proposal; • Getting the proposal approved; • Providing guidance on appropriate literature; • Assisting with the determination of the research design and methodology; • Supporting students in the collecting and analyzing of data and writing up the thesis or dissertation as a final product; • Providing detailed feedback to students; • Meeting reporting requirements on students’ progress; and • Writing a final report on the research process for the external examiners and examining committee. (Postgraduate Research & Supervision, ITL Resource No. 7 2004: 4) Thus, from the perspective of the CHE, the supervision of postgraduate students is a combination of both mentoring and teaching. It encompasses a wide range of activities 38 that culminates not only in a thesis, but also (theoretically, at least) in the production of an independent researcher, who will be able to supervise postgraduate students of her own in the future. This definition comes from a resource document developed by the CHE, designed to support academics in the successful undertaking of their role as postgraduate supervisor. Entitled ITL Resource No. 7: Postgraduate Research and Supervision, and discussed in more depth in a later section of this thesis, it is one of a series of publications published by the CHE aimed at supporting academic staff to achieve the quality assurance benchmarks set by the Higher Education Quality Committee (discussed in depth in section 1.4.2 of the previous chapter)16 Melin Emilsson and Johnsson (2007: 166) take a relationship-centred approach to supervision. In their paper Supervision of Supervisors: on developing supervision in postgraduate education, they critique traditional understandings of the term which tended to focus ‘…on the supervisor’s role and competence, instead of the supervision situation as a whole with the relationship at the centre’ (p, 166). They argue that there are essentially two orientations when it comes to the supervision of doctoral candidates: problem-oriented supervision and process-oriented supervision. The former is focused on the specific tasks that the supervisor undertakes with the PhD candidate – in other words, it revolves around the two working together to solve particular problems emerging from the research process. By contrast, the process-oriented supervision orientation focuses on relationships, ‘…placing both the practicing professional [supervisor] and the client [doctoral candidate], and the relationship they create, at the centre’ (Melin Emilsson & Johnsson 2007: 167). This is an interesting dichotomy . Included in this range of publications are those on programme planning, design and management, programme and course review, access and admissions, student development and support, the assessment of student learning, and staff development. Since they were written by South Africans for the South African context, they all have an intrinsic value for this country’s academics. 16 The HEQC is a permanent committee of the Council on Higher Education. Its chief mandate relates to quality promotion and quality assurance in South African Higher Education. According to the Higher Education Act 101 (1997), its three specific functions are to: • promote quality in higher education; • audit the quality assurance mechanisms of higher education institutions; • accredit programmes of higher education. 39 because it separates what takes centre stage in the relationship: tasks and problems, on the one hand, or the human relationships, on the other. I am not comfortable with this separation of issues. Instead, I would assert that the relationship between supervisor/s and student is intimately intertwined with the research and writing problems experienced by students in the process. Therefore, for instance, any obstacle that the student may experience while undertaking the research could be discussed with the supervisor within the bounds of a supportive and interactive relationship, rather than being dealt with in a mechanistic, problem-focused manner. I deal with this issue of interconnectedness in a later chapter of the thesis, where I use the metaphor of a picture window, consisting of four windowpanes, to illustrate how not only these relationships, activities and events are connected, but how they are in turn linked with the processes and contexts of supervision. Each pane is important to complete the picture window, and each adds a unique feature to it. Park, in a 2007 discussion document for the United Kingdom’s Higher Education Academy, distinguished between what he considers to be the ‘traditional’ definition of supervision, and a more modern understanding: Traditionally, most supervision was based on the ‘secret garden’17 … institutions [having] clearly defined roles and responsibilities of both supervisors and research students and clear criteria for defining who is eligible to act as a supervisor…. [S]upervisors are also expected to take a much more active role than previously in guiding and helping the personal development and skills training of their research students in order to enhance their employability (Park 2007: 29). model… in which student and supervisor worked closely together without a great deal of external scrutiny or accountability… (Park 2007: 29) Park subsequently argues that this ‘secret garden model is no more’, although this is debatable. He argues that, in the UK specifically, transparency, accountability as well as alignment to nationally recognised quality assurance codes now take precedence. These include, but are not limited to 17 Park used this term first in a 1995 article, New Variant PhD: The changing nature of the doctorate in the UK, to describe ‘…an activity that takes place behind closed doors between consenting adults’. In other words, it is essentially a ‘hidden interaction’ between two individuals – the supervisor and the student. 40 In this more contemporary understanding of the term supervision, the focus appears to be on the rules and regulations of supervision – what supervisors ought to do and to be, and what nationally-proclaimed directives they ought to comply with – rather than on the wider processes and relationships inherent in a more holistic view of the term. While the need for accountability and transparency – which Parks favours - is indisputably important for all parties involved in the process, the neglected importance of the human relationships and interactions that make up the act of supervision is problematic in this understanding. It seems to reduce supervision to a bureaucratic, box-ticking exercise, rather than acknowledging that it involves the active development of a new generation of researchers and academics. This is evidenced by, for instance, the United Kingdom’s technicist approach to dealing with the issue, namely ‘appropriate personal and professional development [for] those who supervise’ and ‘self-help books for doctoral students’ (2007: 29). 2.2.3 The pedagogy of supervision Understanding supervision as a form of pedagogy adds another dimension to understanding the supervisor-student relationship. I found it difficult to pin down a single, commonly agreed-upon understanding of the term ‘pedagogy’ in my research. Nonetheless, I have offered a range below, and have taken what I believe to be useful for my research from each. I have subsequently made the connections between the literatures on pedagogy, on the one hand, with the practice of postgraduate supervision, on the other. Kamler and Thomson (2006) offer a useful starting point for understanding pedagogy. They argue that this term, …used in the northern European tradition, encompasses the everyday formal and informal practices of education. This includes the use and development of theory about and in the work of educating, as well as the formulation of policy and the education of those who are the pedagogues. Pedagogy is a relational concept since it refers to what happens between the authorized pedagogues and students (Kamler & Thomson 2006: 18). From this description, a number of key elements of pedagogy are clear: firstly, that it is about practices in a learning environment. Put another way, pedagogy revolves around 41 what both the student and the supervisor do in order to advance learning – thus, learning is not something that happens passively or in a vacuum. Secondly, and leading on from the previous point, pedagogic practices are social in nature because they require the engagement of both parties who bring with them to the supervision relationship their ‘…histories, hierarchies, customs, teleologies and narratives’ (Kamler & Thomson 2006: 18). A third element that Kamler and Thomson’s definition of pedagogy raises is that practice need not be formal – that is, it need not happen at scheduled times or in official settings such as the supervisor’s office, or indeed even with university-appointed supervisors. On the contrary, I would argue that learning may (and often does) occur spontaneously in informal contexts, and without prior planning. This is, I believe, fundamental to understanding pedagogy. Alexander’s (2008) understanding of the term mirrors Kamler and Thomson’s, but explicitly adds another dimension - that of discourse: …[P]edagogy is both act and discourse. Pedagogy encompasses the performance of teaching together with the theories, beliefs, policies and controversies that inform and shape it… Pedagogy connects the apparently self- contained act of teaching with culture, structure and mechanisms of social control (2008: 3) Here, I am following Gee’s (1996) conceptualization of the term discourse: Discourses… are ways of behaving, interacting, valuing, thinking, believing, speaking, and often reading and writing that are accepted as instantiations of particular roles (or ‘types of people’) by specific groups of people… Discourses are ways of being ‘people like us’. They are ‘ways of being in the world’; they are ‘forms of life’. They are, thus, always and everywhere social and products of social histories (1996: viii) In connecting Gee’s concept of discourse with Alexander’s definition of pedagogy, what is clear to me is that, firstly, the latter involves the numerous ways that teachers/supervisors have of speaking about and understanding the practice of teaching. Teachers/supervisors are a specific ‘type of people’ who have a specific job to do within specific contexts. Logically, then, they also have specific ways of ‘behaving, interacting, valuing, thinking…’ and so on about their practices. Put another way: 42 pedagogy is not just the ‘what’ of teachers’/supervisors’ daily practices, but is also the ‘how’ and the ‘why’, as well as the communication of these with one another. Within the sociology of education, there has been a move towards the concept of critical pedagogy, which McWilliam argues is: … a moral and political endeavour. Its social reconstructionist agenda is pitted against any technicization of the process of production and exchange that takes place in the interaction of the teacher, learner and knowledge jointly produced (Lusted 1986, 3). Critical pedagogy draws attention to the conditions necessary to maximize opportunities for effecting appropriate change. In particular, such writing focuses on the power relations within which knowledge is produced (McWilliam 1996: 4) The key element of power, and its imbalances in most educational contexts, is taken up by Gruenewald (1999), who takes his cue from the work of Burbules and Berk (1999). These authors suggest that critical pedagogy is … an effort to work within educational institutions and other media to raise questions about inequalities of power, about the false myths of opportunity and merit for many students, and about the way belief systems become internalized to the point where individuals and groups abandon the very aspiration to question or change their lot in life (Burbules & Berk 1999: 50; cited in Gruenewald 1999: 4). What is striking about this understanding is how relevant it is to the South African political and educational contexts, both past and present. Certainly, the oppressive Apartheid regime (1948 – 1994), with its system of Bantu education extinguished the aspirations of hundreds of thousands of black South Africans, even as they tried to resist its implementation. Ndimande (2009) describes the Bantu education system as one in which ‘…segregated black public schools were required by law to teach a specified curriculum, which sought to impose inferiority on teachers and students with the sole requirement of bureaucratic and political compliance…Thus, Bantu education prohibited equal education opportunities for black people mainly through school funding and inferior curriculum (2009: 125). 43 Even the ‘best of the best’ black students rarely made it to tertiary level – thus realising Burbules and Berk’s observation about the ‘false myths of …merit’. While the issues of unequal funding and the inferior curriculum have been dispatched by the advent of democracy in 1994 – in theory, anyway – new forms of educational inequality seem to have taken their place. Social class, for instance, seems to be the new issue around which inequality in education is organizing. In her introduction to the book Changing Class: Education and Social Change in Post-Apartheid South Africa, Chisholm (2004) describes this shift in race-based to class- based inequality as follows: ‘…even as the stated intent of post-apartheid’s education policy-makers has been to reconcile the interests of competing and unequal social classes, those of a new deracialised middle class have come to predominate…School financing policy has provided targeted funding for the poor but has protected public schools for the middle class. Curricula have promoted philosophies and forms of education which shift the goals of schooling but they also facilitate middle class leadership and its creative self-expression. Decentralization of schools has democratized local control but has given the middle classes the greatest command over how schools are run and what they can buy for their fees. Desegregated, formerly white, Indian and coloured schools have opened ‘the doors of culture and learning’ (ANC 1955) but have integrated only a minority of African children. They are offered the values and practices of middle-class schools which officially promote ‘non-racialism’ and gender equity but in practice are far from race-blind or gender-sensitive. Assessment practices have broadened but hide continuing inequalities of performance. Language policy recognizes African languages but has enabled middle classes, and particularly white, English-speaking middle classes to exercise choice and discretion in selection of the medium of instruction and, in doing so, to maintain their social dominance. The private school section has mushroomed and includes many poorer African children, even as elite private schools have maintained their privilege. Higher education has also been deracialised but simultaneously become more selective and competitive, regulating entry through its portals to the middle class in more stringent ways (2004: 6, italics added). From the perspective of critical pedagogy theory, this situation represents inequality (albeit in a different guise to the more traditional racial variety) and an imbalance in the power relations in which knowledge is produced. Put another way: education has now become an oppressive force, effectively working against the needs of the working and lower classes which constitute the majority of the country’s population. In the context 44 of postgraduate study in South African universities: the criteria for selection to study at this level are stringent. At the University of the Witwatersrand (Wits), for example, a student needs to have achieved an average of 65% across all subjects at the undergraduate level18 Braa and Callero (2006) are explicit about the revolutionary nature of critical pedagogy, describing it as ‘… a radical approach to education that seeks to transform oppressive structures in society using democratic and activist approaches to teaching and learning’ (2006: 357). While I doubt that they had postgraduate education in mind when they wrote this, I can nonetheless see how it is applicable to this context, in the guise of transformation (discussed in the Introduction to this work). In short, a significant part of the transformation process being undertaken at most HEIs seeks to remove the barriers to (postgraduate) education imposed by Apartheid, and replace them with structures that encourage and nurture students at this level, through mechanisms such as improved access to funding for previously disadvantaged (black) students. Through acts of redress like this, transformation aims to democratize postgraduate education. However, there is a danger that the process – with all the best of intentions – ends up over-compensating for the errors of the past, and creates a in order to have a chance at selection for postgraduate study in their chosen faculty. Even this does not guarantee them a place, though, as Schools often restrict enrollment numbers and ‘cherry pick’ the best students for these places. The origins of this predicament lie in a preceding period of rapid expansion and relatively open registration for postgraduate students at the university. From my experience of postgraduate study at Wits, and from the interviews that I conducted for this study, I have the sense that the ‘philosophies and forms of education’ to which Chisholm refers above are as overtly liberal and covertly unequal in this realm as in the primary and secondary schooling system. Evidence of this includes the fact that English is Wits’ medium of instruction, assessment tasks (in this context, the thesis) are so broad that they can hide a multitude of what Chisholm refers to as ‘inequalities of performance’, and finally, the majority of the university’s governing structure – Council – is made up of middle class individuals. 18 This average seems to be the norm across all Faculties at the University, but may be altered at the discretion of the Dean and/or Head of School. 45 generation of white students who are oppressed by the very educational structures designed to bring about democracy. I recognize, though, that while white students may be discriminated against by such race-based practices, they nonetheless still have access to the informal educational practices, and middle-class identities and discourses that black students may not have. Thus, it is possible that class may ultimately come to replace race as a source of discrimination and division in education (at all levels). Henry Giroux (2004), a leading critical pedagogy theorist, confirms this link between the theory and democracy: …[C]ritical pedagogy emphasizes critical reflexivity, bridging the gap between learning and everyday life, understanding the connection between power and knowledge, and extending democratic rights and identities by using the resources of history…At the very least, critical pedagogy proposes that education is a form of political intervention in the world that is capable of creating the possibilities for social transformation. Rather than viewing teaching as technical practice, radical pedagogy in the broadest terms is a moral and political practice premised on the assumption that learning is not about processing received knowledge but actually transforming it as part of a more expansive struggle for individual rights and social justice (2004: 34). This point echoes the argument of one of the ‘founding fathers’ of critical pedagogy Paulo Freire. In his seminal 1970 work, Pedagogy of the Oppressed, he observed that A careful analysis of the teacher-student relationship at any level, inside or outside the school, reveals its fundamentally narrative character. This relationship involves a narrating Subject (the teacher) and patient, listening objects (the students). The contents, whether values or empirical dimensions of reality, tend in the process of being narrated to become lifeless and petrified. Education is suffering from narration sickness (2000: 71) Education practiced in this way therefore has no transformative value – socially, politically or culturally - for either the teacher or the student. Freire refers to this as the ‘banking’ concept of education because it ‘…becomes an act of depositing, in which the students are the repositories and the teacher is the depositer. Instead of communicating, the teacher issues communiqués and makes deposits which the students patiently receive, memorise and repeat’ (2000:72). It is in this uncreative environment that individuals are dehumanized, since it is only through the processes of 46 ‘invention and re-invention (p. 72) that knowledge emerges, something that Freire describes as ‘… hopeful enquiry [that] human beings pursue in the world, with the world, and with each other’ (p. 72). He subsequently offers a list of the attitudes and practices that characterize an oppressed education system (and by implication, an oppressed society). I have reproduced this below because echoes of some of them appear in the supervision experiences of many of the doctoral candidates that I interviewed for this research, and certainly in my own experience with my first PhD supervisor. These practices are: (a) The teacher teaches and the students are taught; (b) The teacher knows everything and the students know nothing; (c) The teacher thinks and the students are thought about; (d) The teacher talks and the student listen – meekly; (e) The teacher disciplines and the students are disciplined; (f) The teacher chooses and enforces his choice, and the students comply; (g) The teacher acts and the students have the illusion of acting through the action of the teacher; (h) The teacher chooses the program content, and the students (who were not consulted) adapt to it; (i) The teacher confuses the authority of knowledge with his or her own professional authority, which she and he sets in opposition to the freedom of the students; (j) The teacher is the Subject of the learning process, while the pupils are mere objects (2000: 73) These practices foreground the importance of students engaging with knowledge in order to transform it, as opposed to being filled up with the (unquestioned) knowledge of the teacher. This has particular relevance for postgraduate supervision because it is in this pedagogic context that students are in the position – possibly for the first time in their educational careers – to interact with, alter, apply, reconfigure and reconstruct knowledge. They have to think critically, come up with solutions and make informed choices about the path of their research. It is the student who decides on her study program in negotiation with her supervisor. It is the student who exercises self- discipline in meeting research deadlines. It is the student who is the Subject of the learning process – she is the one who is actively inventing and re-inventing knowledge, while the supervisor oversees. Moreover, in this context, the teacher/supervisor is not necessarily the omniscient guru but often learns alongside and at the same time as the candidate. Thus, one could use the attitudes and practices listed by Freire as a checklist 47 of what the pedagogy of supervision ought not to be. This is by no means a radical idea, although much of Freire’s work has been described variously as ‘revolutionary’ and ‘politically provocative’; rather, it is a means of challenging the existing status quo of, for instance, the master-apprentice approach to supervision (discussed in more depth in the ‘Communities of Practice’ section later on in this chapter). What Freire is essentially arguing in favour of is a situation in which the student is not understood as an empty vessel, waiting to be filled up by the all-knowing teacher, but is constructed as a sentient being that has the capacity to actively engage with knowledge in order to transform it. Furthermore, she has a consciousness of these facts that may be used to alter that which is oppressive about the knowledge system in which she works. What I have encapsulated in this section is an overview of the literature on postgraduate supervision and pedagogy. It was important to illustrate how the process of supervision can be understood as a type of pedagogy – that is, a combination of face- to-face teaching, as well as the beliefs, policies, and social contexts in which it takes place. Specifically, I have suggested that supervision is a form of critical pedagogy because it valourises the transformation of knowledge, and in the process, the doctoral candidates are themselves transformed and can appropriate societal change through their active application of this knowledge. The tenets of critical pedagogy are for the most part at odds with the communities of practice theory developed by Lave and Wenger in the early 1990s (discussed in more depth in section 2.5), although there is some match when it comes to conceptions of collective and collaborative learning. The primary dissonance between the two lies in how each once conceptualizes knowledge production. While critical pedagogy argues for an empowered learner who is actively engaged in knowledge production, the master-apprentice model of communities of practice advocates that the former teaches the (‘empty vessel’ and largely passive) apprentice in what Freire referred to as the banking model of education. Thus, knowledge tends to flow in a single direction: from the master to the apprentice. Before I discuss communities of practice model in more depth, though, it is a worthwhile exercise to foreground the various discourses of supervision that exist. 48 2.2.4 The discourses of supervision Grant (2005) suggests that, based on an extensive review of the literature, there are four key discourses of postgraduate supervision: the psychological, traditional academic, techno-scientific and neoliberal. Each constructs the supervisor and postgraduate student in a specific way. In the psychological discourse, the supervisor is portrayed as a ‘caring, expert professional (2005: 340), who is ‘first and foremost a source of motivation and support’ (2005: 340) for the student. According to this discourse, the student is ‘… a whole person but one who is inexperienced and uncalculated with regard to the task of independent research and therefore in need of help’ (2005: 341). Thus, the emphasis here is on support, sensitivity, motivation and development, with seemingly little or no room for conflict. By contrast, the traditional-academic discourse of postgraduate supervision is understood ‘… less as an interpersonal relationship and more as an intellectual apprenticeship’ (2005: 341). This discourse echoes Lave and Wenger’s master- apprentice model of supervision because it proposes that there is a master and a disciple in the relationship. In many ways, this discourse implies a Darwinian approach to supervision – only the (academically) fittest survive. The power relations that epitomize this discourse are described by Grant as follows: Like the guru/disciple relationship, proper Trad-Supervision is infused with sovereign indifference from the Trad-Supervisor… alongside grateful, even eager, subjection from the Trad-Student who submits because she/he believes in the extraordinary qualities of the specific person who is her/his supervisor (2005: 342) The techno-scientific discourse – which ‘originated with the rise of research universities from the late nineteenth century and the constitution of the social sciences in the image of positivist science’ (2005: 342) – constructs the supervisor as a ‘trained and expert scientist’, and the student as an ‘inexperienced trainee’ (2005: 342). The discourse portrays the process as a ‘predictable and orderly process of research skills training’ (2005: 342). Power relations in this discourse are characterized by 49 surveillance on the part of the supervisor, whose role it is to ensure that the student is trained in the ‘correct’ methods of research. According to Grant: The malleable and obedient Techno-Student listens, tries and reports; the techno-Supervisor observes, judges, instructs (2005: 343). Grant reports that this is the discourse most favoured by national governments and higher education bodies, and it is obvious why: it posits that supervision has an ‘orderly and predictable trajectory’ (2005: 343), which means that funding bodies can calculate how much money needs to be set aside annually for postgraduate students. By implication, if these students do not graduate in the ‘orderly and predictable’ timeframes set by the proponents of this discourse, then individuals and their institutions can be held accountable – even to the point of losing funding if their postgraduate students go beyond the expected timeframe. Finally, Grant points to the existence of a neoliberal discourse, which constructs ‘…education as a commodity, and education institutions as commercial enterprises’ (2005: 343). In this construction, the student arguably holds significant power because she is the consumer of the commodity (that is, she pays for it), and it is subsequently up to the supervisor to provide the service (of supervision). As Grant points out, ‘… the student, as the service chooser and consumer, has the power of the purchaser and expects value for money’ (2005: 343). Thus, there is what Grant calls a ‘quasi-legal’ element to this discourse, embodied by a ‘specific contract whereby both [supervisor and student] have specified rights and responsibilities’ (Grant 2005: 343). Furthermore, this commercialization of higher education in general, and postgraduate research in specific, has resulted in increased pressure on supervisors to ensure that their students finish on time – that is, according to Grant, ‘…within the period of the ‘contract’ which is the period covered by government funding’ (2005: 343). This discourse, in common with the techno-scientific one, ‘privileges the rational specification of process (services) and product’ (2005: 344) over the intellectual development of the student. Grant acknowledges the existence of what she calls ‘marginal discourses’, too – these are the radical discourses that include progressive, critical and feminist discourse. These are generally ‘gendered, classed, ethnically situated, sexually orientated (and so 50 on) and these social positionings play out in supervision’ (2005: 344). In the South African context, I would argue that these so-called marginal discourses are not that marginal – if anything, the issues that they raise are particularly relevant in a country that was, until 1994, controlled by a white, mainly Afrikaans-speaking, mainly middle- class government. Universities and technikons19 The term ‘communities of practice’ was developed by in the 1990s by Lave and Wenger were largely racially segregated, and hence supervisors and their postgraduate students generally had a great deal in common in terms of race, ethnicity and gender. In the sixteen years since the advent of democracy, however, South African universities have widened access to a far more diverse population of both staff and students, and hence cross-race, cross-gender and cross-class supervisory relations are more the norm than the exception. 2.2.5 Communities of practice 20 In his seminal work, Communities of Practice: Learning, Meaning and Identity (1998), Wenger breaks the term down into its two constituent ideas, namely communities and practice, and examines each separately before re-joining them as a unit. Interestingly, he chooses to begin his analysis by looking at the second idea: practice. This, he as a social learning theory. In his article Communities of Practice and Social Learning Systems (2000), Wenger explains the centrality of communities – of whatever kind – in the learning process: Since the beginning of history, human beings have formed communities that share cultural practices reflecting their collective learning: from a tribe around a cave fire, to a medieval guild, to a group of nurses in a ward, to a street gang, to a community of engineers interested in brake design. Participating in these ‘communities of practice’ is essential to our learning. It is at the very core of what makes us human beings capable of meaningful knowing… [They] are the basic building blocks of a social learning system because they are the social ‘containers’ of the competencies that make up such a system. By participating in these communities, we define with each other what constitutes competence in a given context… (Wenger 2000: 229) 19 Also known as universities of technology. 20 The concept of communities of practice appeared in their book entitled Situated learning: Legitimate peripheral participation, which was first published in 1991. There have been numerous reprints since then. 51 argues, happens when human beings interact with one another in ‘the pursuit of enterprises of all kinds’ (1998: 45). It is during the process of these interactions that learning takes place, ‘as we define these enterprises and engage in their pursuit together’ (1998: 45). Therefore, learning takes place wherever these interactions occur, and with the passage of time, this kind of learning takes on a new role: Over time, this collective learning results in practices that reflect both the pursuit of our enterprises and the attendant social relations. These practices are thus the property of a kind of community created over time by the sustained pursuit of a shared enterprise (1998a: 45). Wenger subsequently reiterates the social aspect of all enterprises. Put another way, enterprises (of whatever kind) do not occur in a vacuum, but rather in discourse with others. He argues that this social intercourse is at the heart of the learning that ultimately takes place. In his words: The concept of practice connotes doing, but not just doing in and of itself. It is doing in a historical and social context that gives structure and meaning to what we do. In this sense, practice is always social practice (1998a: 47). Furthermore, Wenger points out that practice is not just about the explicit enterprises that humans engage in, but includes the tacit ones, too – ‘It includes what is said and what is left unsaid; what is represented and what is assumed’ (1998: 47). The explicit enterprises are, not surprisingly, also the more obvious ones: ‘… the language, tools, documents, images, symbols, well-defined roles, specified criteria, codified procedures, regulations, and contracts that various practices make explicit for a variety of purposes’ (1998: 47). In the context of postgraduate supervision, these explicit enterprises could include such things as supervisory meetings, institutional documents and policies dealing with all aspects of the postgraduates’ tenure at the university, ethics committees and their procedures, the master-apprentice perception of the supervisor-student relationship, and so on. These are the enterprises that both supervisor and student would be aware of as they undertake the doctoral journey together. 52 By contrast, the more tacit enterprises, according to Wenger, would include ‘… all the implicit relations, tacit conventions, subtle cues, untold rules of thumb, recognizable intuitions, specific perceptions, well-tuned sensitivities, embodied understandings, underlying assumptions, and shared world views’ (1998: 47). What struck me as I typed out this list was how these tacit enterprises – upon which the apprenticeship model is premised - are the ones that could potentially alienate the student in the supervision relationship, because they are based on assumptions of a common knowledge or understanding of ‘the way things work’. It is my contention that this assumption – as with many others regarding postgraduate supervision – is a dangerous one. It would seem that these ‘tacit conventions’ and ‘untold rules of thumb’ are used to exclude certain individuals or groups of individuals (such as women, certain ethnic groups, or people with particular political affiliations) from becoming fully-fledged members of any particular community, and in so doing, maintaining the elitist nature of the doctorate. In the case of South African higher education, it has traditionally been people of colour and women who were at the receiving end of these fundamentally exclusionary practices. Although this is changing, the doctoral domain has traditionally been dominated by the White men who developed these unarticulated ‘rules’ which have proven difficult to access by those who are not White males. Thus, although Wenger has highlighted the tacit aspect of practice in a neutral way, I would argue that it is this aspect that has – in the South African context – effectively obstructed access to, and advancement through, the postgraduate years of study of previously excluded groups. Therefore, as a counter-response to Wenger, I would argue strongly in favour of making that which is implicit, explicit, so as to open the doors of postgraduate learning even wider, and for all. While I recognize the idealism underpinning this assertion, I nonetheless believe that opening access – both to postgraduate study itself and to the discourse of doctoral study - is vital if we are to achieve educational equity in a system that historically has been a discriminatory one. From the concept of practice, Wenger moves on to that of community. He argues that ‘…practice is the source of coherence of a community’ (1998: 72), and that the relationship of the two may be described in terms of three dimensions: mutual 53 engagement, a joint enterprise, and a shared repertoire. I will discuss each dimension briefly, relating it to the theme of my thesis. 2.2.5.1 Mutual engagement Firstly, the element of mutual engagement revolves around the fact that within any community, the members are (almost by definition) engaged with one another in some way. Wenger argues Practice does not exist in the abstract. It exists because people are engaged in actions whose meanings they negotiate with one another… Practice resides in a community of people and the relations of mutual engagement by which they can do whatever they do (1998a: 73). By implication, learning happens when two or more individuals interact with one another – even if this interaction does not happen face-to-face. Mutual engagement is particularly visible in doctoral supervision because it almost always involves a student and one or more supervisors, engaging with one another to produce new knowledge. They are a particular community – a community of scholars in a particular disciplinary area, and they are engaged in the kinds of practices described on the previous pages. The more established scholars – the supervisors – (should) also introduce their students to other scholars in their community with whom they can communicate (that is, engage), and through this communication, advance knowledge. Hence, learning takes place through the medium of mutual engagement. Wenger subsequently unpacks mutual engagement into three separate constituent parts: enabling engagement, diversity and partiality, and mutual relationships. All three are a vital part of this aspect of ‘community’. I have attempted to capture the essence of each one of these ‘sub- sections’ below as succinctly as possible without eroding the integrity of their meaning. Enabling engagement Whatever it takes to make mutual engagement possible is an essential component of any practice (Wenger 2004: 74) This sentence sums up the essence of the term ‘enabling engagement’: in order to be an engaged member of a community, one needs to be ‘included in what matters’ (2004: 74). In order to be ‘included’, one needs to do whatever is necessary to facilitate a 54 process of mutual engagement. Wenger suggests that this can be as simple as a group of colleagues chatting together during work, to a family having dinner or cleaning the house together, or even to picking up a telephone to make a connection (2004: 74). What is important is that something gets done in order to make the mutual engagement possible. For research students, especially those in the laboratory sciences who have to be at their bench every day, creating this enabling engagement is a relatively simple task: since most share the laboratory with other research students, it could involve sharing results, asking for advice or even having an informal chat about a non-research related matter. For candidates not located in a laboratory – for instance, those in the mathematical sciences and the humanities – creating an environment that would enable engagement could be a more difficult proposition. Certain schools in these disciplines hold regular research seminars and presentations which they expect the candidates to attend (although, these are not necessarily valuable – see my Prologue); some invite guest lecturers and research fellows to share their work. These are both examples of something being actively done to encourage mutual engagement both between individual students, and between students and academic staff. Diversity and partiality Wenger argues that there does not need to be homogeneity of membership in order for a community to be considered a community of practice – diversity and difference within the membership are more often than not the common feature of any community of practice. By the same logic, homogeneity is not a result of the development of a community of practice, either. Instead, argues Wenger, each participant within the community ‘finds a unique place and gains a unique identity’ (2004: 76) within it. These unique identities of the individual participants become more integrated and defined (while maintaining their essential uniqueness) during the course of the mutual engagement in their specific community. In the case of doctoral candidates: as they advance in their studies, so their identity as a researcher develops, from that of a novice, to that of a more experienced researcher. What is significant is that throughout the process, the candidate’s position in the community of practice remains unique – no 55 other participant possesses exactly the same interests (although some may be very similar). Partiality is, for Wenger, also inherent in mutual engagement. This term refers to the fact that each member of the community of practice has competence in some (but not all) areas. Thus: Mutual engagement involves not only our competence, but also the competence of others. It draws on what we do and what we know, as well as on our ability to connect meaningfully to what we don’t do and what we don’t know – that is, to the contributions and knowledge of others (Wenger 1998a: 76; italics added). In any research community, each member has differing access to differing knowledges, and no single individual holds all the knowledge of that community. Thus, for instance, the doctoral candidate has access to the data that emanates from her research, but may not know how to interpret it. The supervisor, on the other hand, could possess the specialised interpretation skills while not being totally au fait with the content of the research. In this scenario, the candidate is learning how to learn by observing the master. For Wenger, this feature of mutual engagement is key for the effective functioning of a community: Because they belong to a community of practice where people help each other, it is more important to know how to give and receive help than to try to know everything yourself (Wenger 1998a: 76). Mutual relationships At this stage in the description of communities of practice, it should be clear that the concept of mutual engagement revolves around the creation of relationships between people. However, Wenger is very clear in his assertion that these relationships are not always characterized by ‘peaceful coexistence, mutual support, or interpersonal allegiance’ (1998a: 77); rather: Most situations that involve sustained interpersonal engagement generate their fair share of tensions and conflicts. In some communities of practice, conflict and misery can even constitute the core characteristic of a shared practice, as they do in some dysfunctional families. A community of practice is neither a haven of togetherness nor an island of intimacy insulated from political and social relations. Disagreement, challenges, and competition can all be forms of 56 participation. As a form of participation, rebellion often reveals a greater commitment than does passive conformity (Wenger 1998a: 77). I found this reference to the dual nature of communities of practice to be particularly relevant in the context of doctoral supervision. In any research community, there is rarely complete harmony of ideas, and this is precisely what drives the development of new knowledge forward. Even within the fairly limited community of the doctoral candidate and her supervisor/s, there are spaces where peaceful coexistence reigns, and there are also spaces for conflict and challenges to occur. The latter is common where the vision of the candidate and that of the supervisor diverge. What is important is the recognition that participation in a particular research community does not (and indeed, should not) guarantee harmony and consonance; tension and conflict are inevitable, but it is in the resolution of these that progress is made. 2.2.5.2 Joint enterprise The second dimension of communities – the joint enterprise – also has three sub- dimensions to it: the negotiated enterprise, the indigenous enterprise, and the regime of mutual accountability. Each of these represents a ‘building block’ of joint enterprise and is therefore worth examining in more depth. Negotiated enterprise Wenger begins this section by observing that the practices of any community are as complex and diverse as the individuals who constitute it. There will be a constant negotiation of meaning and understanding, although these are not guaranteed to produce agreement – which is not necessarily a bad thing, according to Wenger: In fact, in some communities, disagreement can be viewed as a productive part of the enterprise. The enterprise is joint not in that everybody believes the same thing or agrees with everything, but in that it is communally negotiated (1998a: 78. Italics added) What is striking about this is that agreement is not the holy grail of the community; disagreement is where meanings and understandings are negotiated, and out of this emerge new meanings and understandings – learning, in other words. In the context of doctoral supervision, the knowledge that the student is pursuing is either still 57 undiscovered (and the student is working on discovering it), or is requires further investigation in order to produce a deeper understanding. While the research that doctoral students are engaged in is ideally supposed to be original, nonetheless, it resides in one or more broader fields of study. In the process of her research, the doctoral candidate has to engage with the underpinning theories of her field in order to advance knowledge; these theories are often competing, sometimes even contradictory, and she will have to situate her discoveries somewhere within them, defending them when necessary. Alternatively, her discoveries may add yet another alternative (possibly competing and contradictory) view to existing knowledge in the field. What is important here is not that her research confirms existing knowledge, but that it contributes to it by opening discussions and negotiations of meaning. Indigenous enterprise Communities of practice are not self-contained entities. They develop in larger contexts – historical, social, cultural, institutional – with specific resources and constraints. Some of these conditions and requirements are explicitly articulated. Some are implicit but are no less binding. Yet even when the practice of a community is profoundly shaped by conditions outside the control of its members, as it always is in some respects, its day-to-day reality is nevertheless produced by participants within the resources and constraints of their situations. It is their response to their conditions, and therefore their enterprise (1998a: 79) I have reproduced this lengthy tract because it captures the essence of what Wenger means by the term ‘indigenous enterprise’. It highlights that the enterprise or community need not be formally constituted in a single geographical area, with specific conditions of membership. To a certain extent, there is a spontaneity in which the participants engage with one another whenever, wherever and however they can – the important point is that engagements do occur, regardless of the circumstances. The practice of postgraduate supervision can be described as an indigenous enterprise because, as I have already mentioned, it does not occur in a vacuum, but rather within an educational context that has strong historical, social and political elements to it which influence how the participants are able to interact with one another. The practices that the supervisor and student engage in are largely constrained by these realities – such as access to conference or laboratory funding, availability of a suitably qualified supervisor, or even access to an adequately stocked library or computer 58 laboratory. Despite these constraints (or maybe because of them), the participants adapt their enterprise – and it is their enterprise, according to Wenger - in order to advance the knowledge of their community. This may take the form of, for instance, a student sourcing outside funding for a project, or making use of other experts – not necessarily just her official supervisor – to advance her research. As Wenger points out, In sum, it is only as negotiated by the community that conditions, resources, and demands shape the practice. The enterprise is never fully determined by an outside mandate, by a prescription, or by any individual participant.[T]he power – benevolent or malevolent – that institutions, prescriptions, or individuals have over the practice of a community is always mediated by the community’s production of its practice. External forces have no direct power over this production because, in the last analysis, … it is the community that negotiates its enterprise (1998a: 80; italics added) Regime of mutual accountability As the term ‘accountability’ suggests, being a participant in a community of practices means that one should be able to be held responsible for one’s actions within that community. Wenger is more specific than this, though; he states that These relations of accountability include what matters and what does not, what is important and why it is important, what to do and not to do, what to pay attention to and what to ignore, what to talk about and what to leave unsaid, what to justify and what to take for granted, what to display and what to withhold, when actions and artifacts are good enough and when they need improvement or refinement (1998a: 81). What this rather vague list of both implicit and explicit assumptions points to is a need for agreement about what is acceptable within the community. Put another way, the community needs to be governed by both the formal (rules, policies, standards, and so on) and informal (‘specialised sensitivities, an aesthetic sense, and refined perceptions’) aspects of accountability so that the participants know when there is cause for concern about events within the community, and how to make sense of these. Implicit in this sense-making is the idea that the joint enterprise in which the participants are engaged is not a static, unchanging one, but rather a process. Wenger asserts that this process is 59 ‘as generative as it is constraining. It pushes the practice forward as much as it keeps it in check’ (1998a: 82). Hence the need for accountability21 21 The need for accountability seems to be the common denominator when comparing the work of Lave and Wenger with CHE and University documents, as well as that of Park (2007). All recognize that in any learning relationship, there is a mutual responsibility between the parties to ensure that the integrity of the work, the process and the relationship is maintained. It is only in this context that the advancement of knowledge can be seen to have occurred in an ethical fashion. . Applied to the context of postgraduate supervision, there exists within the supervisory relationship a mutual accountability that goes beyond a formal contract between supervisor and student, in which both commit themselves to the process and to the advancement of knowledge. There is also a ‘meta’ accountability, in terms of which the supervisor inducts her doctoral students into the community in a meaningful way – that is, makes explicit the often implicit practices of the more established scholar. In return, the student undertakes to adhere to the formal and informal ‘rules’ of the community for the purpose of advancing knowledge in an ethical and responsible manner. Ultimately, both parties are not just accountable to each other, but to the community of which they are a part. Although there seem to be no formal mechanisms by which those who break the rules can be disciplined, Wenger believes that mutual accountability is self-regulating and that no such mechanisms are therefore necessary. 2.2.5.3 Shared repertoire The final element of Lave and Wenger’s communities of practice model is that of a shared repertoire. Wenger defines this as the … routines, words, tools, ways of doing things, stories, gestures, symbols, genres, actions or concepts that the community has produced or adopted in the course of its existence, and which have become part of its practice (Wenger 1998a: 83). Put another way, the shared repertoire is a collection of the activities and artifacts of a particular community that give ‘… coherence to the medley of activities, relations, and objects involved’ in that community (Wenger 1998a: 82). According to Wenger, the elements of this repertoire are not necessarily heterogeneous; rather, 60 [t]hey gain their coherence not in and of themselves as specific activities, symbols, or artifacts, but from the fact that they belong to the practice of a community pursuing an enterprise (Wenger 2004: 82). As with the previous two elements of the community of practice theory, this one is also further broken down into two ‘sub-sections’: the negotiated enterprise, and the indigenous enterprise. Negotiation enterprise: history and ambiguity Wenger begins this section by explaining why he chose the specific word ‘repertoire’ to describe this element of the theory: ‘… to emphasize both its rehearsed character and its availability for further engagement in practice’ (Wenger 1998a: 83)22 In the context of doctoral supervision, the idea of interpretation emerging from a history of mutual engagement is pertinent because the process itself can be (and indeed, often is) interpreted in many diverse ways. For instance, in any given research community of practice, there exist numerous supervisory routines and practices. While . In other words, the negotiation of meaning in any given community of practice is not something that happens either spontaneously or in a vacuum; rather, it ‘… reflects a history of mutual engagement’ (Wenger 1998a: 83). Moreover, this meaning is ‘inherently ambiguous’ because it may have more than one interpretation – something that Wenger considers to be a strength: Histories of interpretation create shared points of reference, but they do not impose meaning. Things like words, artifacts, gestures, and routines are useful not only because they are recognizable in their relation to a history of mutual engagement, but also because they can be re-engaged in new situations… The fact that actions and artifacts have recognizable histories of interpretation are not exclusively, or even primarily, a constraint on possible meanings, but also a resource to be used in the production of new meanings… When combined with history, ambiguity is not an absence or a lack of meaning. Rather, it is a condition of negotiability and thus a condition for the very possibility of meaning (Wenger 1998a: 83; italics added). 22 In an explanatory endnote, Wenger unpacks his understanding of the term ‘repertoire’ in more depth: ‘The notion of repertoire typically refers to performances rather than artifacts, but the distinction is not particularly relevant for the negotiation of meaning. The achievement of meaning is always a performance, and a repertoire thus construed certainly includes props’ (Wenger 2004: 288). What he seems to be saying is that meaning cannot be negotiated in isolation from props or artifacts; rather, the latter are vital components in achieving the former. 61 the supervisor calls on her experience of these routines and practices in her work with the doctoral candidate23 This rather lengthy account of communities of practice is important for two reasons: firstly, because it offers a theory of community learning (or learning within a community) which goes beyond the often technicist guides (discussed in more length in Section 3.2. of the following chapter) that have flooded the market in recent years. Secondly, the account captures both the process and the product of the postgraduate supervision process: from the mutual engagement of the supervisor and the candidate (both with each other and with the research), to the negotiation of the parameters of the , she is also aware that these need to be flexible (or negotiable) for each of her students because their research is original and hence requires original negotiation from the supervisor. Resources of mutual engagement Because the repertoire of a community is a resource for the negotiation of meaning, it is shared in an interactive and dynamic sense. In particular, shared beliefs – in the sense of the same mental objects or models – are not what shared practice is about (Wenger 1998a: 84). Clearly, Wenger perceives a community’s repertoire – particularly the inherent ambiguity of interpretations within that community – to be ‘…difficult, in continual need of repair, and always unpredictable’ (p. 83), on the one hand, and ‘…dynamic, always open-ended, and generative of new meanings’, on the other (p. 84). This tension between fixed and dynamic processes is at the root of meaning-making within the community, and hence also of the evolution of the community. Conflict between these processes is, according to Wenger, an opportunity for the development of new meanings. For supervisors and doctoral candidates alike, this element of the community of practice is central because it offers a space for both to negotiate how the research process will unfold, in terms of both existing and future practices (of supervision and research). 23 Examples of these routines might include the bureaucratic aspects of registration, proposal submission and progress reporting to the relevant Faculty. These ‘[h]istories of interpretation create shared points of reference, but they do not impose meaning’ (Wenger 2004: 83). Put another way, these routines or histories offer a common point of departure for the supervisor, but by no means dictate the entire supervision process - the latter is dependent on the nature and scope of the research being undertaken by the doctoral candidate. 62 research project; through the ethical considerations implicit in the regime of mutual accountability, to the dynamism of the shared repertoire of each specific research community of practice. Embedded throughout Lave and Wenger’s theory is the centrality of relationships – mirroring those of postgraduate supervision. Individuals in a community, by definition, act in relation to others, just as doctoral candidates and their supervisors work with each other and the broader community in their pursuit of knowledge. However, I found that in the specific context of doctoral supervision, communities of practice theory took on a different guise than that described by Lave and Wenger: I discovered that the values of each particular community – whether it was the Science laboratory community, the one developed through networking at conferences, or the ones initiated by the candidates themselves – were open to change depending on how the individuals involved experienced them. Thus, for instance, some students valued the interactions that they had with peers in their laboratory community of practice because they provided sounding boards for their own research; Megan was such a student. Others, such as Patricia, were more self-contained and comfortable with a two-person community of practice. What became clear from my data was that the candidates only accessed these communities when they felt they needed to – they were not an implicit or permanent part of their daily research practice. Thus, in the postgraduate context, communities of practice were used selectively by the candidates to fulfill a particular purpose. This is a significant departure from Lave and Wenger’s conceptualization of a community of practice as a ‘constant’ in the lives of its participants. Ultimately, as my study will show, the levels of participation in their specific community of practice differs from candidate to candidate; however, it is never totally absent because as Wenger notes, it ‘… has value to its members’ (Wenger 1998b: 4). What this value is will be discussed in more depth in later chapters. 2.2.6 Beyond communities of practice The idea of communities of practice is a contested one. Scholars such as Lindkvist (2005), Handley et al (2006) and Roberts (2006) have pointed to numerous limitations of the theory; for the purpose of this study, I will restrict my examination of these 63 limitations to the ones that have a direct bearing on postgraduate supervision processes and practices. They are the compartmentalization of practices (that is, the artificial division of practices into discrete parts), the contested nature of the terms community and participation, as well as an overall critique of the master-apprentice model. 2.2.6.1 Compartmentalization of practice Wenger (1998a) recognizes that it is possible for individuals to belong to more than one community of practice at a time, because they exist in more than one context (employee, student, researcher, and so on). However, he is confident that being a participant in more than one community of practice is unlikely to cause conflict between these communities. In his words: … [W]e engage in different practices in each of the communities of practice to which we belong. We often behave rather differently in each of them, construct different aspects of ourselves and gain different perspectives (Wenger 1998a: 159). In other words, his contention is that individuals are able to separate their roles, identities and practices in each of these communities into discrete compartments, which may exist alongside one another, but the contents of which never ‘leak’ into each other. Wenger also seems to assume that these compartments exist in a state of harmony with one another. However, these perceptions are artificial and questionable – they assume that the roles, identities and practices of individuals operate in a complete vacuum from one another, never coming into contact with each other and hence, never presenting the opportunity for conflict. However, it is my contention that participation in an array of communities of practice increases the potential for conflict since it increases the junctures in which conflict could potentially occur. Attempting to compartmentalize individuals’ participation into boxes or assuming that these multiple participations can be ‘managed’ is unrealistic. Handley et al (2006) agree; they argue that conflict is inevitable because it is not possible to effectively manage and separate one’s multiple roles: Potential for tension and conflict exists because, during their lifetime, individuals participate not within one community (or collectivity or network) but within several – each with different practices and identity structures…The important issue here is 64 how individuals manage their roles, actions and relationships within multiple communities (2006: 647, italics added). For the duration of their registration, doctoral candidates usually participate in more than one community of practice: they may be tutors or teaching assistants in one, doctoral candidates in another, and members of an international scholarly community in yet another – and these are just in the academic context. Each context vies for a finite measure of time from the doctoral student, and there may be times when this leads to conflict. For instance, in her role as a tutor or teaching assistant, the doctoral candidate may have to spend significant amounts of time preparing for and facilitating tutorials, marking student assignments and being available for consultations with undergraduate students. At the same time, in her role as a doctoral candidate, she may be required to conduct experiments or interviews (depending on the discipline), and to analyse the data that emerge from these activities. This is a time-consuming and often labour- intensive process. In this instance, time spent marking assignments takes the candidate away from her research work, which has the potential to cause conflict between these two communities (tutor/teaching assistant, on the one hand, and researcher, on the other). An example of such multiple participations from my research interviews was Thabang. He participated in his academic community as a doctoral candidate (which included travel to numerous Southern African states to conduct interviews) and as a part-time lecturer - an assignment that required him to research and prepare lessons, and to set and mark various assessment tasks. He also served as a political analyst for several regional organisations. Clearly, his participation in academia was not confined to a single community, and he often had to juggle this participation when each made competing demands on his time. Furthermore, candidates may, in their personal lives, belong to sports clubs or charity organizations, too. In each, they take on different roles and identities which may conflict with one another. Kenneth, for instance, was a keen mountain-biker who enjoyed participating competitively in the sport. This meant that several evenings a week were dedicated to training, which diverted his attention from his PhD. Neither of these contexts (academic and personal) exists in isolation from one another – nor can they - because the common denominator in both is the doctoral candidate, and 65 therefore the two are ‘connected’ through her. How she manages her participation in both will decide the extent of the conflict. Therefore, Wenger’s contentions that participating in more than one community of practice serves only to positively develop the individual, and that each community is largely cut off from the other, seems to be idealistic at best. Thus, it seems to ignore the fact that different communities are fundamentally interconnected as a result of having shared membership; that is, communities intersect with one another where they share the same members. 2.2.6.2 The contested nature of ‘community’ The contested nature of the term ‘community’ is the source of another limitation identified by critics of the communities of practice theory. Lindkvist (2005), for instance, contends that the concept of ‘community’ has been a ‘…cornerstone of sociological thinking for nearly two centuries’ (2005: 1192). During this period, it has taken on a number of different meanings, all of which reflected the philosophy of the period. For instance, Ferdinand Tönnies (1855 – 1936) , a German sociologist, developed the twin concepts of gemeinschaft (community – ‘natural, organic forms of group existence’) and gesellschaft (society), defined as an ‘… artificial group which is held together by some common, conscious purpose’ (Wirth 1926: 415) to try and explain the term. For Tönnies, gemeinschaft was characterized by a small group of people living in close proximity to one another in a village (Agrawal & Gibson 1999: 633) and in which the community ‘…grows out of the organic relationship of man to his environment and those natural, involuntary bonds that inevitably grow up between human beings and between groups’ (Wirth 1926: 416). It ‘… referred to the old better days, to be found in village and small town settings where enduring social relations of intimacy and solidarity could prevail’ (Lindkvist 2005: 1192). This understanding of the term community bears some similarity to Wenger’s conceptualisation of the term despite the context in which it was written (nineteenth century rural Germany). Specifically, the notion of spontaneously-occurring bonds between individuals as a result of their common interest or proximity echoes Wenger’s description of the bonds that develop between claims processors during the course of their daily interactions at work. 66 By contrast, gesellschaft is an artificial construct in which individuals are drawn together for a specific purpose – the ties that bind them are engineered for a particular end, thus creating only an imitation of a community. It is this gesellschaft that best describes a community of doctoral students, since the individuals involved – both the doctoral candidates and their supervisors – have been brought together in a specific context for a common purpose: research. There is nothing organic about the process. Thus, my contention is that the term community is not a natural coming together of people with similar interests (as suggested by Lave and Wenger), but is rather an artificial formation with a specific goal. Once the goal has been achieved, the community changes, or even dissolves: doctoral students graduate and move on, new students register and different supervisors arrive on the scene. A new community with new dynamics, in other words. Another conceptualization of the term community is offered by Amitai Etzioni, an eminent sociologist, who suggests that Community is defined by two characteristics: (1) A community entails a web of affect-laden relations among a group of individuals, relations that often crisscross and reinforce one another (rather than one-on-one relations or chains of individual relations); and, (2) community requires commitment to a set of shared values, norms, and meanings, and a shared history and identity – in short, a shared culture (Etzioni 1996:5). This understanding contradicts the assertion by Bell and Newby (1974: xliii) that a satisfactory definition of the term community has never existed. These authors, in an earlier work, pointed out that ‘…over ninety definitions of community have been analyzed and… the one common element in them all was man!’ (1973: 15). Bauman (1994), by contrast, offers a definite – if somewhat cynical – understanding of the community: …those larger than life groupings into which people are born, only to be held inside them for the rest of their lives by the dead hand of tradition strengthened by collective surveillance and blackmail. … [E]ach one of us is born into a certain tradition and language, which decides what to see before we begin to look, what to say before we learn to speak, what to consider important before we start weighing things against each other, and how to conduct ourselves before we start pondering the choices. Thus in order to know what we are… we must fathom and consciously embrace that tradition; and in order to be 67 ourselves, to keep our identity intact and waterproof, we must support that tradition with all our heart. In fact, we owe it our complete loyalty; and we ought to offer its demands an unquestionable priority whenever loyalty calls in that society… (Bauman 1994: 24). Thus, for Bauman, individuals do not have a choice when it comes to community membership; rather, it is decided for them at birth. Put another way: the community into which one is born is the community in which one remains for life. His argument that an individual’s identity is interconnected with that of the community is an interesting one because it suggests an inevitability of belonging that does not necessarily exist; that is, that the individual has no control over her membership. However, many people including celebrities, such as Madonna and so-called ‘black diamonds’24 It is not an easy matter to build one’s own identity relying on one’s own guesses and hunches alone. And there is little reassurance to be drawn from the self- made identity if it has not been recognized and confirmed by a power stronger and longer-lasting than the solitary builder. Identity must be seen as such; the dividing line between a socially accepted and an individually imagined identity is one between self-assertion and madness. This is why we all feel time and again an overwhelming ‘need of belonging’ – a need to identify ourselves not just as individual human beings, but as members of a larger entity. That identity-through-membership is hoped to provide a firm foundation on which to erect a smaller and feebler personal identity. As some of the old, once solid, entities underwriting and endorsing individual identities lie in ruins, while such as mining entrepreneur Patrice Motsepe, have chosen to break out of their ‘birth communities’ and to reconstruct their identities according to a different set of values. For Bauman, the fact that individuals are expected to offer their absolute loyalty to this entity is problematic. He seems to question the reciprocity of this loyalty – in other words, does the community offer its members the same level of commitment as it expects from them? Therefore, from this perspective, the concept of community could be understood as something rigid and controlling, subsuming the identity of the individual into a larger group identity. The only way that one can remain ‘intact and waterproof’ is to conform. On the question of individual identity within the community, he makes the following argument: 24 ‘Black diamond’ is the term used in South Africa to describe a new middle class of affluent, wealthy African people. 68 others are fast losing their holding power, there is a demand for new ones, able to pronounce authoritative and binding judgments (Bauman 1994: 23; italics added) This is an interesting line of reasoning because Bauman does not seem to differentiate between the identity of the members at the core and that of those at the periphery – membership (and hence, identity) is undifferentiated. What is key in this understanding of the community is what Bauman perceives as a fundamental human need for a higher authority within that community which can and does make ‘binding judgments’ on its members – some form of regulation and recognition, in other words. Applying this understanding of community to Lave and Wenger’s theory, and to the idea of a community of practice of doctoral candidates would be problematic because the latter is differentiated (both in terms of prior experience and of the stage of research that each member has reached). Furthermore, I would question the need for a ‘higher authority’ to issue ‘binding judgments’ on the grounds that doctoral candidates are sentient adults - while they may not be as self-regulated or self-motivated as supervisors and institutions might like them to be, they are nonetheless capable of making decisions about their research without the need for direction. By the same logic, should they require such direction, they have the agency to ask for it. This understanding of community is in contrast to that of Lave and Wenger; these authors recognize that in any community, there are individuals at the core, at the periphery, and at any number of positions in between these two poles. Their identity is defined by where they are situated at any one time, and it changes as they move towards the core. Thus while identity is a component of community membership, it is neither static nor rigidly defined by the group. In addition, the assertion that the ‘regulation’ of the community is controlled by those at the core, and accepted by those at the periphery is problematic, since it implies that the latter unquestioningly submit to the directions of the core members. I would suggest that the process is far more complex than that, and involves significant amounts of negotiation amongst all members, with those at the core clearly having greater influence. The viewpoints described above represent only a sample of the ways in which the term community has been understood over the last century. Clearly, there is little agreement 69 on a) what constitutes a community, and b) what the role of a community either is or should be in the social interactions of individuals. Further, the understandings of the construct of community that I have presented here reflect only a selection of the scholarship on communities, but should offer an idea of the breadth of debate around it. For the purpose of my research, what is important to understand is this: because there is no universally-accepted understanding of the term ‘community’, there can logically be no universally-accepted understanding of the phrase ‘communities of practice’ either. Stated another way: if one word in a term is contested, then it is reasonable to question the meaning of the entire term. Since ‘community’ can mean so many things in so many contexts, and even though Lave and Wenger has attempted to pin it down, the lack of consensus nonetheless undermines a common understanding of the communities of practice concept. This, in turn, has implications for the use of the term ‘community of practice’ in relation to postgraduate supervision – for instance, does a group of chemistry students, working on different aspects of a central project, represent a Gesellschaft (because the bonds are imposed, not spontaneous) or a Gemeinshaft, since such bonds are a natural, inevitable result of laboratory relationships? Can this context be considered to be a community, simply because there are people involved, as suggested by Bell and Newby? Or is Bauman’s view most appropriate – that these students are metaphorically ‘born’ into the laboratory community, and have no say in their membership? Each of these understandings offers a different angle on what a community of practice in this setting means. This lack of consensus undermines how the concept can be applied to postgraduate supervision, and how doctoral candidates view their experiences of working in a group of like-minded people on a joint enterprise. 2.2.6.3 The disputed nature of participation One of the key elements in Lave and Wenger’s communities of practice theory is that of participation (see section 2.2.5 of this chapter). According to these authors, community members naturally move from peripheral participation – that is, from the margins of the community – to the centre over a period of time and apprenticeship. However, Lea (2005: 184) challenges the assumption that the transition from peripheral 70 to full membership of the community is as ‘simple and smooth’ (p. 184) as Lave and Wenger imply. Rather, she argues that This perspective does not take account of the more contested nature of participation in communities of practice, that is when participants are excluded from full participation in the practices of a community; for example, where students struggle to engage in the unfamiliar discourses or literacy practices of the academy, always feeling excluded and on the margins (Lea 2005: 184). This critique echoes a point made earlier in this chapter that black students – both undergraduate and postgraduate – are often unable to access the required discourses and literacy practices of their disciplines, and therefore remain largely on the periphery, unable to participate in their core activities. If one constructs an academic discipline as a community of practice – as I do – then it is clear that in this context, there is no natural or inevitable transition from the periphery to participation in the core. In other words, the marginalized students on the periphery do not magically ‘come to know’ the required discourses of each discipline as they move between semesters and years of study, as Lave and Wenger’s theory suggests. Furthermore, and implicit in the act of participation, is the role played by power – specifically, that which is situated in the hands of ‘old-timers’ (those who are already at the core). In this respect, Handley et al note that within any community: … full participation may be denied to novices by powerful practitioners… Constraints on newcomers may be strongest if the latter threaten to ‘transform’ the knowledge and practices of the extant community, since that knowledge is important or ‘at stake’ to the full participants who have invested in it (Handley et al 2006: 644). This situation is a familiar one in South Africa higher education, where the full participants of the academic community (generally the minority white population) have in the past denied full participation rights to (generally black) members on the basis that the latter represent a threat to established knowledge in the community. Thus, there has been for instance, strong opposition to the transformation of education, which has included the Africanisation of curricula25 25 The concept of Africanisation of education has been around since the 1960s, when decolonization of the continent began. It is described by Makgoba (1997) as ‘… the process or vehicle for defining, interpreting, promoting and transmitting African thought, philosophy, identity and culture. It encompasses an African mind-set or mind-set shift from the European to an African paradigm’ (1997: and the recognition of indigenous knowledge 71 systems26 as legitimate elements of a post-apartheid education system. Much of this resistance has originated from those who wanted to maintain the status quo – apartheid’s ‘old guard’/old-timers who seem to embrace ‘mental colonialism’ (Muendane 2006: 54) and ‘knowledge apartheid’27 A further critique of the notion of participation in a community of practice is that it is neither linear nor inevitable – that is, students/apprentices do not necessarily move from the periphery to the core as a matter of course. As Roberts (2006) observes, ‘… peripheral community members may not necessarily develop beyond a position of peripheral participation’ (2006: 627). This may be as a result of historical causes (such as the limitations imposed on black learners by apartheid education) or an absence of commitment to the community on the part of the participant. While Lave and Wenger (1991) recognize the former - that access to learning opportunities (and by implication, to participation in the core activities of the community) is ‘…subject to structural constraints and control’ (Fuller et al 2005: 54) – they fail to address it in any substantive way. This anomaly undermines the broader theory because of the implicit assumption that all participants a) have equal access to opportunities for movement ( Odora Hoppers 2005:14). In other words, those who wish to maintain the dominance of ‘white’, Western knowledge at the expense of African ways of knowing. 203). Horsthemke (2006) broadens this understanding to include Afrocentrism. In his words: ‘Afrocentrism does not simply mean teaching students about Africa, its history, cultures, philosophy and values. It means ‘placing Africa at the centre’, historically, culturally, philosophically and morally (Schiele 1994, 152; Ani 1994). It encompasses the view that Africa is the cradle of humankind and the locus of the first great civilizations from which all others derive (Asante 1980, 45; Asante 1987, 170; Van Sertima 1999; Seepe 2000). It teaches that Africa is the birthplace of technology, metallurgy, astronomy, mathematics, agricultural science and medicine (Asante 1980, 45; Van Sertima 1999; Seepe 2000).’ (Horsthemke 2006: 450). Essentially, then, Africanisation in the context of education demands that the contribution to knowledge made by Africa and its people must be recognized, and should not be subsumed by dominant Western philosophies of education which valourise the contributions to knowledge of the global North. 26 In a 1999 discussion document entitled Indigenous knowledge and the integration of knowledge systems: toward a conceptual and methodological framework, Catherine Odora-Hopper defines indigenous knowledge systems (IKS) as follows: ‘…[T]he word indigenous refers to the root , something natural or innate (to). It is an integral part of culture. By indigenous knowledge systems (IKS) are meant the combination of knowledge systems encompassing technology, social, economic, philosophical learning/educational, legal and governance systems. It is knowledge relating to the technological, social, institutional and scientific development including those used in the liberation struggles’ (Odora-Hopper 1999: 2). 27 Odora-Hoppers uses this term to describe a situation ‘…when textbooks and formal institutions designated to produce and legitimize knowledge become cognitive regimes that acknowledge only the victor, and defeated knowledges are erased and condemned as unscientific’ (2005: 17) 72 from the periphery to the core, and b) inherently want to move from one to the other (Handley et al 2006: 644). Furthermore, argues Lea in a later article, what the theory fails to recognize is that participation often takes place in a ‘… multiplicity of communities of practice within the academy’ (Lea 2004: 741), choosing instead to focus ‘…on the novice student acting as an apprentice moving towards full membership in the wider university community’ (Lea 2004: 741). Thus, instead of a ‘contextually-based approach’ (Hirst et al 2004: 67) that involves ‘…the introduction of students to the conventions and genres of particular disciplines as an integral part of teaching within that discipline’ (Reid & Parker 2002: 23), students are exposed to a ‘one-size-fits-all’ approach. In the latter, the assumption appears to be that the provision of a generic set of skills will equip each student with the ability to fully participate in the academic community. However, since the academy consists of numerous communities in which students can participate – not just a single monolithic one – this assumption is clearly flawed. Furthermore, participation in any community of practice implies the existence of power, which Lave and Wenger recognize. They note the potential for the misuse of this power in the community, particularly by the old-timers: ‘Hegemony over resources for learning and alienation from full participation are inherent in the shaping of the legitimacy and peripherality of participation in its historical realizations (Lave & Wenger 2003: 42). Despite this, Contu and Willmott (2003) have questioned how …adequately Lave and Wenger conceptualise power, and how adequately they incorporate their understanding of power into the analysis of learning as a situated practice (Contu & Willmott 2003: 286). While Lave and Wenger have recognized that their understanding of power ‘…is left largely as an intuitive notion’ (Lave & Wenger 2003: 42), the absence of a systematic and clear understanding of term problematises the entire notion of communities of practice because it weakens the foundation upon which the theory is built. In other words, since the concept of participation in communities of practice is at least partially premised on the element of power, and since this element has seemingly been glossed over by Lave and Wenger, it is possible to question the validity of communities of 73 practice theory with respect to participation. This is particularly evident in any community of researchers, at least one of whom is the designated supervisor. Thus, there is a level of disparity in the sharing (and understanding) of knowledge in the community. As Northedge (2003) notes, Typically, within academic debate, well established members of the knowledge community exchange views, while others, participating vicariously, construct the flow of meaning in terms that work for them. Of course, these different understandings are not equal in status. The way leading members understand and speak the discourse sets standards for the community as a whole. Better established members will tend to converge more closely to the community’s dominant usage and understanding, while more peripheral members, including newcomers, are more variant in their usage and understanding (Northedge 2003: 21, italics added). Thus, the more established that one is in one’s community – in terms of discourse knowledge – the more powerful one is considered to be. Foucault (1975), too, recognizes the fundamental connection between power and knowledge. For him, the two concepts are intertwined, and one cannot talk about one without talking about the other. In Discipline and Punish: The Birth of the Prison, he argues that We should admit… that power produces knowledge (and not simply encouraging it because it serves power or by applying it because it is useful); that power and knowledge directly imply one another; that there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations… In short, it is not the activity of the subject of knowledge that produces a corpus of knowledge, useful or resistant to power, but power- knowledge, the processes and struggles that traverse it and of which it is made up, that determines the forms and possible domains of knowledge (Foucault 1975: 27) Thus, wherever there is knowledge – be it Western, African or any other ‘variety’ – there is also, by implication, a power relationship and vice versa, between those who possess it and those who do not. My interpretation of Foucault’s argument is that knowledge does not (and indeed, cannot) exist in isolation from power and power relations. If an individual has undertaken research in a particular field over a period of time, and is thus au fait with the works of key thinkers in it, then she can be seen as being powerful in that field because she is able to transform her knowledge into something more advanced. It is in this ability to transform (and add to) knowledge that 74 she wields power. The reverse also holds true: where there is power, it will inevitably reside in an individual’s knowledge about something. Thus, those who have power – the community’s ‘old-timers’ described by Lave and Wenger – do so because of the accumulated mass of knowledge that resides in them. Put differently, they hold the power to direct the activities of the community by virtue of the fact that they have, over the years, developed an extensive knowledge network through their participation in that community. What Foucault is arguing – and which seems to contradict Lave and Wenger’s conceptualization of participation in communities of practice – is that anyone who possesses knowledge also possesses power. Thus, even those who participate on the periphery of a community who have only a fundamental knowledge of the workings of that community would, according to Foucault, have a small measure of power. By contrast, Lave and Wenger relegate those on the periphery to a position of relative powerlessness which does not change until they have served their period of apprenticeship. 2.2.6.4 A critique of the master-apprentice model The literature that looks beyond the communities of practice theory is also highly critical of the master-apprentice notion put forward by Lave and Wenger. At the most fundamental level, argues Lindkvist, Relevant knowledge resides in practice, not in the master… Accordingly, there is an underlying notion of a coherent, communal frame of reference and value system that has to be acquired by the individual in order to become a full member… Vital knowledge resides in practice, in the system of activities and the tacit, communal background knowledge contained in the practices and narratives of the community (Lindkvist 2005: 1196, italics added). Brown and Duguid take this one step further by suggesting that knowledge is ‘… not only revealed in practice. It is also created out of practice’ (Brown and Duguid 1998: 95). Thus, when individuals work together in a community – for instance in a chemistry laboratory – they possess a background knowledge that they have developed over time, and during interactions with others in the same community. They all have stories of their experiences in the community, as well as what they have learnt from these experiences. It is arguably in the dissemination of these narratives that knowledge is passed on. Such dissemination can be bi-directional: from the supervisor to the doctoral 75 candidate, and vice versa. Thus, although the research frame of reference and value system may remain the ‘property’ of the supervisor, the exchange of knowledge between supervisor and candidate flows in both directions. Thus, the supervisor can (and often does) learn as much from the candidate’s research as the candidate herself. Another criticism of the master-apprentice model is offered by Grant (2008, 2009), who argues that the traditional configuration of supervision as an apprenticeship fails to ‘…shed light on the troublingly asymmetrical institutional architecture of supervision’, nor does it sufficiently elucidate the ‘… ideas of layered relations and competing discourses’ (2009: 129). In other words, the master-apprentice model does not adequately describe the relation between the two individuals. Instead, she suggests that German philosopher Hegel’s account of the master-slave relation is a more apt way of viewing …the struggle for recognition that accompanies the emergence of self- consciousness. Premised on the ‘existence of conflict in the ongoing process of the determination of social identity (Burns 2006: 101), the trope of master and slave shows how two consciousnesses become bound together in an ambiguous and contradictory relation of domination and subordination. Yet, productively, knowledge of the self and the world is motivated by the inter-subjective desires mobilized through this relation. Thus, in Hegel’s view, the master-slave relation is a necessary form of mutual struggle and dependence (Grant 2008: 12). What Grant appears to be suggesting is that the apprenticeship model is an inappropriately tame representation of the supervision process. Rather, following Gurevitch, she argues that the dialogic nature of the relationship means that it contains ‘elements of opposition, coercion, fear and struggle…’ (Gurevitch 2001: 89). It is only from the interaction of these elements that self-consciousness (or knowledge) can be produced. Grant explains it as follows: The emergence of self-consciousness requires this difficult relation as knowledge of the self and the world is motivated by its very desires and struggles. In theorizing supervision as analogous to the master-slave relation, I draw a parallel between the processes of emergence for a general state of self- consciousness and those for the disciplined self-consciousness of the scholar/researcher (Grant 2009: 129). Furthermore, she argues, the master-slave relationship is not linear (that is, only between master and slave). Rather, it is a ‘… complex, triangular relationship: the 76 master’s relations with the slave are mediated by the things of the world, and his relations with the things of the world are mediated by the slave’ (Grant 2008: 12). Applying this triangular relationship to postgraduate supervision, Grant points out that the latter consists of the supervisor, student and thesis, ‘… and likewise, the supervisor’s relations with the student and the thesis are not direct but mediated’ (Grant 2008: 13). Thus, the relationship is not just between individuals (as the master- apprentice approach seems to imply), but includes the written text as a mediating influence. Grant concludes by suggesting that the domination, subordination and conflict that exist within a master-slave relation is beneficial for the supervision process because it produces ‘…in both supervisor and student the committed (obedient and diligent) work required to get the research and thesis finished’ (Grant 2009: 130). While she recognizes that the conceptualization of supervision as a master-slave relation might be offensive to our ‘modern liberal consciousness’ because of its connoted links to the repression of the colonial era (2009: 130), she is adamant that supervisors need to confront the ‘troubling ambiguities of supervision’ that are represented by this master- slave model of postgraduate supervision (2009: 130). Grant’s conceptualization contrasts sharply with the master-apprentice approach which does not appear to recognize the potential for conflict between the two parties. Rather, the apprentice/student is conceptualized as willing and able learner, and the master/supervisor as a benevolent source of knowledge and skills. My interview data will demonstrate that this is not always the case. Grant’s perspective also echoes that of Foucault, who understood power to be ‘relational and productive’ (Fox 2001: 858). In the context of postgraduate supervision, this could be interpreted to mean that the power-laden interactions between supervisor and students could produce the kinds of conflict that in turn result in knowledge construction. 2.3 Conclusion The aim of this chapter was to introduce the contextual and theoretical framework of the thesis. Commencing with an exploration of the thesis’ key elements (the doctoral 77 qualification and the process and practices of supervision), it proceeded to situate the latter in the sphere of pedagogy. However, I argued that pedagogy alone – as both an act and a discourse – was not a sufficient descriptor of the act(s) of postgraduate supervision. There was a political and moral element to it, too, which was best encapsulated by the concept of critical pedagogy. This variation asserted that there is a connection between knowledge and power which needs to be harnessed in order to achieve transformation. In the context of doctoral supervision, candidates have the opportunity to construct and develop knowledge at a high level, knowledge that can be used to elicit this change. I subsequently reviewed some of the literature on discourse, and in particular, on Grant’s (2005) four key discourses of postgraduate supervision, namely psychological, traditional-academic, techno-scientific and neoliberal. Each represents a particular construction of the supervision process and challenges candidates, academic supervisors and institutions alike to reflect on whether (or how) their ways of thinking and speaking about supervision have a bearing on their experience of the process. The chapter also highlighted the tensions that exist between proponents of the communities of practice model of understanding postgraduate supervision, on the one hand, and those who contest the efficacy of the model, on the other. Adherents of communities of practice contend that as a social learning theory in which apprentices (doctoral candidates) learn at the metaphorical feet of the knowledgeable master (supervisor), the model accurately captures the essence of the postgraduate supervision experience. Its critics, by contrast, assert that communities and the individuals that constitute them are far more complex than Lave and Wenger present them. They propose that the term ‘community’, the nature of participation and the idea of communities of practice as sites of learning need to be critically explored and theorized in more depth before the model can be validly applied to the learning context. I recognize the veracity of both sides of the debate, but do not favour one over the other. They both add value to my research data, but in different ways. 78 CHAPTER 3 LITERATURE REVIEW (Part 2): The ‘how’ and the ‘who’: mapping the route and destination Chapter overview Much of the work on postgraduate supervision is contained within so-called ‘How-to’ manuals; that is, texts that offer both supervisors and postgraduate students a step-by- step directory for each stage of the supervision process. The first section of this chapter will survey a range of these texts and explore the contribution that they have made to the field of postgraduate supervision. Since these books do not exist in isolation from broader national guidelines, the chapter will subsequently review the guiding principles developed by South Africa’s Council for Higher Education (CHE) for the process of supervision. Finally, the issue of the identity formation of postgraduate students within (and beyond) these texts will be examined since the contribution to knowledge made by these academics is inextricably linked to their sense of who they are and where they locate themselves intellectually in their specific fields of study. 3.1 Introduction In this section of the literature review, I will explore the role of the thesis/dissertation manuals in advancing ‘good practice’ (however this is defined) in supervision. These texts often support both supervisors and students through the general research degree process, but offer little with regard to the unique relationship-specific dynamics that characterize each individual process. Thus, while they offer some useful, if somewhat generic, tips for the practice of supervision, they cannot be held up to be the definitive source of wisdom about a process that is essentially about unpredictable relationships between two or more unique individuals. In this chapter, I will examine a number of these texts and the extent to which they contribute to improved supervision skills. More specific than these manuals – although still reminiscent of the ‘one-size-fits-all’ approach alluded to earlier - are the guidelines developed by both my institution, the University of the Witwatersrand (Wits), as well as the national Department of Education (DoE). Both of these role-players have a vested interest in providing comprehensive support structures for postgraduate supervisors because it is the latter group of individuals who are effectively preparing the next generation of academics and, by implication, contributors to the international knowledge economy (cf. section 2.3 in Chapter One). Wits’ response to this state of affairs was to develop and disseminate an extensive set of institution-specific guidelines for its supervision staff. 79 Entitled Strategies for Postgraduate Supervision, and co-written by various senior professors and staff development practitioners at the University, it serves as a first ‘port of call’ for all supervisors at the institution. While some of its content has already been discussed in the previous chapter (section 2.2), I will use this chapter to scrutinize how this document differs from the At the national level, numerous documents such as the Draft National Plan for Higher Education in South Africa and ITL Resource No. 7: Postgraduate Research and Supervision have been produced by the DoE with the intention of establishing principles for good supervisory practice at South African universities. I will survey the substance of these and other national documents in order to determine whether their stated intentions are consistent with the existing realities of postgraduate supervision in this country. Whether they realize it or not, a significant part of postgraduate students’ identities are defined by these institutional and national documents. However, there is more to scholarly identity than simply what HEIs and government departments project. Identity formation, particularly at the doctoral level, is mediated by texts, and participation and engagement with other scholars. It is thus a social construction. The final section of this paper will consider how doctoral candidates develop their scholarly identity beyond and outside the boundaries of technicist documents (such as those mentioned in the previous paragraphs) in their pursuit of the advancement of knowledge. 3.2 Supervision guidelines As mentioned in the introductory chapter, much of the literature on postgraduate supervision is under-theorised. It is mainly concerned with the mechanics of postgraduate supervision (Bartlett and Mercer 2001; Booth, Colomb and William 1995; Craswell 1996; Delamont, Atkinson and Parry 1998; Evans and Green 1995; Gullan 1996; Hart 2001; James and Baldwin 1999; Knowles 2003; Leder 1995; Murray 2002; Phillips & Pugh 1989; Remenyi & Money 2004; Shannon 1995; Singh and Martijntje 2003; Walker & Thomson 2010; Wellington 2010; Wisker 2005; Yeatman 1995; Zhao 2003). These authors mainly dwell on the technical aspects of how one ought to progress through the postgraduate degree. For instance, they discuss how the student 80 ought to go about developing a research topic, how s/he ought to put together a literature review, how the student ought manage her/his time effectively, and so on. These guidelines are often accompanied by checklists against which both students and their supervisors can compare their progress (or lack thereof). Phillips and Pugh’s 1989 book How to get a Ph.D, is a classic example of such a guide. The first four chapters of the book deal with the bureaucratic and psychological issues associated with doctoral study: how to become a PhD student; how to choose the appropriate institution and field of study; the meaning of a doctorate; and how not to get a PhD. The latter includes such issues as a lack of commitment to the study, under- or overestimating what is required for a PhD, and not having a supervisor who understands the requirements of the degree. Subsequent chapters deal with the more structural or systemic issues related to the thesis: how to do research, format the thesis, define/refine short and long term goals, manage one’s supervisor, deal with the bureaucracy, and survive in a ‘predominantly British, white, male, full-time academic environment’ (1989: 131). While the book does provide some useful information for PhD students, it is not an appropriate resource for South African students for a number of reasons. Firstly, it works from the assumption that the majority of PhD students at any university are white males. In the British context (from which the book was written), this may well be the case. In South Africa, however, the demographics are substantially different, with female, black and foreign (mainly African) students making up a large portion of the doctoral student body28 Secondly, because this book was written by Britons for Britons, the conceptualization of the role of a PhD in the life of the university is significantly different from that of the South African context. According to the authors, a Master’s degree is a professional qualification - a ‘licence to practice’ that ‘marks the possession of advanced knowledge in a specialist field’ (1989: 16). The PhD, on the other hand, is a ‘licence to teach’ (italics added) at a university as a faculty member (1989: 17). In South African . 28 According to Bawa (2007), 23% of doctoral students at South African universities are foreign nationals, most frequently from other African states. Furthermore, from 2003 data, he points out that ‘…of the 8112 doctoral students, some 30 per cent were African South Africans, 5 per cent were ‘coloured’ South Africans, 9 per cent were Indian South Africans, and 56 per cent were white South Africans’ (2007: 212). In terms of gender, Bawa notes that ‘African South African women’ constituted approximately 9% of total doctoral enrolments in 2003, compared to 21% of African men (2007: 213). 81 universities, this distinction has not yet been made in all Faculties. At the University of the Witwatersrand, for instance, I am aware of a number of academic staff members with only an Honours degree who are teaching students up to the third-year level. The reasons for this are complex, and fall outside the scope of this study. Suffice it to say that since higher levels of quality education were denied to the majority of the South African population for so long, the country has a dearth of PhDs; if we were to adhere to Phillips and Pugh’s (rather dated) distinction, the future of many of South Africa’s tertiary institutions would be threatened. Delamont et al (1997) identify the primary aspect of supervision as being the development of a good working relationship between student (‘supervisee’) and supervisor. They argue that: Relationships have to be worked at, and discussed, because most of the problems stem from a failure to set out the expectations both parties have for the relationship…As the needs of the student will change over time, the ground rules of the relationship may need to be renegotiated periodically, but it is most important to set up guidelines early on so the student knows what to do, and how to work with you. (1997: 15) The authors subsequently offer neatly packaged ‘guidelines for a good relationship’, which include ‘the best time of day to meet’, ‘scheduling the meetings’, ‘an agenda’, ‘mechanics: confirmation and cancellation’, ‘the annual cycle’ and ‘mutual availability’. While these guidelines offer some structure for a productive relationship, I would argue that they are overly prescriptive because they fail to offer flexibility in terms of the research process. For example, there are arguably times when an agenda for the supervision meeting is superfluous: either the supervisor and student are both already focused on the same thing and hence have no need for an agenda, or the issue/s to be discussed are not of a strictly academic nature; that is, personal issues impacting on the research process, such as divorce or a death in the family. These should not have to be formalized on an agenda. This is not to say that such guidelines are not important. However, by enforcing a strict adherence to them, one might end up stifling the student’s creative process by denying her the opportunity to introduce new ideas into the conversation. Further, one needs to bear in mind that each student has a different relationship with her supervisor; similarly, one supervisor may use different techniques 82 of supervision with different students, depending on the level of guidance required by each individual. Thus, the prescriptive guidelines set out by Delamont et al, while a good idea in theory, may be too inflexible in practice to be effective. In subsequent sections of the book, and in keeping with their implicit position that the supervisory process is a fixed, linear one, the authors propose a ‘checklist for early supervision sessions’ (1997: 37): There are six main criteria to bear in mind, and you should discuss your topic, bearing all six criteria in mind, with your supervisor as soon as s/he is allocated to you. These six criteria are enjoyment, timetabling, thesis length, feasibility, methods and theoretical perspective. Each of the six criteria includes a description, as well as advice about how that particular criterion can be achieved. For instance, concerning the topic of enjoyment, they encourage the ‘good supervisor’ to probe the student’s interest and motivation in the research of a particular topic. Furthermore, the student is expected to undertake a self-evaluation of her/his commitment to the research topic (1997: 39). Supervisors are also alerted to the following: It is important not to confuse your own enthusiasm for a research topic with a student’s. Beware of attributing your own intellectual and personal agenda to your graduate students. Equally, one’s own agenda should not be allowed to pour cold water on students’ research plans inappropriately. (1997: 39) From the perspective of a novice supervisor, this is a useful book. It contains important explanations of how one goes about supervising the postgraduate student as effectively as possible, although the requirements are largely administrative in nature. However, and in common with the other literature reviewed in this section, the content is not grounded in any theoretical tradition. While Delamont et al do mention the concept of apprenticeship, it is used to describe the relationship, rather than to theorise about the concrete practices engaged in by the master and the apprentice during their time together. This critique is equally true of Wisker’s The Good Supervisor: Supervising Postgraduate and Undergraduate Research for Doctoral Theses and Dissertations 83 (2005). Like Delamont et al, Wisker describes the technicalities of supervising postgraduate (and undergraduate) students, in a theoretical vacuum. As a result, she is able to make such sweeping observations as: In order to develop students as researchers, we need to help them focus on asking research questions and exploring fields of study in critical, problem- solving, creative ways, matching this with a dedication to organization and planning and achieving coherence and clarity of expression (p. 5) Research is the fundamental human learning activity, involving enquiry, problem-solving, diversity, flexibility and decision-making. It encourages and enables the development of creative thinking, problem-solving strategies and abilities which in turn help others to approach everyday life as well as professional, political, local, national and international questions and issues. (p. 5). The structure of the chapters suggests that the supervision process is a purely linear one, without any recursive stages: starting with the establishment of a good working relationship with one’s student, continuing with the development of research processes and practices (such as crafting a viable research topic, developing an understanding of the relevant conceptual frameworks and methods as well as the design of the study and research ethics). Wisker subsequently examines issues that emerge from the relationship between the students and the supervisor, including dealing with difference (such as gender, race and learning style), encouraging students to interact with one another in formal or informal seminars, conferences, peer groups and so on, and developing a relationship with distance or international students. In addition, Wisker devotes an entire chapter to the discussion of how one supervises one’s colleagues or individuals currently engaged in professional practice. The chapter is aptly titled “A Little Too Close to Home: Supervising your Colleagues and/or Other Practice/Professional-based Research”. Again, however, this chapter deals only with the technicalities of the process, and not with the broader range of practices and experiences that characterize the supervisor-student relationship. The book concludes with a section on the actual writing up of the research, once the data has been collected. This includes helping the student to prepare for his/her viva, something rarely practiced in South African universities. 84 Within the genre of manuals that deal predominantly with the structural requirements of the thesis or dissertation, there is a sub-category of works that go beyond just ‘tips for supervisors’. Rather, they provide models for supporting postgraduate students in their research journeys. One such book is Kamler and Thomson’s Helping Doctoral Students Write: Pedagogies for Supervision (2006), which offers well-designed advice for guiding the doctoral candidate through the thesis-writing process. Their point of departure is that writing is not something that happens when the research is finished, but is rather an integral part of the whole research process, right form the start: Researchers keep notes, jot down ideas, record observations, summarise readings, transcribe interviews and develop pieces of writing about specific aspects of their investigation. These writings are not simply getting things down on paper, but are making meaning and advanced understandings through these various writings. Then there are public texts – conference papers, articles, and the thesis itself – all of which do productive work. It is through these writings that researchers produce knowledge and become members of their various scholarly communities (2006: 3) The remainder of the book provides valuable guidance (to both the student and supervisor) on the writing process itself. For instance, it has a chapter entitled ‘Reconsidering the personal’, which demonstrates how students can (and should) write their chapters in the first person (that is, using ‘I’ as the authority). Referring to the work of Giltrow (1995), Kamler and Thomson make a useful distinction between the personal ‘I’ and the discursive ‘I’. Both are linked to identity (discussed in more depth in section 3.4 of this chapter): The discursive I describes the writer in her capacity as writer/researcher. It often occurs with verbs that refer to some discourse action such as : ‘I want to suggest’; ‘I intend to begin’; ‘I shall focus’; ‘I begin with a discussion’; ‘I explore’; ‘I examine’; ‘I evaluate’; ‘I close’; ‘I draw on evidence’; ‘I provide’; ‘Let me conclude’ (Giltrow 1995: 252). That is, while writers of scholarly genres often refer to themselves, the identity signaled by the textual ‘I’ is limited (2006: 64) 85 Insights such as the above are useful for both postgraduate students and their supervisors because they remove a large amount of the uncertainty associated with composing an academic text. This is particularly true when the student is unsure about how to express her intellectual ownership of the research findings. The remainder of the book continues in a similar vein, offering practical advice on research writing. It is a powerful reference for both supervisors and their students when it comes to the writing stage of the research process. Another work that models good practice in supporting postgraduates through the research writing process (as opposed to just tips) is that of Badenhorst. Among her numerous books, one stands out as being most valuable for postgraduate students grappling with the thesis: Dissertation Writing: A Research Journey (2008). In this work, Badenhorst uses the metaphor of an extended journey, filled with milestones, mazes and mountains, to describe the sequence of events that students live through in order to complete their dissertations. She begins the book with reflections on both her own doctoral experience and, once qualified, working with postgraduate students on the writing of their dissertations. This is useful because it foregrounds the specific adversities that doctoral candidates experience along this journey, particularly with reference to the writing. It is also comforting for the reader to know that she is not alone in her struggle to write the thesis, but that others (including a published author) have also laboured hard to achieve this end. According to Badenhorst (2008: 2): I wrote this book for two reasons. First, because I grappled with writing for many years as an undergraduate and graduate student, and second, because over the last eight years of working with Master’s and PhD students, I see how many struggle to negotiate countless explicit and hidden requirements. It is a rare doctoral student who, on completion of a dissertation, says ‘That was a fabulous process, I enjoyed every minute of it!’. Comments are more likely to be: ‘That was like climbing a huge mountain with rocks in my backpack and chains on my ankles.’ Or ‘Those were the worst five years of my life’. Doctoral students can take comfort from these words because they explicitly lay out the complexity of the writing process, and the often negative impact that this has on them as individuals. Badenhorst subsequently develops the argument that in order to write a dissertation – or, indeed, any extended piece of research - one needs to just write, 86 period. One of her key principles is that ‘writing begets writing’, something that she raised in an earlier work, Research Writing: Breaking the Barriers (2007). In the latter, she recommends that authors (in this case doctoral students) undertake regular free- writes in order to get started. She defines this practice of free-writing as … writing non-stop for a period of time (five or ten minutes) without lifting your pen from the paper (to scratch out or think) and without editing or censoring your thinking. This way you by pass your critical, logical left side of the brain and tap into the creative right side of the brain. (2007: 73) By starting the process in this way, the student is able to make some discernible progress, even if this is only at the preliminary, draft stage. Thereafter, the doctoral candidate can revise her writing to fit the specific genre of the thesis, editing out the parts that are irrelevant and adding in evidence and explanation where necessary. Quoting the storyteller Dorian Haarhof’s theory that ‘… you need to look at [writing] revision as an owl eats a mouse. Eat everything, then spit out the bits that you don’t need’ (2007: 118), Badenhorst subsequently takes the reader through the writing revision process to the final product, providing encouragement to the student along the way. At a meta level, it seems that Badenhorst is using her books to supervise the readers through the writing process. Instead of focusing on supervisory relationships or processes – the realm of university-appointed supervisors - she has instead concentrated her efforts on ‘breaking the barriers’ associated with academic writing through the use of focused tasks at the end of each chapter. The aim of these is to guide students through the writing process and to help them overcome any obstacles which might be impeding their progress. The student experience is at the centre of Badenhorst’s book, Dissertation Writing, in which she discusses the stories, or narratives, that students construct for themselves about their research. These stories provide some insight into how the student experiences the work, and what her perception of her progress is; from this, the supervisor can ascertain what (if any) interventions to make in order to support the student. Based on the work of Ylijoki (2001), Badenhorst suggests that four key narratives emerge: the heroic, tragic, businesslike and penal. She suggests that it can be 87 useful to get doctoral candidates to situate their experience in one of these narratives (discussed in more length below) so that they can reflect on what it says about their attitude towards the dissertation: Without fail, when I ask students if their story is similar to [any of the four] there is some rueful shuffling of feet and sheepish grins. Now they begin to open up and explore their attitudes to dissertation writing with a frankness lacking before. I think this self-reflection is invaluable at the beginning of the research so that students can adapt their story to reflect the journey they wish to have. (Badenhorst 2008: 13; italics added) Candidates who tell the heroic tale are ‘convinced that the dissertation is a quest for scholarship and intellectual wisdom’ (Badenhorst 2008: 12). They believe that their dissertation will make a definitive contribution to knowledge, and the ‘heroism’ of the title of this narrative involves attempting to live up to a high standard (or ‘holy grail’) of scholarship, toiling against the criticism of their supervisors, peers and authors that they have read (and with whom they disagree). According to Badenhorst, ‘Students on this path are often stuck for long periods, sometimes paralysed by the task they have set themselves’ (p. 12). The tragic narrative is one in which the student lives in fear of the dissertation because it has the potential to lay bare the student’s biggest fear: ‘The student will be exposed as a fraud’ (p. 13). This feeling of unworthiness often translates into students not completing the degree because they are convinced that they are not good enough to undertake research at this level. Badenhorst suggests that these students rarely ask for help because they perceive this as a sign of intellectual weakness on their part, and so ‘They tend to struggle on alone and eventually drop off the programme’ (p. 13). The tragedy is that this perception may be easily corrected with the right kind of supervision. As its name suggests, the businesslike narrative is one in which the student simply puts down her head and does what she needs to do. In the minds of these candidates, ‘the dissertation contains no mystical features – it is something to be understood and constructed’ (p. 13). In contrast to the tragic narrative, there is no fear in the process for ‘businesslike’ students, nor is there any misconception that their work will be a magnus 88 opus, or widely hailed as such by their scholarly community. Their ‘bottom line’ is the benefits that will accrue to them upon completion of the degree, such as a promotion or a salary increase. Hence, ‘They focus on a goal and are not delayed by emotional or self-esteem issues’ (p. 13). Finally, there is the penal narrative, and as the name suggests, at its core is the notion of punishment. Students who tell this narrative when talking about their dissertation generally do not have an intrinsic interest in undertaking the research, but are ‘externally motivated and need to complete a dissertation for job requirements or because they feel they can’t progress in an academic context without it’ (p. 13). Hence, the dissertation is constructed as being a nuisance and the student does the bare minimum in order to complete it. Almost inevitably, the quality of the work suffers. Badenhorst encapsulates their view of writing the dissertation as follows: [S]uch students view the whole process negatively – the research, the supervisor and the university all fall under this negative gaze. The supervisor doesn’t understand them, the university rules are petty and obstructive, and the whole trip seems pointless. They often get stuck, mired in negativity (2008: 13) Badenhorst argues that whichever one of these broad narratives the student identifies with will also be the way in which she approaches the writing process. It could therefore be argued that it is imperative that supervisors to be aware of their students’ perceptions of their dissertation – heroic, tragic, businesslike or penal – and to question the source of these. Subsequently, they could open up a dialogue with their students about their ‘chosen’ narrative, and how this could be re-written into a more constructive, positive one. I would add a fifth narrative to the four presented by Badenhorst – that of the activist. At the core of this is the candidate’s need for the dissertation to be meaningful and widely-used in the discipline, rather than being relegated, unread, to a dusty library shelf. Candidates who tell the activist tale tend to focus their energies and enthusiasm on the analyses and recommendations of the dissertation because they consider these sections to hold the most potential for effecting positive change in the practices of their discipline. The dissertation thus becomes the means to a much broader end than simply 89 the awarding of a degree: it becomes a catalyst for disciplinary transformation and advancement. All five narratives involve the process of scholarly identity formation. This is the opportunity that doctoral candidates have to create an academic niche for themselves. As Rose and McClafferty (2001) point out, ‘Graduate students… begin to form their own scholarly identities through their choices about what they research, whose work they cite, and how they communicate their own ideas’ (2001: 30). For many, it is a risky undertaking because it has the potential to expose their self-perceived weaknesses. Kamler (2008) encapsulates the sense of risk for the student as follows: Doctoral writers need to learn the sophisticated genres and textual conventions of their fields – but they also need to learn how to take an authoritative stance in a field of expert others, and to assert their contribution to that field before they feel authoritative. Students find doctoral writing difficult because texts and identities are formed together, in and through writing. They feel vulnerable when their work is made public because the text is an extension of the scholar and scholarship; it literally puts the self and the work ‘out there’. (Kamler 2008: 286. Italics added). Doctoral candidates who tell the heroic tale feel up to the challenge of developing a significant scholarly identity: they are prepared to engage with authorities in the field, even if this means disagreeing with established disciplinary canon. By contrast, those who identify with the tragic tale – who are convinced that they will be exposed as an academic fraud – will often have trouble asserting their scholarly identity because they are convinced that they have nothing of value to add to existing debates. Narrators of the businesslike tale are goal-oriented – that is, they are more interested in the degree and its related rewards than with issues of identity. For them, the piece of paper is more important than the impact that their scholarly contribution has had on the academic community. Conversely, those who identify with the penal narrative recognize that an explicit scholarly identity is vital for advancement in their professional contexts, but has little value to them beyond this. Finally, the teller of the activist tale, by definition, is eager to develop a scholar identity that is associated with transformation and action. This person wants to take on the identifying mantle of a ‘doer’, as opposed to a ‘thinker’. 90 My research journey (described in the Prologue of this thesis) has been characterized by an activist narrative with occasional traces of the tragic, businesslike and penal. Which narrative I most identified with depended on at what stage of the process I was. For instance, at the proposal-writing stage, I felt like a complete fraud and unworthy of even considering undertaking a PhD. I was convinced that my work would never be good enough – key features of the tragic narrative. As I began the data collection, though, this was replaced by a businesslike narrative: there was nothing mystical about the hard work of transcribing interviews. I just sat behind my computer and did what needed to be done. Towards the end, as I made my final revisions, I began to feel uneasy about the idea of the thesis mouldering on a library shelf; I wanted to summon academic supervisors to meetings to discuss my findings and how they could best be applied in their contexts. I wanted to write journal articles and present at as many conferences as humanly possible to share what I had learnt – the activist narrative. Recurring questions throughout the research and interviews have included ‘So what? Why is this important? How is it going to change the way that supervisors work with postgraduate students, or how doctoral candidates experience their PhD process? How can I present the data so that people who read it will be galvanised to change their existing supervisory practices, and as a consequence, improve the process for doctoral candidates?’ This approach to the writing of my dissertation could be interpreted as being both positive (a call to action) and negative (advocacy is not always considered to be synonymous with academic rigour). I have chosen to construct it as the former. In the next section, I will examine the guidelines provided to academics by both the institution (Wits) and the national Department of Higher Education and Training (DoHET). Those published by the latter have encompassed a more general one-size- fits-all approach, while Wits has attempted to customize (as far as possible) what it considers to be good practice in the sphere of postgraduate supervision. 3.3 Institutional and National Guidelines The institution at which the research was undertaken – the University of the Witwatersrand – plays a pivotal role in defining the parameters of postgraduate study for its students. These include defining admission requirements, dictating the format of 91 thesis proposals, listing the accepted structure of theses, and so on. A number of individual Schools and Departments also disseminate their own discipline-specific information to doctoral candidates. These resources are designed with the intention of supporting students through the research process, although they seem to be written with the unspoken assumption of a shared understanding of what is understood by terms like ‘doctorate’ and ‘research’. Terms such as these originated centuries ago in Europe and have evolved over time; whether or not this evolution has been reflected in institutional documents is questionable. The most widely-used set of guidelines at the University are to be found in a staff handbook designed for the Wits context, Strategies for Postgraduate Supervision (see section 2.2 of the previous chapter for more on the content of this text). As mentioned in Chapter 2, it contains a skeletal description of what the supervision process ought to look like. It covers much of the same substance as the manuals referred to in the previous section, but in less detail. Where it differs from these texts, however, is that it directs supervisors to supporting documentation and information that specifically pertains to postgraduate study at Wits. These include the University’s Statement of Principles for Postgraduate Supervision, various formal policies including the Grievance Policy, as well as information on the specific support services (such as accommodation and healthcare) available to students during their registration. It is this information, I would assert, that separates this text from those discussed in the previous section and makes it a more appropriate source for staff to draw on for their immediate supervisory needs. Nationally, there has been a concerted effort in recent years to raise the profile of (and hence, the number of enrollments in) doctoral study. This has largely been driven by the steady decline in South Africa’s contributions to international research output, and the parallel dwindling of research funding from overseas. According to Habib and Morrow (2006: 9), … South Africa’s share of global research output has been declining for over a decade, from 0.8 per cent in 1990 to 0.5 per cent by 2001. Independent assessment of total publications by South Africa’s public research sector 92 suggests that the country’s scientific output has been stagnating for the last decade and a half. Put another way, South Africa is not sufficiently competitive in the international research arena to justify foreign injections of funding. Doctoral research (and the publication thereof) is seen as being one potential means of reversing this trend. Hence, the National Department of Education (which since 2009 has been sub-divided into the Department of Basic Education, and the Department of Higher Education and Training) has issued a number of documents, the purpose of which was to improve the standard of postgraduate study in the country – partly through improved supervision practices – and consequently, to raise the country’s international research profile. These documents, including the Draft National Plan for Higher Education in South Africa (2001) and Education White Paper 3: A Programme for the Transformation of Higher Education (1997) will be discussed in some detail in the pages that follow. While neither of them is particularly current, their core principles are as applicable in 2010 as they were in the late 1990s when they were written. The key Higher Education document of the national Department of Education (DoE) is the Draft National Plan for Higher Education in South Africa, published in February 2001. This National Plan consists of sixteen key outcomes29 29 These outcomes relate to higher education in its entirety, and are not solely focused on the postgraduate cohort. that the DoE believes will ‘…give effect to the vision for the transformation of the higher education system outlined in the Education White Paper 3 – A Programme for the Transformation of the Higher Education System’ (p. 4). Essentially, this vision …provides an opportunity and challenge to chart a path that locates the higher education system as [the] key engine driving and contributing to the reconstruction and development of South African society. (Draft National Plan for Higher Education in South Africa February 2001: 4). Of the sixteen outcomes, it is Outcome 14 - Increasing Graduate Outputs at the Masters and Doctoral Levels - that is most relevant for this thesis. Outcome 14 - Sustaining and Promoting Research - responds to Goal 4 of the White Paper, which states that one of the key national goals at is 93 [t]o secure and advance high-level research capacity which can ensure both the continuation of self-initiated, open-ended intellectual inquiry, and the sustained application of research activities to technological improvement and social development (Education White Paper 3 – A Programme for the Transformation of the Higher Education System 1997: 10) The substance of Outcome 14 thus relates to an ‘… increase [in] graduate outputs at the masters and doctoral level’ (Draft Plan 2001: 65) in order to advance the numbers of individuals in this ‘highly qualified group’ (p. 65). The reason for this is a sluggish higher education system that currently produces ‘approximately 4600 masters graduates (or 5% of total enrolments) and 750 doctoral graduates (or 0.8%) annually” (p. 65). More recent figures supplied by the CHE indicate that almost 7900 masters students graduated in 2005, while 1176 doctorates were awarded in the same year (Postgraduate Studies in South Africa: A Statistical Profile January 2009: 10). In essence, both documents point to a graduate study system characterized by relatively small enrolment numbers, high drop-out rates and slow completion rates (Draft Plan 2001: 65). This is further exacerbated by what the CHE refers to as the ‘pile-up effect’: We use the term ‘pile-up’ to refer to the state of affairs where students remain enrolled for their degree for much longer than ‘expected’. When the number of ‘recurring’ students becomes too large, this inevitably puts strain on the resources and affects the efficiency of the postgraduate system in general as it leads to increasingly larger numbers of students who need supervision. We constructed two indicators to measure this pile-up effect: Ongoing enrolments as a percentage of total enrolments and Graduates as a percentage of ongoing enrolments. When there is an increase in the value of the first indicator, it shows that more students are remaining, or ‘piling up’ in the system, while a decrease in the value of the second indicator means the system is producing fewer graduates. (Postgraduate Studies in South Africa: A Statistical Profile January 2009: 11) This ‘pile-up’ of Masters and Doctoral students has a negative knock-on effect for supervision practices due to the overloading of supervisors. It is expressed by the CHE in the following tables: 94 MASTERS (Headcounts) 2000 2001 2002 2003 2004 2005 First enrolments (X) 14162 15888 18062 19352 18279 17398 Graduates (Y) 5795 6426 6871 7396 7536 7881 Ongoing enrolments (Neither first enrolment nor graduate) (Z) 9556 9642 11648 13091 14671 15105 Total enrolments (X+Y+Z) 29513 31956 36581 39839 40486 40384 INDICATORS 2000 2001 2002 2003 2004 2005 Ongoing enrolments as % of total enrolments [Z/(X+Y+Z)]*100 32% 30% 32% 33% 36% 37% Graduates as % of total enrolments [Y/(X+Y+Z)]*100 20% 20% 19% 19% 19% 20% Table 1: Pile-up effects of Masters postgraduate students (Postgraduate Studies in South Africa: A Statistical Profile January 2009: 11) DOCTORAL (Headcounts) 2000 2001 2002 2003 2004 2005 First enrolments (X) 1897 2122 2480 2519 2693 2692 Graduates (Y) 822 843 981 1031 1087 1176 Ongoing enrolments (Neither first enrolment nor graduate) (Z) 3236 3495 4307 4829 5323 5566 Total enrolments (X+Y+Z) 5955 6460 7768 8379 9103 9434 INDICATORS 2000 2001 2002 2003 2004 2005 Ongoing enrolments as % of total enrolments [Z/(X+Y+Z)]*100 54% 54% 55% 58% 58% 59% 95 Graduates as % of total enrolments [Y/(X+Y+Z)]*100 14% 13% 13% 12% 12% 12% Table 1: Pile-up effects of Doctoral postgraduate students (Postgraduate Studies in South Africa: A Statistical Profile January 2009: 12) Translated, what these tables indicate is that more than half of the country’s doctoral students and approximately a third of its Master’s students have remained registered beyond their institutionally-projected completion dates, causing a bottleneck in demand for supervision at this high level. Furthermore, a worrying trend is that these percentages are increasing annually. The long-term implications of this for the growing postgraduate sector in South Africa are not good – high supervisor-student ratios that emerge from this situation could lead to supervisor burnout, a decline in research output (as the supervisor puts her own research on hold in order to accommodate increased supervision loads) and poor quality graduates. The Council on Higher Education (CHE) has also published some guidelines for postgraduate supervision in the form of ITL Resource No.7: Postgraduate Research and Supervision. As mentioned in Chapter 2 (section 1.4.2), this resource is one of a set, aimed at supporting academics in their role as educators. In the words of Mala Singh, the Executive Director of the CHE at the time It is hoped that [Resource No. 7] will be adapted creatively for a number of purposes and used by higher education practitioners individually and in teams in the process of improving the quality of teaching and learning. Improvements in teaching and learning are essential to give effect to the transformation objectives in the restructuring of higher education, especially in relation to redress and equity and to the responsiveness of higher education to national goals and challenges. (ITL Resource No. 7: Postgraduate Research and Supervision 2004: 2) The Resource is aimed primarily at research-intensive institutions – that is, those for whom a strong research focus is core to the institutional mission30 30 Generally speaking. The South African higher education landscape is divided into two ‘types’: research-intensive universities, and comprehensive institutions. The latter integrates ‘traditional’ university and technikon-type programmes and are aimed at addressing ‘… a range of goals, which are central to the Government’s human resource development strategy, in . It needs to be read 96 in conjunction with two key HEQC documents, entitled Criteria for Institutional Audits and Criteria for Programme Accreditation. While the substance of these two documents is not directly relevant to this thesis, it is worthwhile to note that both recognize that there is an element of research in all HEIs – even those not designated as being research-intensive – and that this needs to be quality assured. The Resource also offers a number of ‘evaluative questions’ against which HEIs can self-evaluate the quality of their postgraduate supervision practices. These questions are discussed in more depth in the remainder of the document, and the authors include ‘suggested good practice descriptors’ to guide institutional thinking about what it means to offer high quality supervision and research training to postgraduate students. The questions are reproduced in Figure 1 below: particular, access to higher education; enhanced articulation between career focused and general academic programmes, thus promoting student mobility; strengthening of applied research; and enhanced responsiveness to regional and national human resource, skills and knowledge needs’ (Department of Education 2004: i). EVALUATIVE QUESTIONS The following questions may be adapted for use in self-evaluating the quality of postgraduate supervision and research training: 1. Do the institutional mission and strategic plan emphasise research and postgraduate niche areas? If so, how is this aspect of the institutional mission realised through its planning and resource allocation procedures? 2. How are postgraduate studies and supervision managed and quality assured at the institution? 3. Does the institution have clear policies, regulations and procedures for each stage of the postgraduate research process? How does the institution ensure accountability for the implementation of these policies and regulations? 97 Figure 1: Evaluative questions for measuring the quality of institutional postgraduate supervision and research training (ITL Resource No. 7: Postgraduate Research and Supervision 2004: 8). 3.4 Identity and the doctoral candidate Kamler and Thomson (2006: 16) have argued that text work – the writing done by academics – is inextricably linked to their identity as scholars. They point out that Nowhere is the connection between identity and text as clear as it is when scholars get together to debate the relative merits of particular texts. There is continual slippage between the person and the text. The text is an extension of the scholar, a putting of ‘self’ out there which is either successful – or not. (2006: 15) 4. To what extent are guidelines, acceptable to the local research community, communicated to postgraduate students on their rights and responsibilities with respect to the supervision of these policies and regulations? 5. How does the institution/faculty/school/department provide a research-conducive infrastructure and environment for its postgraduate students? 6. How does the institution provide appropriate induction to research and research skills training and support for postgraduate students, particularly for underprepared students? 7. How does the institution provide access for funding opportunities for postgraduate students? 8. To what extent are guidelines and criteria, acceptable to the local research community, provided for the selection and appointment of postgraduate supervisors? How does the institution provide guidance and staff development for postgraduate supervisors? 9. To what extent are guidelines and criteria, acceptable to the local research community, provided for the appointment of internal and external examiners and for the examining process? 10. How does the institution regularly monitor and review its postgraduate provision, completion rates, outputs and the employability of its graduates? Does this quality management system include eliciting student feedback on supervision, and the review of postgraduate policies and procedures? 98 An individual’s scholarly identity is a social category, in much the same way as age, gender, class, race and so on, are. It is concerned with how a scholar is produced and reproduced (2006: 16) through the process of writing. They subsequently offer six propositions about their particular understandings of this social identity. The most pertinent of these propositions for this work is that identity is not fixed but is always in formation (Kamler and Thomson 2006: 17). The authors elaborate on this point as follows: Supervisors often expect doctoral students to move seamlessly from student to fully-fledged scholar. However, this shift involves a change in identity, self-narrative and behaviour … [a]nd is not sudden or abrupt, but rather occurs in a series of moves (Kamler and Thomson 2006: 17) The implications of this proposition are that a student’s scholarly identity does not emerge, fully formed, upon registration for the degree. On the contrary, it develops over time and during the process of undertaking a variety of different ‘types’ of writing – from summarizing readings, to the specific discursive practices of writing the proposal, to the development of thesis chapters, to the different writing practices required for a journal article, and so on. In a broader context, though, the formation of these writerly identities does not develop in isolation, but rather as a result of interaction with the supervisor as well as with one’s peers - for instance in the form of reading groups/communities of practice (cf. section 2.2.5. of Chapter 2) . The latter reads the students’ written work, and makes suggestions – often textual - for improvement. In this process, (or ‘series of moves’ referred to by Kamler and Thomson), the student develops a more scholarly identity and becomes more firmly entrenched in the academic community in which she is working. Kamler and Thomson’s point of view that one’s scholarly identity is essentially a socially-constructed one is echoed by Hyland in his book Disciplinary Discourses: Social Interactions in Academic Writing (2000). He argues that when academics write, they do so with a particular audience in mind, namely that of their colleagues from a particular discourse community. He argues that …academic texts do more than report research that plausibly represent an external reality, they work to transform research findings or armchair 99 reflections into academic knowledge. This knowledge is not a privileged representation of non-human reality, but a conversation between individuals and between individuals and their beliefs. (2000: 6; italics added) In other words, the meaning of an academic text is socially mediated between the author of the text, and its reader(s), and is “… influenced by the communities to which [these] writers and readers belong” (Hyland 2000: 12). From this, one can extrapolate that the scholarly identity of postgraduate students are a) mediated between the student and the texts (and by implication, the authors) with which s/he interacts; and b) this mediation happens in a specific context, using the specific discourse of the specific community to which the student aspires to belong. Thus, Hyland argues that Texts are the actions of socially situated writers and are persuasive only when they employ social and linguistic conventions that colleagues find convincing (2000: 8). Wenger (1998a) has also recognized the importance of identity, devoting an entire section of Communities of Practice to the concept. Essentially, he argues that identity has a ‘social character’ (1998a: 145), which is developed in conjunction with practice. In his words: Developing a practice requires the formation of a community whose members can engage with one another and thus acknowledge each other as participants. As a consequence, practice entails the negotiation of ways of being a person in that context (Wenger 1998a: 149; italics added). He argues that an individual’s identity is negotiated through participation in the community, even when this participation takes place at the periphery of the community. At this point, they explain, the individual’s position is ‘…an opportunity for learning’ (1998a: 166), and it is in the process of learning that identity is formed. This is borne out by Lave and Wenger’s assertion in Situated Learning: Legitimate peripheral participation that Viewing learning as legitimate peripheral participation means that learning is not merely a condition for membership, but is itself an evolving form of membership. We conceive of identities as long-term, living relations between persons and their place and participation in communities of practice. Thus, 100 identity, knowing, and social membership entail one another (Lave & Wenger 2003: 53). Thus, for the two authors, identity formation does not happen in a vacuum, but rather is socially constructed. It evolves from the apprentice/novice identity on the periphery of the community, to a more established version as an individual’s participation increases. Thus, becoming a legitimate participant in a disciplinary community of practice is the final point of a doctoral candidate’s journey towards a scholarly identity Identity is something that may also be defined in terms of gender. Weedon, in Feminist Practice & Poststructuralist Theory, describes feminism as a form of politics, the basic starting point of which is the patriarchal structure of society (1991: 2). She argues that Feminism questions the assumptions about women which social theories posit as true, pointing to their irrelevance to women’s experience or highlighting the frequent absence of women from them (1991: 6) This suggests that academic literature is a predominantly masculine one, having been written and developed by men (1991: 13). It is one in which the ‘feminine’ is either overlooked or misrepresented. She points out that the theories that have help to structure poststructuralism have been shaped by theorists such as de Saussure, Benveniste, Marx, Althusser, Freud, Lacan, Foucault and Derrida, and that It is no coincidence that these theorists are all men; this is a consequence of the gender relations which have structured women’s absence from the active production of most theory within a whole range of discourses over the last 300 years. It is also a mark of the particular conditions under which prestigious and powerful bodies of knowledge were and are produced. This is manifest in the professional institutions of science, social science, medicine and humanities which exclude alternative forms of knowledge, in particular those produced by women under different social conditions of knowledge production (1991: 13). Under these conditions, I would argue that female postgraduate students come to academe at a disadvantage. Their identity as women means that they have little or no epistemological access to the theory written predominantly by men, and have to adapt their ways of thinking and understanding to this masculine theory. Furthermore, I would assert that even in their own writing, women have to adopt masculine discourses 101 in order to have their work recognized by the academy, although there have been shifts in the terrain. Arguably, this way of thinking may be extended to include other marginalized groups within the academy, such as English second or third language speakers (in an English medium university) and African students registered at an overtly Eurocentric tertiary institution These thoughts are echoed by Leonard (2001), who argues that, certainly in the British system, there has been a concerted attempt to homogenize students - “They can be any age, culture or gender so long as they resemble a young British man” (2001: 43). She points out that the new managerialism that is sweeping through institutions of higher education worldwide is inimical to women because: …academic professionality, in both its old, professorial and new, managerial forms, and the entire Enlightenment project of Science, is gendered. They have embedded within them an antagonism, a project of masculinity, of (super) rationality, of scientism, of independence which attempts to keep safe and secure and strong by keeping or driving out or denying elements associated with femininity (emotions, bodies, acceptance of the diversity of humanity, personal interconnections). These ‘mind-forged manacles’ makes such projects lopsided and finally counterproductive and destructive” (2001: 43) In such a context, where the masculine identity (rationality and scientism) is valued over that which has ‘feminine’ characteristics, it is difficult to maintain one’s identity as a ‘female scholar’ – this is almost a contradiction in terms in light of Weedon’s and Leonard’s arguments. Thus, I would argue that a woman’s scholarly identity needs to be taken into account when researching postgraduate pedagogic practice. However, the focus on identity and its bearing on pedagogic practices within the academy cannot and should not be limited to gender; race and class as forms of identity play an equally important role. 3.5 Conclusion In recent years there has been a marked increase in the number of thesis-writing guidelines that have been published, the target market of which has been postgraduate supervisors. The stated purpose of these texts has been to support supervisors as they travel alongside their students in their research journeys. At one level, they have succeeded in this: they have foregrounded the ‘big issues’ and offered generic solutions 102 to common problems experienced during the research process. As this chapter has attempted to show, however, these guides are not without their problems. They generally present a ‘one-size-fits-all’ account of the process of supervision. In other words, they are acontextual – they fail to sufficiently acknowledge that each supervision relationship is unique, occurring within a singular set of gender, racial, institutional and disciplinary conditions. Thus, while they offer a useful point of departure for supervisors – especially novice supervisors – their efficacy is limited. Within this technicist genre of writing on postgraduate supervision, however, are a handful of texts that go beyond the technicist, check-list approach and offer instead models for writing that go to the heart of the research process. The authors of these works – such as those of Badenhorst (2007, 2008) and Kamler & Thomson (2006) – tackle the issue of research writing and how postgraduates can break the intellectual barriers often thrown up by their anxieties associated with the writing process. The chapter also delved into institutional and national guidelines for postgraduate supervision. These documents are arguably more relevant to this thesis than the above- mentioned thesis-writing publications because they have the South African higher education context as their point of departure. Wits’ Strategies for Successful Postgraduate Supervision was presented as this institution’s input to supporting academic staff in their capacity as postgraduate supervisors. In common with the various thesis-writing manuals, this document provides a generalized overview of the supervision process; where it differs, however, is in its provision of Wits-specific information, relating to issues such as relevant policies and student services. South African national guidelines for postgraduate supervision were also discussed in this chapter. These documents – including the Draft National Plan for Higher Education and A Programme for the Transformation of Higher Education – call for an increase not only in postgraduate enrollments, but in graduations too. ITL Resource No. &: Postgraduate Research and Supervision takes this call one step further by offering self-evaluation questions which HEIs can use to identify internal limitations in their current practices, and thereafter begin the process of engagement about how these should be addressed. What was interesting to note in this section was that while these 103 documents are not recent, they nonetheless remain relevant to the contemporary context. The final section of the chapter – dealing with the issue of identity – was included here because I wanted to juxtapose the students, as individuals and as scholars, against the voluminous guidelines that have been written, supposedly with them (the students) in mind. With the exceptions of Kamler & Thomson (2006), and Badenhorst (2007, 2008), I found little evidence that the texts recognized that the student’s identity is in constant flux throughout the research process. The postgraduate scholarly identity develops with time, and in interaction with the supervisor, peers and texts. Since it is socially constructed between these elements, and since in each relationship the elements are unique to that relationship, it follows that there can be no single academic identity common to all doctoral candidates. 104 CHAPTER 4 METHODOLOGY Chapter overview In this chapter of the thesis, I focus on the research data and the methodology I chose for its analysis. Since my qualitative approach involved human subjects, a significant section is devoted to the ethics of the research, as well as to the reliability and validity of the results. 4.1 Introduction In the preceding chapters, I have looked at the ‘what’ of the thesis, namely the process and practices of postgraduate supervision. In this chapter, I will establish the ‘how’. It commences with an account of qualitative research methodology, and why I considered this to be most appropriate to the research that I conducted. Flowing from this, I examine the narrative method of inquiry and the interview technique, both of which I used in order to gain insight into how the doctoral candidates experienced their process of supervision. The selection of these candidates was undertaken with the intention of getting as wide a range of experience, race and gender as possible – of both candidates and supervisors. A description of this selection process will be presented. Subsequently, I will consider the issues of validity, reliability and generalization, and how my chosen research methodology responds to each. Since no work on qualitative research that includes human subjects is complete without acknowledging the role of ethics and ethical conduct, the penultimate section of the chapter will examine these issues. Lastly, I will render an assessment about the appropriateness of the research design to my research aims and resources. 4.2 Qualitative research Qualitative [research] seeks to understand human and social behaviour from the ‘insider’s’ perspective, that is, as it is lived by participants in a particular social setting (Ary, Jacobs & Razavieh 1990: 445). Qualitative research is a situated activity that locates the observer in the world… This means that qualitative researchers study things in their natural 105 settings, attempting to make sense of, or interpret, phenomena in terms of the meanings people bring to them (Denzin & Lincoln 1994: 3). These two quotes lay the foundation for understanding the nature of qualitative research. At its core is an attempt to understand phenomena as they are experienced by people, as well as the context in which they are experienced. Bogdan and Biklen (2007) explain it more elegantly, as ‘…understanding behaviour from the informant’s own frame of reference’ (2007: 2). This differs from quantitative research, which exhibits a ‘…preoccupation with operational definitions, objectivity, replicability, causality and the like’ (Bryman 1984: 77). Bogdan and Biklen argue that there are five key elements to qualitative research: firstly, it is naturalistic, that is, ‘… the researcher frequents places where the events he or she is interested in naturally occur. And the data are gathered by people engaging in natural behaviour: talking, visiting, looking, eating and so on’ (Bogdan & Biklen 2007: 3). Put another way, naturalism is an attempt to understand ‘… social reality “as it really is”, using methods like prolonged observation and open-ended observation’ (Silverman 2000: 197). Thus, it seeks to understand human behaviour as it occurs, not how it ought to occur. The second element of qualitative research identified by Bogdan and Biklen is that it is descriptive. It uses words and pictures as media of understanding the data. However, these words and pictures are not simply superficial renderings of an event or experience. Rather, they present ‘…such things as gestures, jokes, who does the talking in a conversation, the decorations on the walls, and the special words we use and to which those around us respond’ (Bogdan & Biklen 2007: 5). Thus, their role is to record the social reality of the research subject. Thirdly, qualitative research is concerned with process rather than product/outcomes. In other words, it focuses on the unfolding of events and experiences, rather than on their ultimate outcomes. Despite this, Maxwell (1996) argues that qualitative research does not rule out the importance of products, but instead places an emphasis on ‘… getting at the processes that led to these outcomes, processes that experimental and survey research are often poor at identifying’ (1996: 20). 106 The fourth key element of qualitative research, according to Bogdan and Biklen, is that it is inductive (2007: 6). Babbie (2005) describes inductive reasoning as a form of thinking that ‘… moves from the particular to the general, from a set of specific observations to the discovery of a pattern that represents some degree of order among all the given events’ (2005: 22). This pattern subsequently develops into a theory which Bogdan and Biklen argue results from ‘…the bottom up (rather than from the top down)’ (Bogdan & Biklen 2007: 6). Finally, qualitative research is primarily concerned with meaning: ‘Researchers who use this approach are interested in how different people make sense of their lives’ (Bogdan & Biklen 2007: 7). In other words, how do people understand their lives? How do they interpret the events and experiences that they encounter every day? How do they perceive their own contribution to their social reality? It is in the process of discovering answers to questions such as these that the understanding of the social world is advanced. Bogdan and Biklen recognize that not all of these elements will be present ‘to an equal degree’ (Bogdan & Biklan 2007: 4) in every qualitative research context. This research has all of these key elements, in different degrees: it is naturalistic in that I interviewed most of the doctoral candidates at the University itself – the context in which their supervision occurred. In these interviews, the candidates were encouraged to describe their experiences - including anecdotes and accounts of their supervision meetings – and the processes that formed a part of the supervision experience. From the pages and pages of interview data that I collected, I was able to distill certain themes through the process of inductive reasoning. Finally, the primary concern of the research was to discover how doctoral candidates made sense of their supervision experience; that is, what meaning did they draw from their experiences of the practices of their supervisors? Thus, my research is clearly situated within the qualitative paradigm. In the section that follows, I will examine the narrative method of inquiry and how it develops on the qualitative methodology of this research. 107 4.3 The narrative method of inquiry The primary method of inquiry that will be used in the research is that of the narrative. According to Connelly and Clandinin (1990: 2) The main claim for the use of narrative in educational research is that humans are storytelling organisms who, individually and socially, lead storied lives. The study of narrative, therefore, is the study of the ways humans experience the world. (Italics added) The centrality of the lived experience as a rich source of research data is one of the dominant themes of a later book by the same two authors, entitled Narrative Inquiry: Experience and Story in Qualitative Research. In this book, they argue that it is neither productive nor necessarily accurate to reduce research, particularly in the social sciences, to mere statistical correlations. This is because such numbers-based analyses fail to take account of the lived experiences of those being analysed (2000: xxiv). They define narrative quite simply as: … a collaboration between researcher and participants, over time, in a place or series of places, and in social interaction with milieus. (2000:20) Based on the theories of John Dewey, they assert that narrative inquiry takes place in a three-dimensional space. These dimensions are situation, continuity and interaction (2000). By ‘situation’, Clandinin and Connelly mean ‘place’ or ‘context’; that is, where the event being narrated took place and under what circumstances. As the authors succinctly describe it ‘… they occur in specific places or sequences of places’ (2000: 50). ‘Continuity’ suggests that the narrative has a past, present and future. In other words, it is not simply the outcome of a series of events (for instance, ‘He died’) but rather a culmination of episodes that led to it (‘He drank and smoked a lot throughout his life; this led to a series of strokes and heart attack over the years. Ultimately, no one was surprised when he had a fatal heart attack yesterday’). The ‘future’ of the narrative is often the lessons learned from the event (‘Having seen what drinking and smoking can do to the human body, his son made a commitment to never touch alcohol or cigarettes, lest he end up like his father’). 108 Finally, ‘interaction’ refers to the personal and social contacts that are the source of the narrative (2000: 50). Since individuals do not live in a vacuum, interaction between them is inevitable. From such contact, narratives emerge as the individuals try to make sense of what came to light as a result. Kohler Riessman’s Narrative Analysis takes the theoretical description of narrative inquiry put forward by Clandinin and Connelly above, and contextualises it. She succinctly defines the purpose of the narrative method as: … to see how respondents in interviews impose order on the flow of experience to make sense of events and actions in their lives. The methodological approach examines the informant’s story and analyses how it is put together, the linguistic and cultural resources it draws on, and how it persuades a listener of authenticity. Analysis in narrative studies opens up the forms of telling about experience, not simply the content to which language refers. We ask, why was the story told that way? (1993: 2) Thus, her point of departure, like that of Clandinin and Connelly, is that of lived experience. Unlike these two authors, however, she also delves into the linguistic and cultural sources of the story. In turn, this leads to the issue of human agency in the narrative – after all, it is the individual who decides what is included in the narrative and what is not, its sequence of events, and what these are supposed to mean to the person listening or reading the narrative (1993: 2). In other words, ‘Individuals construct past events and actions in personal narratives to claim identities and construct lives’ (1993: 2). Human agency is represented at various levels in the research process: attending, telling, transcribing, analyzing and reading. This is conveyed by Kohler Riessman in the diagram below (1993: 10): 109 Primary experience Figure 2: Levels of Representation in the Research Process (Kohler Riessman 1993: 10) Kohler Riessman uses Figure 1 to break down the narrative into its constituent elements so that when it is rebuilt, it provides a rigourous rendering of the narrative. The first level, attending to experience, refers to the reflections, memories and recollections of what she calls the narrator’s ‘stream of consciousness’ (1993: 9). In other words, it is what the narrator notices and selects from ‘the primary experience’ – the event that is narrativised. Thus, argues Kohler Riessman, “I actively construct reality in new ways at this first level of representation, to myself, by thinking” (p. 9, italics added). Telling about the experience is the second level of representation that she identifies. She labels this ‘the performance of a personal narrative’ (p. 9) because this is when the narrative is recounted to others. However, the performance does not simply end at the 110 initial telling of the story; it is elaborated upon as those listening to it ask questions, and the narrator responds. Thus, the narrative is not produced by only one person, but rather by a number of people to whom it is related (p. 10). Furthermore, in the telling of the narrative, there is inevitably a gap between the experience as it was lived by the narrator, and his/her telling of it. Language and meaning are intertwined, as the narrator uses the former to construct the latter. Thus, the narrator’s choice of words decides the meaning that she wishes the listener/reader to have of the experience. Hence, she asserts that: In telling about an experience, I am also creating a self – how I want to be known by [my listeners/readers] … My narrative is inevitably a self- representation (1993: 11). The third level of representation, transcribing experience, is concerned with transforming the oral narrative into written text. This is an important element of narrative because it emphasizes the importance of language in the telling of a story. Specifically, it demonstrates that language is not transparent: Qualitative researchers now ask themselves how detailed transcriptions should be. How, for example, could they best capture the rhythm [of the language]? Should they include silences, false starts, emphases, nonlexicals like ‘uhm’, discourse markers like ‘y’know’ or ‘so’, overlapping speech and other signs of listener participation in the narrative? Should they give clauses separate lines and display rhythmic and poetic structures by grouping lines? Not simply technical questions, these seemingly mundane choices of what to include and how to arrange and display the text have serious implications for how a reader will understand the narrative. (1993: 12) Clearly, the act of transcribing a narrative is an interpretive one. It is the transcriber who decides what will be told, and how it will be presented. In so doing, the transcriber allows the reader to see only what s/he wants there to be seen. Furthermore, different conventions of transcription result in different interpretations of the text, not to mention different ideological positions and understandings on the part of the reader. Quoting Mischler (1986), Kohler Riessman (1993) points out that “Meaning is constituted in very different ways with alternative transcriptions of the same stretch of talk” (quoted on page 13). 111 Analysing experience, the fourth level of representation identified by Kohler Riessman, takes place across a number of transcripts. Typically, the transcripts are read with a view to identifying the similarities and differences across the stories. Subsequently, the researcher cuts and pastes these into a final text (which could be a book or a dissertation). The format of this document, the style in which it is written, and the order in which events are depicted as happening are all very thoroughly considered by the researcher. As a result, argues Kohler Riessman, “… the analyst creates a metastory about what happened by telling what the interview narratives signify, editing and reshaping what was told, and turning it into a hybrid story, a ‘false document’” (1993: 13) The final level of representation is the reading of the narrative. In light of what the author discussed about the nature of the previous levels, it is clear that the reader has only the representation of the analyst (p. 14). In other words, the reader only reads what the analyst has written; for instance, the analyst may have chosen to include certain events at the expense of others, thereby skewing the text in a certain direction. The reader will therefore never know if there is any more to the story than that which appears in the text, and if there is, what it is. These five levels of representation are useful in the proposed research because they embody the multidimensional qualities of the narrative method. They emphasise that narrative is not simply about the telling of a story, but rather about the many ways in which a story can be told, as well as the multiple different ways that it can be interpreted depending on elements such as language use, the transcription method of the interviewer, authorial voice, and so on. Kohler Riessman subsequently turns her attention to practical models of presenting narrative, although she does warn that there is “… no single method of narrative analysis but a spectrum of approaches to texts that take narrative form” (p. 25). She emphasizes that whichever approach the researcher chooses to use, it will inevitably raise the following questions about the representation of the experience (1993: 25): 1. How is talk transformed into a written text and how are narrative segments determined? 2. What aspects of the narrative constitute the basis for interpretation? 112 3. Who determines what the narrative means and are alternative readings possible? The first model that Kohler Riessman describes is that of the ‘life (hi)story’ model. The focus of this model is the sequential representation of events described by an interviewee, even if the interviewee has not told the story in chronological sequence. Thus, the interviewer will (re-)arrange the events in the story along a timeline so that the reader can make out what happened first, second, third and so on. The result is a ‘birth to death’ flow of information. The conventions that are used in this model include a combination of brief, direct quotes taken from the interviews, much longer summaries of the general content of the interview, as well as the author theorizing about the issues and themes that emerge from the interview (p. 27). Kohler Riessman warns, though, that this model tends to favour the interpretative voice of the author over that of the interviewee. In other words, the author of the text can selectively use quotations from the interview to further her/his own arguments, and the reader would be none the wiser. Thus, since readers do not have the same access to all of the raw data that the researcher has, they are constrained from interpreting the text in any other way than that presented by the author. The potential for author bias is, therefore, ever present in this model. The second model of narrative that Kohler Riessman addresses is that of linked stories and meaning in conversation. In this model, the interviewee’s spoken communication is broken down “… into clauses, lines are numbered, and the parts of the narrative are identified by their function (to orient, carry the action, resolve it, etc)” (1993: 35). In addition, those sections of the discourse that are not directly related to the interview topic, such as asides and interactions between the teller and the listener, are deleted (p. 35). This is because researchers using this model are not only concerned with the content of the narrative, but with “… the structures and language narrators [choose] in collaboration with a listener to represent the experience of change in political consciousness” (p. 36). This model could be used by researchers who wish to track the changes in consciousness, and hence in identity, of their interviewees. The markers of this change are the specific linguistic structures used to construct the story. 113 In a similar way to the life story model, the caveat that is attached to this model relates to authorial control of what the reader ultimately gets to read – “The analyst controls meaning to the extent that she selects what features of the discourse … will be the subject of her text, and she interprets the discourse” (1993: 42). However, the reader can be a part of the meaning-making process if the full narrative is attached to the analysis. Another concern about this particular model (which could be applied to the others) is the fact that nonverbal aspects of communication are not captured in the text (1993: 40). Gestures, facial expressions, eye contact with the interviewer (or the lack thereof) arguably tell one as much about the experience being described as the words themselves. In their absence, the thoroughness of the text is questionable. Finally, Kohler Riessman describes the poetic structure and meaning model of narrative inquiry. In her words, It involves reducing a long response , parsing it according to a set of rules into lines, stanzas and parts, examining its organizing metaphors, and creating a schematic to display the structure” (1993: 50). The work of James Gee (1985, 1986, 1991), which emphasizes the poetic features of language and units of speech, partially informs this model. For Gee, it is the oral, rather than the text-based models of language that are important when examining narrative segments. One reason for this is that changes in the interviewee’s pitch and intonation, for instance, provide the interviewer with insights that are absent in a written text (Gee 1991: 23). Another reason is that the structures of the spoken narrative (such as stanzas, metaphors and key words) reduce the potential for multiple interpretations of the narrative (Kohler Riessman 1993: 52). Kohler Riessman, who uses this model in her own research, argues that “This is an ideal realization of the text, because it excludes interactions between teller and listener, false starts, pauses, discourse markers, nonlexical expressions, and other features of spoken language” (1993: 44). In common with both Clandinin and Connelly (2000), and Kohler Riessman (1993), Behar-Horenstein and Morgan (1995) believe that narrative inquiry is based upon lived experience. They define narrative as “…story ...with very special syntactical shape … 114 and subject matter which allows for, or encourages the projection of human values upon this material” (1995: 143). I would argue that it is this definition that best underscores the importance of narrative as a productive research tool. It maintains that narrative is not simply the telling of stories for the sake of entertainment, but rather, for the purpose of reflection and effect. Further, it “…places an emphasis on the connections between what humans think, know, and do as well as the reciprocal relationships between the way that human thinking shapes behaviour and knowing shapes thinking. As an entity with a beginning, middle, and end, [narrative] has a dynamic that is created and interpreted through the lived experiences of a person or a participant observer … A synthesis of the lived experiences results in a narration that occurs through a process of active construction” (Behar-Horenstein and Morgan 1995: 143; italics added) Thus, narrative is not simply about the telling of a story or an experience, but is also dependent upon how the listener dynamically constructs meaning from what s/he has experienced. Despite this, however, there is no guarantee that the listener’s understanding of the story is the same as that of the narrator. Thus, Behar-Horenstein and Morgan point out the need for an agent or a mediator that would allow the narrative to be interpreted in ‘understandable ways” (1995: 143) – these agents or mediators are the accepted codes and conventions that are used to guide listeners to construct the intended meaning of the narrative. Using an analogy to the construction of meaning when listening to a musical composition: Behar-Horenstein and Morgan argue that the codes and conventions that an individual would use to understand it would include melody, harmony, metre, tempo, and so on (1995: 143). Similarly, when listening to a narrative, conventions used by the narrator are significant. For instance, s/he might make use of an honorific like ‘sir’ to denote (in the mind of the listener) a person of high status. This type of mediation helps the listener to construct the meaning intended by the narrator. Narratives may be analysed in a number of different ways, depending on the discipline of the researcher. Cortazzi (1993: 2) proposes four disciplinary models of analysis: sociological and sociolinguistic; psychological; literary; and anthropological. Each of these models has a number of theories about how narratives should be analysed. For 115 instance, sociological and sociolinguistic models include Conversational Analysis (in which talk is examined as a means of understanding how the speaker views his or her world); narrative as a replaying of personal experience (which deals with the ways in which speakers organize face-to-face interactions with their audience); Labov’s Evaluation model (one that examines the ‘formal structural properties of narratives in relation to their social functions’ (Cortazzi 1993: 43)); as well as interviews and narrative Performance (focused on developing the interview narrative into a more substantial, fleshed-out story). I chose the sociological and sociolinguistic methods of analysis because of their emphasis on the social context of the speaker; in other words, her or his attitudes, beliefs, culture and perceptions of their postgraduate supervision and supervisory experiences. However, elements of other models might be relevant and to the extent that they are useful, they will be used. Despite the seeming efficacy of narrative as a tool for gathering research data, it has also been the focus of criticism: … Among these is the argument that it encourages a narcissistic representation of events by those telling their stories (Hargreaves 1994). Inevitably, argue the critics, the voice of the storyteller is morally-laden and prescriptive, and not an empirical representation of events (Clandinin & Connelly 2000: 37). Furthermore, Behar-Horenstein and Morgan point out that researchers who are unable to use experimental design methods and statistical techniques for data analysis tend to use narrative instead, bringing into question the ‘scientific’ and quantifiable validity of their findings (1995: 147). This implies that the only ‘correct’ methods for analyzing data are quantitative in nature. Clandinin and Connelly, following Oakeshott, term this perspective ‘technical rationalism’, defined as “… the assertion that there is no knowledge which is not technical knowledge” (quoted in Clandinin and Connelly 2000: 36). Another criticism is that narrative inquiry “… is portrayed as an indefensible form of inquiry owing to the fluidity that characterizes the context in which data is collected” (Behar-Horenstein and Morgan 1995: 147). This is a legitimate argument if one is simply using narrative to tell a story (or an anthology of stories). However, I would 116 argue that these criticisms are less persuasive when multiple narratives are collected from individuals who have gone through the same process (in this case, the process of researching and writing up a doctoral thesis), and common themes and trends are extrapolated from them for the purpose of the identification of inhibiting and supporting practices. While narrative will never be ‘scientific’ in the positivist sense of the word, this does not mean that the truths and experiences related by individuals are any less true or authoritative than the results obtained from a scientific experiment or a statistical analysis of data. If anything, they are arguably more authoritative because they reflect more holistically the human experience of a process or practice, something that cannot be quantified. For instance, if a student were to evaluate her postgraduate supervision experience as having been 7/10 on a scale of 1 to 10 (1 being awful, and 10 being brilliant), this does not tell us much. It certainly does not tell us which part/s of the supervision was effective and with which s/he was not happy. In other words, it fails to provide the context of the student’s experience and thus reflects only a small part of the experience. By contrast, asking a student to talk about her experiences of supervision will generate much richer data from which common themes can be extrapolated for use in future. In the context of this research, therefore, the use of the narrative methodology is an appropriate method of gaining insight into both the supervisory practices in existence as well as the overall educational experiences encountered at the University of the Witwatersrand. However, the caveat is that interviewing supervisors would open up the possibility of non-Wits experiences of supervision. More importantly for this study, however, is the way in which the interviewees experienced these practices, both cognitively and affectively. Postgraduate student narratives will thus be used to extract the supervision experiences that they have encountered, as well as the impact that these experiences have had on their time as novice researchers trying to develop their own scholarly identities. 4.4 The interview technique An interview is a purposeful conversation, usually between two people but sometimes involving more (Morgan, 1997), that is directed by one in order to get information from the other (Bogdan & Biklen 2007: 103). 117 Mischler (1986: 96), a key theorist in the field of research interviewing, conceptualizes the interview technique as one that produces ‘jointly produced discourses’. In other words, … interviews are not neutral tools of data gathering, but active interactions between two (or more) people leading to negotiated, contextually based results (Fontana and Frey 2000: 646) Thus, the interview situation is not decontextualized, and the context is rarely a neutral one. Interviewers have increasingly come to be seen as active participants in the interview process, as opposed to simply the asker of questions and subsequently, the passive recipients of information provided by the interviewee (Fontana and Frey 2000: 663). The interview itself has become a ‘negotiated accomplishment’, shaped by the context and situation in which it occurs (Fontana and Frey 2000: 663). Put another way, It has become increasingly common in qualitative studies to view the interview as a form of discourse between two or more speakers or a linguistic event in which the meanings of questions and responses are contextually grounded and jointly constructed by interviewer and respondent (Schwandt 1997: 79). Bogdan and Biklen (2007) agree, offering the following practical insight: Good interviews produce rich data filled with words that reveal the respondents’ perspectives… Good interviewers communicate personal interest and attention to subjects by being attentive, nodding their heads, and using appropriate facial expressions to communicate. The interviewer may ask for clarification when the respondent mentions something that seems unfamiliar, using phrases such as, ‘What do you mean?’, ‘I’m not sure I am following you’, ‘Could you explain that?’. The interviewer also probes the respondent to be specific, asking for examples of points that are made (Bogdan & Biklen 2007: 104) . Interview techniques cannot be pigeonholed into a single, homogenous and unvarying model. Rather, there is a wide range that may be used by the researcher, depending on the focus of the research. Mischler (1986) identifies the following models: stimulus- response (p.13), structured and unstructured (p. 27), medical (p. 54), moral- development (p. 58), life-history (p. 87), and oral-history (p. 99). The aim of each model is to collect a specific ‘set’ of information from those being interviewed. For 118 instance, the medical interview is designed to elicit a patient’s symptoms, as well as the possible disease pathology. By contrast, the stimulus-response interview aims solely at obtaining a verbal reaction to questions posed; these reactions “… meeting particular response requirements posed by the questions” (Brenner, quoted in Mischler 1986: 14). Wolfson (1976: 192) makes an important distinction between the narratives elicited from interviews and those from ‘spontaneous’ conversations. Her research, based on the analysis of 150 taped narratives occurring in interviews and 400 taped narratives from more spontaneous conversations, reached the following conclusion: …interview narratives are usually in the form of a summary, short and to the point. Respondents know they are answering questions and so details in such narratives are directed to the question. These narratives must be characterized as answers. Conversational narratives, in contrast, are more detailed and are performed. Speakers choose their own topics, elaborated in their own time. Bogdan and Biklen also highlight the distinction between interview and conversational narratives: ‘Too often, people involved in conversations do not concentrate intensely on what the other person is saying’ (2007: 105). As a result, their focus is not on the subtleties of both the physical and body language of the other, nor on the nuances of the argument being put forward. Under these conditions, rich descriptive data is lost. This is an important factor to take into consideration during the proposed interview process. It reinforces the fundamental point made by Fontana and Frey, above, that the interview is (and indeed, must be) a negotiated process in which meaning is jointly constructed. An interview that simply records the respondents’ words has no research value because it is presented as a decontextualised piece of writing. Hence, a caveat for this research will be to actively engage with the interviewees so that their responses are not simply answers as suggested by Wolfson, but are in fact detailed stories. Therefore, the interview questions that I used (see section 4.5) were designed to elicit detail and stories, as opposed to mere one word or one-sentence answers. 119 Much has been written about the strengths and weaknesses of the interview methodology as a means of gathering meaningful data (for instance Kvale 2006; Cassell and Syman 2004; Greenhalgh, Russell & Swinglehurst 2005; Nunkoosing 2005; and Burman 1997). These will be outlined below – not only for the purpose of information, but also to demonstrate that all the angles were carefully considered before I chose this methodology. Kvale (2006) argues that while qualitative interviews can (and often do) advance the understanding of people’s lived experiences, and in so doing ‘give voice to the many’ (2006: 481), they are not necessarily any better than positivist, quantitative methods such as questionnaires and laboratory experiments. He recounts that when qualitative interviews came into general use in the 1980s, proponents of the methodology suggested that interviews provided a ‘personal alternative’ to the ‘harsh manipulation’ of quantitative methods (2006: 481). However, Kvale cautions that this does not mean that interviews are free from manipulation by the researcher. Drawing on the work of Burman (1997), he contends that the qualitative interview is ‘a fantasy of democratic relations’ that ‘masks the basic issue of who gains materially and symbolically from the research’ (2006: 482). This exposes the methodology to the possibility of manipulation by the researcher. Burman (1997) encapsulates this ostensible dichotomy between the personal and qualitative, on the one hand, and the callous quantitative on the other as follows: The current impulse toward qualitative research is … a humanitarian one: to be “nicer” to people; not to exploit or dehumanize them by treating them as objects. Talking to and with people, and treating their accounts as informative (and in some cases authoritative) warrants for their actions, rather than measuring their behavior, was crucial to the humanist rejection of behaviorist psychology. “Being human” is therefore counterposed to an ethic of manipulation (Burman 1997: 790, italics added). Burman’s critique of the qualitative interview revolves around three areas in which the ‘discourse of democracy’ that is often associated with qualitative research methodologies (such as the narrative interview) tends to shroud the ‘…maintenance of the traditional direction of power in researcher-researched relations’ (1997: 790). These 120 areas are: relationship, equality and participation. In respect of the first of these – relationships - she argues that facilitating a positive research relationship with participants is merely an ‘insidious form of manipulation’ on the part of the researcher because it gives her (the participant) a false sense of empowerment: she believes that she has volunteered some unique data, when in reality it is the interviewer who has directed their contribution by developing a rapport with her and subsequently using this to elicit data. Burman describes this more emphatically: Known in the trade by the cozy epithet “rapport”, the social features of the research relationship easily become quantified and commodified in some accounts into devices to promote disclosure (Burman 1997: 791) Thus, she rejects the suggestion that rapport is innocuous or lacking in deceit, but rather views it as a tool for manipulating the participants. Kvale (2006) is similarly unconvinced by the argument that the rapport developed between the interviewer and the participant is either untainted or untaintable. He argues Creating trust through a personal relationship… serves as a means to efficiently obtain a disclosure of the interview subjects’ world. The interviewer may, with a charming, gentle, and client-centred manner, create a close personal encounter where the subjects unveil their private worlds. A quasi-therapeutic interviewer role, building on emotional rapport and therapeutic knowledge of defense mechanisms, can … serve as a “Trojan horse’ to get behind defense walls of the interview subjects, laying their private lives open and disclosing information to a stranger, which they may later regret (Kvale 2005: 482). I am unconvinced by this argument, especially in light of the stringent ethical requirements to which qualitative researchers are held (at Wits, anyway). Part of the ethics process is an ‘opt-out’ clause for subjects in which they can withdraw the results of their participation if they are unhappy about it (see Section 4.8). While this is by no means a fool-proof way of protecting the subject, it is nonetheless an important safeguard. The second area of qualitative interviews critiqued by Burman is that of equality – the idea that the interviewer and informant possess an equal amount of power. She dismisses this, asserting instead that: [R]esearch relationships mobilize and reproduce within their process the structural positions that exist outside research contexts (or indeed may well 121 have motivated the research). Thus, researchers have to confront the fact that, except where the research question is explicitly arrived at upon the request of (and not even merely in consultation with) the researched, the fantasy of democratic research is belied by the basic issue of who gains (materially, symbolically) from the research (Burman 1997: 791). This might be interpreted to mean that ultimately, even with the best of intentions on the part of the interviewer, the power within the interaction remains firmly with the interviewer. From a structural perspective, this may be true. However, I would argue that the informant is endowed with a significant amount of power – the fact that she possesses the data that the interviewer needs, which she can withhold or disclose it at her discretion. Thus, the interviewee could be understood to exert a degree of power over the researcher. Nankoosing (2005) seems to support this stance: he argues that throughout the interview, power constantly shifts between the interviewer and the interviewee: “In this dance... of power, both the interviewer and the interviewee are constantly seeking to (dis)equalize their respective authorities’ (2005: 699). Thus, while the researcher undoubtedly gains both materially and symbolically from the interviewee’s contribution, she also has to surrender some of her power to the interviewee in order to gain access to the data that she requires. Finally, Burman offers a critique of the role of participation in the qualitative interview, although she has less to say about this area than the previous two. For her, the term ‘participation’ implies a certain level of investment in the process on the part of the informant, but one that stops short of full ownership of it. This is because ‘[t]o participate often means to join, or be enjoined into, a (someone else’s) project’ (1997: 792). This could be interpreted to me that the informant plays a subordinate role in the process: the contribution of each to the wider data set is peripheral and therefore, each can make no significant knowledge claims on the final product. I disagree. In my experience, each individual that I interviewed added a subtle layer to my study. While there were some overlaps between the experiences of some, there were just as many variations, and it was the latter that added depth to my findings. Thus, while it is possible that participants may experience the process of being interviewed as less than 122 a democratic one, this is perhaps more likely to be due to an oversight on the part of the interviewer in including them in the final product, rather than with the process itself. A more structural critique of the qualitative interview is put forward by Cassell and Symon (2004). They argue that this methodology is time-consuming at many different levels, from the development of the interview guide, to the conducting of the interviews, and finally to their analysis (2004: 21). The latter is described as ‘data overload’, resulting from the ‘huge volume of rich data produced by even a moderate- sized study’ (2004: 21). This critique is a valid one. The thirty interviews that I conducted left me with mounds of data, not all of which I could include in the thesis. It was very difficult for me to sift through the data and emerge with only a handful of themes, especially when I knew that there was so much more I could have presented. It feels that the data that I have not used in this thesis has been wasted, even though I hope to use it for future journal articles. Nonetheless, I felt that even though the methodology was time-consuming, it allowed me to immerse myself in the process and more importantly, in the substance of what the interview subjects were saying. Thus, for example, the transcription of the interviews allowed me to ‘re-hear’ the subject responses many times; each time I listened, I heard something new that added value to my understanding of their experience. Thus, the ‘data overload’ could be constructed as a useful part of the study, rather than a burden. Despite the aforementioned weaknesses, there are also a number of advantages associated with the use of qualitative interviews. Firstly, they allow the researcher to probe any given context in order to understand it in more depth. Quoting Strauss and Corbin (1990), Hoefl argues that qualitative interviews can be (and are) used to understand phenomena about which little is known (1997: 48). Similarly, they can be ‘used to gain new perspectives on things about which much is already known, or to gain more in-depth information that may be difficult to convey quantitatively’ (1997: 49). Put another way, interviews allow the researcher to explore a phenomenon in a significantly more multifaceted manner because they are not constrained by the need to 123 quantify data (although this is possible, too). This ability to describe and explore the data more fully also makes it more accessible for the reader. Citing Lincoln and Guba (1985) – ‘If you want people to understand better than they otherwise might, provide them information in the form in which they usually experience it’ (1985: 120 – Hoefl argues that qualitative data elicited from interviews may resonate with readers’ epistemological preferences, and hence it becomes more meaningful (1997:49). Bogdan and Biklen (2007) suggest that the value of the qualitative interview lies in the fact that it allows the researcher to ‘gather descriptive data in the subjects’ own words so that [she] can develop insights on how subjects interpret some piece of the world’ (2007: 103). Denzin and Lincoln (1994) articulate a similar idea: ‘[Q]ualitative researchers study things in their natural settings, attempting to make sense of, or interpret, phenomena in terms of the meanings people bring to them’ (1994: 3). In other words, the qualitative interview allows the subjects to express their understanding of a phenomenon, and from this, researchers can develop an authentic theory based on the lived experiences of these subjects. Finally, qualitative interviews ‘give voice to the many. For example, the marginalized, who do not ordinarily participate in public debates, and in interview studies have their social situations and their viewpoints communicated to a larger audience.’ (Kvale 2006: 481). Greenhalgh, Russell and Swinglehurst (2005) concur: they argue that the value of the qualitative interview ‘comes into its own when considering a quality improvement initiative from the perspective of disadvantaged groups such as the socially excluded...’ (2005: 444). Thus, by giving the interview subjects a voice, the qualitative interview offers an opportunity for change to be effected. Kvale offers two instances where this has been the case: interviews conducted by anthropologist Oscar Lewis in 1964 foregrounded the living conditions of exploited groups in Mexico; similarly, interviews undertaken in 1990s France by Bourdieu, highlighted the poverty and marginalization of the lower classes (Kvale 2006: 481). Thus, the argument that the qualitative interview can be used as a instrument for change is a powerful one. While this thesis has no such lofty aspirations, it nonetheless seeks to allow the participants the opportunity to tell their stories of the experience of doctoral supervision; if, in doing so, supervision practices change, so much the better. 124 Based on the above weighing up of the disadvantages and advantages of the qualitative interview as the most appropriate methodology for this thesis, I was convinced that I had made the correct choice. I wanted to hear the stories of the PhD candidates, in their own words. I wanted to hear how they had experienced the process, as well as what they had chosen to foreground and background. By allowing the candidates to give voice to their experiences, I hoped to develope a better (not to mention more authentic) understanding of these experiences. For the purpose of the research, I made use of the unstructured interview technique, specifically the open-ended, ethnographic (in-depth) interview. ‘Open-ended’ does not imply that the questions posed during the interviews are either unsystematic or disordered. Rather, I sought the establishment of a human-to-human relation with the respondent and the desire to understand rather than to explain (Fontana and Frey 2000: 654) In other words, I posed the questions (listed in the next section) without having what Fontana and Frey (2000: 653) term an a priori categorization in mind; that is, I had no preconceived ideas nor did I ‘filter out’ any information. Mischler (1986: 69) seems to favour this approach to interviews, arguing that a researcher is more likely to …find stories reported in studies using relatively unstructured interviews where respondents are invited to speak in their own voices, allowed to control the introduction and flow of topics and encouraged to extend their responses. Since I was keen to encourage the ‘voice’ of the doctoral candidates that I interviewed, the open-ended approach seemed to be the most appropriate. In the next section of the chapter, I will address the specific interview questions that I used, and advance reasons for my choice of questions. 4.5 Interview questions The interviews that I conducted were structured around the following questions, the aim of which was to generate narratives of supervision: 1. Tell me the story of your PhD supervision experience, from the point when you decided to undertake this degree until now. Specifically: 125 a. What were the milestones, triumphs and obstacles? Who were the people that were central to the experience? b. Please describe a typical supervision meeting. c. What kind of feedback do you get from your supervisor/s (written, oral, a combination of both)? Is it constructive, or superficial? d. The issue of power between the supervisor and the doctoral candidate is one that is frequently mentioned in the literature on supervision. How would you describe the power relations between you and your supervisor? e. Ideally, the doctoral candidate should continue to undertake research and publication after the awarding of the degree. How (if at all) has your supervisor developed your research skills for this purpose? As each interview unfolded, I found myself asking secondary questions (not included in the above list) as a means of extracting richer descriptions from the doctoral students. Obviously, not each interview candidate required such questions, and I rarely used the same secondary question in more than one interview. Instead, I followed the lead of the interviewee: where they were sparse on detail, I probed further; where they were not, I simply steered them along. It seemed to me that this flexibility is one of the key features of the unstructured interview; that is, actively listening and adapting the questions as the situation demanded. In reflecting on the use of the interview technique in general, and these questions in particular, I was aware of the fact that I was navigating my way through a potential minefield. The nature of data that I was attempting to elicit from these candidates could be perceived of as sensitive, particularly in those cases where the supervision process had not been a smooth one. Since I did not know what kinds of narratives31 31 See section 3.2., in which Badenhorst (2008) describes the various kinds of narratives that postgraduates use to describe their experience of the research and supervision processes. I was going to get when I began each interview, I had to ‘read’ the participants’ body language and specific verbal cues to ensure that I was not alienating or offending them with my questions. At the same time as I was maintaining this delicate balance, I was also fascinated by the mosaic character of the information that was imparted during these face-to-face interactions. Here, I am not only referring to the overt substance of what was said, but also to what was not explicitly stated: the candidates’ body language and non-verbal cues were often more telling than the narratives themselves. My field notebooks contain many observations such as ‘laughed nervously’ and ‘frowned at the 126 memory of x experience’ which added richness to the data. This kind of information added a dimension to the data that I suspect would have been absent in any other form of qualitative research method. At a meta-level, I was also keenly aware of comparing my own experiences of doctoral supervision to those of the interview participants as they spoke, and again when I transcribed their interviews. I was initially concerned that I might be over-identifying with the candidates’ narratives. After some reflection, however, I realized that such a comparison was inevitable, given that I was not only the interviewer, but also a peer of the interview participants. However, this dual perspective allowed me some unique insights into not only the supervision practices themselves, but also the way in which these were experienced by the students. 4.6 Sample selection My interview sample came from a random selection of names from official University printouts of students registered for PhD studies in the Science and Humanities faculties at Wits. The printouts were provided after I had met with the university Registrar to request permission to interview students. On the advice of my first PhD supervisor, I pulled out every fifth name from each faculty list, and contacted the candidates via email. Where I was unable to locate their email addresses, I contacted their supervisors to request this information. In the first round of selection, I got a number of responses; some were willing to be interviewed, while others declined because they did not feel that they had anything substantial to add to my study. Three were geographically too far away for me to interview face-to-face, and although they were open to the idea of filling in a questionnaire, I was reluctant to use this method of data collection. I felt that it would not have allowed me to interrogate answers immediately or to have witnessed the body language of the respondents as they engaged with the questions. These two activities were, in my mind, vital to obtaining a more rounded picture of the candidates’ experiences than could be expected from a one-dimensional written text. In addition, I wanted comparability of data when it came to the analysis. That is, I wanted to be able 127 to compare like texts: interviews with interviews, and not interviews with questionnaires. Any analysis of the latter could have produced dubious and misleading results, which in turn could undermine my findings. The first ‘round’ of this random sample did not yield the thirty candidates that I needed for my research. Therefore, I returned to my lists and started again: I selected the second name on each, and then each following fifth one again (as I had done in the initial ‘round’). I ended up repeating this process four times in all before I reached my target. The fact that my interviewees were randomly selected in this manner meant that I was able to interact with people of different ages, races, genders and disciplines, all of whom were at different stages of their research. Some had recently submitted, others were tackling their proposals and the remainder populated the area in between. Thus, the range of data that I was able to collect added significant depth to my research. 4.7 Validity, reliability and generalization In all research, it is important that the cornerstones of validity, reliability and generalizability of the data are observed. Without them, the research loses credibility. In this section, I shall engage with all three in relation to the research interviews and the narrative compositions that I undertook. The term ‘validity’ has been traditionally used in quantitative research to describe ‘… the extent to which an empirical measure adequately reflects the real meaning of the concept under consideration’ (Babbie 2005: 148). Put another way, it is ‘… concerned with the extent to which an instrument measures what one thinks it is measuring’ (Ary et al1990: 256). It has its roots in the positivist tradition, in which ‘…universal laws, evidence, objectivity, truth, actuality, deduction, reason, fact and mathematical data’ (Golafshani 2003: 597) are key to establishing the credibility of data. Since qualitative research is not concerned with numbers but rather with the subjective interpretation of social interactions, such measures clearly cannot apply. However, there has been little agreement in the literature about what constitutes ‘validity’ in a qualitative study (Creswell & Miller 2000; Winter 2000). Despite this, Creswell and Miller (2000) suggest that at its most fundamental level, theorists agree that ‘…qualitative inquirers 128 need to demonstrate that their studies are credible’ (2000: 124; italics added). They flesh this definition out to explain …how accurately the account represents participants’ realities of the social phenomenon and is credible to them (Schwandt, 1997) … [and]… refers not to the data but to the inferences drawn from them (Creswell & Miller 2000: 125). Winter (2000) is wary of narrowing down the notion of validity too much in qualitative research because this could ‘…confound and obscure the more general purposes of the research and analysis’ (2000: 3). This is because it is not a ‘unitary concept’: There is no single form, construct or concept that can universally be claimed to define or encompass the term. Neither, however, can validity be said to be a discretely identifiable element of any research project, which is capable of being located at multiple and specific stages within the research. The concept of ‘validity’ defies extrapolation from, or categorization within, any research project (2000:3). He subsequently argues that validity can be applied in different ways, depending on both the beliefs of the researcher and what part of the research she thinks needs validation (2000: 3). Thus, there does not appear to be a single understanding of the term to which I could ‘hook’ my research. I was reluctant to obsessively hone in on the accuracy of every single word uttered by my interviewees, but at the same time wanted to ensure that the interview transcripts and narratives reflected an accurate and credible representation of doctoral candidates’ experiences. Ultimately, I resolved the validity issue through what Denzin and Lincoln describe as ‘member checks’ (1998: 50). These are opportunities that participants have of reviewing the material that they provided. For this research, once I had transcribed each interview, I sent a copy to the interviewees, asking them to peruse it for accuracy. Only two-thirds of my sample responded; of these, six added additional comments or requested certain deletions. I assumed that those who did not respond either did not care about the accuracy, or assumed that I would interpret their silence to mean that they were happy with the transcripts. This interaction with the participants reduced bias in the data, which Cohen, Manion and Morrison suggest emanate from the following sources: • the attitudes, opinions and expectations of the interviewer; • a tendency for the interviewer to see the respondent in her own image; 129 • a tendency for the interviewer to seek answers that support her preconceived notions; • misperceptions on the part of the interviewer of what the respondent is saying; • misunderstandings on the part of the respondents of what is being asked (Cohen, Manion & Morrison 2002: 121) Reliability is a second key aspect of ensuring the rigour of research. Like validity, it has a different conceptualization in quantitative and qualitative research. According to Merriam (1995): Reliability in the ‘hard’ sciences revolves around repeated measures of a phenomenon. Typically investigators dissociate themselves from the phenomenon being investigated by using ‘objective’ measures. The more times the findings of a study can be replicated, the more stable or reliable the phenomenon is thought to be (1995: 55). Kvale and Brinkman (2009) have contextualized the term with regards to the reliability of interview data: Reliability pertains to the consistency and trustworthiness of research findings; it is often treated in relation to the issue of whether a finding is reproducible at other times and by other researchers. This concerns whether the interview subjects will change their answers during an interview and whether they will give different replies to different interviewers (2009: 245). Clearly, when working with human subjects and their experiences of social phenomena, there can be no replicability: no two individuals will experience the same phenomenon in the same way. In a later work, Merriam elaborates on this: ‘Reliability is problematic in the social sciences simply because human behaviour is never static nor is what many experience necessarily more reliable than what one person experiences’ (Merriam 2002: 27). This was a concern that I had about my interview data, and I spent many hours reflecting on how I could extrapolate themes when each candidate had had a unique experience of being supervised. Thus, those who had had a positive experience related the causes of this to a varied set of reasons. Further, what some found to be positive (for instance, exposure to disciplinary conferences), others were indifferent to. Instead of viewing this as a problem, however, I chose to use it to strengthen my argument: that there is no ‘one-size-fits-all’ approach to doctoral 130 supervision. In other words, it is in the absence of replicability that conclusions about supervision can be drawn. Thus, as Bogdan and Biklen (2007) point out: Qualitative researchers tend to view reliability as a fit between what they record as data and what actually happens in the setting under study, rather than the literal consistency across different observations (Bogdan & Biklen 2007: 40). This was the conceptualisation of reliability that I used for my study. As I have already mentioned, because there was little replicability across my interview data, it would have been impossible to ensure the literal consistency to which Bogdan and Biklen refer. While there were some similarities, these could not be construed as anything more than just similarities. As I analysed my data, I became increasingly concerned that I would be unable to demonstrate its reliability and that this would undermine the value of the research. I subsequently came across the following argument by Lincoln and Guba (1985): ‘Since there can be no validity without reliability, a demonstration of the former [validity] is sufficient to establish the latter [reliability]’ (Lincoln & Guba 1985: 316). In view of the fact that I have been able to assert the validity of both my data and its analysis, I would argue that its reliability flows can therefore be assumed. The term ‘generalizability’ is explained as ‘the extent that findings in one situation can be transferred to other situations’ (Kvale & Brinkman 2009: 324). In common with validity and reliability, it is a term that is more appropriate to quantitative than to qualitative research. Myers (2000) points out that generalizations are not always possible in qualitative research, especially when it is largely irrelevant to both the aims of the research and to the context in which the research took place. My study was never intended to be generalizable – my aim was to investigate current practices of doctoral supervision at the University of the Witwatersrand. The fact that it focused on two specific faculties at this institution meant that I never anticipated that its results would be applicable to other contexts. On the contrary, I wanted to know what the existing practices were so that I could adapt my work as an academic staff developer to addressing particular areas of concern, as well as to strengthening existing good practice. Therefore, I am satisfied that even though my results may not be generalisable outside the university, they are nonetheless valid and reliable within this context. 131 4.8 Ethical issues An interview inquiry is a moral enterprise. Moral issues concern the means as well as the ends of an interview inquiry. The human interaction in the interview affects the interviewees, and the knowledge produced by an interview inquiry affects our understanding of the human condition. Consequently, interview research is saturated with moral and ethical issues (Kvale & Brinkman 2009: 62). When working with human subjects, researchers are often required to make decisions about how to use the data provided by these subjects in such a way that it does not end up harming them. Christians (2000) suggests four guidelines for ‘… directing an inductive science of means toward majoritarian ends’ (2000: 138). These are informed consent, deception, privacy and confidentiality, and accuracy. While I am aware that other authors (such as Bogdan & Biklen 2007; Cohen, Manion & Morrison 2002; and Silverman 2000) have offered a broader range of ethical guidelines than this, I am satisfied that Christians’ four adequately address the issue in the context of my research. I will examine each in some detail, and will report on my efforts to ensure that I upheld ethics at the highest level. Whether or not I was successful remains to be seen. 4.8.1 Informed consent As the name suggests, this guideline relates to the freedom that research subjects have to participate or not participate in the research, based on their understanding of the risks involved. Kvale and Brinkmann (2009) expand on this definition as follows: Informed consent involves informing the research participants about the overall purpose of the investigation and the main features of the design, as well as of any possible risks and benefits from participation in the research project. [It] further involves obtaining the voluntary participation of the people involved, and informing them of their right to withdraw from the study at any time (2009: 70). From this, it is clear that providing the research participant with a thorough and clear overview of the project, as well as her contribution to it, is essential to the integrity of the research. She needs as much information as possible to decide whether or not her participation would represent an unacceptable risk to her – either physically or emotionally. This subsequently makes the space for her to volunteer. However, 132 informed consent goes further than just the initial interaction between researcher and participant: it also includes how the data provided is analysed and reported (Babbie 2005: 64). In other words, participants who read the final work and who are able to identify themselves in it, should not find characterizations of themselves that are harmful, destructive and possibly far removed from the truth. As Babbie notes: ‘At the very least, such characterizations are likely to trouble them and threaten their self images’ (2005: 64). This is a form of harm, even though it may not have been deliberate. For my study, all my interviews were preceded by an explanation of my research: what I was studying, how I was approaching the research and why I believed it to be important. The interviewees filled in two consent forms: one that gave me permission to interview them, and one that gave me permission to tape record the interaction. The University’s Ethics Committee had instructed me to separate these two forms – instead of combining them into one – because it felt that the interview and the recording thereof were two distinct processes and that the participants needed to know that they could refuse the latter if it made them uncomfortable. I also made it clear that they could decline to answer any of my questions, and that they could withdraw from the interview at any time, without prejudice. I have included a copy of each on following pages for illustrative purposes (Figures 3 and 4). 133 Figure 3: Letter of consent for the interview Letter of consent: interview To: Doctoral Students in the Faculties of Science and Humanities at the University of the Witwatersrand My name is Jenny Hadingham and I work at the Centre for Learning, Teaching and Development (CLTD) at the University of the Witwatersrand. I will be undertaking research for a PhD degree, focused on doctoral students’ experiences of their postgraduate research education, including supervision. I will collect data from past and present doctoral students about their experiences as postgraduate researchers and would like to invite you to participate in the study. This will involve an interview which should take between 30 and 45 minutes to complete. This interview will be transcribed and a copy given to you for verification purposes. Participation in this interview is entirely voluntary, you may decline to answer questions and you may withdraw at any time, without negative consequences. Please be assured that should you agree to an interview, your identity will be known only to me and when reported, a pseudonym will be used. Furthermore, all features that may identify either you, your supervisor or your discipline will be removed in order to protect the anonymity of all parties. You will also have right to use your interview data on request. If you are interested in taking part, please sign below that you give your consent to take part in this study. Please be aware that even if you do consent to take part, you have the right to withdraw from the study at any time without prejudice. Yours sincerely, Jenny Hadingham Centre for University Learning, Teaching and Development University of the Witwatersrand _______________________________________________________________________________ I agree to take part in this study on postgraduate supervision experience. I am aware that all information provided to the researcher will remain anonymous, and that I can withdraw from the study at any time, without prejudice. Name: Signature: Address: E-mail: Phone number: 134 Figure 4: Letter of consent for the transcription of the recorded interview Letter of consent: transcription of recorded interview To: Doctoral Students in the Faculties of Science and Humanities at the University of the Witwatersrand My name is Jenny Hadingham and I work at the Centre for Learning, Teaching and Development (CLTD) at the University of the Witwatersrand. I will be undertaking research for a PhD degree, focused on doctoral students’ experiences of their postgraduate research education, including supervision. I will collect data from staff and students about their experiences as postgraduates and would like to invite you to participate in the study. This will involve an interview which should take between 30 and 45 minutes to complete. This interview will be transcribed and a copy given to you for verification purposes. You will have the freedom to delete any features in the transcript that you feel might identify either yourself, your supervisor or your discipline. Should you wish to, you may withdraw your transcript from the study without prejudice. Please be assured that should you agree to an interview and for your transcript to be used in the thesis, your identity will be known only to me and when reported, a pseudonym will be used. You will also have right to use your own data upon request. If you are interested in taking part, please sign below that you give your consent to take part in this study. Please be aware that even if you do consent to take part, you have the right to withdraw from the study at any time without prejudice. Yours sincerely, Jenny Hadingham Centre for University Learning, Teaching and Development University of the Witwatersrand _______________________________________________________________________________ I agree to have my interview recorded. I am aware that all information provided to the researcher will remain anonymous, and that I can withdraw from the study at any time, without prejudice. Name: Signature: Address: E-mail: Phone number: 135 4.8.2 Deception As with many concepts in social science, the term ‘deception’ is understood differently in different contexts. Bröder (1998), for instance, defines it as ‘concealing or camouflaging the real purpose of an experiment’ (1998: 806). By contrast, Ortmann and Hertwig (1998) feel that ‘… not telling participants the purpose of an experiment is not necessarily deception; telling participants things that are not true necessarily is (1998: 807). Soble (1978) – predating both of these definitions – draws on the Articles of the Nuremburg Tribunal and the Declaration of Helsinki32 In the second place, deception seems to imply allowing participants only partial access to the information that they would require in order to give informed consent. As the definitions above suggest, some consider that this is acceptable, while others (such as Soble) adopt an ‘all or nothing’ attitude; that is, the participants are either given all the information that they require, or the experiment should not take place. However, he also acknowledges that ‘…certain bits of knowledge cannot, for logical reasons alone, be obtained without the use of deception’ (1978: 40). Haggerty (2004) provides the following example of this: ‘[R]esearchers deliberately misrepresent their identity in order to enter otherwise inaccessible social situations’ (2004: 405). The moral dilemma that this state of affairs poses is arguably at the centre of ethical conduct. I am not convinced that by withholding vital information, researchers will get a better quality of as sources for arguing that ‘the subjects must be told the duration, methods, possible risks and the purpose or aim of the experiment…[I]nformed consent has not been obtained if there has been any element of deceit or fraud’ (1978: 40). There are two pertinent points that can be drawn from these three conceptualizations of deception. Firstly, all three mention experiments, which suggest quantitative research. However, I would consider that any interaction between human beings could potentially expose participants to deception – whether these interactions are (qualitative) interviews or (quantitative) drug trials. 32 The Nuremburg Tribunal (also known as the International Military Tribunal), was a legal process following World War II to bring to book the perpetrators of crimes against humanity during this conflict. The principles of the Tribunal were later adopted by the Law Commission of the United Nations Organisation (http://deoxy.org/wc/wc-nurem.htm). Adopted by the World Medical Association in June 1964, the Declaration of Helsinki ‘…as a statement of ethical principles for medical research involving human subjects, including research on identifiable human material and data’ (http://www.wma.net/en/30publications/10policies/b3/index.html) 136 data. Prior to each of my interviews, I spent at least 10 minutes explaining the context and reasons for my research to the doctoral candidates. Many had questions, such as ‘Why is this important?’ and ‘How can I be sure that my supervisor will not hear about what I have told you?’ I erred on the side of thoroughness when I answered them – I wanted there to be no doubts in their mind about how participation in my study would affect them. 4.8.3 Privacy and confidentiality Confidentiality in research implies that private data identifying the participants will not be disclosed (Kvale & Brinkman 2009: 72). The essence of anonymity is that the information provided should in no way reveal their identity … A participant or subject is therefore considered anonymous when the researcher or another person cannot identify the participant or subject from the information provide (Cohen, Manion & Morrison 2002: 61). [Privacy] extends to all information relating to a person’s physical and mental condition, personal circumstances and social relationships which is not already in the public domain. It gives to the individual or the collectivity the freedom to decide for themselves when and where, in what circumstances and to what extent their personal attitudes, opinions, habits eccentricities, doubts and fears are to be communicated to or withheld from others (quoted in Cohen, Manion & Morrison 2002: 60). I have chosen to juxtapose the definitions of the terms confidentiality, anonymity and privacy to demonstrate that while they have similar meanings, they are not the same thing at all. All three are vitally important in protecting the participants in any study. In my mind, they constitute the three sides of a triangle: one side, confidentiality, means that no one but the researcher knows who participated in the research (including things such as names, age, gender and race). The second side, anonymity, suggests that anyone reading the research will not be able to identify who provided the data. The final arm of the triangle, privacy, allows the research participant to choose what aspects of her input can be used, and in what way. Together, they should ensure that the participant is protected from harm – in an ideal world. However, in reality, there are caveats, because they can: … serve as an alibi for the researchers, potentially enabling them to interpret the participants’ statements without being gainsaid. [They] can protect the 137 participants, but … can also deny them ‘the very voice in the research that might originally have been claimed as its aim’ (Parker, 2005, p. 17)…. (Kvale & Brinkman 2009: 73). I worked hard to ensure that all three were present in my research. I changed the names of the students (and their supervisors), as well as that of the disciplines from which they came in order to protect them. However, I remained true to the race and gender of the candidates and their supervisors in these name changes so that I could maintain the character of their contributions. Furthermore, I did not reveal the real names of my interviewees to either of my supervisors, and the only copy of a list with both sets of names on it is safely stored at my personal residence. As I have mentioned, I also sent a copy of the transcript to each candidate for their perusal and confirmation. While not all responded, those who did had clearly read the transcript carefully, and offered some suggestions which I felt added some depth and richness to the text. 4.8.4 Accuracy In some ways, it almost goes without saying that accurately representing the data is vital. As Christians argues ‘Fabrications, fraudulent materials, omissions and contrivances are both nonscientific and unethical. Data that are internally and externally valid are the coin of the realm, experimentally and morally’ (Christians 2000: 140). However, as Babbie (2005: 144) notes, accuracy can be measured in degrees – there is no such thing as absolute accuracy. He elucidates: There are no conditions under which imprecise measurements are intrinsically superior to precise ones…. Exact precision is not always necessary or desirable… The operationalization of concepts, then, must be guided partly by an understanding of the degree of precision required (2005: 144). To my mind, it makes sense for the researcher (in conjunction with her supervisor/s) to decide what degree of accuracy is most appropriate for her particular research. Furthermore, there is a subjective aspect to determining how much accuracy is ‘enough’. I would therefore argue that a joint candidate/supervisor agreement on what constitutes a fitting level of accuracy should satisfy the institutional ethical requirement. 138 4.9 Conclusion While there may have been other research methodologies that could have been used to answer the research questions (such as questionnaires), I am satisfied that the qualitative interview was appropriate for this work. It provided access to doctoral candidates’ narrative accounts of their supervision experience and allowed me to draw out the themes that were important to them, as well as to the research. As a result, I was able to get to the crux of not only how the candidates were supervised, but also how they experienced the various practices of their supervisors. Having said this, I also recognize that there are limitations. My interviews were confined to the doctoral candidates and did not include the perspective of their supervisors. With hindsight, I realize that the latter would have added another dimension to the thesis, namely, how the supervisors experienced their students’ reflections of the supervision process. Since they represent the other ‘half’ of the research relationship, their perspectives are also important to research in this area. Despite this, it is my assertion that the student ‘voice’ on the practices of doctoral supervision has traditionally been subordinated to that of the supervisors, and this work has attempted to rectify the gap. In the chapters that follow, I present the narratives distilled from the interview transcripts of thirty doctoral candidates at Wits. They represent the student voice alluded to above. It is their perspectives on the supervision process that provides significant insight into the factors that both enhance and impair progress. 139 CHAPTER 5 Tales from the Humanities Chapter overview In this chapter, the Humanities doctoral candidates that I interviewed share their experiences of their doctoral supervision. In each narrative, I changed the names of the candidate and supervisor, and altered the name of the discipline in order to protect those involved; however, I remained true to the letter and the spirit of each interview throughout. 5.1 Introduction This chapter contains the narratives of the Humanities candidates interviewed for the thesis. Each is a reworking of the interview transcripts, and encapsulates the essence of the interviewees’ experiences as doctoral candidates. I chose not to present them in the same sequence in which the interviews took place. Instead, it made sense to me to present the profiles according to faculty because in this format, the reader would be able to identify, amongst other things, the particular models traditionally associated with each Faculty, as well as any deviations that occurred within these Faculty models. 5.2 The narratives 5.2.1 Hannah33 Supervision model: Single supervisor Race and gender - student: White female Race and gender - supervisor: Coloured male Graduation status: Graduated November 2009 Hannah’s introduction to doctoral supervision by Professor Pieter Pieterse was a positive one: he was intellectually engaging and ‘had interesting ideas’. In addition, as the research process unfolded, she found his style of supervision to be compatible with her style of learning. He allowed her the space to ‘do my own thing’ and did not micro- manage her at every step of the process. As a mature student, Hannah possessed much 33 The names of both the supervisor and the student have been changed to protect their identity, as has the name of the specific discipline. However, I have maintained the race and the gender of the participants even though neither of these was used as a unit of analysis in the thesis. Given the racialised and patriarchal nature of South African society – despite it being in its second decade of democracy – I felt that it was important to preserve these designators. 140 more independence and internal motivation than her younger counterparts in the department, and she was allowed more freedom to do her research. That is not to say that she did not need some input from Professor Pieterse from time to time. On the contrary, when questions did arise, she would simply phone him for advice and the subsequent informal interaction would suffice to deal with the issue at hand. During the early phases of the process, supervision meetings were generally ad hoc and only took place when particular issues, such as the use of software for the analysis of her interview data, arose. In these cases, Prof Pieterse’s input was thorough and thoughtful, and contributed to her progress. However, despite these constructive practices, Professor Pieterse was not always the model of a good supervisor. For instance, one of his expressed ‘conditions’ for supervising her was that she would be an active participant in the intellectual life of the department, including making a contribution to various departmental research projects over and above her specific PhD research. Hannah agreed to do this, on the counter- condition that this participation would lead to learning opportunities for her. However, this never happened. Often, Hannah and her peers in the programme were required to design individual projects (which the leaders of the programme had won tenders for), undertake research and analyse the data from it, but this was never done in a group and so many social learning opportunities were lost. In addition, these projects took Hannah’s focus away from her PhD research, often for months at a time, thus hampering her progress. Hannah expressed anger at the ‘exploitation’ of students that this kind of work represented. In order to contextualize Hannah’s reactions to Professor Pieterse as a supervisor, one needs to look at her background (both academic and personal). Twenty years after her first (failed) attempt at a PhD, Hannah, a middle-aged single mother, returned to academia determined to succeed this time in her PhD studies. Her first attempt had been at another South African university, in a different Faculty (Science), and had been brought to an end by a number of factors, including that her chosen topic was rejected by the supervisor in favour of one that was ‘practical and logical and workable’. A back operation that went wrong left her bed-ridden for many months. A painful divorce followed swiftly by a pregnancy and a move to another part of the country ensued. 141 During a teaching stint at a university, her Head of School made it clear to her that her PhD research would not be supported by the department, and she subsequently resigned in favour of working in industry – where she stayed for the following two decades. Over time, she came to regret that she had never managed to complete her PhD, and began to investigate the possibility of returning to academia. One possibility was to study in Australia, but the financial implications were beyond her means, especially given that she was a single mother. Thus, when she came upon an advert in the newspaper recruiting PhD students for a research programme in the Faculty of Humanities, she decided ‘Why not’? However, the move to a different Faculty from the one in which she had earned her previous three degrees was not a smooth one, initially at least. The intellectual leap from the discourse of Science to the discourse of Humanities was a huge one for her. Problems first emerged in the writing of the proposal: while in Science, no more than half a page was required, covering the broad topic to be researched, the Humanities proposal was a different creature altogether. It had to be a far longer document (about 6000 words, according to the School’s website), and due to the nature of knowledge in the Humanities (in her words ‘messy’, ‘non-linear’ and ‘won’t fall into nice, structured boxes’) it required a broader approach to research design. Throughout the proposal- writing process, Professor Pieterse was erratic in his feedback to her: ‘He asked me to change things and then three versions later, he asked me to change them back, and it depended on his whim and what he had for breakfast…’. As a result, she spent a great deal of time stuck in this stage, feeling disempowered. This was exacerbated by Professor Pieterse’s poor administrative skills; for instance, he failed to complete the necessary procedures when he finally got around to submitting her proposal to the Faculty Higher Degrees Committee. During the six months that it was in limbo in this Committee, he never queried the reasons for the delay. Since Hannah had chosen to be a full-time student for the duration of her PhD studies and therefore did not have a regular income, this had negative financial implications for her. Intellectual frustration also plagued her during this time. 142 While Hannah never co-authored papers with Professor Pieterse (he was usually too busy or away from the office) during the course of her PhD research, she published a number on her own, passing them on to Prof Pieterse for comment but rarely receiving any. Having come from an academic background, she was very aware of the value of these publications for her future career in the academy – something that she feels was not sufficiently highlighted at the University. She noted: [For students] the university needs to be clearer on the whole process of accredited journals and getting your stuff recognized and what reward they get for it because as far as I know, no matter how many journal articles I produce, I don’t get any reward for it. And it would be nice if the university could clarify that at a particular stage. In terms of the power relations between her and her supervisor, Hannah was not over- awed by Prof Pieterse’s high academic status at the University or his role as her supervisor: on the contrary, her years in industry had imbued her with sufficient assertiveness to insist that he performed his part in the relationship. Thus, for instance, she gave him four months to read the final draft of her thesis – if he did not to meet this deadline, she made it clear that she would submit it for examination without his permission. Fortunately, it never came to that. Overall, it appeared that Prof Pieterse’s approach to supervising Hannah was one of benign neglect. While he did provide feedback and support when requested to do so by Hannah, he rarely offered it without her prompting him. Fortunately, Hannah was confident in her own ability as a researcher, and was able to adapt to this style. She was thus prepared to submit her thesis for examination without his approval if he did not read her final draft timeously. This position of personal power and ownership was in stark contrast to her initial position of confusion and disempowerment. 5.2.2 Thabang Supervision model: Single supervisor Race and gender – student: Black male Race and gender – supervisor White male Graduation status: Submitting at the end of 2010 143 Thabang, a research fellow in International Politics, is a thirty-something Swazi national. His experience of being supervised by Prof Jack Baldwin began at Master’s degree level in 2002. Upon completion of this degree, and with the intention of becoming a full-time member of the department’s academic staff, Thabang registered for his PhD at Wits. Thabang perceptions of Prof Baldwin as being ‘super’, ‘knowledgeable’, ‘brilliant’ and a ‘hard worker’ meant that there was never any question in his mind that Prof Baldwin should be his supervisor at PhD level, too. The almost reverential awe in which he held this supervisor has played a big role in his own identity formation as an academic. Thabang portrayed Prof Baldwin as the kind of supervisor who required his students to focus on their research and academic identities, rather than work on projects outside the immediate ambit of their PhDs. Consequently, he was unhappy with Thabang’s many forays into consulting work during the course of his PhD. He urged him instead to concentrate on activities such as tutoring, lecturing and the publication of papers based on his research. In the case of the latter, Prof Baldwin was receptive to the practice of co-authoring research papers with Thabang, essentially modeling for him the processes that an academic should go through in the development of conference papers. This included using ‘positive criticism’ to guide Thabang’s ideas away from ‘naïve’ thinking, towards something more reasoned and do-able. Supervision meetings were flexible, unscheduled affairs, with Thabang requesting the meeting via email, and this meeting taking place within a day or two of the email. They were relatively informal affairs and generally centred on a piece of written work that Thabang had submitted to Prof Baldwin several days in advance. During these meetings, Thabang never had to take notes because Prof Baldwin had made a point of putting all of his comments in writing. As a result, Thabang was able to use the meeting time to actively listen to and engage with what his supervisor was saying. He subsequently attached Prof Baldwin’s comments to his computer CPU, where he could frequently make reference to them. The fact that he was flexible in his meeting time was not to say that Prof Baldwin dropped everything to see Thabang when the latter needed to meet. At one point, Prof 144 Baldwin was forced to take on numerous additional administrative tasks when the Head of School retired, and during this time, Thabang had to make formal appointments to see him. However, even then, Prof Baldwin made a point of including Thabang in the day-to-day affairs of the academic department – one that Thabang would later join as a lecturer. For example, at this time, the Centre at which Thabang was a research fellow was in need of a major upgrade of its computer labs – Prof Baldwin assigned Thabang the task of writing the funding proposal for this project, and kept him informed of its progress as it entered the donor’s review process. Thabang related how this one thing made him feel less isolated and bored, and more like he was making a contribution to the department. In terms of his style of supervision, Prof Baldwin was able to engage with Thabang at both a professional and intellectual level, while at the same time recognizing that there was more to his identity than just the academic aspect. Thus, for example, the two would often have frequent (and often heated) discussions about topics outside Thabang’s specific area of research. In addition, the two of them (along with Prof Baldwin’s children) would often play tennis together, or attend football matches at which Bidvest Wits34 Thabang identified a number of specific practices undertaken by Prof Baldwin that ensured that the PhD research process was a productive one for him (Thabang), namely constructive feedback on his written work, and within a relatively short space of time; co-authored research papers with him; acting as a sounding board for Thabang’s ideas; introduced him to key thinkers in the field, and encouraged him to engage with them for several chapters in his thesis; provided him with access to funding for his field was playing. Thus, Thabang described Prof Baldwin as being like a father or brother to him during the course of his PhD studies – someone who cared about him, and about whom he in turn cared. This father/brother wielded immense authority over Thabang, so he was never able to get away with being lazy or falling behind in his studies. To do this would mean disappointing Prof Baldwin, and this was something Thabang was unwilling to do – “I would be quite cross with myself if at any point I would find myself in a situation where I have to disappoint him”. 34 Bidvest Wits is the University’s soccer team, which participates in South Africa’s Premier Soccer League. 145 research; and encouraged him to improve his time management skills. Each of these practices played an important role in his development, according to Thabang, and added to the final value of his thesis. 5.2.3 Nkosinathi Supervision model Two supervisors/co-supervision Race and gender: student Black male Race and gender: supervisor White female and black male Graduation status Graduated April 2010 Nkosinathi is an earnest young researcher in his late thirties. His positive experience of PhD supervision has derived from the fact that he had two supervisors: his primary supervisor, Professor Evelyn Smith, headed up the research project of which he was a member and Professor Thabo September was his secondary supervisor. Professor Smith’s area of expertise related to the overarching theme of the research project, while Professor September provided the guidance on a very specific sub-theme that Nkosinathi wished to pursue in his doctoral research. Nkosinathi felt that he benefited from joint supervision because each of his supervisors brought something different, but equally valuable, to his work. Professor Smith ‘would encourage a kind of structure’ that his research needed, while Professor September contributed a ‘conceptual rigour’ and ‘thematic bias’ to the process. This situation was in stark contrast to his Master’s experience, in which he had a single supervisor, who was not only impatient with him (‘sometimes she would shout at me, she would be very horrible, more than once’), but also failed to give him the intellectual feedback that he required. Thus, for instance, the first time he discovered that the theoretical framework that he had used in his thesis was inappropriate was when he read her internal examiner’s report, once the work had already been sent out for external examination. During supervision meetings, Professor Smith would spend a great deal of time guiding Nkosinathi through the process of structuring his arguments, modeling for him how to ‘develop ideas from one to another until you get to that point where you are taking your readers along… until the reader understands fully what … what you are arguing about’. She was also able to redirect him when, for instance, he lost sight of the historical 146 development of certain ideas and theories, and attempted to analyse an event that occurred in the 1960s using an idea that only emerged in the 1980s. In addition to these constructive, content-focused practices, Professor Smith also possessed a ‘sensitivity’ which Nkosinathi described as the awareness of how far she could push him and when, as well as when she needed to hold back in order to prevent him from getting anxious about his work. She thus seemed to instinctively know what she would work with him on, and when, recognizing that his approach to research was unique. Professor September, on the other hand, provided Nkosinathi with the opportunity to think freely about his research. He recognized that students often fall into the trap of getting so ‘locked up in theory that they forget the text’ and so he encouraged Nkosinathi to think widely and freely and to ‘go for the text, focus on it’. Thus, it seemed that the two supervisors complemented each other. Since Nkosinathi was a member of a seven-member research entity in Humanities, he was already a member of a particular community of practice. This community had a physical space – a set of adjoining offices as well as a ‘staff room’– and within this space, Nkosinathi and his project colleagues would often chat informally about their research, sharing their findings and debating theory. However, this never translated into joint publications. He reported that he had not attended or presented at many conferences, a valuable opportunity for expanding his learning. His primary reason for this was that most conferences that he knew about had themes were ‘not so relevant to what I’m doing’. However, there was a sub-reason, too: he argued that once he started working on a chapter (or even a section of a chapter) he got so absorbed in the work that he was reluctant to get ‘side-tracked’ by the preparation of a conference paper. He seemed to be very focused on completing the thesis, and anything (even a conference paper) that might impede this was anxiety promoting. Fortunately for him, neither supervisor forced the issue which alleviated some of his anxieties. Nkosinathi’s greatest obstacle in the path to his PhD was his command of the English language. Since English was not his ‘mother tongue’, nor had he taken it as a first language at high school, he found it challenging to write in this language. This problem was amplified by the requirement for the thesis to employ the relevant disciplinary 147 discourse35 Supervision model . The fact that the discourse that he needed to use was a highly specialized and complex one only added to this problem. At an even more fundamental level, he struggled ‘knowing how to put facts correctly’. This was often evident in workshops that he co-facilitated with Professor September: ‘I would have to present and I feel frustrated by the fact that the language is not readily available to people...’. Although both supervisors have been supporting his language development, and he believes that he has improved, it seemed that it remains an anxiety for him. In terms of the power relations between himself and his two supervisors, Nkosinathi was content to identify himself as an apprentice to their master, and ‘the reason is simply that one’s experience is not so much, uh, as one’s supervisors’ experiences’. Having said that, he argued that there was nonetheless room for ‘conversation, and for understanding the implications of taking certain routes in one’s PhD’. This had become evident to him as the research process had unfolded. When he began the process, for example, ‘I would take the models… that I’m given so that I’m on the straight’. Once he had reached a certain point in the process, he reported that ‘the supervisor would say “Now it’s yours to take … to do whatever you want to do”’. What he seemed to be telling me was that the power relations had been scaffolded: it was more directive at the start of the process, but gradually balanced out as he advanced. 5.2.4 Marcus Single supervisor; upgraded Masters to PhD Race and gender: student White male Race and gender: supervisor White male Graduation status Graduated November 2008 Marcus is a slightly built lecturer in the Social Sciences, who completed his PhD in 2008. His experience of being supervised was complex, for two reasons: firstly, he had upgraded his Master’s degree to a PhD, which had extended the process by a 35 I have already defined the term ‘discourse’ in an earlier chapter. However, it is worth repeating here. Gee (1996) defines it as the ‘… ways of behaving, interacting, valuing, thinking, believing, speaking, and often reading and writing that are accepted as instantiations f particular roles…’ (1996: viii). In a similar vein, Lea and Street (…: 370) define the terms as ‘…the ways of talking, writing, thinking, and using literacy that typified members of a disciplinary or subject area community. 148 significant period of time; and secondly, he was a lecturer in the same Department in which he was being supervised, and in this capacity he had been given the responsibility to coordinate this Department’s postgraduate programme. Marcus felt that he had been very fortunate in his choice of supervisor – Professor McNeil – because he knew the area in which Marcus was interested, as well as the literature. This was very affirming for Marcus, particularly because his research interests are ‘obscure’, by his own admission. Furthermore, in order to ‘keep up’ with Marcus, Professor McNeil ‘read up’ even more on the topic. The feedback that Marcus received from Professor McNeil was provided within a relatively short period of time, and was ‘very high-level’. This written feedback was discussed at length during the supervision session, and was supported by oral input: ‘…he would kind of say, like, “Well, what about this angle, what about that angle? Have you read this? You’re going too far here, cut off there, you’re gonna make this thing too big” ‘ Power did not seem to play a big role in the relationship between Marcus and his supervisor. He was allowed the space to disagree with Professor McNeil, which he frequently did. Marcus reflected in the interview that ‘…he was very open-minded. I criticized him a lot in my thesis, I criticized his arguments, his articles, and he didn’t penalize me for that’. He attributed this relative power balance to the fact that he was ‘confident in my abilities. I mean, I wouldn’t have had a problem disagreeing with him, and saying “You’re wrong” ‘. Furthermore, at the same time as he was being supervised by Professor McNeil, he was also working on various projects alongside him – ‘as a sort of a peer’ – such as organizing conferences, co-writing articles (the order of authorship was based on alphabetics – and hence, Prof McNeil’s name appeared as first author - not power), and co-editing various journal and book articles. Marcus used the specific word ‘protégé’ to describe this relationship. These two identities – Marcus as student and Marcus as peer – seemed to blur the boundaries of power between supervisor and student. 149 5.2.5 Faith Supervision model Single supervisor Race and gender: student Black female Race and gender: supervisor Black male Graduation status Graduated November 2009 Faith is a young African woman who registered for a PhD directly after completing her Master’s degree three years ago. She reflected that this may not have been a good decision because it meant that it did not allow her the space to develop a ‘maturity and familiarity’ with the field or a ‘clarity of “I want to do this because of this”’. What swayed her to begin her PhD was the powerful and positive experience that she had had with her Master’s (and subsequently doctoral) supervisor, Prof Simon Nzimande, as well as their shared interest in the field of Kenyan and British colonial literature. In her mind, they would both benefit from her research in this area. The downside to this familiarity with one another was that he would ‘forgive stuff that other people wouldn’t forgive’. This ‘stuff’ was specifically about writing style: I believe in a variety of opinions and … that’s the benefit of working with a different person at PhD level from the person you worked with at Master’s level, because you learn other ways of doing things, [which you would not learn] having worked with this same person from Master’s to PhD. Sometimes this person gets too familiar with your style of working…. Prof Nzimande knew from his supervision of Faith’s Master’s degree that she was not a student who worked well with supervisor-imposed timelines and deadlines. By her own admission, Faith felt ‘overwhelmed’ by these, and had therefore developed a system of setting (and largely sticking to) deadlines for herself. She recognized that the danger of this system was that the responsibility for adhering to her deadlines lay with her – any breach on her part would upset the ‘delicate relationship’ of trust between her and her supervisor: ‘trust is always delicate because you don’t want to create a scenario where your word would not be taken seriously’. Finally, she and Prof Nzimande agreed that she would work at her own pace, on the proviso that she kept him updated on her progress, and gave him sufficient notice when she needed to meet with him (especially if he was required to give feedback on written work). 150 There were also the occasional ‘impromptu’ meetings, when Faith would bounce some ideas off the professor, and he would give her oral feedback (including recommending the works of relevant authors). In addition, there would be a great deal of written feedback from him, which was always supported by discussion. It usually dealt with ‘big picture’ issues in her writing. For instance, after she had submitted her first two chapters, he sat her down and ‘…told me twice “Listen, you’re saying great stuff, but there’s no argument here”’. At first, this kind of feedback frustrated her, but over time she realized how important it was to get it sooner, rather than later: ‘…you only get what you are doing in your last year… the other years are just trial-and-error years’. During the process of undertaking her PhD research, Faith has published ‘just a few miserable publications’ (two chapters from her thesis, in fact). She credited Prof Nzimande with alerting her to various publication opportunities in journals. On one occasion, he was asked to edit a special edition of a particular journal and he told her: ‘Listen, I’ve read this chapter and I think it fits in very well with ... what I’m doing with this journal’. Her facial expression (a broad smile) and the amount of times she laughed at herself while relating this anecdote, suggested that she had recognized the intellectual affirmation that Prof Nzimande was giving her by offering her these opportunities: ‘I guess we are sort of children to a certain degree, just knowing that listen, my supervisor honestly thinks that I can submit a paper for publication in this journal, makes a huge difference’. Interestingly, though, the two of them had never co- written anything, but she was upbeat about the future possibility of this happening. Faith was very critical about the absence of a community of practice in her research experience at Wits. Aside from alerting her to publishing opportunities, Prof Nzimande did little to introduce her to the community of scholars working in the same field as her. She recognized this gap quite early on in the PhD research process, particularly after she had attended graduation ceremonies and Graduate School workshops: ‘Hell, that’s in Humanities, its right down my path but I didn’t know someone was working on this field’. She felt quite bothered by it because she was conscious of the fact that had she not attended the workshop or graduation, she would never have known that there were other people working in a similar area to hers. Ultimately, ‘… you’ve got to 151 go out of your way and build your own personal relationships with people…’. It also served to heighten her sense of isolation within the Faculty. The issue of power in the supervision process was something that she was keenly aware of, and that she raised before I had even asked the question. Clearly, she was an assertive student who knew where she wanted to take her research. However, she did feel from time to time that Prof Nzimande was diverting her from the direction she wanted her research to take. She was quite candid about her determination to not ‘… end up writing the PhD thesis that [my] supervisor wants’. Hence, when situations like this arose, she had to draw on ‘some degree of diplomacy’ to deal with them. For instance, when he suggested that she investigate the work of Foucault, she knew that she couldn’t ‘dismiss it off hand’ – thus, she did the reading and prepared a strong case for not using his theory in her work ‘… because that gave the sense that I have actually considered this suggestion’. Prof Nzimande was comfortable with this, and allowed her the space to disagree with him. Her perception of whether or not Prof Nzimande had developed her as a researcher was telling: ‘I’m not sure, but I don’t think… probably inadvertently, he probably developed me, but mainly inadvertently’. She explained that while he had developed her understanding of the expectations of PhD research (such as the contribution of new knowledge to the field – ‘Listen, everybody knows about this, you need to move, you need to tell us what you are telling us about this that is new’) and had firmly slashed the ‘padding’ that she had added to her chapters when she had nothing to say, he had not actively given her research guidance. Since Faith was a self-starter student, this was probably all that was needed. 5.2.6 Patricia Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White female Graduation status Still in progress 152 Patricia is a married mother-of-one in her mid-thirties who lives in Cape Town and commutes to Johannesburg three or four times a year to meet with her supervisor, Professor Kerry Pinker. Prof Pinker also had a house in Cape Town, and makes a point of seeing Patricia when she has gone there to holiday. Thus, the two do not see each other very often, and Patricia has to make the most of each session. Working in academe has long been a dream of Patricia’s, one that she cut short when the politics at the University (not Wits) at which was working simply became overwhelming and she resigned her post there in 2003. Her desire to earn a PhD, however, never wavered. She knew what she wanted to research and what methods she wanted to use in her research (anthropological), and she was very clear about the caliber of academic with whom she wanted to engage. Sinceshe knew that her former employer could not offer her the latter, she began to cast around for alternatives. Throughout this time, Patricia had stayed in informal contact with Professor Pinker – ‘… she was always there in the wings, … being interested in how I developed my academic career, and of course, I continued to follow hers…’. When Patricia resigned from her job, and decided to register for her PhD at Wits, Prof Pinker expressed a strong interest in supervising her, with some tongue-in-cheek conditions: ‘She advised me not to get married during the time, she advised me not to have children during the time!’. Patricia went on to do both. Initially, Prof Pinker attempted to steer Patricia into a broader project that she was overseeing, and for which she had extensive funding. However, Patricia already had a topic of research in mind and after some non-confrontational ‘back and forth, back and forth’, her supervisor ‘allowed me to do what I wanted to do’. This was despite the fact that Prof Pinker was running a number of research projects with ‘huge amounts of funding’ and ‘so it was very much in her interests, for both hers and mine, to be a part of the larger study’. Further, Patricia’s interest in a specifically anthropological study of her chosen research subject was not shared by her supervisor, ‘but she still trusted me and she let me get on with it’. And ‘get on with it’ Patricia did – she compared Prof Pinker’s guidance to someone steering a ship: ‘she uses her very breadth of knowledge in various fields to suggest a direction to me, and then it’s up to me to take that direction and develop it’. 153 The location of a typical supervision session for Patricia and Prof Pinker was somewhat different from the norm – ‘often it’s not [in an] office, sometimes it can be on the floor of the canteen at [University], sometimes it can be at her house in Cape Town on a Saturday morning, with my baby running around’. The content of these session, however, seem to have a ‘regular’ format: Patricia emails her supervisor some written work in advance of the meeting, so that she has time to prepare. During the supervision session, Prof Pinker talks her through her impressions of the work in a way that ‘…she doesn’t engage in a critique of the material, she sees where it can be developed and extended”. In addition, she puts Patricia in contact with academics from around the world that she has met at various conferences, and who are experts in their fields, and suggests relevant readings where there have been recent developments in the field. These readings come from across the spectrum, ‘right from literature through to medicine’, allowing Patricia a very broad scope for learning. With few exceptions, she has walked away from a supervision meeting feeling very positive about her progress, ‘that my material is appreciated and that it’s worthwhile’. Where she has felt frustrated after a session, it is usually because Prof Pinker has suggested something highly impractical: ‘for example, she’s asked me to video record my data collection. I work in a squatter camp!’ Thus, her irritation is usually at the level of practicalities, rather than the actual substance of the research. Feedback from Prof Pinker takes the form of both oral and written comments, and is variable in its usefulness. There are times when it is clear that she has given the written work her ‘utmost attention’ and her comments have provided extensive guidance for Patricia. On other occasions, the rather unhelpful comment at the bottom is ‘Beautifully constructed, pleasure to read’. Patricia identified this as a source of frustration for her, but recognized that because the process had been a protracted one for her (5 years at that point), ‘I cannot expect that at any point in time, she doesn’t know if I’m going to hand her something in March, in April, in May, in June…’. In addition, Patricia made use of her own notes that she jotted down during her supervision meeting: ‘often, you know, I’ve said to her “Just stop, I want to write down that sentence word-for-word”, and she’s okay with that’. Despite these fluctuations in the quality of feedback, one 154 thing that has remained constant has been Prof Pinker’s excitement at Patricia’s findings (even when Patricia feels that she is mired in a ‘swamp of data’). Patricia felt that, during the research process, Prof Pinker was developing her into being an even better researcher: ‘[She’s] helped me to become a very astute thinker, a very critical thinker and she’s helped me to be able to understand very, very, very broad fields of information’. She also credits her supervisor with guiding her beyond simply giving an answer, but ‘to seek meaning in something, to totally reserve judgment’. Patricia went as far as to say that even if she never got her PhD, she could still continue to develop both in knowledge and as a practitioner. Power does not play a very large role in the relationship between the two women. Patricia attributes this to two main factors: 1) That she is working independently, and is not a part of one of her supervisor’s large research projects (Those students who are a part of these projects generally end up undertaking administrative and organizational tasks for Prof Pinker) and 2) That she is situated in a different geographical location to her supervisor, which puts a distance between them. In addition, having worked in academe herself, and currently working as a practitioner in the field, Patricia considers herself to be less of a ‘student’ (with the power imbalance that this implies) and more of an ‘equal’. This more-or-less equality plays itself out in the realm of writing and publishing together. While the two women have never physically sat down and co-written, they have developed presentations together. In these cases, Patricia would write up the research, while Prof Pinker would correct it for ‘editorial errors’. Invariably, Prof Pinker would insist on being second author, if she even put her name on the research. Patricia is in two minds about the latter situation: ‘I thought “Doesn’t she want her name on my paper? Or is it because she feels that it hasn’t been hers?”’. 155 5.2.7 Zodwa Supervision model Single supervisor Race and gender: student Black female Race and gender: supervisor White female Graduation status Still in progress Zodwa is a forty-something doctoral student and university lecturer from the Eastern Cape. Her first attempt at a PhD – at her home institution – was stymied by a disengaged supervisor. This man (who she refused to name) was a colleague of hers, whose office was right next door to hers. She had high hopes for the relationship (‘I had the greatest opportunity because we were in the same institution and his office was right next to mine…’), describing her supervisor as ‘good and venerated’. However, he never took the time to sit and work through the various drafts of her proposal with her. Zodwa recalls how ‘he would take whatever I had written, go to his office or wherever, go through it and mark it, and come back and say “Here’s your proposal, your draft proposal, there are corrections that you need to make”’. She felt increasingly disaffected with the fact that there were no one-on-one interactions between the two of them. To add insult to injury, his comments were mostly about grammatical errors. He did not make any comments on the actual content of the proposal, which was a source of concern for her. Furthermore, when she got her proposal back from the Higher Degrees Committee, the title had been changed without her consent. She was unconvinced by the reasons that the committee provided for this change, and when she confronted her supervisor about it, ‘he was also wishy washy’. Zodwa was very angry at this point – the supervisor had seen (literally) thirteen drafts of the proposal, and at no point had he indicated to her that the study (or its title) might be rejected by the Committee. On the advice of a colleague, she withdrew her proposal from her home institution, and began to send copies of it to a range of other South African universities, including Wits. Prof Diane Bellarosa, who later became Zodwa’s supervisor, went through the proposal with nine other colleagues from the Department, and within a week had sent her constructive feedback. This included an affirmation that her topic (that had been 156 changed by her home institution’s HDC) was ‘researchable, original and [had] never been done in South Africa’. Prof Bellarosa invited Zodwa to register at Wits, and asked her for a copy of her Master’s thesis. She felt uncomfortable about this because ‘I knew that it wasn’t a good thesis’, but sent it anyway. In the end, though, allowing Prof Bellarosa to read the Master’s in conjunction with the PhD proposal had a useful spinoff because from these two documents ‘they had identified that one of my major problems was the methodology’. As a result, they encouraged her to attend a six month methodology programme run by a School in the Faculty ‘where they help you with different methodologies and also a lot of readings that you do, summarise those readings and make comments as to how you would use them in your topic…. The lightbulb was really switched on’. Zodwa met with Prof Bellarosa on a weekly basis, usually off-campus. There were times when they met twice a week. Zodwa usually emailed her supervisor some written work two or three days prior to the meeting so that there would be something of substance to discuss. In contrast to her previous PhD supervisor, Prof Bellarosa was very engaging: ‘We discuss what I had written… I have to give her a presentation of what my chapter is all about, what I am saying here, and then she gets the opportunity to ask questions and then seek clarity on certain issues, and then we discuss the comments that she has made on the paper’. Since Zodwa is a second-language English speaker, there were times when her arguments were unclear. Prof Bellarosa dealt with the issues of language and meaning by asking guiding questions, such as: ‘What do you mean by this? How do you understand it?’ Zodwa felt supported by these questions, and remarked that they had helped her to ‘unearth’ meanings that had been obscured by her poor English skills. Prof Bellarosa also introduced Zodwa to a new and completely different way of working with feedback, designed to help her view it through a different ‘lens’: ‘She’s taught me something which I’ve never thought about, like to sit in a coffee shop, go through what I have written and she likes to say “You must take yourself away from your study, from what you are reading, so just print it out, go to a restaurant or go somewhere, you know, on the lawn, sit on the lawn and read it through and edit it”’. 157 Feedback on Zodwa’s work did not only come from Prof Bellarosa, though: ‘she’d organise, just out of the blue, like her husband and other friends, just people to sit and then I’d talk about my study…’. Their feedback was very encouraging for her: ‘It made me realise [that] I’m going somewhere’. This group became her informal community of practice, brainstorming with her and through this, generating new ideas for her research. That it included internationally renowned experts in the field was also very motivating for her. Beyond this group, Prof Bellarosa also shared Zodwa’s work with (unnamed) people in the same field in order to solicit further feedback. This was always done with Zodwa’s permission. The issue of power in the supervisory relationship between Zodwa and Prof Bellarosa was a variable one. Zodwa explicitly stated that there was a balance of power between the two women: she described Prof Bellarosa as a very humble person, despite being the author of numerous books and internationally recognised in the field. However, the Professor was also very conscious of the gaps in her knowledge in respect of several of the theorists that Zodwa was using to frame her central argument. In this context, the student taught the supervisor. Furthermore, since Zodwa is not only a doctoral student, but also a lecturer and a member of an NRF-funded project with Prof Bellarosa, the power ‘boundary’ is further blurred. In spite of the fact that Zodwa made a case for there being an equal balance of power between her and the Professor, she later contradicts this by describing her supervisor as being her ‘mentor’ – a term that is laden with connotations of unequal power. This seemed to echo the master-apprentice notion of Lave and Wenger, discussed in the literature review. In other words, she appeared to have constructed herself as an apprentice, learning at the metaphorical feet of the revered master/old-timer. This was in spite of the fact that she (Zodwa) had introduced her supervisor to new theorists in the field of study, and by implication had a better understanding of them. Zodwa has an impressive record of conference attendance and publication, although none of the latter has thus far been undertaken with Prof Bellarosa (although it was her supervisor who encouraged her to look for and attend conferences). While both women have contributed a (separate) chapter to a book on Identity – in conjunction with colleagues at Munich University – they have yet to collaborate on a publication. Zodwa 158 expressed an interest in working with her supervisor on some research papers in the future, and remarked that Prof Bellarosa would probably be open to this. 5.2.8 Xolani Supervision model Single supervisor Race and gender: student Black male Race and gender: supervisor Coloured male Graduation status Graduated April 2010 Xolani, an African man in his late forties, works for a local education NGO. He submitted his PhD thesis for examination at the beginning of 2009. He attributes his success in the completion of this degree to his supervisor, Professor Pieter Pieterse (also Hannah’s supervisor) because ‘he has got a passion, generally, for research and writing and publishing, and the general professional development of students’. Prof Pieterse’s approach to supervising Xolani was unique: once Xolani’s proposal had been accepted by the Higher Degrees Committee, the Professor gave him the ‘green light to do a complete draft the way I had planned it’. During the writing up of this draft, Xolani received some recommendations ‘in terms of additions and subtractions’ from his supervisor, but for the most part he had carte blanche to develop it as he wished. He explained that he had been allowed to do this so that Prof Pieterse could have a ‘clear picture’ of the direction in which he had taken his research. Once the professor had the complete draft, ‘he revamped the whole thing and gave it to me to rework’. This is really when the substantive supervision began. During the process of revising this draft, Xolani would meet with his supervisor once a month, and would have access to him (via his ‘open-door policy’) whenever he needed it – including on weekends. A few days prior to this meeting, Xolani would email Prof Pieterse some written work so that he had time to read it and provide feedback. This feedback was in the form of ‘comprehensive written comments’ which were then discussed during the meeting. Xolani admitted that on a number of occasions, he had pressurised the professor when he had needed feedback ‘like… like yesterday’ and in these cases, the feedback had been oral. That Professor Pieterse was willing to provide 159 feedback at such short notice is significant because it seems to suggest that he attached enough importance to the student’s work to make the effort. The issue of power, and where it was located within this relationship, played itself out at a number of different levels. In their relationship as supervisor and student, Xolani acknowledged that the bulk of the power lay with Prof Pieterse because ‘he is more senior, he is more knowledgeable’. However, there was space for Xolani to query some of the professor’s suggestions and to defend positions that he had taken in the writing up of his research: ‘He was very tolerant in terms of what a student would say and if it made sense to him, he would give in and he would accept his student’s point of view’. At the professional level, the power relationship between the two men was far more equal. For instance, Xolani co-taught several courses at the University with Professor Pieterse, and the two have given a number of joint presentations at conferences. At yet another level, the two had a friendship which extended to having lunch together and getting to know one another’s families. In this context, the power differential seemed to be non-existent. It was not clear how Xolani kept these three ‘relationships’ separate from one another, though. Professor Pieterse facilitated Xolani’s entry into the world of conferences and publications by taking him to the annual Brighton conference, held on the east coast of South Africa. At the first two of these conferences that they attended together, the two men gave a joint presentation (one of which as later published as an article in an accredited journal). Thereafter, Xolani generally went on his own to present, and has since become a member of the conference’s coordinating committee. Prof Pieterse funded every conference that Xolani attended. The Brighton conference was one of the communities of practice that Prof Pieterse inducted Xolani into. However, there were two other communities that he was also a part of: the Windsor PhD research programme (with Hannah), and the more informal School community. The Windsor programme included PhD students from a number of South African universities who met twice a year at the so-called ‘Windsor Schools’ to share and discuss their research. The informal, School-based community, on the other hand, held regular lunch-time and weekend research sessions at which postgraduate 160 students would ‘meet and make presentations about their work’. Xolani felt that all three communities had played a beneficial role in his development as a scholar, since they allowed him a space in which to talk about his research and to get ideas from others. 5.2.9 Sophia Supervision model Single supervisor (although she started out with two) Race and gender: student White female Race and gender: supervisor White female Graduation status Graduated November 2008 ‘It’s the biggest thing that personally I have ever lived, psychologically. We face our demons, we are unsure of ourselves, we can be inspired, we can be passionate, we can be devastated thinking that this is never going to end’ – Sophia on her PhD Sophia is a young woman in her early thirties who was born and raised in Eastern Europe. She was appointed as an associate lecturer at Wits about four years ago. Her employment contract with the University stipulated that she had to complete her PhD within three years. This was easier said than done: she could not find what she termed a ‘suitable supervisor’ because of the specific nature of her research interest. In addition, her School ‘was going through a lot of staffing changes, and I was the only person who was a constant here’. Thus, she was forced to take on an overwhelming teaching load. Eventually, she approached the Head of School of a cognate discipline, Dr Phillip Anderson, and asked him to supervise her PhD. He was willing to take her on, and in the months that followed the two worked well together. Unfortunately, soon after her proposal was accepted by the Higher Degrees Committee Prof Anderson was offered (and accepted) a post at a European university. He initially offered to continue supervising her via email, but Sophia sensed that this would not work. She subsequently approached Professor Deborah van Nieuwkerk, a world-renowned researcher in the Faculty, to take on the role of supervisor. Prof van Nieuwkerk agreed 161 to supervise Sophia, and was upfront about the fact that she was about to go on sabbatical to another country. Since Sophia had experienced two years of PhD uncertainty already, she decided to work with the professor, even if the supervision had to occur via email. Fortunately, Prof van Nieuwkerk was ‘good on email’, so the interactions that ultimately took place between the two of them were very constructive. Nevertheless, Prof van Nieukerk was concerned about leaving Sophia alone, and so she nominated a co-supervisor who would be physically present should Sophia require any support. The relationship between Sophia and her co-supervisor soon fizzled out, however, because ‘She basically corrected typos, spelling mistakes and stylistics, but nothing of substance at all’, and this feedback took up to four weeks to be given. Ultimately, Sophia simply stopped sending her work to the (unnamed) co-supervisor, and this seemed to suit them both. Feedback from Prof van Nieuwkerk, on the other hand, was far more ‘instant’ and substantial, and included some highly relevant references and readings. The professor’s feedback was far ‘gentler’ than Dr Anderson’s36 Sophia was very aware of where the power lay in her relationship with Prof van Nieuwkerk. She argued that it was more or less balanced, and that this was because the professor was consciously walking the intellectual path between mentoring (and thus exerting a certain amount of power) on the one hand, and treating Sophia ‘like an adult, and not a kid at school’, on the other. Sophia seemed to suggest, though, that she was one setting up the power relationship – it was not coming from the professor. The example she gave to illustrate this was an occasion when she was waiting for feedback , although Sophia remarked that the professor seemed to take her (Sophia’s) hard work for granted: ‘You sit on your own for days and days and hours and work and someone reads it and goes “Ah, its okay”, not “It’s good!”’. In between this rather demotivating feedback, the professor would surprise Sophia with a comment like ‘Very well articulated, really good work, keep going’ – this empowered her to want to work through the night on the rest of the research. She subsequently reflected that she would have liked more encouragement from Prof van Nieuwkerk. 36 Sophia is an assertive young woman, and yet she admitted t that Dr Anderson has ‘intimidated me much more than she [Prof van Nieuwkerk] did, so with him it was hard to progress because I always felt that the stuff that I hadn’t been able to see or find an angle on… it always made me feel that I was not good enough and that didn’t come from him, it came from me’. 162 on a chapter from the professor (who was overseas), who was usually very prompt. She waited in vain for nearly three months for Prof van Nieuwkerk to send the feedback. She recalls that eventually ‘I emailed her and I can’t remember what happened, but I think she had done corrections and I had never got them, or something as ridiculous as that, but I didn’t dare ask her for that. And when I did, it was a very hard thing to do’. In other words, Sophia did not feel that she had the right to remind her supervisor to send feedback. This suggested a self-imposed power dynamic that constructed Sophia as the powerless student who felt that she had no right to prod her supervisor for the feedback due to her. Prof van Nieuwkerk, by contrast, had the upper hand by virtue of the fact that she had what Sophia wanted – the feedback. Sophia was a very independent student, and she admitted to enjoying working in isolation from others, although she liked the idea of being ‘in contact with people and then having informal coffees, lunches, emails’. She had, on her own initiative, contacted and subsequently interacted with one of the theoretical thinkers whose work she had used in her research. Further, she suggested that the professor’s failure to introduce her to some of the key people in the field had nothing to do with poor supervision, and everything to do with some untested assumptions about Sophia’s existing capabilities and community: ‘I think she probably thought that I wasn’t completely a students and not completely a researcher. I was somewhere in between, and I think at times she might have wrongly assumed that I was very comfortable in that world’. Professor van Nieuwkerk and Sophia have not had the opportunity to publish together, and there is some doubt about whether this will ever happen. The main reason for this is that their specific research interests are too divergent for this to be a reality. However, Sophia also pointed out that publishing articles requires very different skills from the writing of a PhD: ‘I don’t really feel like a researcher in some ways. … I think its going to take me a while of quite a lot of work and probably needing even more mentoring to move from this huge abstract project to now researching a focused, smaller way of publishing articles’. It would seem, then, that a gap in her doctoral learning process was that of being taught how to write shorter pieces of research, in the format of articles or book chapters. 163 5.2.10 Bridget Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White female Graduation status Submitted October 2010 Bridget is an American woman in her late thirties, who is currently working on her PhD. She is also a full time member of staff at Wits. Her Master’s degree supervision – at an American university – had been a negative one. Her supervisor had presented Bridget’s data (with out her knowledge or permission) at an overseas conference, and without acknowledging its source. In addition, this supervisor had tried to coerce Bridget into delaying her graduation by six months in order to allow for further analysis of the model that Bridget had developed – something that this student felt was unnecessary. Fortunately, Bridget is an assertive woman, and she petitioned her Head of School and Dean on the grounds that she had more than met the requirements for the degree. They supported her petition and she graduated as planned. Bridget admitted that this experience had caused her to feel ‘apprehensive’ about undertaking a PhD. Fortunately, however, her PhD supervision has thus far been very positive, and there is no reason to assume that this will change. When Bridget began her PhD, she was not yet employed in the country, as her work visa had not been approved by the Department of Home Affairs. Quite coincidentally, she popped into the Department to chat to its Head about her PhD on the same day that Dr Edith Martins (who was more or less the same age as Bridget and who was to become her supervisor) returned to the Department from a six-month sabbatical in the United States. Bridget felt that the convergence of these two events was not a coincidence. Over the following six months, the two woman worked together to develop Bridget’s literature review and background of the study. This one-on-one relationship was also supported by the Department’s Postgraduate Forum monthly meetings, at which students presented their work and got feedback from their peers. Bridget felt that these were useful opportunities for learning and for contact with other 164 postgraduate students. As a community of practice, she found the input of her peers at this Forum to have been invaluable to her, especially during her proposal-writing process. When the Forum fizzled out37 37 By this stage, Bridget was a full-time member of the Department’s teaching staff. , Bridget recalls how she ‘… insisted they [Forum meetings] come back and so our new postgraduate coordinator – under duress, I believe - has had to bring them back, and I find it to be a very, very powerful motivator when you sit in a room with peers and formulate your ideas and the fact that we are not required to present, but you can bounce off ideas with people who understand the field’. This was a powerful argument in favour of a postgraduate discussion space. A somewhat smaller, but no less significant, community was the two-person relationship that Dr Martins set up between Bridget and another colleague in the Department who had extensive experience in applying for (and getting!) National Research Foundation (NRF) grant money. This colleague – Professor Raines – also introduced Bridget to the significant players in a national support group that dealt with the issue that she was researching ‘in case I wanted participants from there’. Thus, Dr Martins did not play the role of omniscient supervisor, but rather gave Bridget the widest access she could to people who were in a better position to help this student in advancing her study. Dr Martins was the ideal supervisor for Bridget. During the proposal process, the two women met once a month. Each meeting would begin with Bridget giving her supervisor an update on her progress, which they would then discuss in depth. Dr Martins would then give Bridget a typed ‘to do’ list for the next month – she gave Bridget a copy of this, and kept a copy for herself. Bridget showed me one of these lists: each bullet point generally began with ‘Consider this’, rather than ‘Do this’. Invariably, the last item on the list would read ‘Enjoy life’. This balance of encouragingly worded directives and emotional/psychological support was described by Bridget as ‘a gift’. Further, Dr Martin’s feedback on Bridget’s thesis chapters was thorough: ‘content editing, punctuation editing, reference editing….’, and was always accompanied by verbal feedback. 165 In mid-2008, Dr Martins and her husband relocated to another province in South Africa, but she maintained a sessional post at Wits. This meant that face-to-face time between supervisor and student was curtailed, although Dr Martins was required to be at the university for a few days every two months. Fortunately, according to Bridget, the two had ‘ironed out so much of the big stuff’ before Dr Martins left and therefore most of the remainder of the work could be handled via email and monthly telephone conversations. However, when Bridget needed to analyse the data that she had collected from a pilot study, Dr Martins invited her to stay with her, and ‘I did the data analysis at her house in the spare room and had her there to ask questions’. Bridget also describes her supervisor as being ‘intuitive’ – in terms of knowing where the possible pitfalls were and whether Bridget was in danger of falling into them – as well as very nervous: ‘I think she’s a first-time supervisor and she doesn’t want to mess it up’. This nervousness was also somewhat exacerbated by the friendship that sprang up between the two women. Bridget, however, assertedthat she had too much respect for Dr Martins to allow a friendship to impinge on the academic project. She used the term ‘mutual regard’ to describe how they balanced the two roles, suggesting a virtual absence of a marked power imbalance. Bridget has yet to publish aspects of her research in journals ‘because I don’t know that I’ve been in a place where I’ve been ready to write’. Dr Martins had suggested that Bridget write an article about the findings of her pilot project, but backed down when Bridget demurred in favour of focusing on her thesis chapters. However, Dr Martins did ask her to be a reliability coder on some of her most recent research, and also ask her to teach a second-year course that Bridget was more knowledgeable about than her supervisor. These two examples of collaboration may have been in lieu of the two of them co-writing articles. Bridget’s perspective on the role played by Dr Martins in developing her as a researcher was emphatic: I came in as a blind researcher, with a negative Master’s experience and she’s made me a better supervisor of our undergraduate researchers’. The key example of this developmental approach was the one mentioned earlier: Bridget analyzing her data at Dr Martins’ house: here, Dr Martins guided her through the analysis of the data (quite a bit of which was quantitative) and was available to 166 answer any questions that arose as they arose. Bridget described this as a ‘powerful’ experience. 5.2.11 Rebecca Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White male Graduation status Still in progress Rebecca is a highly intelligent, erudite young lecturer who registered for her PhD despite a difficult MA experience. Her Master’s supervisor, Lizanne Shale, was …an incredible intellectual, an extraordinary intellectual and my … imaginative and intellectual horizons shifted so dramatically in the time that I worked with her. However, there were issues around her own relationship with her PhD which she projected onto her students, and it really was a kind of pathology… By Rebecca’s account, Lizanne was a very driven personality who placed a great deal of pressure both on herself and her students to produce a magnum opus. As a result, Rebecca and her peers were ‘petrified’ of submitting work to Lizanne because “…you’d get almost five times as much script back than what you had written, in these tiny little notes in the margin and it was just overwhelming”. Subsequently, Rebecca applied for (and was granted) extension after extension on the submission date for her Master’s dissertation, until she landed a full-time teaching post at the University. One of the conditions of this post was the completion of the dissertation within the first three months of her appointment, which she did. Partly because of this experience – and partly due to her overwhelming teaching load in the department - it took Rebecca several years to register for a PhD. Once the decision to undertake a PhD had been made, and recognizing that she needed to access funding for teaching relief during her proposed sabbatical, Rebecca applied for (and was granted) a lucrative grant which would fund her for the duration of her research as long as she met all of the work plan objectives for the six-year period. She also received an 167 internationally-sponsored grant in 2008 which allowed her to take six months off to “… really do a lot of writing”. Rebecca’s PhD supervisor – Prof Glen Peterson – was exactly what she needed in a supervisor, even though he was situated in a different division of the School, and was not an ‘expert’ in the content area that she was working in: “However, he is probably the most astute and subtle thinker in the School… and has a very good profile internationally for engaging with contemporary theory and philosophy and political philosophy”. He was also a colleague in the School but never allowed the two roles (supervisor and colleague) to mix. Prof Peterson did not “…hold my hand and monitor everything I’m doing” because that was not what she needed from him. As self- confessed ‘competent’ student who asserted her own agency over the research, what she wanted from her supervisor – and what she got – was someone who challenged her “intellectual horizons” by playing Devil’s Advocate, who helped her to negotiate the “politics of publishing” and at the same time, who had a “…gentle and careful approach to how he engages you in a conversation, and he really does listen”. A typical supervision session – of which there are only three or four per year – took place in Prof Peterson’s office. He defused any potential for an overt power imbalance in the room by the way he configured the seating arrangements for the meeting: there was no table between student and supervisor, and the two sat facing one another for the duration of the meeting. Rebecca described this configuration as being “… like a therapist’s office”. In addition, Prof Peterson allowed her to drive the sessions’ direction. In return, what she got from him was “… an extremely intellectually curious and independent thinker who is always excited about the things I have to say”. She derived a great deal from these sessions and she invariably left each one with a powerful sense that “… he recognizes that I am an intellectual and our values around thinking and intellectual labour and practice are in alignment”. However, despite this attempt by Prof Peterson to even out the power differential between him and Rebecca, she nonetheless recognized both the existence and the role of power in the relationship. She argued that it resided predominantly in the intellectual realm, rather than the emotional/affective one. Thus, she describes Prof Peterson’s 168 feedback on her written work as “… the most pithy, extensive, and subtle commentary” that she had ever received. She felt that the quality and depth of this feedback indicated that he had engaged with the text at a very high level. It was “… always about pushing you to think in a more complex way about your own assumptions as you are operating in the field; and so, as difficult as it can be when you have to confront that, it pushes you to the next level”. While Prof Peterson and Rebecca have neither published together nor attended conferences together, this is not an area of concern for her. The reasons for this are twofold, and tend to reflect Rebecca’s independence and sense of agency within the research process. Firstly, as already mentioned, supervisor and student were situated in two different divisions of the same school, and thus Prof Peterson was not a ‘content expert’ in Rebecca’s field of study. The second reason related to her specific field of research, and the fact that there are very few academics in South Africa who are actively engaged in it. In other word, the ‘pool’ of conferences and research article opportunities was a small one. A much less compelling reason, but which was present on the interview radar, related to School politics: … there was no-one to help me here, there was a politics playing itself out in the division that made it impossible to go to anyone who may have had some expertise, and so I just had to run the gauntlet and have a baptism of fire myself. Rebecca ran this ‘gauntlet’ successfully: she ended up being awarded an internationalgrant, a separate doctoral grant to attend a Summer School in Brazil, as well as funding to attend a conference while there. 5.2.12 Mojalefa Supervision model Joint supervision Race and gender: student Black male Race and gender: supervisors White females (one in South Africa, one in Belgium) Graduation status Graduating in November 2010 Mojalefa is an experienced researcher who returned to academia after several years of working in the research industry. As a result of this background, academic writing – 169 from proposal to ethics to final draft – was never a struggle for him. In addition, the topic of his PhD thesis was an extension of what he had done in his Master’s degree, so he was familiar with and comfortable in the field. His primary supervisor, Professor Hazel Liebowitz, was not his first choice, although she is an internationally renowned academic. She was, however, the head of the South African component of the international working group in his area of interest. The international component was based in Eastern Europe. Prof Liebowitz possessed a fierce reputation as being a very tough supervisor “… who doesn’t allow you to get away with poor work”. He also had an (unnamed) co-supervisor located in Minsk, and it was in this city and with this supervisor that the bulk of the initial supervision took place. Indeed, Mojalefa’s supervision experience was, to all intents and purposes, a relay event between the two supervisors. In his words: My Minsk supervisor was more active in the development of the proposal, but as soon as I left, I did feel that the responsibility shifted to this side, so my supervisor here [Prof Liebowitz] basically took over the whole process of supervision. Supervision meetings were always organized by Mojalefa, and most took place during the period that he spent in the field collecting data. In the first few months of this fieldwork, the purpose of these supervision meetings was to “…sharpen what I was trying to do, because normally the field work is very exploratory at the beginning when you are trying to find your way.” At this point, Prof Liebowitz contributed enormously to Mojalefa’s research, because she was able to ‘organise’ him, and keep his attention focused on the crux of the research. At the same time, however, she never tried to impose her ideas about the direction that the research should take. In the single instance when he perceived her to be “…pushing me towards a particular direction”, he felt sufficiently confident to assert himself. The supervision meetings were generally very constructive sessions. Prof Liebowitz was described as being “engaging” throughout, although this engagement often served to “destabilize” him. He constructed this sense of destabilization in terms of feelings of intimidation: 170 I always felt intimidated, especially when she’s disagreeing with me, and I find it intimidating, and I felt intimidated to talk and argue this point because obviously she’s got more knowledge … and a better way of putting things. Over time, however, and as he got to know her better, he began to feel more “settled”. Knowing Prof Liebowitz was something that Mojalefa identified as being very important to him – despite having previously known only of her, and of her formidable reputation. However, “…I have always felt that our interaction was just too impersonal, it’s just academic. We talk about my work, but that’s it”. He described in some depth how this lack of a more personal interaction was “…the biggest weakness in our relationship”. As he put it: It’s good to have a kind of cosy relationship with your supervisor. I mean, we have never even had tea together or a drink together. I don’t know her house. I don’t know anything about her. I’m not saying we should be close friends, but at least there should be a level beyond merely being a student. He contrasted this separateness to his relationship with his supervisor in Minsk, whom he describes variously as “friendly”, “down-to-earth”, “approachable”, and “…you don’t feel intimidated at all by her. She’s also a well-known scholar but she has this attitude that makes it very easy to relate”. He “knew her house”, and was often invited around for a drink or a meal. He fondly recalled how “I would browse through all of the books in her house, and I would pick up the book I’m interested in, read it”. His primary observation from this stark contrast between his two supervisors was that one often learnt more during such informal interactions than through any formal supervisory channels. However, in order to mitigate his criticism of Prof Liebowitz’ style of supervision, he offered a reason for the perceived social distance between them – the general culture of such interaction in the project to which they both belonged. He related how “There is no culture of free interaction, with people engaging freely with their students”. This distance between supervisor and student is also evident in this particular case in terms of the absence of joint conference attendance and the co-publication of papers. At the time of the interview (early 2009), Mojalefa had attended five international conferences (sans supervisor), all of which had been funded by the University of 171 Minsk. Two years ago, he and Prof Liebowitz were tasked by the University with writing a review of another research entity at the University, but this collaboration never materialized – the review was recently published by Prof Liebowitz. Furthermore, Mojalefa expressed deep dissatisfaction with the nature of Prof Liebowitz’ feedback. The feedback was always oral, and hence he spent the better part of the supervision sessions taking notes, and then expanding on them in more depth after the meeting. He recalled how she rarely criticized his work in the early period of their supervisory relationship, choosing instead to tell him how “good” he was. Subsequently, and nearing the end of the writing process, “… it’s as if she was sleeping, somehow, and decided that this is not good enough, and if it goes through, it won’t pass”. Mojalefa described how bitter this about-face made him feel, especially since he was so emotionally and intellectually invested in the work. He chose not to challenge Prof Liebowitz on this U-turn because “I found it difficult to question her, that ‘You’ve been happy all along, what has gone wrong?’ and I didn’t find myself in a very comfortable mood, I was very upset”. He used the term ‘intimidated’ several more times when describing this situation, suggesting that the power imbalance between the two was as strong nearing the end of the process as it has been at the beginning. 5.2.13 Ntombi Supervision model Single supervisor Race and gender: student Black female Race and gender: supervisor White male Graduation status Still in progess Ntombi is a softly spoken African woman who, at the time of the interview, was working on the analysis of her PhD research data. She is also a full-time member of staff at the University, and has had to juggle the demands of her employer with those of her PhD and family. Her experience of supervision at the doctoral level has been a profoundly positive one, despite the funding difficulties that she experienced near the beginning of the process. 172 Ntombi began her PhD studies as a full time student with Dr Gould, with whom she had a shared interested in a specific field of research. However, since Dr Gould himself was a newly minted PhD and did not have access at that time to any grants, he was unable to help Ntombi with funding for her degree. Thus, she was left with a dilemma: leave Dr Gould and join a funded PhD programme (whose research focus was somewhat different from hers), or stay with Dr Gould and scrape together some money. Ultimately, she managed a compromise: she signed up with the funded programme, and persuaded Dr Gould to stay on as her supervisor. He was happy to do this, on the proviso that she find an additional supervisor who possessed the skills that he did not. This she did – her co-supervisor was based in the US, and affiliated to the funded programme. The role of this individual was constructed as an informal supervisor, given the geographical distance between them. The relationship between Ntombi and Dr Gould was a very profound one, right from the start. At the proposal stage, she struggled with the demands of qualitative research, having come froma background in undertaking quantitative research. Dr Gould was sensitive to the huge differences between the two types, and spent numerous supervision sessions guiding her through the process. He insisted on three-weekly meetings so that he could assess her progress, and ensure that she was not wandering off. At these meetings, he would provide her with written feedback on her written work (which she had provided to him a few days prior to the meeting, so that he had had time to read it). This feedback was invariably about the content of the writing, not the quality of the grammar, which she raised as another positive feature of his supervisory style: ‘… he is not fussy about the language, the English, you know? I am not an English first language speaker, but he is not fussy about things like that. He points them out to me, “You need to look out for those”, but he doesn’t focus on [them]…’. Thus, his focus is more on the substance of her arguments, as well as on the importance of defining terms so that the reader will be able to follow her arguments. She recalled how she often teased him about this insistence on definitions. The power relations between Ntombi and Dr Gould, while not evenly balanced, were not an obstacle in their relationship. For instance, at one point, Ntombi felt very strongly about including a certain area in her research, arguing that it would add value 173 to her argument. Dr Gould argued that she would be extending herself too far by doing this. They discussed it in some depth, and ultimately, Ntombi decided to continue working on this particular area regardless of Dr Gould’s concerns. It was only when she began collecting her data a few weeks later that she realized that Dr Gould had been correct – ‘I should have listened to the words of wisdom!’ – but had allowed her the space to make this discovery for herself. She described this experience as being ‘more empowering than anything else’. However, she did recognize that there were contexts in which a power imbalance in his favour was not only necessary, but appropriate. For instance, because ‘he’s the expert’ in her field, she argued that it was only right that she deferred to him. In addition, because of his experience as an ‘administrator’, she felt that it was appropriate that he took the lead in issues of administration. Dr Gould did, however, make a concerted effort to close the power gap by ‘treating me as a colleague’. He did this by co-authoring a number of papers with her, and insisting that she take first authorship. Further, he raised money to take her to conferences with him (three thus far), at which he introduced her to ‘… the big authority in this, this, this that you’ve been talking about’. As a result, she has ‘… met almost all the people whose theoretical frameworks I’m using!’ He also encouraged her to send her work to key thinkers in the field, and provided her with the necessary introductions for her to be able to do this. However, Ntombi was not just a student to Dr Gould: in 2007, her father passed away suddenly and Dr Gould was a source of great compassion for her. This death meant that her financial situation would change dramatically, since her late father had been her only source of income. She contemplated dropping out of her studies. Dr Gould was very understanding, but argued that she had gone too far to deregister, and so suggested exploring other options such as alternative sources of funding, as well as various job opportunities in the School. He even went as far as to discuss the pros and cons of the various jobs for which she had applied. During this time of bereavement, she put aside her PhD research in order to allow herself some time to recover. Dr Gould supported this, but encouraged her to continue to work on articles for publication so that she would not get out of the habit of writing. 174 5.2.14 George Supervision model Single supervisor Race and gender: student Coloured male Race and gender: supervisor Coloured male Graduation status Still in progress George is a garrulous middle-aged man who came to academia late in life. His formative years were spent in the Cape Flats38 Thus, the combination of full-time employment and part-time doctoral studies presented a number of challenges for him, not least of which was “…getting my head into this research space – how do you become a researcher?” It took him two years to put together his proposal, a period characterized by “depths of depression”. The feedback that he received at this stage – both from his supervisor and from his peers in the scholarship programme – was both empowering and destructive. While the questions asked of him at proposal seminars were the kinds of questions that by his own admission he needed to engage with, nonetheless “….they really took my theory , and his parents had been unable to afford to send him to university after high school. In his early twenties, he spent some time studying theology and philosophy at a local seminary and it was here that he “… discovered that I really enjoyed this academic stuff”. He began his first degree in the mid-90s, and followed this up with a ‘crappy’ Master’s degree at another South Africane University. After this, he decided to take a break from his studies to focus on his marriage. However, before long, the urge to continue his studies resurfaced. He noticed a newspaper advert offering scholarships for PhD candidates at Wits and on a whim, he applied for one. To his surprise, he was selected. At the time, he was employed as a programme director by the national higher education advisory body, but was very aware of his own shortcomings in terms of his research capacity, something that he attributed to his Master’s degree experience. This negative self-consciousness manifested itself in the number of disparaging remarks that he made about himself, such as “I was a bad student”, “I was always a very poor student”, and “I must say that I have been a poor student”. 38 A Coloured township created by the Apartheid government, situated just outside Cape Town. 175 apart”. Even at this early stage, he described how the doctorate began to “consume” him and how he interpreted his failure to progress as a “black mark” against him. Fortunately, an interaction with a family member who had recently completed a PhD from Harvard precipitated a “mental shift” “… where I saw this more as a journey than a pain”. This was further consolidated by the positive and constructive feedback that Prof Pieterse provided on his work once the proposal had been approved by the Higher Degrees Committee. George’s supervisor, Prof Pieterse, was his “guiding light” throughout this journey. Despite this, George recalled how he would often get frustrated at Prof Pieterse’s refusal to direct his research, choosing instead to ask probing questions: I was presenting to him my work and I was hoping that he would say “This is the way you do it” [and] I wanted to say “For God’s sake just tell me what I need to do so that I can do it!”, and he kept on “But you must look at that and you need to think about this…” … but I think I also see now in hindsight why he led me into the desert – to go and find myself and try and unravel some of these things. That’s the self-discovery. It’s not self-discovery if you’re given a map. The flipside is you really just get thrown into the deep end and its sink or swim, but here it was really as if the invisible hand that was there … grabbed hold of you when there was a potential hazard. Without you realizing the gentle way that he re-directed you… George also supplemented Prof Pieterse’s supervision with input from others in the field. For instance, a former colleague and prominent academic at a Cape university gave him valuable advice on how to separate the vital data from the peripheral: I remember William saying to me “You take everything and you put it into this sieve… You gooi39 In addition, he identified two recognized experts in his field from abroad and asked them to act as critical readers for his work (they agreed), as well as a statistician from his office who was also a quantitative research expert. He ‘cleared’ these people with Prof Pieterse before approaching them, and assured him that they were not “…in everything in and then as this thing is distilled, then you will see your focus becomes narrower and narrower, until you get ‘Aha! This is what I’m trying to do! This is what I’m trying to investigate!’”. 39 Gooi is the Afrikaans word for ‘throw’. 176 competition with you or trying to second-guess what you’re doing”. Prof Pieterse allowed him the freedom to consult these individuals. George suggested that the reason for this was because … he’s as much of a professional as I am, and in that sense, I think that the playing fields in terms of power relations are level. I don’t think [he] sees me as a traditional student who he has to mollycoddle. George also asserted that within the supervisory relationship, Prof Pieterse was learning as much about George’s area of interest as George himself was, and that “…I think he appreciates the fact that I brought a different dimension to his world, and a different challenge”. Interestingly, George subsequently contradicted this claim of a balanced relationship of power between the two of them when he chose to compare Prof Pieterse’s supervisory role to that of a coach: He’s [Prof Pieterse] saying “I’m the expert, I know what’s best. I will tell you how you need to train and what you need to do if you’re going to play. If you make the wrong moves, I’m going to tell you” This was reinforced when he described how Prof Pieterse was perceived by other students at the national seminars that they attended as part of the scholarship programme: “It was interesting that the other students from these other universities spoke with huge respect of Prof Pieterse as a scholar, not just as a supervisor”. In this context, George seemed to (willingly) construct himself as an apprentice to Prof Pieterse’s master. 5.2.15 Alphons Supervision model Joint supervision – three supervisors Race and gender: student Black male Race and gender: supervisors White male, black male and black female Graduation status Still in progress Alphons is a Senegalese academic in his mid-thirties, who was the recipient of a PhD fellowship grant at one of Wits’ most prestigious research entities, WRITE. He came across as being a highly independent and self-motivated PhD candidate who felt that 177 the role of his three supervisors - Professors Graeme McKenzie, Mbuyu Nematandeni and Nonceba Matabane – should be limited to that of ‘guidance’. He understood this to mean that they would only offer him advice when he requested it, and that he could choose whether or not he wished to follow it. This is what he had done during his Masters degree studies in Senegal and he saw no reason why things would be different for his doctoral studies in South Africa. However, he was quick to point out that he had the highest regard for all three of his supervisors – his independent stance was a personal preference and not a reflection of any disrespect for them. He began his fellowship at WRITE in January 2007. Alphons and the other fellows in his cohort were immediately taken on a retreat at which they presented their research topics to the WRITE staff, and received numerous ‘relevant’ inputs from them. Although Alphons appreciated the staff input, he felt that they were trying to force him to do ‘…something that was overtly political’. His Masters degree had had a political theme, and he was reluctant to revisit this theme in his PhD: I’ve done that before – can I find something completely new, something that I’m not aware of? This topic is actually tired, it’s not exciting from me, it’s not working. Thus, he asked his supervisors for time to ‘look for something different’. They agreed. He subsequently spent a number of weeks in the University libraries, reading widely in the hope that he would be able to ‘… figure out what it was that I wanted to do’. Fortuitously, he stumbled on a topic that grabbed his attention immediately, and, having read as widely as he could around it, he presented it to Prof Nematandeni. This supervisor said ‘Ja, there is something that you can work on here’, and Alphons set about deepening his research, and subsequently writing a proposal. When he presented the latter to all three of his supervisors, their response was: ‘Ja, there’s something here. Maybe you should add that, read more on this, read more on that’’. They were very, very encouraging. All three lived up to his expectation of only providing guidance. In his words: Basically, how I expect a supervisor to work with a student is, rather than imposing something and saying ‘This is what you must do!’, I welcome 178 suggestions and guidance: ‘Maybe try that, maybe approach that from a different angle…’. This independence extended into other areas of his doctoral studies, too. For instance, he sourced and contacted important scholars in his field of study, and identified conference which he needed to attend. WRITE paid for two local and one international conference in the first two years of his fellowship. He also took advantage of WRITE research meetings, at which the PhD candidates presented their latest research to their peers and the WRITE academic staff. Oral feedback was provided at these meeting, which Alphons found to be enormously valuable. A typical supervision session usually consisted of him and at least two of his supervisors sitting down to discuss his progress (or lack thereof). These sessions would be arranged by Alphons when he felt that he needed some guidance, or by his supervisors when they had had time to read the written work that he had submitted. He related how feedback from each professor often contradicted that of the others. For instance, while one of the supervisors felt strongly about Alphons focusing on Senegalese literature, instead of comparing it to South African writing, the other two were emphatic that a comparative study would be more advantageous. Ultimately, though, the choice was his: They would allow me to take what I want. They’ll not impose anything on me. They’ll simply make those suggestions. So I’d leave them to fight amongst themselves, and at the same time, I’m figuring out ‘Okay, which one is the best that I want to take from here?’. So, from there, I simply take hold of what it is that I want to take. Far from finding these disagreements between his supervisors disconcerting, Alphons welcomed them because they presented him with a range of options for taking his research forward. He argued that such rich opportunities would not exist had there been a single supervisor. The fact that they allowed him to decide which feedback to use was also an indication of the relatively balanced power relations between Alphons and the three professors: [I]n my experience, I respect their positions, I admire them, … but its balanced because what I like about it is eventually, … they still come back to you and say 179 ‘Okay, now, Alphons, tell us, from what we’ve said, what it is that you would like to do’. The feedback that he received in these sessions was invariably oral – he took extensive notes, though, as the supervisors spoke. He also described how he had developed as a researcher by merely watching the interplay between the three professors, all of whom had solid international research reputations. He felt that they led by example, and by so doing, had helped him to approach his research ‘boldly and with confidence’. 5.3 Conclusion The fifteen narratives presented above offered a range of experiences of supervision, and sometimes the same supervisor was experienced very differently by different students (for instance, Prof. Pieterse’s supervision was described in different ways by Hannah, Xolani and George). In addition, even though the Humanities model is traditionally embodied by a single supervisor for each doctoral candidate, this was not true of all of my interviewees. Nkosinathi and Mojalefa are examples of where the scientific model of co-supervision was utilized with varying levels of success. The related themes of power and agency over the research (cf. sections 4.1. and 4.2 of Chapter 8) were also differently understood and managed by the candidates – generally speaking, the more agency the student exerted, the less power she or he conceded to the supervisor. Supervision meetings were unique in each narrative, from formal office interactions (Rebecca), to telephonic conversations (Bridget), to electronic discussions (Sophia). What was interesting was that candidates were not wedded to formal interactions, but were open to engaging with their supervisors in numerous different and often informal contexts – more important than the context of the meetings, it seemed, was their content. My overall sense of the combined narratives was that each experience of supervision was unique; although there were a few overlapping incidents, these were irregular and did not form any significant pattern from which a firm conclusion could be drawn. In the next chapter, I will present the narratives of the Science Faculty candidates in the same format as those of their peers in Humanities. 180 CHAPTER 6 The scientists tell me their stories Chapter overview This chapter contains the profiles of the Science Faculty candidates interviewed for the thesis. As was the case in the previous chapter, each profile is a narrative version of an interview transcript, and encapsulates the essence of each interviewee’s experience as a doctoral candidate. 6.1 Introduction The doctoral candidates that I interviewed from the Science Faculty provided me with some fascinating insights into the processes associated with scientific research. As a Humanities researcher, I came to appreciate how different this kind of research was to my own. In the following fifteen narratives, the doctoral candidates offer their experiences of being supervised at this level, as well as their reflections on the process. 6.2 The narratives 6.2.1 Ace Supervision model Single supervisor Race and gender: student Black male Race and gender: supervisor White female Graduation status Still in progress Ace, an earnest young African lecturer, was supervised by his Head of School, Prof Janice Silvermann. This was not ideal, since institutional demands on her time meant that she had little to spare for her students. The fact that she was supervising 32 other postgraduates only added to Ace’s concerns that he was not getting as much support and feedback as he felt he needed. Since he saw her every day in the department, he could see the huge pressure that she was under to meet all her obligations, and as he remarked ‘then you try and avoid adding more pressure’ by ‘letting my issue take second stage’. In addition, although she was a well-known researcher in South Africa, hers was not the area of research in which Ace was interested. Nonetheless, Prof Silvermann had ‘recently developed an interest’ in it. 181 Supervisory meetings, when they happened, were generally very relaxed affairs: ‘Sometimes, we are talking, we are literally laughing all the way, you know, but we are making progress’. However, feedback was slow in coming due to Professor Silvermann’s overloaded schedule: ‘So, sometimes, you expect … delays or where you are expecting it [feedback] tomorrow, it may actually come a month after’40 In terms of the power relations between Professor Silvermann and Ace, they seemed to be fairly balanced – ‘’I’ve never felt inferior’. Nonetheless, it was evident that she was someone to whom he looked up. His description of her being ‘more like… a mother’ implied (intentionally or not) a distinct hierarchy, in which she held authority over him. Thus, in constructing himself implicitly as the ‘son’, he seemed to be suggesting that the power relations between them were, in fact, not that balanced. . Furthermore, the substance of this feedback fell well below his expectations: ‘[I]t’s clear that she puts a lot more emphasis on the grammatical errors, the spelling… the commas, which are quite easy to pick’. Since he was in the early stages of his PhD, Ace was not too concerned by the slowness or the quality of the feedback, but acknowledged that later on in the process it could become a problem. Despite the fraught context in which the supervision was taking place, Professor Silvermann showed a great deal of confidence in Ace as an academic. Her almost constant refrain when she saw him (which was every day) was: “Ace, how far are you? How far have you got?” He interpreted these comments as encouragement, rather than as harassment. From the beginning, she guided him as he attempted to develop a research topic, and helped him to whittle down his initial, very broad, topic to a more specific one. The process was difficult for him because it was the first time in his academic career that he had had to develop a research project from scratch. He has read for his Master’s degree in Europe, where the supervisor defined the topic and the student simply did the research (described by Ace as being ‘smooth sailing’ compared to the PhD). 40 Just prior to this interview, four of Prof Silvermann’s PhD students (excluding Ace) had laid a grievance with the Faculty regarding her non-availability for regular consultation and the long turnaround time in getting feedback from her. She subsequently resigned as Head of School, and later from the University. Ace now has a new, less experienced supervisor, and Prof Silvermann continues to provide him with informal supervision. 182 Ace submitted his proposal in March 2008, but has made little progress on the research since then due to his extensive teaching load. 6.2.2 Emily Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White male Graduation status Graduated in 2009. Emily is a young woman in her mid-twenties whose experience of being supervised by Professor Lambson began in her Honours year. She remained with him into her Master’s and PhD degrees, not only because he was one of only a handful of experts in her field of interest in South Africa, but also because of his particular style of supervision. According to Emily, this was characterized by his strong communication and interpersonal skills. She described every interaction with him as a ‘discussion’ in which he ‘makes you think’ without ‘babysitting you’. Honesty was a key feature of these interactions, even when ‘I did something wrong’ – ‘As long as you’re honest with him, that’s all he cares about.’ She described how Professor Lambson was always available for consultation, even without an appointment. Emily would chat with him on an almost daily basis about the experiments that she had conducted that day, and the results that they had yielded. Professor Lambson would also frequently make a point of dropping into the laboratory to check on her progress, so ‘he knows what I am doing every day’. During their interactions in these contexts, he was able to provide immediate verbal feedback, ranging from ‘Well done’ when she got a good result, to ‘And when you did this, did you try this? When you did that, did you try this?’ when the experiment did not yield the expected result. His developmental style of supervision also played itself out when he witnessed her making a mistake: instead of correcting her, he would ask ‘What are you doing?’ which served as a hint to Emily that she was on the wrong track. If she was unable to figure out where she was going wrong, and continued on this track, he would say something like ‘Yes, but why did you do this?’, forcing her to think more critically 183 about what she was doing41 All of these positive supervision practices were kept in place by clear, yet often implicit, boundaries between Professor Lambson and Emily. At the core of these boundaries is the sense of respect: ‘He’s, like, a very friendly supervisor and we get on well and stuff, but I know… I know he’s still my supervisor and I’ll never disrespect or cross that line’. The professor also laid down some very clear rules for Emily and her fellow students when they entered the PhD Laboratory for the first time; for instance, their working hours would be from 8am until 4pm – they could not simply come in and leave as and when they pleased. If they needed time off, they had to make a formal . As a result of this constant attention and feedback, Emily felt supported throughout the course of her research. She also felt that she was being intellectually stimulated, resulting in personal growth and development. Professor Lambson has also had a profound impact on the writing up of her research. He allowed her (and his other students, for that matter) to maintain their own style of writing, while directing them as far as possible in terms of the discourse in which their research should be written. This was important for Emily becauseshe recognised that her style of writing was an expression of her academic identity (‘…because some people, you can read, it’s so-and-so’s supervisor writing, it’s not their own style…’). In the Sciences, conference presentations usually centre on both posters and academic papers. At the time of this interview, Professor Lambson had attended two conferences with Emily. At both, she had presented a poster, although she would have preferred writing a paper with him. Unfortunately, she had not had sufficient data to develop into a paper, and had hence chosen to do a poster instead. Prior to both conferences, she had had to submit an abstract, which he had read through and supported. In both cases, he insisted that she be first author and that his name come afterwards. He made a point of always upholding his students’ right to put their names down as first author, as a form of recognition that they had done the bulk of the work. 41 Emily expressed a strong liking for the way Professor Lambson made her think: ‘Obviously, I had to have my own thinking, but like, he’s taught me how to think and I … I mean, I’m much better than I was in Master’s and when I was in Honours and I see the difference now that I’m in PhD. Now I can think much better on my feet and, like, I can analyse things better.’ 184 request to him (which he usually granted). This instilled in Emily and her peers a sense of discipline and routine for working in the lab. Aside from his support of Emily’s academic research, he also took on a pastoral role during the period of her PhD research. She cited several instances where his input was more like that of a friend or family member than that of an academic supervisor. For instance, at one point in her studies, she had begun to feel taken-for-granted by her friends, and one day arrived at Professor’s Lambson’s office in tears. He gently probed her about what was wrong, but when it became clear that she was unable to talk about it just then, he agreed to her request for two hours off so that she could collect herself. When she returned, he again asked her what was wrong, and she was able to confide in him. He listened, and gave some advice, which he summarized as ‘You have to realize, you are important too’. She left his office feeling ‘cheered up’, and was able to continue her work in the lab. 6.2.3 Indira Supervision model Joint supervision – one principle supervisor and one co- supervisor Race and gender: student Indian female Race and gender: supervisor Both white males Graduation status Still in process Indira is a young Chemical Science student. She had two supervisors, but relied on the main supervisor (Professor Pierre Plaget)for the bulk of the research input; she only approached the co-supervisor for help when she experienced specific technical problems in the laboratory, since this was where his expertise lay. In the first year of her PhD, she was a part of a cohort of four other PhD students, and Professor Plaget had attempted the ‘group thing’, in which he met with the students as a group. Each student would get an opportunity to present her most recent progress, and then would get feedback from both her peers and from the professor. With time, however, this set- up fell away as the students’ research interests diverged. Subsequently, he had no choice but to see them on a one-on-one basis. These supervisory meetings were 185 formally arranged, and took place weekly, monthly or bi-monthly, depending on need. Prof Plaget would email all of his students the 30-minute timeslots during which he would be available for consultation, and they would pick the one that suited them. Once she had a timeslot, Indira would email Dr Plaget the written work that she wanted to discuss at that meeting. The meetings themselves were always fruitful: he would ask specific, guiding questions about the work that she had submitted, and would also discuss each step of the methods that she had used in her laboratory experiments. Together, they would work through any problems that may have emerged during these experiments. A great deal of informal supervision took place in between these formal meetings. Thus, ‘if I need help now, I … know I can go to him and I will get help there and then or if I can’t get help at that particular instant, I will get a time when I can come back later’. These informal sessions seemed to consolidate the intellectual input that he provided during their formal meetings. Furthermore, Professor Plaget was always up- to-date in his knowledge of her research area. However, when he did not know the answer to something, he told her so and then made a point of working with her towards finding an answer. Indira also reflected that throughout all supervisory interactions, ‘he teaches us at two levels. If there’s something pertinent that I need to know how to do now, he’ll teach it to me directly, but there are other things that are a lot more subtle, [like] the way you write or, you know, a procedure in the lab about thinking… That’s the layer that just happens in the background’. In other words, the supervisor chose to foreground certain learning processes over others. Furthermore, ‘the thing that I find interesting is that he’ll always sit on the same side [of the desk] as I’m sitting… that helps to make me feel that I am on the same level as him’. This need to treated as an equal and therefore to take equal responsibility for the success of the research emerged at various points later in the interview, too. It suggested that Indira was a motivated, responsible student with a powerfully developed internal locus of control when it came to her research. She related howe ‘I don’t feel that I’m an idiot when I go speak to him if I have a question... He will always address the question in such a way or his questions to me [sic] in such a way that I don’t feel like I’m an idiot. And I appreciate that because it builds my confidence as a researcher 186 and I feel like I want to contribute’. In terms of whether she felt that Professor Plaget was developing her as a researcher, she stated: ‘I … don’t think that it’s fair to say that it all depends on him because it also depends on my thinking and my understanding and what I’m willing to grasp from it’. As with many of her peers, Indira experienced a number of obstacles during her PhD research, the main one being that of time: ‘I’m three years on and I still feel that I’m not far enough’. By her own admission, she was the one pressuring herself, and not Professor Plaget, but despite having recognized this, she still felt that she had very little of substance to show from her three years’ of research. A substantial part of this impatience with herself seemed to have derived from the fact that she had yet to publish a paper with her supervisor – ‘I haven’t reached the stage of papers yet’. This is in contrast to her Master’s degree experience, at a different university, during which she published four papers with her supervisor. She has, however, done a poster for a conference with the input of Professor Plaget, on which she listed herself as first author. Indira displayed a distinct indifference to the concept of power within her relationship to Professor Plaget. She recognized that both he and her co-supervisors were experts in their respective fields, that they were both confident in the breadth and depth of their knowledge of the subject, and that they had ‘nothing to prove’ – ‘I feel like [they] know what [they’re] talking about’. In a written comment that she made on the interview transcript sent to her for her approval, she noted ‘if there is [sic] indeed any power struggles with [Professor Plaget], then he does it very elegantly and imperceptibly – I don’t see it’. 6.2.4 Seamus Supervision model Single supervisor. Race and gender: student White male Race and gender: supervisor White male Graduation status Graduated April 2008 187 Seamus is an angry young man in his mid-twenties, who had upgraded his Chemical Science Master’s degree to a PhD. His reason for converting was telling: “…because I pretty much didn’t know how I could write it [the MSc] up. I didn’t think that there was anything to write up, so I thought ‘Oh well, I’ll just carry on and hope for the best’, sort of thing”. Despite this lack of direction (apparently from both supervisor and student), Seamus chose to continue working with his Master’s supervisor, Dr ‘Monty’ Montgomery, for his PhD. However, the relationship was far from ideal, and Seamus expressed a sense of neglect by his supervisor: “We started the project, and I kind of felt that the supervisor kind of lost interest, and I must just wander on, you know?” Seamus felt that this lack of interest in his research was at least partially the result of Dr Montgomery being “bogged down in administration”, which effectively curtailed the amount of time that he had to spend with his postgraduate students in the laboratory. Seamus also expressed strong frustration with Dr Montgomery’s lack of vision (“Plan B”) for when experiments did not yield any usable results – the supervisor would make him ...do one thing and it wasn’t working, and [he] just kept on “Well, just try it again, just try it again, just try it again” as opposed to “Well, okay, let’s think about this and find a way around the problem”. As a result of this repetitive – and seemingly fruitless – series of experiments, Seamus argued that he ended up spending eighteen months more than was necessary to complete his PhD thesis. It was his contention that Dr Montgomery ought to have set him goals and timeframes in which to achieve these because such structures would have prevented him from conducting experiments indefinitely, without writing up their results. While he did not abdicate full responsibility for this to Dr Montgomery, it was clear that Seamus had an external locus of control when it came to his research. Thus: “I understand that during the PhD, they do push you, and you’re supposed to be doing the research on your own, but I still believe that you’re paying for good supervision”. Seamus indicated that he did not get much feedback – either written or oral – from Dr Montgomery, which he attributed to the poor communication that existed between the two of them: “I found that he only knows how to criticize… he can’t ever say ‘Oh, well 188 done! That is good!’ you know? He said ‘Well done!’ when I finished, but that’s obligatory”. When his supervisor was forthcoming with feedback, though, he was quick and thorough. Seamus recalls: “I did my introduction, gave it to him, he got it back to me in a couple of days, and then I handed the rest of the body of it, which was about three hundred plus pages, and he gave it back to me fully corrected in three days”. Nonetheless, the main source of feedback for Seamus and his postgraduate peers, however, seemed to come from the distinct hierarchy that existed in the laboratories that overtly excluded the academic supervisors and which Seamus described as follows: Essentially, we have a system where you have your PhD students who are nearing the end, or within a year and a half from the end, and they basically run the group in the labs. You take all the Master’s students in and you supervise them, as well as the Honours students and you supervise them, and you teach them the lab skills and obviously the technical expertise, and how to do research. Thus, Seamus felt that he was largely self-taught when it came to the development of his research skills. Although he acknowledged the input of other Master’s and PhD students in the lab – they “…gave me a hand and showed me the ropes” – he maintained that he learned research skills by himself. When he first started his Master’s degree, the department held monthly progress meetings for all Masters and Doctoral students and their supervisors. At these meetings each student would give the supervisors a “run-down of what you did for the last month”, and occasionally some useful feedback was forthcoming. However, for the most part, according to Seamus, “I was under the impression that no-one really wanted to be there”. These monthly meetings petered out over time, but were resuscitated (on a weekly basis this time) towards the end of Seamus’ studies. He chose to not participate, but felt that they could be a useful space for students seeking feedback: What happen is… one person just presents their work to the group and they just talk to the group about it, and the group can give them criticism, constructive criticism, about it. The supervisors run these and that’s probably the most likely place that you’re likely to get any ideas or supervision of the project. 189 “I haven’t got papers out of him yet” was how Seamus described his relationship with Dr Montgomery when it came to the publication of his research results. He offered two reasons for this: firstly, that Dr Montgomery did not trust his students’ ability to write, and secondly, that since he was “nearing retirement age” and was no longer actively engaged in research, he was “just coasting towards the end”. By contrast, Dr Montgomery provided Seamus with constructive input prior to the latter giving oral presentations at conferences, which “… teaches you to write up scientific documents correctly, and how to present properly so you don’t look like a fool when you go and present”. This contradicts Seamus’ earlier assertion that he was largely self-taught when it came to the development of research skills. Nonetheless, these perceptions seem to point to a mismatch between supervisor and student. 6.2.5 Nomsa Supervision model Single supervisor Race and gender: student Black female Race and gender: supervisor White male Graduation status In process Nomsa is an engaging young woman who came to Wits from a neighbouring university, having completed an MSc there. Poor supervision at this university had led her to the ‘greener pastures’ of Wits. In the final phase of writing up her Master’s dissertation, she had been looking forward to continuing her research through a PhD at Wits but had had no idea about which supervisors oversaw research students in her field. Her first encounter with the man who ultimately became her PhD supervisor – Professor Joseph MacDonald – was fortuitous: I remember my first encounter with him. I hadn’t made an appointment. I just made a few calls and I said ‘Who’s mainly involved in this field, and who’s supervising students in this field?’ and they gave me his office number and his name… and I woke up one day and I just decided that I’m going to see this Prof MacDonald. Just like that…. I just came here to Wits to see him straight away, without making an application, nothing… Even at this early stage, Nomsa instinctively knew that she could work with Prof MacDonald. Upon arriving at his office, he welcomed her in immediately and asked for 190 a few minutes to complete the experiment that he had been working on in the adjacent office. This done, he spent the next hour chatting to her about herself, who she was, why she wanted to undertake a PhD, and so on. For Nomsa, ‘the reception was just so wonderful, it was almost like he was representing Wits, he was just showing how Wits receives strangers’. This welcome was in stark contrast to her experience at her previous university, where she had felt isolated and as if she did not ‘fit in’ because of her race. She described her relationship with her Master’s supervisor as being quite hierarchical: ‘it was more of a lecturer-student relationship than a supervisor- postgraduate student’. This had played out very clearly in the supervision sessions: Obviously, his door was open, but then the attention wasn’t really there, the type of attention that says ‘Do you understand what I’m saying? Would you like me to go over it? Is there anything else?’ The attention that says ‘Oh, I won’t be here, I’ll be gone overseas for the next week, so let’s make an appointment before then…’, so I knew very little about his comings and goings, what he’s doing now, you know? It was therefore understandable that Nomsa decided that she wanted to undertake her PhD with Prof MacDonald. However, at their initial, unscheduled meeting, he encouraged her not to start on a PhD directly after completing her Master’s. Rather, …he said to me ‘I would like you to go home and think about it, and we can schedule another appointment and you can come and tell me what it is [that you want to do]… I cannot tell you because it’s your PhD’. Two months later she returned to his office – this time with an appointment – and together, they filled in and submitted the application forms. At this stage, Prof MacDonald had also strongly encouraged her to apply for funding for her studies so that she could focus all of her energy on her research. He gave her access to his professional networks in order to source this funding, ‘…and that’s how it began. It was a very good start’. Nomsa’s supervision sessions with Prof MacDonald were made up of a combination of conversations. He began each one by ‘checking in’ with Nomsa, enquiring after her health and general well-being. He seemed to understand the negative impact that any personal problems could have on the progress of her research, and thus gave her a 191 space to raise them with him: ‘So, it will be more of a general conversation, like as a friend’’. After ten or fifteen minutes of this type of conversation, he would focus her attention on her research. They would discuss articles that she had read, the results of experiments that she had done, conference papers that she was writing, and so on. Throughout, …he gives me an opportunity to think for myself, he challenges you to be more creative, to bring out your own ideas. So he nurses your enthusiasm, he allows you to bring everything on the table, and if he sees that you’re going offline, he brings you back. So I was given an opportunity to explore, to bring up my own ideas, to say ‘This is what I think we should do’. And when he brings up a suggestion, he doesn’t say ‘Do it this way’, he’d say ‘I’m suggesting it, go and think about it’. In addition to this support, Prof MacDonald introduced Nomsa to his professional network, both in academia and in industry. He also arranged for her to have access to an informal supervisor, Dr Nicola Stavros, who worked in KwaZulu Natal, but who came to Johannesburg twice a year to meet with Nomsa. Dr Stavros brought a sense of urgency to Nomsa’s research: ‘She takes the approach that says ‘You need to get going! You haven’t done enough!’ She keeps you on your toes’. The fact that her two supervisors communicated regularly with one another about her work also meant that she never got contradictory feedback from them. The focus of feedback from both of her supervisors was always on the content of her work, rather than on her use of scientific language, even though the latter was where she admitted that the bulk of her problems lay. She described her struggle to write in ‘scientific English’ as having presented a major obstacle to her progress. Far from correcting each line of her work to ensure that it contained the correct usage of the scientific discourse, Prof MacDonald chose to encourage Nomsa to read more scientific articles, conference proceedings and books – by doing this, she would be able to get a sense of how to use scientific English appropriately. She appreciated this kind of feedback because it encouraged her to develop her writing skills, instead of being ‘spoon-fed’ a supervisor-corrected text which she would not be able to replicate on her own. Prof MacDonald also kept himself updated on her lab progress by visiting her there frequently. In her words: 192 He keeps himself updated so when you come with a problem, already he will know that the sample has been contaminated, [and] he would say ‘But then have you checked that this lab hasn’t been cleaned? The utensils are not properly washed’, so he’s kind of ahead of me, but be doesn’t put it as your fault – he allows you to grow. 6.2.6 Kenneth Supervision model Joint supervision – two supervisors Race and gender: student White male Race and gender: supervisor Principle supervisor: white female Co-supervisors: white male and white female Graduation status Graduated July 2010 Kenneth, a slightly-build man in his early forties, was a PhD by publication candidate42 42 A PhD by publication consists of a collection of articles authored by the candidate around a particular topic. These articles should appear in Department of Education-accredited journals, and hence are peer- reviewed for quality. Wits’ Institutional Standing Orders on Higher Degrees seems to be deliberately vague on a specific definition of the term, presumably to allow space for individual Faculties to develop their own understanding and application of it (although I have found no evidence that this has been the case). Only a passing reference to it is made in these Standing Orders, where the term ‘thesis’ is defined as ‘… an extended piece of writing based on research that makes an original and significant contribution to knowledge, that may incorporate creative work or publications integral to the overall argument, and is submitted in fulfillment of the requirements for the Doctor of Philosophy qualification’ (Senate Standing Orders on Higher Degrees 2009: 1; italics added). An amendment to the Standing Orders was tabled in March 2009 by two senior professors at the University. As a part of this proposal, they argue that individual Faculties should formulate Faculty-specific criteria for a PhD by publication, including how ‘…publications need to be integrated into the thesis so as to maintain its overall coherence’ (Revised Draft: Senate Guidelines for a PhD thesis that includes publications 2009: 1).Under the heading ‘Publishable manuscripts’, the professors propose the following: Each Faculty is required to specify the number of publications required for a doctoral thesis that includes publications. The number, publication status, method of peer review, and the status of the journals should be stipulated in the Faculty Standing Orders. In addition, Faculties should stipulate the requirements for authorship (e.g. How many independently authored papers, how many papers should have the candidate as first author, etc.) Faculties should define their own minimum criteria in accordance with practices for knowledge production in their different disciplines (Revised Draft: Senate Guidelines for a PhD thesis that includes publications 2009: 2). These proposals have yet to be approved by the University’s Senate. in the Archaeological and Paleontological Sciences. His path to a research topic was a circuitous one, characterized by two failed attempts at a proposal. His first proposal (which had been accepted by the Higher Degrees Committee) was derailed when his supervisor, Professor Michael Heath, invited him to undertake research on a different 193 project to the one in his proposal. Although the work in both projects was similar, the geographical locations were vastly different. After giving the idea some thought, Kenneth agreed to work on the new project, and accordingly put together a new proposal. Just before he submitted this proposal to the Committee, Kenneth was asked to take a group of American students on a field trip, and fortuitously struck up a conversation with Professor Eleanor Rawinowitz, who was in charge of the early Stone Age tools that would be used as artifacts during the field trip. She realized immediately that his proposed research had already been done – by one of her Master’s students. This left Kenneth at a crossroads: he could either return to his original topic, or he could explore something completely different under the auspices of a renowned scholar in the field of Paleoanthropology, Dr Quinn McFadden. Before making a decision, he approached the University’s Deputy Vice-Chancellor of Research for advice. Armed with this advice, he signed up with Dr McFadden and drafted what was to be his final PhD proposal. As the writing of this proposal unfolded, it became clear that it was more interdisciplinary than he had anticipated it would be, and thus he required two co- supervisors in addition to Dr McFadden. One was Prof Elmari Pollock from Biological Science, while the other was employed at the South African Institute of Scientific and Archaeological Research (SAISAR), Dr Olaf Wenger. At the time, Kenneth was employed by SAISAR, which worked out well for him because it meant that he could combine his work with his research. At SAISAR, he also had access to the top-level scientific technology that he required to analyse samples collected from the field. Dr Wenger provided the requisite training on the use of this technology. Supervisory meetings were regular – at least once a month. Even though Dr MacFadden was semi-retired at the time, and only came to her office once a month or so, she always made a point of letting Kenneth know when she would be there so that they could meet; and even though Prof Pollock was extremely busy, she nonetheless also made herself available for these meetings. Dr Wenger was not always able to attend, but since he saw Kenneth in the lab at SAISAR every day, this was not a problem. Kenneth described the supervision by all three as ‘superb’, although he was 194 reticent about describing what a supervision session entailed. He was more forthcoming about the feedback that he received from his three supervisors: When I get stuff back from Elmari, I can see that she’s read through it and gives me good criticism in the fields that she’s capable of giving me feedback, so the nice thing is that I’ve got feedback on my Paleoanthropology from Quinn, I get biological stuff from Elmari, and Olaf’s brilliant – he’s helped me a lot with my writing… Thus, each supervisor made a contribution to his research, but in their own areas of specialty. He juxtaposed his PhD supervision at Wits with his Master’s supervision at a Cape University: I’d give stuff in and wait, wait, wait, wait, wait, wait, wait and then, like six weeks later, you’d get it back and it’s like, it looks like you’ve read the first page and the last page, but you haven’t read anything in between. This lack of feedback on his written work was exacerbated by a lack of follow-up support – the supervisor would give him plenty of ‘brilliant’ ideas, but no ‘clues or pointers’ as to how he could use these to further his research. The issue of his supervisors’ relative power over both him and his research varied between supervisors. Since Kenneth and Dr Wenger worked together and had a similar research background, a power differential (in the traditional ‘master-apprentice’ interpretation, discussed in the literature review) did not play a major role in their relationship. Kenneth also felt that Dr McFadden did not wield excessive power over his research – that is, she did not call the proverbial shots. He had the final say about the direction and the content of his research. Since she was semi-retired, and thus spent limited periods of time at the University, the focus of her time with Kenneth was on the provision of feedback and suggestions about his research. She recognized his independence as a researcher and sought to foster this. By contrast, he defined his relationship with Prof Pollock as having a ‘big’ power imbalance. Although he did not explicitly explain the reason for this, he mentioned a couple of times during the interview how many other students she was supervising (‘forty gazillion’) as well as 195 her stature, both within South Africa and internationally, which may have been causal factors. In terms of the publication of his research, Kenneth acknowledged that this was an area that required a great deal of consideration, not least of all because his PhD was by publication. He described it as ‘A lot more work, a bit more intensive, a bit nerve- wracking, but it’s good’’. Furthermore, It gets your research out, chop chop, instead of working for three years on a thesis, then trying to write papers afterwards. So, it teaches you a lot about writing skills, reading, and so on. He has insisted on being first author of all the papers that he has written towards the degree because he felt that as the main researcher, he had earned this right. He only added the names of his supervisors when they had made a significant contribution to the research and writing process. This was partially in response to the very specific criteria for this type of PhD, but also because he clearly resented previous experiences of writing in which ‘like, forty people get tagged on’. 6.2.7 Karabo Supervision model Joint supervision – two supervisors Race and gender: student Black female Race and gender: supervisor Both white males Graduation status Still in progress Karabo is an independent and self-directed PhD candidate in the Chemical Sciences. Her undergraduate degree was completed at a historically disadvantaged institution (HDI) in the Cape. She came to Wits to undertake an Honours degree, and ended up staying for a Masters and PhD. From the beginning of her MSc, she felt supported by the Department and her supervisors, Profs Frank Hurry and James Prentiss: ‘I was very lucky, because we work as a team’. Thus, when she needed input and Prof Hurry was not around, she simply went to talk to Prof Prentiss, and vice versa. In addition, she was also able to call on the assistance of the PhD students who were working in her area should neither of the 196 Professors be available. Karabo constantly referred to this teamwork model, highlighting the value that support from multiple sources had added to her research experience. This was further consolidated by weekly meetings of the group at which research updates were presented in the form of oral reports. Interestingly, neither of the two professors used the existence of a group to avoid their academic responsibilities towards Karabo – ‘Ja, when I need to see [them], every time, they are there… they make sure that they are there…’ Thus, when Prof Prentiss took a six-month sabbatical in the United States towards the end of Karabo’s MSc, she had plenty of support to fall back on in order to complete her qualification. In addition, the group bonded together socially: We have activities as well, we go out as a team with our supervisors, we go out and have fun… We go to one of my supervisors’ place, have a braai, play together, so it’s like we are a family… A typical supervision session (every Friday morning) began with Karabo providing an update of her latest experimental results and analyses. If she was uncertain about the accuracy of either, she expressed this at the beginning of the meeting. Prof Prentiss would then ask probing, challenging questions to guide her thinking. If she remained confused, they would work together to analyse the results and to ‘try to figure out what’s happening’. Ultimately, once they had identified the ‘problem’, they would advise her which specific techniques to use that would help her to make sense of her results. Since she was a very independent student, she preferred this approach to one in which her supervisors simply gave her the answers. Prof Hurry’s role in Karabo’s research appeared to be somewhat different to that of Prof Prentiss: Frank [Prof Hurry]… like, he gave me a diary… I have a diary here – he just gave it to me. It’s like my father… the way we interact, it’s like he’s my father… She subsequently described the respect that she held for Prof Hurry as being akin to the respect that she had for her own father: ‘When it comes to my school work, he knows what he wants, and I make sure that I do what he wants’. In return, by her account, he treats her like a daughter: ‘There are no limitations’ when she needed chemicals or 197 expensive tests run on her results. In addition, on all of the conference papers and posters that they have prepared together, Prof Hurry insisted on putting Karabo’s name first, and even went as far as patenting some of her work so as to protect her intellectual ownership of it. Karabo was satisfied with the level and type of feedback that she received from her supervisors, although she did not disclose whether it was written, oral, or both. She described it as being ‘useful’. The types of feedback that she particularly enjoyed were the ‘challenging questions’ that her supervisors asked her – ‘They don’t spoon feed you, and I like it [that way]!’ Since the specific nature of her work was somewhat removed from the rest of the team, she found herself working on her own in the laboratory for long periods of time. This is in contrast to the ‘traditional’ model of laboratory research, in which a number of researchers work together, on different aspects of the same problem. Karabo was quite happy with this virtual isolation, though: I was enjoying it – you work on your own. You just do whatever you want to do and then bring the results and sit [the supervisors] down. Most of the time, you do your own work… it’s good, really, I like it so much because you grow! 6.2.8 Leigh Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White male Graduation status Graduated July 2010 Leigh is an Ecological Scientist in her early thirties, who works full-time at the National Institute of Research Into Ecology (NIRIE) while undertaking her PhD by publication. She is a strongly independent researcher who chose Dr Alex Rathbourne as her PhD supervisor because ‘…I’ve quite enjoyed the new ways of looking at things that he’s showed me … it was a different philosophical approach to ecology…’. Further, 198 …he’s very easygoing- and he basically just wants the best for me… He’s aware that it’s my life and what I want […] and being happy for me is the most important thing. Despite this, she acknowledged that Dr Rathbourne was ‘…quite a difficult guy’ who had a reputation of being intimidating and difficult to work with. In addition, …some people criticize him [because] he’s got this art on focusing on the little stuff when you need to but mostly just glossing over the details. Leigh was ‘comfortable’ with these character traits, and chose to ignore the negative impact that these traits could potentially have had on the focus and progress of her research. Thus, she pooh-poohed the idea of any power dynamic in the relationship: I think you can learn a lot from people and show your respect for them without having some silly power dynamic… I think … everyone’s got something to offer… and I knew practically nothing when I first started [but] I think you can acknowledge all that stuff without having some strange power play. Leigh has presented a number of conference papers and posters with Dr Rathbourne. In many cases, it has been Dr Rathbourne who has alerted Leigh to imminent conferences in her field of research, and has encouraged her to attend them. This has been accompanied by the necessary funding. In addition, he always insisted that her name appeared first on any presentation, followed by his – but only when he had given input: ‘I think that he would prefer to feel that if his name’s on the paper that he’s contributed in some way’. Feedback has been central to Leigh’s research progress but not all of this has emanated from Dr Rathbourne. She is a member of a research group that meets once a week for seminars, at which peers present their latest findings in an open and non-threatening environment: ‘We’re quite a nice little group in which everyone is just interested in what each other does’. In addition, she is engaged in research with a variety of other individuals at NIRIE, all of whom supply feedback when necessary: I’m working with a whole lot of people, so the one paper I’ve already got in actually had five names on it, so the other people actually gave a lot more input than Alex did because they were more involved with the work… so I’m getting a lot of feedback but not necessarily all of it from Alex. 199 However, Dr Rathbourne remained an important source of feedback for Leigh. Since both were employed at NIRIE – and, in fact, had offices across from one another at the Institute – it was a simple exercise to elicit responses to problems as and when they arose. Leigh described the following typical interaction: If I have a question, I’ll just pop in and ask him, and often I’ll pop in with just one little detail question, and then if he has the time, we’ll end up talking more generally… like last week, I just wanted to know one little thing … and we ended up chatting for half an hour about [it]. These unplanned interactions seemed to represent standard supervisory sessions, and Leigh thrived on this ostensible lack of structure. This was arguably because as an independent, self-motivated and proactive student who was employed as a researcher, she believed that she did not need as much support as other students. Leigh contrasted this rather autonomous stance in relation to her PhD with her MSc experience at a Cape University, during which ‘I was a lot more needy, I suppose because I didn’t know how to do research’. The latter experience – during which she spent six months wandering off her topic and onto an entirely different one – was a developmental one for her because it taught her how to develop her research ideas in a more focused manner. Leigh also mentioned an opportunity that she had in the first months of her PhD to undertake some research at Princeton University in the United States. This experience made a lasting impression on her and influenced her perception of how postgraduate students (and PhD students in particular) should be considered by Universities: It’s really nice to have your own space and [feel] validated. I mean, … an amazing thing about Princeton is that they’re so organized about that sort of support, there’s this amazing support… you arrive, and before you know it, you’re on the system and you’ve got a cubby hole and you’ve been told how to do this [research] and the IT people have fixed you up and it’s really easy to be a postgraduate student there… It helps that they have a huge amount of money but that department was really dynamic and had a good atmosphere. And once a week they’d organize these really quite high-profile speakers to come and talk, but they’d arrange that in the morning they’d be available and any postgrad student who wanted could book an hour’s time. So once a week, there was some hotshot scientist and they were expected to spend the day talking to postgrad students who wanted to consult with them. 200 While there was nothing comparable to this at Wits, Leigh was quick to point out that initiatives such as these were ‘department-specific’, and she balked at the idea that they were ‘…something you can control at a university level’. Ultimately, she argued that it was up to the student to be proactive in engaging with the leading thinkers in the field – the role of the supervisor in this engagement was minimal because ‘…just because he’s the supervisor on paper, doesn’t really mean anything … [Other] people are remarkably helpful’. 6.2.9 Günther Supervision model Single supervisor Race and gender: student White male Race and gender: supervisor White female Graduation status In process – after being held up by gunpoint at his fieldwork site, he left the country immediately. According to the Head of School of Geosciences ‘[He] is planning to return to complete a revised project (he will only work… where there is less likelihood of a recurrence)’ Günther is a young Geosciences student from Sweden. He met his PhD supervisor, Professor Karen van Onselen, while he was working on his Masters degree in that country: …she was walking around [Sweden] when I was there, that’s how I got to know her … she was part of my circle of people [to whom] I dropped an email: ‘I’m looking for a PhD [topic], would you happen to know anything?’ Prof van Onselen has more than thirty years’ experience in Geosciences research, and for this reason, it seemed that she could teach Günther a great deal. However, her supervision style fell short of Günther’s expectations. Firstly, she ‘…believe[d] in the free rein’, which meant that he was given a great deal of freedom to do what he wanted to do: … she let me go and she let me do things on my own, and she would be there for a consultation and when I needed to contact people, she could be of hand 201 [sic], but most of the project has been almost semi-independent already from the start. However, which large amounts of freedom comes large amounts of responsibility and this posed a very specific set of problems for Günther, given that he was trained in Europe and was working in Africa for the first time. The African continent, he ,argued, has a ‘better quality of rock to Europe’s’. By his own account, for instance, his fieldwork skills were lacking. Therefore, when tracking a specific geographical line to get geoscientific data, he was often de-railed by the absence of the very rocks that he needed for his study – ‘The problem, of course, with Geosciences, is that the rocks are not always where you want them to be, they’re unpredictable’. Despite the absence of rock-data, Prof van Onselen insisted that he not alter his geographical line. At the same time, however, she did not offer him any solutions to his question ‘What if there’s nothing there?’ – all of this seemed to be in keeping with her ‘free rein’ style’. This had profound implications for his research and resulted in a great deal of frustration for him. Ultimately, he reflected that If you are on a leash, you have to come back frequently and give progress reports – the Professor will quickly step down and say “No, you don’t do it like that’, but when you are on free rein, you have to decide when do you come to your Professor?.. I like my independence but I’m starting to feel that I may lose out because I’m too free. A second area of frustration for Günther was Prof van Onselen’s seeming indifference to inducting him into the Geosciences community of practice by alerting him to conferences that would interest him and to which he could contribute. He gave the example of a local conference of which ‘I wasn’t aware… until it began, and by then it was too late’. Prof van Onselen had been scheduled to both attend and present a paper at this particular conference, but had not told him about it. Günther remarked that he would have liked his supervisor to have notified him about it and encouraged him to attend, instead of simply talking to him about it after the fact. This may have been the result of mismatched expectations, though: while Günther maintained that ‘…she doesn’t do that, she still expects me to find out everything…’, he gave no indication that there had been an initial meeting between himself and Prof Onselen at the 202 beginning of the relationship, during which their goals and mutual expectations were discussed. Intriguingly, Günther subsequently portrayed the power relations between him and his supervisor as being ‘in the middle’ of the power continuum. He described her as being ‘like a mentor’, which he defined as being someone … in authority [and] she has the upper hand, but we still talk with each other as equals, and I consider her a friend as well. A typical supervision session is in the form of a dialogue between the two of them: It alternates with who’s talking. First of all, she will ask me what’s happening and I give her the brief on what I’ve been doing, and then I may generally prepare as well […] with a small amount of notes or questions on how to proceed or who to contact. And she generally contacts people as well, in fact quite quickly, even during the talk, she will pick up the phone and then make phone calls. It is always in a relaxed environment, that is very important for her. This description contrasts, though, with earlier descriptions of Prof van Onselen as being ‘hard to pin down’, ‘distracted by a lot of stuff’ and having ‘a hundred thousand things to do, and all her students to attend to’. Günther also noted some interesting parallels between his experience with Prof van Onselen in South Africa, and his MSc supervisor in Sweden: He [Swedish supervisor] also gave me free rein. We hashed out a programme and then he buggered off to supervise a field trip which was six [or] seven weeks, then came back and flew off to the United States for his family there, so I was completely on my own, which was very liberating and empowering, but very frustrating. He was always supportive with emails, but when we came to the writing part, he clamped down on me and I actually really liked that. Finally, Günther raised the issue of how his PhD research has had a negative impact on his personal, romantic relationships, all of which have had to be long-distance. His first girlfriend came to South Africa to live and study so that she could be near him, but ultimately decided that she ‘didn’t really like South Africa’ and the relationship ended. Subsequently, he met someone else who was also Swedish but whose career was very firmly situated in Europe. This relationship ended too. Günther was now ‘really sick and tired of long-distance relationships’ and thus wanted to complete his PhD so that he 203 could settle down. In order to do this, though, it was clear that he needed a supervisor who would rein him in and enforce strict deadlines – Prof van Onselen was, however, not this supervisor. 6.2.10 Bill Supervision model Joint supervision – two supervisors Race and gender: student White male Race and gender: supervisors Principle: white male Co-supervisor: white female Graduation status In progress – has taken five months leave to work in Canada, and hopes to complete his remaining chapter by the end of 2010. Bill is an energetic, highly self-motivated young man in his early twenties, who upgraded his Masters degree to a PhD once it became obvious that his work was of doctoral quality. He described himself, his project and his relationship with his two supervisors, Profs Bronwynne Stewart and Douglas Oxbridge, as follows: I chose my project quite specifically and I have what I feel is complete control over what I’m doing, so I know exactly what I’m doing and I know exactly what I want out of it, and that’s the kind of person I am. So, my supervisors, I use them to guide me but I don’t expect them to give me too much. I feel it’s up to me to do what I want to do, because this is for me Bill came across as being an almost defiantly empowered student. Since his third year in the School of Petrochemicals, he ‘decided’ that Prof Oxbridge was ‘someone I wanted to learn from’ and hence he undertook his Honours project with this professor. When he registered for his MSc, several staff members in the School (including Prof Stewart) offered him spaces on their projects, but none of these really appealed to Bill. In his mind, no matter which project he ended up in, ‘I knew that I would have [Prof Oxbridge] as my other supervisor’. Ultimately, he accepted the offer made by Prof Stewart on the proviso that Prof Oxbridge would jointly supervise him. Bill was emphatic that this model of joint supervision was ‘a much better idea’ than that of single supervisors because ‘it means that things are a bit more transparent’. 204 Supervisory meetings were rarely very formal affairs, which suited Bill. He describes a typical session as follows: We all meet in someone’s office. It is usually myself or [Prof Stewart’s] – [Prof Oxbridge’s] is full of papers – and then I usually tell them what I’m doing, and what results I have, and what I want to do next and then they will discuss it and give me the go-ahead or say ‘Rather, you need to do this’. It’s usually not very often that we have a meeting of all three of us… Usually, if I need to get some advice or something, I’ll just pop down to [Prof Oxbridge] or [Prof Stewart], whichever one I need to speak to. Most of the time, I won’t even phone, I’ll just run down to their office and say ‘Listen, can I ask you about this?’ and then they’ll talk about it for fifteen minutes and then that’s the end of it. Bill acknowledged that he received very different kinds of feedback from each of his supervisors. Prof Stewart tended towards general editing comments, such as ‘Change this word, change that word’, whereas Prof Oxbridge’s feedback was ‘very, very thoughtful’. Bill described it as guiding, rather than directing his work. It seemed that he venerated all of this professor’s inputs, even when he felt overwhelmed by them: I think he takes his job as a supervisor very seriously, so he really puts effort into really thinking about what needs to be done, and it’s very helpful, and sometimes it’s a little tiresome when you’re a student and you send something off and you get pages of corrections back, but at the end of the day, I think that it’s constructive, and he’s never harsh about anything. From this, it was clear that power was not a central issue, certainly in Bill’s relationship with Prof Oxbridge. He described the power relationship between himself and his supervisors as ‘pretty balanced’, mainly because he perceived himself to be a very confident student. However, he conceded that the supervisor ‘always has a level of power over a student’ which cannot be escaped from: ‘You’re kind of at the mercy of your supervisor [because] they pay you and they’re paying for your [degree] and they decide when you can submit’. In general, though, Bill seemed unconcerned about any power differential which may have existed in his relationship with his supervisors. For their parts, the two supervisors made sure that Bill had access to the kinds of resources that he needed in order to conduct his research. For instance, they motivated the university to provide him with funding for an EU-led course on Isotopes in Petrochemistry. This course provided him with some fundamental skills for his field 205 research. The supervisors also both alerted and sent him to conferences in Spain and Australia, as well as to a number of local ones. Further, when Wits’ laboratory facilities fell short of Bill’s needs, the professors pooled their resources to send him to a New Zealand university whose facilities were world-class. Bill acknowledged that for all of his confidence and hard work, he was by no means an ideal student. He cited the following incident as an example: There’s been a couple of times when I’ve gone off, just dropped what I’ve been doing on my thesis to go on holiday. I think Prof Oxbridge was quite annoyed the time I went to the Vic Falls for two weeks, and I kind of, just decided on the spur of the moment and I think that he was a bit [annoyed], but he never took issue with it at all, he [just said] ‘Okay, did you have a good trip?’ and then that was that. He gives people space for that kind of thing. Finally, Bill offered some fascinating insight into why he thought that certain supervisor-student relationships worked, while others did not: I think it depends on the student. I think there are three kinds of postgrad students: I think there’s the ones that want to learn something and really want to do that, and then there’s the kind that wants to get a piece of paper and I don’t think that’s quite as good as wanting to learn something but its still not bad because you’re generally motivated to finish, and then I think that there’s the kind who are not really sure of what to do and are a bit afraid of getting out into the ‘real world’ and so they figure they can buy a few more years in Mommy and Daddy’s house before they have to go face the real world, […] and those people always take the longest to finish. Bill definitely fell into the first category: he had an unmistakable thirst for learning as well as a hunger for the degree itself. 6.2.11 Dumisani Supervision model Single official supervisor; another informal supervisor Race and gender: student Black male Race and gender: supervisor Black female Graduation status Graduated April 2010 206 Dumisani is an ambitious young man, who lectures in the Biological Sciences at Wits. Despite the status that comes with being a both a lecturer and a doctoral student at one of South Africa’s leading universities, he confessed to possessing a lack of confidence as a scholar and as a researcher. One of the reasons for this was that his area of PhD research was significantly different from that of his MSc, and hence, he had to learn a great deal of the disciplinary content from scratch. From the outset, his PhD supervision was not ideal: his supervisor, Dr Ella Mswati, did not guide him through the proposal-writing stage at all, nor did she provide much support during the research and thesis-writing processes. Dumisani did not let that become an obstacle – early on in his registration, he had identified and invited an additional, external supervisor to guide him through the PhD. Although this external supervisor (Prof Basel) was based in Australasia and the two could therefore not meet on a regular basis, Dumisani was nonetheless able to draw on a great deal of expertise from him. This combination of a disinterested university-appointed supervisor, and a supportive, unofficial and long- distance supervisor forced Dumisani to become an independent researcher. He pointed out that in his mind, the key role of the PhD process was to expose the student to a wide range of what he called “stuff”, and to subsequently become an independent worker in this “stuff”. Supervision sessions between Dumisani and Dr Mswati lacked the level of intellectual debate that Dumisani craved. They rarely met on a one-on-one basis, but rather as a group (consisting of Dr Mswati and all of her postgraduate students). In preparation for these meetings, the students were required to provide Dr Mswati with weekly or monthly progress reports. Each student was required to present these reports to her in the presence of her other students. Dumisani felt that the only good thing that came out of these sessions was the fact that he learned how to present his research in front of peers. However, Dr Mswati was not forthcoming with feedback during these sessions and thus Dumisani was forced to look elsewhere for the kind of guidance that he felt he needed for his research. He reported: “You’ll give her the chapter, and instead of looking at the chapter, she would just… perhaps, with a red pen, then just mark it perhaps you forgot to check one word, she would just mark that word only and say “Go… go back and redo it again” and without even checking the rest of the stuff”. He 207 also felt that her reading of his work was superficial – hence the dearth of constructive feedback. He recalled one specific incident in his writing in which he knew something was missing, but could not quite put his finger on what. He hoped that Dr Mswati would be able to shed light on this ‘missing link’, but “that was not the case”. For this, he turned to Prof Basel, Stanley (another one of Dr Mswati’s students who was au fait with issues of content that were new to Dumisani) as well as an unnamed colleague at a Cape Town university. Dr Mswati was also resistant to change, which frustrated Dumisani because he recognized the evolving nature of the discipline, and hence the need to evolve with it in order to stay “credible”. He described an incident where he had learned a new and more up-to-date laboratory technique at a conference. He was keen to apply this method to his own PhD research and thus “…give us credibility in terms of when we have to publish the work”. Dr Mswati refused, arguing instead that the “old way” was the best way. Eventually, Dumisani resorted to introducing the new techniques to Dr Mswati’s other postgraduate students, whose en masse acceptance of the new methods persuaded her to acknowledge their efficacy. Dumisani’s experience of co-publication with his supervisors was equally fraught with problems. He admitted that there were times when writing with Dr Mswati that he had wanted to simply remove her name from the paper because of the constant obstacles that he perceived her to be throwing in his way. For instance, she would force him to rewrite entire sections of the paper (“Rewriting for the sake of rewriting”) without giving him a valid reason for these re-writes. As a result of this, it took over a year for one particular article to be published. Writing with his Australasian, Prof Basel, supervisor was no less frustrating: even though Dumisani did most of the research and writing, and was listed as first author on the paper, his Australasian supervisor insisted on being listed as the corresponding author. This is because Prof Basel was employed by a research entity in Australia and his continued employment there meant that he had to deliver a certain amount of published articles annually – the only way that he could justify working with Dumisani to this entity was by publishing articles listing him as the corresponding author. In both cases, Dumisani felt like “… I wasn’t there in the paper” and that he was alienated from the article even though he had done most of the 208 groundwork for it. Furthermore, Wits has calculated his research subsidy for the latter paper at only 25% because Prof Basel is listed as corresponding author. Dr Mswati had a singlestrength as a supervisor, according to Dumisani: the fact that she regularly took her students to conferences so that they could keep up to date with the latest trends in the field. She paid for the first conference that Dumisani attended as a PhD student, and thereafter helped him to put together NRF funding proposals so that he would be able to fund his own conference attendance. This money also paid for the equipment and chemical reagents that he required for his laboratory research. Once he had access to these funds, the balance of power between him and Dr Mswati evened out because he no longer had to rely on her for the provision of the laboratory materials that he required to undertake his research. Dumisani’s narrative of his PhD supervision experience contrasts sharply with that of his Master’s degree, which he undertook in KwaZulu Natal. His Master’s supervisor, Professor Johann Maritz, was described as being “world class” because he took Dumisani through the research process, step-by-step, right from the proposal stage, to the final writing of the dissertation. This included taking Dumisani into the laboratory and literally demonstrating each technique that was needed for his research, and then allowing Dumisani to practice these techniques for himself. Prof Maritz’s feedback on Dumisani’s written work was far from superficial: he spent a great deal of time challenging the content of what Dumisani had written: “Why did you say this? Why did you say that? How is this related to your research?” By doing this, he effectively modeled for Dumisani the kinds of thinking that he needed to adopt in articulating his research and more importantly, in “writing scientifically”. The result of this intense, one-on-one supervision was that Dumisani completed his MSc in eighteen months, and not in the standard two years. 209 6.2.12 Megan Supervision model Joint supervision – two supervisors, although the co- supervisor is ‘in name only’ Race and gender: student White female Race and gender: supervisor White males Graduation status Still in progress Megan is a Chemical Scientist in her late thirties, who returned to academia after a stint working in industry. She immediately registered for a PhD in the department with Prof Zakinsky as her supervisor. She was also given a co-supervisor, but he was a ‘…supervisor in name [only]’. Just over a year later, Prof Zakinsky unexpectedly resigned to take up a position at another university in the province. He took with him all the scientific equipment that he (and Megan) had been using at Wits, as well as most of his postgraduate students. Despite this, Megan chose to maintain her registration at Wits, and keep him as her supervisor, albeit a long-distance one. The two kept in contact via email, but ‘…he’s not a great emailer. I email him but he doesn’t reply’. Despite Prof Zakinsky’s move and his lack of regular email interaction, Megan argued that he had not been a bad supervisor. In the first place, she described how he took a genuine interest in her work, and the results that came out of her experiments. His input at supervisory meetings was described as being “wonderful”. Even when he sent her back to do work on aspects of the experiments that she later discovered to be tangential to the crux of her research, she still felt that this was because he was truly interested in the outcomes of the research. Secondly, “If you had problems with instruments, he’d go and he’d actually sit in with you and help you …”. Megan valued this because one of the instruments in question was one which had been built up – not bought – and thus it did not come with an instruction manual. Even after he had moved universities, Prof Zakinsky would offer telephonic suggestions about what Megan should try doing to get the instrument to work. Her co-supervisor, on the other hand, was of little use to her in this context because “He’s not familiar with the field, really… he’s only familiar with one side of it…”. Fortunately, Megan is a “fairly independent worker” who feels that 210 she does not “…need somebody to tell me what to do next, I’ve got my own ideas of what I want to do, and I’ll do it”. A typical supervision session was undertaken in a group: Prof Zakinsky and his students would meet all together, and each one would list what they had done since the previous meeting, describe results from their experiments, raise questions that emerged from these experiments, and so on. Prof Zakinsky would then give his interpretation of these aspects of the work. Although these sessions were time-consuming, Megan found them to be very valuable: “So, you’d not only learn from your own work but you could learn from others’ as well, which was nice. It was very time-consuming but I think it was worthwhile especially in the beginning [when] you get people that are further along and you actually learn from what they’ve done”. This made up for the fact that Prof Zakinsky did not often provide written feedback on her research work. According to Megan, Prof Zakinsky’s most valuable written feedback was on her proposal for NRF Thuthuka funding – “He’s the one that said to me ‘No, no, no, no, you’ve got this all wrong, you must learn how to play the game’”. Thus, feedback from Megan’s supervisor could be described as being ‘uneven’ and ‘patchy’. Even though supervision took place in a group, each of the students in this group worked in rather disparate areas of Chemical Science, and hence there was no real opportunity for Megan to become part of a community of practice in her specific area of research. She nonetheless found that being a part of a group to be reassuring, since it seemed that the other students were experiencing similar problems to her. This was particularly important given that she is works for a research unit that is housed in a separate building to the Chemical Science department, and thus often feels isolated from the rest of the department. This changed in early 2009 when she began working on a collaborative research project with a colleague, Garth. Even though Garth was not in the same field of research as Megan, “… we sort of overlap in where our ideas are going…” Furthermore, she has been assigned an MSc student to supervise whose area of research is in line with her own. Although this is not a community of practice in the 211 purest sense of the term43 Prof Zakinsky did take Megan (and his other students) to conferences, albeit only the local ones. Megan enjoyed them, especially when she got the opportunity to present an oral , it does nonetheless represent a point of connection with other researchers in her field. Despite this bumpy PhD supervision experience, Megan argued that it was still a step up from her Master’s supervision. She had undertaken a part-time Master’s degree with a neighbouring university during her time as an industry employee. She did the practical experiments at work (when she had time) and submitted work to her supervisor – an individual whom she had thought was a ‘guru’ in her field. Unfortunately, this perception did not translate into reality: at one point, she wrote up and forwarded to the unnamed supervisor a large chunk of her experimental results, organized a meeting with him and arrived with a list of questions. However, “I came out of there feeling like ‘This was a waste of time’”. This supervisor was simply not “forthcoming with anything” and thus added no value to Megan’s experience. She argued that “…when you’ve got problems, you go and you talk, but there’s no supervision per se”. She commented that she doubts that this supervisor even read the MSc before sending it out for external examination. 44 Megan reflected that the power relationship between her and Prof Zakinsky was one that he had initially tried to “enforce”. He delineated a clear intellectual boundary between himself and “his students” (which she described as being “hierarchical”), even or a poster. Prof Zakinsky gave her input by means of ‘practice runs’, but never insisted on being acknowledged for this. Furthermore, on the only paper that the two have submitted for publication, Megan listed the authorship alphabetically – which meant that her name appeared first. She argued that this was appropriate because she had done most of the ‘donkey work’. Prof Zakinsky did not dispute this order of names when he submitted the paper for review, and Megan describes feeling “chuffed” and somewhat surprised by this. 44 In the context of Chemical Science conferences, an ‘oral’ was simply a presentation – according to Megan “We don’t publish the stuff that we present”. 212 when student Megan was also colleague Megan. She described the following incident to illustrate this point: … he had organized a seminar with … they were trying to get a kind of international collaboration going on a project, and there were a whole lot of people who had come together from all over, and they were talking about how they were going to do this and so on, and he said won’t we take notes and stuff like that… and you think “Okay, we might be part of it”… but we were essentially the donkeys doing the work and then said they’re going off for lunch, so I said “Oh, aren’t we going with?”. He said “Oh, I didn’t book enough places”, so I said “Oh, well, fine!”. At least the next day, he organized lunch for us, but you know, if I hadn’t said anything, he would keep that distance there. However, Megan conceded that Prof Zakinsky appears to have reached the point where he has acknowledged that she is the expert in her area of research, and the boundary between the two seems to have diminished somewhat as a result. 6.2.13 Bruno Supervision model Joint supervision – two supervisors Race and gender: student White male Race and gender: supervisor Main supervisor: white male Co-supervisor: Indian male Graduation status Still in progress Bruno is an intense, serious young man, who registered for a PhD in Mathematical Science mainly because it was a requirement for his permanent employment in this department. However, he observed that the PhD was something that he had always been working towards, since his undergraduate studies. Bruno’s PhD supervisor, Professor Stone, was also his Master’s supervisor. The department insisted that he also have a co-supervisor – Prof Shriva – because “…Prof Stone… we took a long time to finish my Master’s45 45 Note the use of pronouns: we took a long time to finish my Master’s. so they gave me Prof Shriva as well, who was known to finish PhDs quickly…”. However, there have been two major obstacles in Bruno’s path to completing the doctorate: firstly, the issue of time (or rather, the lack thereof) to work on his research – he teaches several classes in the first 213 semester of every year, and hence these months are effectively ‘written off’ in terms of PhD work. A second obstacle was that of keeping up to date with the latest literature in the field, “All the reading that goes into doing a PhD… so you need to know the literature, so you’re trying to do two things at the same time – do the research and do the reading”. Thus, despite having a PhD topic in mind, and despite knowing that the method he has chosen to test the problem “… has to work, because it can’t not work”, Bruno still feels constrained by the context in which he finds himself. Supervision sessions do not seem to be important for Bruno. This is because he is rather possessive about his research: while he will ask both his supervisors for advice when he is stuck, he is reluctant to engage with them too much because “I don’t want to do someone else’s research – I want to do my own”. Hence, he describes his meetings with Prof Stone (his ‘main’ supervisor) as lasting for about five minutes, and occurring only when he has submitted written work and is awaiting feedback. For the most part, however, Prof Stone “Just lets me do my research, which is fine”. On two separate occasions, Bruno commented on how he believes that his supervisors are simply there to “sign the papers, for the admin”, because the research work itself belongs to him (he emphatically describes his research as ‘My problem’). In addition, the feedback from both supervisors is mostly oral – only on research articles that Bruno wishes to submit for review does he receive written feedback. He describes this feedback as being ‘editorial’. Bruno’s PhD experience of supervision is significantly different to that of his Master’s only in terms of the source of the research topic. While for his PhD, Bruno identified a problem to research, his Master’s topic came from Prof Stone. He described the process as follows: I …knew that Master’s was one step closer to a PhD, that I had to get through, so I just went looking for something and oddly enough, I went to Prof Stone and I said “I’m looking for a Master’s… something to do for a Master’s [but] I don’t want to do boundary layers”, … well, he gave me something but it ended up being boundary layers, and that’s why it took me the best part of four years to finish my Master’s. I just didn’t understand it properly” 214 Interestingly, it was during the period of Master’s research that his PhD topic ‘cropped up’, and thus when it came to undertaking doctoral research, he already had a problem to solve. Bruno’s perceptions about conferences and publications were rather ambivalent. On the one hand, he commented that he found them to be a ‘waste of time’ – both the local and the international conferences. While his supervisors supported his attendance at these gatherings, and he was encouraged to present and publish papers, he felt that all conferences were the same because “You know what you’re talking about [but] the audience doesn’t know what you’re talking about… most people say that you just go to conferences to make contacts”. On the other hand, however, he constructs himself as having little to offer the field since he is a relative newcomer: “It’s difficult to go to a conference and say that the work they’ve been doing for the last eighty years is wrong… they know more than I do in that they’ve been in the field longer than I have…” Even though he acknowledges that conferences are excellent opportunities for networking with experts in the field, he tends not to engage at this level, because “…they always appear to be on a different level than what I do, so I’m looking up to them, rather than seeing them as an equal”. In terms of publications, and in keeping with his strongly individualistic approach to his research problem, Bruno has not published with either of his supervisors, although he has asked for their input on papers that he has submitted for publications. He stated quite baldly that they were ‘not co- authored”, although he does acknowledge them both at the end. Neither Prof Stone nor Prof Shriva seem to be concerned about this – they seem to recognize that Bruno is a independent student, with no real interest in collaborative work or in being a part of a community of practice. The issue of power has been a non-issue in Bruno’s relationship with his supervisors. In this context, he described the three of them as “being on the same level, almost”. Thus, while they have taught him a great deal, his research has also fed into their knowledge and understanding of the field. 215 6.2.14 Shanti Supervision model Joint supervision – two supervisors Race and gender: student Indian female Race and gender: supervisor Principle supervisor: white male Co-supervisor: white female Graduation status Graduated in July 2008 Shanti is a young, slightly built woman who does not fit the traditional profile of the term ‘scientist’. Her PhD supervisor – Prof Joseph MacDonald – also supervised her Master’s degree. His role expanded in the second year of her PhD, when Shanti joined the University’s ‘Developing Young Researchers (DYP)r46 46 ‘Developing Young Researchers, or DYPas it was commonly known, was a staff mentorship programme aimed at developing academics from previously disadvantaged groups so that they could advance through the academic ranks of the university. One of the main aims of the programme was to effect the transformation of the University’s academic demographic profile, particularly at the level of senior lecturer and above. ’ programme, and he was appointed as her mentor. She reflected that having the twin roles of supervisor and mentor situated in the same person was not difficult, but rather reinforced the existing relationship between her and Prof MacDonald: ‘I think it was fine… I suppose you do relate on a slightly different level because you start talking about other things aside from just science, you know…. So we got to know each other very well.’ The supervisory relationship between the two was, according to Shanti, a ‘very easy’ one, mainly because they had become used to what Shanti described as each other’s ‘work ethic’. By this, she meant ‘way of working’, inferring that they understood each other’s approaches to research, laboratory work and interactions at supervisory meetings. This was consolidated by the fact that their offices were close together on the same corridor, as was the laboratory that they shared. They combined formal and informal meetings according to what Shanti needed from Prof MacDonald at any given time. Thus, for example, during the data collection stage when Shanti was conducting experiments in the laboratory, ‘…as soon as you get some interesting results, then you’d rush off and go and show him what’s happening, and then he’d come to the lab…’ 216 The formal supervision meetings occurred towards the end of each set of experimental results; at this point, she would tell him that she needed to have a meeting with him that was ‘going to take more than twenty minutes or so’, and he would block off some time in his diary. This happened once every six to eight weeks. During these meetings, Prof MacDonald took on the role of a sounding board for Shanti’s ideas, except when he perceived that she was “off the mark”. When this happened, ‘…he would just, sort of, steer you in another direction but without being overt about it. He’d just give you an idea….’ By the same token, when she decided that she no longer wished to pursue a particular route of experimentation because it did not seem to be leading anywhere significant, he would ‘more often than not’ say to her ‘Well, try it, see how it goes, and then we’ll adjust accordingly…’ Shanti found this to be a very productive and generative way of working, since it allowed her the opportunity to develop as a scientist within a supportive environment. Prof MacDonald’s feedback on Shanti’s work was generally oral ‘… because [in terms of the science behind it] by that stage we had established pretty clearly what’s going on…’ Any written feedback that she did receive from him was usually about ‘… the writing style, or the techniques or that kind of thing…’ This stood in contrast to the feedback that she received from her co-supervisor, Dr Anita Perreira [who was only mentioned for the first time approximately a third of the way into the interview]47 47 Dr Perreria was based in a coastal city in South Africa, and was described by Shanti as being “…sort of like a mother”. Their relationship began when Shanti read for her Honours degree, and was supervised by Dr Perreira. Furthermore, Dr Perreria co-supervised Shanti’s Master’s degree alongside Prof MacDonald. She was also the link with the company that commissioned Shanti’s postgraduate research. . Dr Perreira ‘…went a little bit more in depth in terms of her feedback on the written work” and while this was useful from one perspective, Shanti reflected that “…you sort of get the feeling that it’s starting to become someone else’s work when there’s too much feedback’. Shanti would always show Prof MacDonald her written work first before sending it to Dr Perriera who would invariably return it with a “whack load” of feedback. From this perspective, Shanti felt that having two supervisors was more trouble than it was worth: ‘Joe and Anita have very different writing styles … I would hand in something with Joe and he’d be happy with it and then Anita would say ‘Well, why don’t you change this?’, so I’d do it and then you give it back to Joe to read again 217 and he says ‘Why did you change this? It was fine the way it was before’. So we had a couple of iffy moments there, and we’d go back and forth and eventually I’d just discard one of the suggestions and go with whatever felt better to me.’ Shanti was aware that this often undermined her sense of control over her research project. Nonetheless, she admitted that Dr Perreira’s comments were presented as being “suggestions”, rather than as instructions. Power seemed to be a non-issue in Shanti’s relationship with both of her supervisors. This was particularly true of Prof MacDonald. He was a constant source of affirmation for her, especially when she began writing up the thesis. At this stage in the process, she was usually the one who came up with new insights, whereas at the beginning, Prof MacDonald had been the source of these insights. Shanti described the absence of a power relationship as follows: ‘I remember [Prof MacDonald] telling me when I started writing up … and I would ask him questions and he says “But you’re the expert in this, you have to tell me now… because you actually know more about this now than I do!’. The lack of a marked power relationship was also reflected in their collaboration on conference papers and posters. Shanti and Prof MacDonald attended ‘plenty’ of conferences together – both local and international. Funding for these came from both the School and from the DYP programme, and they often stayed in an international city for a few days after the conference, to relax and sightsee. In this context, their relationship took on the guise of two friends enjoying a holiday together. Shanti was always listed as first author on all of their co-publications, with Prof MacDonald’s blessing. In her words: ‘…whenever we write papers, if you’ve done the work, you get first authorship, and then the others get listed. It never really occurred to me that it would be challenged…’ Both Prof MacDonald and Dr Perreira were credited with Shanti’s development as a researcher, but she argued that this was done more through a process of modeling of good practice on their parts, rather than through explicit instruction. This was balanced with the recognition that Shanti was not only a PhD student, but also existed as a daughter and wife, and needed time away from her studies. The interview excerpt below describes this well: 218 “..towards the end of my PhD, I’d been furiously writing up, from about August to the end of November, and … I told Ron [Head of School] “I’m not doing any teaching here, I need this block off, just let me do it”, and, uh, so he did, so I was at home and I got it done and I wanted to …have that final draft in by I think it was the first couple of weeks of December, in fact, no, I gave it in in the first couple of weeks of December, the first week I think, and I told Joe “Well, see if you can have a look at it, maybe we can submit it first thing in January”, and he said “No!” [laughs], so I said “But why? You know, I want to get this done. I’ve been pushing so hard, just let me finish it”. He says “Just leave it, you need to take a break, I mean look at you“, he says, “You’re wasting away! You need to take a break and I’m not going to do this now. When we come back, I’ll look at it in January and we can submit it in February”, and I was cross, I was like [laughs] “How can you do this? Let me finish it!”, and he said “No, just leave it. Go and have a good break, enjoy yourself, forget about this completely” and when I came back, it actually did make a huge difference, you look at it with such different eyes and he was right, and I felt a lot healthier after that, I mean, I was wasting away, literally..” 6.2.15 Kristelle Supervision model Single supervisor Race and gender: student White female Race and gender: supervisor White male Graduation status PhD placed in abeyance for 2010 Kristelle is a highly accomplished young woman from the Mathematical Sciences – she has Master’s degrees in two separate branches of this discipline. At the beginning of 2009, she took the decision to put her PhD studies into abeyance. She was quick to point out that this had little, if anything, to do with her PhD supervision experience. Rather, it was her packed teaching schedule that forced her to take this step. Kristelle described 2008 as a year of ‘serious staff shortages’ in her department at Wits. The timing of these shortages affected her work on her PhD proposal, which had begun in January 2007. However, as the interview progressed, it became clear that her teaching workload was only the overt reason for her decision to walk away from the PhD. The proposal itself was a sticking point for her: she argued that Wits expected students to write a broad and all-encompassing proposal in six months, even though 219 PhD ‘…research is about going into uncharted territory…’. To support this point, Kristelle related a story from her first Master’s degree (at another university): When I did my Master’s in Numerology, … the Faculty expected us to have a proposal that needed to be written and my supervisor said ‘Don’t worry about it, you’re not going to write a proposal because you are not going to know what you are doing until you’ve finished it’. For Kristelle, the proposal should therefore only be 6 pages long, ‘…to say a brief outline of what has been done and this is how we’re going to attempt to do it. That’s it!’ At the point that she withdrew her candidature, Kristelle’s proposal was about a hundred pages long ‘… because there are so many things that you need to define and say what has been done and so forth for you to be able to define where the gap is and where you need to go. That’s not what it should be.’ She was also very critical about the pressure placed on students to register for their PhDs before they had gotten a sense of the field of study they wished to pursue – and especially the gaps in this field. For this reason, her advice for budding PhD candidates is Before you have a proposal, don’t even think about registering for a PhD! Before you’ve done the reading that you need to do, in order to know what the gap is, don’t even go to registering for a PhD! Kristelle’s PhD supervisor, Professor Maxwell, whom she absolves of all responsibility for her decision to put her PhD in abeyance, is a half-time appointment in her department. Thus, he spends 50% of his time in the department, and the other 50% running a private consultancy. Even for a self-starter like Kristelle, Prof Maxwell was a bit too ‘hands-off’ for her liking: ‘I sometimes felt like he was a little too afraid to ask [about my progress], like almost walking on eggs, also not wanting to offend on my side…’ She admitted that there were certain times when she felt that Prof Maxwell needed to be ‘a little bit more forceful and said ‘You know what? You’re not working as you should be’’. When he did encourage her to work harder on her PhD, it was usually couched in terms of her continued employment at Wits: ‘You need to start looking at your PhD as part of your job.’ This did not sit well with Kristelle, who pointed out that she is employed on the tutor track which does not require one to complete higher research degrees. 220 Despite this seeming lack of alignment between what Kristelle needed in a supervisor and what Prof Maxwell actually provided, he did support her when it came down to the practical skills that she needed for her research. Specifically, because she did not have a programming background, and because her research problem had so many variables and measurements that needed to be coded and calculated, he sat down with her at his computer and took her through the coding steps that she needed to follow. According to Kristelle ‘And after that – it was probably an hour, hour and a half session, after that hour and a half session – that click you needed? It finally happened’. On other occasions, he helped her to ‘interpret’ some of the articles that she had read – generally, those written in unnecessarily dense English. Supervision sessions were relatively regular events, even when ‘… there wasn’t necessarily that much to discuss.’ At these meetings, Kristelle would discuss what readings she had done and identify areas of interest that had emerged from them. Prof Maxwell gave her both written and verbal feedback. His comments were generally constructive, and Kristelle recognized the encouraging nature of this feedback. However, he did not provide her with a way of envisioning the ‘big picture’ of her PhD research - in this sense he was also quite hands-off in his approach to supervising Kristelle. She recounted how she often felt that the two of them had very different ways of viewing her research and, instead of communicating his views with her (as she did with him), Prof Maxwell tended to tell her ‘Okay, go and do what you think you need to do.’ Kristelle attributed this approach to two main factors: firstly, that “I don’t think he realized exactly how much literature there was in the area”; and secondly, ‘… this is not necessarily one of the areas that he is specifically interested in.’ These two factors seemed to have supported Kristelle’s decision to put her PhD into abeyance, although she did not verbalise this herself. Kristelle is considering re-registering for her PhD in 2010, but is anxious about her ability to complete it given her historical experience as a PhD student. 6.3 Conclusion As was the case with the Humanities candidates, there were a range of supervision experiences related by the Science candidates. While some were co-supervised – in 221 accordance with the traditional science model – others such as Ace and Nomsa undertook their research with the support of a single supervisor. Yet others – Dumisani and Leigh are cases in point – supplemented their official supervision by interacting with other authorities in their disciplines. These variations on the traditional model of supervision (discussed in more depth in Chapter 7, section 2.1) arguably represent a new direction in doctoral supervision in this Faculty, a point that I will pick up on again. Power and agency appeared to be almost non-issues for the Science candidates. With the exception of Seamus, who projected an image of abject disempowerment and anger, the other candidates seemed to have asserted a significant amount of control over their research work to the point that many were dismissive of the suggestion that it was the supervisor who was ‘in charge’. Supervision meetings took place in offices or laboratories, and were rarely formal affairs. As was the case with the Humanities candidates, the location of these meetings seemed to be unimportant to the students – they were far more interested in the engagement with and feedback from their supervisors. This is telling because it foregrounds the fact that these novice researchers are not as concerned with the priorities emphasized by the thesis-writing publications (discussed in Chapter Three) as the authors of these texts appear to think they ought to be. In other words, these narratives reveal that doctoral candidates are more interested in the substance of supervision – feedback, co-publication, engagement – than they are with the administrative niceties such as meeting in a quiet office with no interruptions. The only main difference between the Humanities narratives and those of Science related to how two candidates who shared a supervisor experienced this individual. Nomsa and Shanti both related similar stories about the supportive, nurturing and encouraging nature of Prof MacDonald’s supervision style. While he demonstrated these qualities differently for each student, the nett result was the same: doctoral candidates who felt validated and empowered. In the chapters that follow, I will analyse this narrative data in order to surface some meaning from it. 222 CHAPTER 7 ANALYSIS Through the looking glass: students reflecting on their supervision experiences [My PhD is] the biggest thing personally I have ever lived, psychologically. We face our demons, we are unsure of ourselves, we can be inspired, we can be passionate, we can be devastated thinking this is never going to end…(Sophia, Humanities PhD candidate). Chapter overview: In this section of the thesis (and the chapter that follows), I present an analysis of my interview data. I begin by developing a framework through which the data can be understood. This framework is represented by a metaphorical picture window composed of four panes. Each pane – representing context, process, relationships and activities - offers a medium through which to understand the data. The context pane will be presented in this chapter. It focuses on the faculty in which my interview subjects conducted their research, as well as what their experience of Master’s research had been like, and whether or not this had swayed their decision to undertake doctoral study. 7.1 Introduction As mentioned in the Introduction, I interviewed 30 PhD students, who were at various stages of their PhD research – some were still in the proposal-writing phase, while others had submitted their theses for examination and were awaiting the result. A handful had recently graduated. Fifteen of these students were registered in the Faculty of Science, and fifteen in the Faculty of Humanities. Fourteen members of the cohort were male, and the remainder female. In terms of racial categorization, fifteen were white, twelve black, two Indian and one coloured. While I recognized that the lived experiences of my interviewees might have been racialised and genderised, and that these are unquestionably significant issues, especially in the South African historical context, the intended focus of my research was on those issues that the candidates themselves felt were important. Where they raised race and gender, so did I. I sought to pinpoint what their expectations of doctoral research had been, what their individual encounters with their supervisors (and other contributors to their research) had been like, whether they had been changed by the supervision process and whether, in turn, they might have changed it for future PhD candidates. I wanted to probe the 223 experiences that are often not spoken about, that hover just below the surface – the assumed knowledge that PhD candidates are understood by supervisors to bring to both the supervisory relationship and to the research itself. In my own personal experience as a PhD candidate (described in the Prologue to the thesis), this prior knowledge included the ability to develop a topic on one’s own. The assumption that a doctoral candidate is able to do this is clearly problematic, given the fact that, as the literature points out, the first time that many students may be required to develop a topic for themselves is at doctoral level (Badenhorst 2008; Wisker 2005; Glatthorn and Joyner 2005; Walker & Gold et al 2008; Golde & Walker (eds) 2006; Kelly 2003). This is in contrast to the Honours and Master’s levels where the supervisor often provides the research topic (although this is more prevalent in the Sciences than in the Humanities). In addition, I wanted to probe whether or not the PhD candidates in this study understood the fundamental differences between a Master’s research report and a PhD thesis – or whether there had simply been an assumption on the part of the supervisor and/or institution that these differences were ‘common knowledge’. I wanted to explore the inherent contradiction in the substance of the process – that the expectations of a PhD are much higher than those of the preceding levels, and yet these are rarely fleshed out for doctoral candidates in any meaningful way. What is ‘common knowledge’ for a supervisor/institution is not necessarily the same for the student, who is expected to generate new knowledge without the benefit of this ‘common knowledge’. As I mentioned in Chapter 4, I identified five categories from a data-coding scheme developed by Bogdan and Biklen (2007: 176) with which to analyse the data from my interviews: setting/context, the process, relationships, activities, and events. I conflated the latter two categories because I found a great deal of overlap between them. For example, conferences attendance by students and their supervisors may be considered to be both an activity – in that it is a pursuit undertaken by both – and an event. Separating them out into two different categories would have diminished my arguments about the critical impact that conferences (and conference attendance) have on the candidates’ experience of the supervision process. 224 Before I begin to apply these categories to my research, a description of each of them is in order. At one of our supervisory meetings, one of my supervisors, illustrated a point (related to a different part of my thesis) using the metaphor of a picture window, consisting of four frames. As I began to think about how I could present the data in the above categories, I realized that the metaphor of individual windowpanes making up a bigger, picture window would make an ideal organising principle for my analysis. The transparency of the glass provides a broad appreciation of the landscape beyond it (the data), while each unique pane in the larger window reveals a specific facet of the bigger picture. One can either choose to stand back and take a panoramic view of the data, or one can examine it pane by pane, with each adding to the other until the reader has a mastery of the ‘bigger picture’. No pane is more significant than any of the others; they all contribute to the broader landscape view of the students’ experiences of being supervised (See Figu The conceptual framework of postgraduate supervision pedagogy and practices are comprehensively captured within these analytical panes. Figure 5: The picture window analysis structure Context/Setting Process - Faculty - Supervision meetings - “Model’ of supervision - Supervision practices - Master’s experience - Feedback Relationships Activities/Events - Issues of affect - Conferences & networks - Power - Co-publication 225 The first pane through which I directed my analytic gaze was that of context or setting – in this case, the backgrounds of the PhD candidates who were interviewed for this research. Included in this pane was their home faculty (Science or Humanities) as well as the ‘model’ of supervision generally attributed to this faculty in the literature; the site of their Master’s degree (Wits or another university); as well as their experiences of being supervised at Master’s degree level. By this, I am referring to the impact (if any) that the supervision at this level might have had in their decision to register as PhD candidates. I was interested in finding out whether this experience had influenced their decision to pursue a doctorate, or whether their reasons for embarking on a PhD were independent of these experiences. Only six of the thirty participants in my study had undertaken their Master’s research at Wits; three more had upgraded their Master’s degree to a PhD (also at Wits). The remainder had come to this university from other institutions in South Africa (15) and from what I have chosen to call ‘international’ universities (6), even though this term includes institutions in neighbouring states such as Lesotho and Swaziland48 My interest in the candidates’ Master’s experiences was largely informed by my awareness of both national and institutional imperatives for increasing the number of PhD candidates in the South African educational system (discussed in more depth in Chapters 1 and 3). At both levels, burgeoning numbers of PhD students represent an increased South African participation in, and contribution to, the highly competitive (international) knowledge economy. The word ‘economy’ implies money; in this . Thus, for the majority of this group, the Wits context was a new one. At the time that I was conducting the interviews, I did not probe why this cohort had migrated from their Master’s institution to Wits, but as I analysed the data, I wondered whether there was any significance to this migratory pattern, or whether this inter-institutional attrition was a normal part of general trends in higher education, both nationally and internationally. While finding the answer to this question is not the focus of this thesis, it would be an area that a future researcher may wish to probe. 48 Both Lesotho and Swaziland are recognized under international law as being sovereign, independent states, despite their relatively small size and the fact that they are both landlocked by South Africa. Their education systems remain aligned to those of their former colonial power, Great Britain, and include A and O level qualifications. By contrast, South Africa’s education system consists of an outcomes or competence-based system (although this will be scrapped at the end of 2010), culminating in a matriculation certificate upon the completion of grade 12. Learners may leave the system at the end of Grade 9 with a General Education and Training (GET) certificate. 226 particular instance, it alludes to the lucrative ‘market’ for new knowledge, inventions and discoveries. Ergo – although this is not necessarily always an accurate representation of the practice – the more knowledge that a state/institution can generate, the more it stands to benefit financially. Status and prestige are by-products for both. Thus, I wanted to identify the factors experienced by students at the Master’s level that encouraged them to register for a PhD, and in so doing, to contribute to the knowledge economy. The second pane of analysis through which I will analyse my interviewees’ responses is that of process. This category of analysis will be used to examine the doctoral supervision process as experienced by these candidates, and will encompass aspects such as descriptions of supervision meetings, the kinds of feedback that supervisors provided to students and how (if at all) this feedback supported further development of the research. What became clear as I developed this section was that there was no single ‘correct’ path that students and supervisors followed when it came to the doctorate; rather, there were numerous different paths, each defined by a confluence of factors such as personality, motivation and previous experience – of both the supervisor and the student - to name but a few. Thus, creating a ‘checklist’ for the process seems to be pointless since it would not have been able to adequately capture these nuances. The third analytical windowpane presents a view of the relationships that existed between supervisor/s and student during the course of the research process. For some, these relationships pre-dated the PhD research – either the students had worked with the same supervisor since their Master’s degree (or even earlier), they had upgraded this qualification to a doctorate, or they had worked with their supervisor in a professional setting prior to their registration as a PhD candidate. These relationships are often built around the perception of power and authority, and with whom they reside. For some of the interviewees such as Nkosinathi and Günther, it was the supervisor who decided the direction that the research would take, which could be considered as being disempowering for the student because they effectively had no agency over or ownership of their research. In other relationships – the majority, in fact – it was the candidate who either overtly controlled the direction of their research, or who chose to incorporate suggestions from their supervisor when they deemed it 227 appropriate. It seemed that the manner in which the student experienced the power in the relationship defined how she or he experienced the research. In other words, where the supervisor’s influence on the direction of the work prevailed (as in the case with Günther), the research did not seem to excite the PhD candidate to the point that it became the focus of his life, both academic and personal. On the other hand, when there was a more or less equal balance of power between supervisor and student, or where the student possessed a strong sense of agency over the research project there seemed to be more excitement, both about the content of the research, and about the process of collecting, analyzing and writing it up. This was the case for Rebecca and Shanti. Thus, it is not possible for this particular pane in my window of analysis to disaggregate the issue of affect (that is, the emotions that students associate with their research) from the issue of power. The final pane of my window explores the particular activities and events that my interviewees identified as having been pivotal in their development as competent researchers. These included conference attendance with the supervisor, co-publication and in some cases, and the introduction of students to key thinkers in their discipline, thus giving them the opportunity to establish professional networks in their field. Just as interesting were the instances where these events and activities were completely absent, because they did not always result in a negative student experience (as might be anticipated). These four panes provide the framework of the chapter. However, there is a caveat attached: there are times when the ‘view’ from one pane may seep into another, thus blurring the boundaries between each of them. This is the nature of qualitative research in general, and of my interview data in particular – it resists attempts to be boxed into a single pane. This is to be anticipated, given that human narratives – as provided by my interviewees - are themselves impossible to limit to set parameters such as those developed by Bogden and Biklen (2007: 176). While I have tried, as far as possible, to prevent this seepage, where it has occurred I have allowed the data to dictate the way forward, thereby maintaining its integrity – these instances have been highlighted in the text. 228 7.2 Setting 7.2.1 Context The literature on postgraduate supervision identifies two broad ‘models’ of practice (for instance Delamont et al 1997; Johnson 1995; Neuman 2007; Conrad & Phillips 1995): firstly, the Science model, which is characterized by students undertaking their research in laboratories, often working with a team of researchers on different aspects of the same project, and having access to multiple sources of support and feedback through this teamwork structure. In this model, a single supervisor is rarely assigned to each student, but team co-supervision or supervision by committee is the norm (Edwards 2002; Becher 1994; Becher & Trowler 2001). Furthermore, Brew (2002) argues that learning in this context …comes not from explicit instruction but from a process of ‘enculturation’ during which newcomers (i.e. PhD students) learn the socialized skills of laboratory work, and through which research problems are transmitted. The process is one to which various participants contribute over time, so that the continuity is from the process, not from the individual participants who come and go (Brew 2002: 141). The second model of practice in postgraduate supervision is the Humanities model in which the student generally works in isolation (from both her supervisor and her peers), and whose sole source of support is her supervisor. There is no laboratory-like structure in which peers can interact with one another, share problems or brainstorm solutions. As a result, alienation of the student from both her research and her peers is recognized as being one of the key features of this model (see for instance Manathunga 2005 and Whittle 1992). Co-supervision is rare (although not unheard of) in the Humanities model, seemingly as a matter of tradition. The experiences of the participants in my study did not adhere to this ostensible binary of the ‘pure’ Science or the ‘pure’ Humanities model: while some of the Science candidates interviewed were part of a bigger research project, led by their supervisors, the majority did not find themselves in this situation. However, co-supervision – even at the nominal, ‘unofficial’ level – was the norm for this Science cohort. Among the Humanities students interviewed, and in contrast to the traditional Science/Humanities Faculty binary described above, the practice of co-supervision seems to be increasing. 229 The numbers from this study are telling: eight of the Science candidates reported having single supervisors (although they all reported having made use of additional supervisory resources beyond their university-appointed supervisors), while five Humanities students related that they had had two or more supervisors. What this suggests is that these models – which may be more accurately understood as non- categorical modes of supervision, rather than as absolute types – are organic in that they can be used and adapted in either Faculty, depending on the nature of the research involved. While it might provide a generalized framework in which to attempt an understanding of what kind of supervision is provided in which context (that is, Science or Humanities), it can nonetheless not be held as an absolute descriptor because of the current nature of knowledge and research. In recent years, there has been an increase in the multi- and inter-disciplinary nature of knowledge and its creation – in other words, the boundaries between disciplines have become progressively more blurred to the point that it is not unusual for ‘hybrids’ to exist, combining such diverse disciplines as law, medicine, anthropology and chemistry, for instance. Inevitably, perhaps, this has led to a corresponding move away from a two-model system, towards one in which there are a number of variations. This became clear as I analysed the data from my interviews, and a third, non-specific49 49 By ‘non-specific’, I mean that it does not have any identifying characteristics besides the fact that it is neither ‘pure’ Humanities nor ‘pure’ Science. In other words, in any given context, this model may have some of the characteristics of one (or both) of the pure models, but these have been offset by the unique aspects of each individual case, such as the cross-disciplinary nature of the research being conducted, or the narrowness of the research topic, or any number of other variables that may exist in each. As this chapter will demonstrate, such permutations often serve to strengthen the PhD candidate’s experience of supervision because they do not conform to a ‘one size fits all’ model, but rather recognize the uniqueness of each supervision context. ‘model’ emerged: those who, for instance, despite having been assigned a supervisor (or supervisors) by the university, had made contact with additional experts in the field, and had consulted extensively with these individuals during the course of their research. Interestingly, these ‘unofficial supervisors’ were often of more value to the students than their university-appointed supervisors. Dumisani, a Molecular Biologist, for instance, actively sought out the help of colleagues in Australasia and in the Cape for the purpose of augmenting the input (such as it was) from his formal, university- appointed supervisor. His criterion for choosing these individuals was the ‘credibility 230 of the person’ in the specific field of research. In other words, he actively sought help from academics whose knowledge and reputation would enhance his research, and who would ‘fill in the gaps’ left by his official supervisor. In contrast to Dumisani’s self-directed search for outsiders to provide him with additional supervision, it was Zodwa’s supervisor, Professor Bellarosa, who would frequently assemble internationally renowned experts in the field of fine arts, drama and music for Zodwa to discuss her study with, to brainstorm with and to provide additional feedback on her work. The professor also made a point of sharing aspects of her student’s written work with other experts in order to solicit further feedback and insights. Far from being isolated and virtually dependent on the input of a single supervisor for the development of her research, Zodwa became an active participant, engaging with a community of practice much broader than simply her and her supervisor. She was given access to a range of experts in her field, all of whom became sounding boards for her research, thus allowing her to enrich her work beyond the confines of single-supervisor input – as would have been the case had her supervisor adhered to the ‘pure’ Humanities model of supervision. Thus, in this specific case, even though she did have the (institutionally sanctioned) single supervisor required by this model, she also had the perspectives of additional experts at her disposal, both in terms of face-to-face contact and electronic communication. These exercises effectively translated into her induction into a specific research community of practice (elaborated on later in this chapter). Wenger’s description of the mutual engagement required between the individual (Zodwa) and her specific community of practice (local experts in the field of fine arts) for this induction is an apt one: Mutual engagement involves not only our competence, but also the competence of others. It draws on what we do and what we know, as well as on our ability to connect meaningfully to what we don’t do and what we don’t know – that is, to the contributions and knowledge of others (Wenger 1998: 76). For Zodwa, It’s good, I mean, when you’re presenting your work to internationally known [academics] and they make such good comments about your work – it really makes me feel good. It really made me realize I’m going somewhere. 231 Thus, it would seem that this third model – of informal co-supervision – was not just used by students to supplement the university-sanctioned supervision that they were already receiving (especially if this was deemed to be inadequate by the student). Rather, it also seemed to have been employed by supervisors themselves in order to augment what they were able to provide to their PhD students in terms of intellectual and professional support. While these informal relationships may not strictly be defined as ‘supervision’, nonetheless the input that they provide to students represents a powerful and valuable contribution to their growth as researchers and academics. The informal co-supervision model therefore seems to rather incongruously meet a specific need between both parties for advancing knowledge, by virtue of its generality. Put another way, because it does not dictate what practices ‘ought to belong’ to a particular Faculty or discipline, it allows for the recognition of a range of such practices that are unique to each individual supervision context. As research becomes more inter- disciplinary, and as knowledge becomes more socially (as opposed to individually) constructed, I suspect that the model of informal co-supervision will become more prevalent. This is especially possible in the Sciences, where much of the knowledge base increasingly resides in industry, rather than in academia. The growth of the international knowledge economy – discussed in section 1.4.3 - and the enormous amounts of money attached to technological and scientific advances in fields such as medicine and information technology has resulted in many multinational corporations setting up research and development laboratories in situ for these purposes. Consequently, supervision as an academic practice may no longer be confined to the academy, but may be spread into industry where access to the informal supervision discussed above is likely to be situated. In summary, the models of supervision experienced by the doctoral candidates that I interviewed were not always the ‘pure’ versions generally associated with their respective disciplines. There emerged an interesting array of adaptations, seemingly as a combined result of increasingly blurred inter-disciplinary knowledge boundaries, and candidates who actively sought out supplementary guidance beyond that of their ‘official’ supervisor/s. What this suggests is that the traditional practices of supervision in each Faculty have been fundamentally eroded and replaced with knowledge-driven 232 practices. That is, the ‘kind’ of knowledge being developed determined the mode of supervision provided. 7.2.2 Master’s experience As mentioned in the previous section, the majority of the participants in this study (21) had undertaken their Master’s research at other universities before registering at Wits for their PhDs. For some students, such as Nomsa, Nkosinathi and Bridget, the change of institution was due in part to the difficulties that they had experienced with their supervisors at their previous institutions. Nomsa, a vivacious young Biology PhD candidate, described how she had experienced her Master’s supervisor’s indifference to both her work and to the development of a productive, professional relationship with her: Obviously, his door was open, but then the attention wasn’t really there, the type of attention that says ‘Do you understand what I’m saying? Would you like me to go over it?’ I knew very little about his comings and goings. So it was more of a lecturer-student relationship than a supervisor-postgraduate student. Far from being a ‘needy’ student, Nomsa’s expressed desire for intellectual attention from her MSc supervisor stemmed from her desire for professional interaction with him. In other words, she wanted to engage with him at a scientist-to-scientist level, working together in the laboratory, sharing knowledge and making discoveries together. Her natural curiosity about the area of biology was neither encouraged nor nurtured by this particular supervisor. Thus, her (unspoken) expectations of her Master’s supervisor were never met, resulting in her experiencing a degree of detachment from her research. Furthermore, in terms of her development as a researcher-scientist, her growth was severely constrained by his indifference, and thus a valuable learning opportunity was missed. What was also conspicuous by its absence was the sense of a personal relationship between supervisor and student, in which the former showed interest in Nomsa’s life outside her academic work. By contrast, Professor MacDonald (her PhD supervisor at Wits) made a point of ‘checking in’ with her regularly because: 233 ...he realizes that that area [personal life] really needs to be fixed first, so he helps whenever he can to make sure that you’re comfortable [so that] you’ll be able to pay attention to your research. It was clear from her description of her Master’s experience that it had been a disappointment for her, at the academic, professional and personal levels. Fortunately, the reverse was true of her doctoral candidature. For Nkosinathi, an African Literature PhD candidate, the Master’s experience was characterized by distressing behaviour on the part of his supervisor. This was manifested in a number of different ways. In the first place, the supervisor – Dr Agatha Miller - had been prone to frequent emotional outbursts during their supervision sessions. Nkosinathi explained: My supervisor was going through some horrible time, I could see – personal problems, you know. It was like some kind of a trauma that she was going through, so sometimes she would shout at me, she would be horrible – more than once – and apologise afterwards – more than once. By his own admission, Nkosinathi has a great respect for authority (which was clearly how he had constructed this supervisor), and was thus reluctant to ‘go to the higher echelons’ to remedy the situation. Understandably, he was anxious to complete the degree and ‘get it over with’. Any complaint against his supervisor could have delayed the process, and so he chose to suffer through her outbursts. However, he encountered a further supervisor-related obstacle later on in the process, in which his supervisor seemed to have deliberately undermined his work, instead of offering him timeous and accurate guidance. As Nkosinathi tells it: I’m very stimulated in terms of wanting to access theory and I had ventured into something that I hadn’t ventured into before – theoretically, that is – and I was trying to make it part of my Master’s process, but [the supervisor’s examination] report showed that she knew from the start that [this] theoretical take had some problems. She only said that in the report. She didn’t tell me that while I was writing my Master’s, so you can imagine how that feels. All along you’ve been told that ‘No, you’re okay’, and when her report comes, it says ‘No, because you have mixed two theoretical traditions that are not really going together’. 234 Despite these obstacles, Nkosinathi managed to graduate with his Master’s ‘because of my tenacity’. However, it took him several years before he returned to academia and registered for his PhD. Bridget is a PhD candidate in Speech Therapy, who completed her Master’s degree at an American university. Her experience of the supervision process there was not a happy one, and she confessed to feeling ‘apprehensive’ about undertaking a PhD – even if it was in a different context to her Master’s. Firstly, her supervisor in the USA had plagiarised her research data and used it in a paper that she had presented overseas – without Bridget’s knowledge or permission. The supervisor never acknowledged the source of the data. Like Nkosinathi, Bridget chose not to confront the supervisor about this, focusing instead on completing the degree. However, she had to fight to do this as well: [The supervisor] wanted me to continue for another six months after [I was due to] graduate [with my Master’s degree]. I had already done a therapy model when most of my peers had only done surveys – she then wanted it to be hand analysed and computer analysed to check the validity across the two. I petitioned and said there was reverse discrimination going on and I went to my Head of Department and the Dean and said ‘You will sign me off [in order to graduate] – I have adequately and then some met the requirements for a Master’s thesis, and I will not be [held back by my supervisor]. It was clear that this experience had coloured Bridget’s perception of the supervision process, and she was understandably concerned that there might be a ‘repeat performance’ of this kind of behaviour during her PhD. This use of her data by the supervisor represented a very specific kind of misuse of Bridget. Even thought this was not verbal, as experienced by Nkosinathi, it nonetheless represented a deviation from the accepted norms of ethical supervision. It also represented a poor example of the kinds of practices – tacit though they may be in many contexts - which researchers must adhere to. In other words, supervisors are supposed to model good research behaviour for their students – this includes, at the most fundamental of levels, the practice of academic honesty and ethical behaviour towards their students. Obviously, this would also apply in their own research. Bridget’s MA supervisor proved to be a 235 very poor role model. Fortunately, her PhD supervisor at Wits has directed her through the research process with integrity. The Master’s degree experiences of some of the other PhD candidates in this data set, such as Shanti and Emily from the Faculty of Science, and Thabang from the Humanities, were by contrast much more favourable. In all three examples, the students credited the supportive relationship that they had had with their supervisors as a reason why they had decided to undertake doctoral studies. Shanti completed her doctorate in mid-2008. She described Professor MacDonald (who supervised both her Master’s and her PhD) as being ‘hugely supportive’. The two developed an ‘easy relationship’ during the course of her research, which was most clearly evident in the informality of many of their interactions: We didn’t really have any set meetings – it wasn’t like we had to diarise ‘I’m going to meet you every month’ – but because the labs are right here and his office is right down the corridor, it would be an up and down kind of thing… [A]s soon as you get some interesting results, then you would rush off and show him what’s happening, and then he’d come to the lab and [see]. In this case, the relationship was bolstered by the fact that Professor MacDonald was also appointed by the University as Shanti’s mentor in an initiative to support and encourage young, mainly black, academic talent. Shanti felt that the two roles – supervisor and mentor – were more complementary than contradictory: It was fine. I suppose you do relate on a slightly different level because you start talking about other things aside from just science, so we got to know each other very well. Emily’s story followed a similar trajectory to Shanti’s. She had been supervised by the same person – Professor Henry Lambson – over the course of her Honour’s, Master’s and PhD studies. She reflected that the reason why she had remained under his tutelage for this extended period of time was that his approach to supervision was developmental and growth-oriented, rather than ‘baby-sitting’. In Emily’s own words: He’s a great supervisor and he’s always been there for me, he’s always helped me. If I struggle in the lab or I don’t understand, he’s always helped me out. You know, sometimes … my research goes up and down and doesn’t work as 236 nicely as they should but I always discuss it with him and he always has the time to sit down with me and go through the research when I get to that point. But he doesn’t baby-sit – he makes you think about it and only once you come up with plans [does] he help you out because he wants to see what you’re thinking and then if you’re really struggling, he’s there for you. The theme of her supervisor encouraging and developing her skills threaded throughout the interview. Emily was adamant that it was Prof Lambson’s commitment to helping her to think and problem-solve for herself that provided the biggest learning curve for her as a young scientist: He does make you think. It’s hard to, because a lot of people say that no one can teach you how to think, but I’ve learnt… to think. Obviously, I had to have my own thinking, but he’s taught me how to think and, I mean, I’m much better when I was in Master’s than I was in Honours, and I see the difference now that I’m in PhD. Now I can think much better on my feet, and I can analyse things better. What was interesting about Emily’s description of her combined postgraduate experience was the fact that, in reflecting on her experience across all the years of her postgraduate study, she was able to identify how she had developed intellectually at each level of study. She was able to pinpoint the growth in her thinking and analytical skills, and attributed this growth to her supervisor’s input. While she did acknowledge that she had come to the academy as someone who was able to think and reason for herself (which is arguably true of all postgraduate students), she was nonetheless aware that these skills did not develop in a vacuum, but rather were honed by an external source – her supervisor. One could speculate that the ability to reflect on her own growth as a researcher was an implicit skill that she picked up from Professor Lambson – it is certainly not something that can be actively taught. Instead, the Professor seemed to have chosen an indirect method of challenging Emily’s thinking: by asking her probing questions about various aspects of her research, he was effectively modeling for her how a doctoral student should critically engage with her research (as well as with any emergent problems thrown out by this research). In so doing, he developed her questioning and reflective skills without explicitly teaching them to her. 237 Thabang, a Swazi International Politics PhD candidate in his mid-thirties, shared many of the supervision experiences that Shanti and Emily described. What was unique in his case, though, was the almost deified esteem in which he held his supervisor, Professor Jack Baldwin. This was made clear right from the beginning of the interview: I have found him to be super in many ways. Number one, what I have learnt about my supervisor is that he is a knowledgeable guy, he’s brilliant, he’s a hard worker … I’m never in the dark, to say the least. Then, as far as the presentations, exposure, stuff like that [is concerned], I think he’s beautiful… I like the fact that with him, if [he] says to me ‘This paper is good’, I know it will stand anywhere in the world… If he says it’s good, then you know it’s good. So, back then when he was still only a [PhD], we used to refer to him as ‘professor. There’s a lot of professors out there that are not half as good as he is. So that just goes to indicate to you how we hold him academically – he’s a really revered man among those students who interact with him. Thabang returned to the theme of the ‘revered supervisor’ later on in the interview, too. Interestingly, this was in response to my comment that many students select their supervisors based on the supervisor’s access to resources such as money and networks. While Professor Baldwin had facilitated a financial aid scholarship for Thabang, his response to my comment suggested that his choice of supervisor had nothing to do with research funds – or networks: I didn’t know about the money thing at first, though that was not really a factor in choosing the supervisor. For me, I have this tendency to rate my instructors – I’ve done that since Honours: ‘This guy is not that good, this guy is super’… I do that a lot. So, he came and he just blew me away. I said ‘This guy, he is super… this guy is educated’, and since then, I thought ‘This is the guy who should supervise me’. Professor Baldwin had also supervised Thabang’s Master’s degree, and had undertaken the kinds of activities that are described as being ‘good practise’ in the literature on postgraduate supervision (cf. Section 3.2). For instance, he had presented papers with Thabang, co-taught with him at the government’s training school for politicians and diplomats, encouraged him to take opportunities for lecturing and tutoring within the School so as to develop his academic profile, tasked him with writing funding proposals for a School computer laboratory, and so on. Another project that Professor 238 Baldwin got Thabang involved in was a university-wide study on institutional culture that the former had been asked by the university’s management to lead. While this study had had nothing to do with the content of Thabang’s Master’s and later, PhD, research (or, indeed, anything to do with International Politics), it had taught him a great deal about common research techniques – from choosing the most appropriate technique, to applying it, and finally, to coding and writing up the data that emerged from it. By being involved in this project, Thabang was exposed to the research process from a different perspective, which allowed him the opportunity to observe the mechanics of the process from a non-discipline-specific angle. I suspect that this was Professor Baldwin’s intention – what better way to teach the ‘nuts and bolts’ of research than by getting your postgraduate student to engage in a low stakes study outside his own research area? What was common in each of these individual cases was the fact that the candidates had had the same supervisor for at least two of their postgraduate qualifications. This raises the questions: does continuity in supervision from Honours to Master’s to PhD guarantee a positive research experience at PhD level? In other words, does a prior supervisory relationship at Honours and Masters’ level necessarily translate into a positive PhD association? If the student has a bad experience at the Honours or Master’s level, and if continuity of supervision across the years of postgraduate study becomes an established norm in higher education, will these students be likely (or willing) to undertake PhD research? In other words, while having the same supervisor for two or more of their postgraduate degrees seems to have facilitated high levels of growth in these three students, can these positive (and highly desirable) outcomes be anticipated in all cases and across the board? I suspect that the answer to the last question is a resounding ‘No’, for a number of reasons. Firstly, and perhaps most obviously, each supervisory relationship is unique. No two are exactly alike, even with the same supervisor in charge. By implication, where an ongoing relationship between Honours, Masters and PhD degrees might work in some cases, it may well be untenable in others. Secondly, locking students and supervisors into such extended relationships could become academically incestuous, and lead to a watering down of the student’s 239 development as a researcher. In other words, the two parties could become so accustomed to one another’s ways of working that they stop exploring, questioning, criticizing, arguing. While this was not the case in with any of my research subjects, Faith, a PhD candidate in the School of African Literature and Languages, expressed her concern about it succinctly: I believe in a variety of opinions… and that’s the benefit of working with a different person at PhD level from the person you worked with at Masters level, because then you learn other ways of doing things, and then there’s a way in which, having worked with this same person from Master’s to PhD – somehow, the person gets familiar with your style of working and sometimes, they even forgive stuff that other people wouldn’t forgive. So I think that it helps to work with two different people, purely from an academic [perspective] – it definitely helps you to have other views. Faith’s point that a supervisory relationship over a succession of degrees could become too comfortable, thus allowing a potential lapse in standards, is a valid one. It raises the possibility of an ‘unspoken shorthand’ developing between supervisor and student in which important issues are swept under the carpet or not sufficiently unpacked on the assumption that ‘my student knows me well enough by now to know that when I say x, I mean y…’. These unspoken understandings may, in fact, result in misunderstandings. No such instance of this assumed awareness emerged in my data, and hence I have no evidence to support this claim. Nonetheless, it is a possibility that needs to be considered in supervision relationships that extend over more than one postgraduate qualification. One way around this problem would be to appoint a different co- supervisor in each degree (if the principal supervisor were to remain the same) to offset any challenges caused by an extended supervisor-student relationship. Whether or not such a solution would ultimately work is hard to predict, given the uniqueness of each relationship. A third reason why the scenario of the ‘continuous supervisor’ may be undesirable is that could set up a set of false expectations between student and supervisor. Students, for instance, may expect to get the same levels of support from their supervisors at PhD level as they did at the Master’s level, even though the two degrees make different contributions to existing research and knowledge production. The PhD is expected to 240 be an original piece of work that makes a significant contribution to knowledge and understanding in the field (Heathcott 2005; Park 2007; Leonard 2001; Mouton 2006; Wisker 2005; Walker & Golde et al 2008; Adkins 2009; Usher 2002). By contrast, the Honours and Masters degrees require students to work within existing knowledge frameworks, and no original contribution per se is expected. By the same token, supervisors may expect that any of the students whom they have supervised through their Master’s degree ‘automatically’ understand that PhD research is of a more advanced nature and therefore may not make the specific requirements explicit. A few candidates in this study had upgraded their Master’s degree to a PhD50 50 A Master’s degree can be upgraded to a PhD when certain criteria are met by the student. According to rule G12.2. of the General Rules of the University: a) A person who has been admitted as a candidate for the award of master may, in exceptional circumstances, at his/her request and on the recommendation of the supervisor and of the Head of the School concerned on the basis of work towards the dissertation be allowed, by permission of the Senate, to proceed instead to the award of Doctor of Philosophy. Provided further that the award of master shall NOT be conferred on him/her in the event of his/her – i) withdrawing his/her candidature for the award of Doctor of Philosophy; or ii) having his/her candidature for the award of Doctor of Philosophy in terms of G5.7; or , which resulted in a somewhat different experience from those who had done the degrees sequentially. As a general rule, students undertake an upgrade when the quality and breadth of their Master’s research make it eligible to be advanced to a higher level. The research may, for instance, have started out as a relatively small-scale study, but the results obtained indicate that it would make a much greater contribution to knowledge than originally anticipated. In such cases, it is fitting that the degree be re-registered and examined at the doctoral level. Only three students in my study had taken this iii) failing to satisfy the requirements for the award of Doctor of Philosophy. b) A person who has completed the requirements for the award of master, at his/her request and on the recommendation of the Head of School concerned, may be permitted by the Senate not to have the award conferred on him/her, but to prosecute, for not less than one academic year of further full-time study, or not less than two academic years of further part-time study, additional research for the award of Doctor of Philosophy, which shall be a significant extension of the research already completed by him/her: Provided that the period of additional research may be waived or reduced in a case considered by the Senate to be exceptional. Provided further that the award of master shall be conferred on him/her in the event of his/her- i) withdrawing his/her candidature for the award of Doctor of Philosophy; or ii) having his/her candidature for the award of Doctor of Philosophy cancelled in terms of G5.7; or iii) failing to satisfy the requirements for the award of Doctor of Philosophy. The rulebook also contains a caveat in terms of which national imperatives are taken account of: ‘Conditions for conversion may change in light of the Higher Education Qualifications Framework’. General Rules, p. 17. 241 route. Such a small number is not unusual, because the scope of a Master’s degree is relatively straightforward to define – it is only in rare cases that the scope expands beyond that of the initial Master’s proposal. Marcus, a political activist and lecturer in the School of Social Sciences, was one candidate who had managed to effect an upgrade. Interestingly, however, he did not raise this as having been particularly advantageous to him because it had been an inordinately lengthy process: ‘I was on the thing in one way or another from about [19]97 until 2007’. That said, he nonetheless described his relationship with his supervisor throughout this decade-long research journey in positive terms: I was very lucky to have him in the sense that, first, he knew the area, and my interests are often quite obscure so I was very, very lucky to get a guy like him… And [second], he also read up a lot on the stuff, and even his own interest changed. So, he knew the literature, he did the extra stuff, he gave feedback very rapidly, … he disagreed and he was very open-minded – I criticized him a lot in my thesis, I criticized his arguments, his articles, and he didn’t penalize me for that. The issue of the power relations within the supervisory relationship will be examined in more depth in Chapter 8. However, in this particular example, it appeared that the duration (and success) of the relationship was attributable to the positive balance of power between the two men that had existed right from the start of their relationship – in other words, the supervisor’s power had not diluted with every year that he supervised Marcus. It had simply never been a factor. It appeared that the absence of a significant power differential between supervisor and student stemmed from a mutual devotion to the advancement of knowledge through the twin processes of self- and supervisor-led critique. Further evidence of this was to be found in the absence of a judgmental, punitive aspect to Professor McNeil’s supervision style. However, upgrading his Master’s degree meant extending the period of time that he would be supervised by Professor McNeil – this is a variation of the ‘continuous supervision’ theme discussed earlier. In this case, while the continuity had had its benefits – in terms of strengthening the relationship between the two – it had also had a detrimental effect on Marcus. The extended length of time-to-degree had hampered 242 opportunities for career advancement in his School, since promotions are typically awarded to individuals who already possess a PhD, not those who are ‘in process.’ Marcus was an exception to the norm, in that very few Humanities students upgrade their Master’s degrees. This may be due to the nature of the knowledge being developed in the Faculty51 This particular experience raised the issue of whether doctoral students choose their own supervisor/s – based on their prior experience and knowledge of these individuals – or whether supervisors are departmentally assigned according to topic or area of research interest. While this was not a question that I posed to my interviewees, it might nonetheless be relevant to investigate whether or not doctoral processes supervised by individuals of the student’s choice are more likely to be successful than those supervised by departmentally-appointed supervisors. In other words, when doctoral students can control the process of deciding with whom they would prefer to work at this high level, does it make the process smoother and more comfortable for the student? Or is the source of supervisor selection irrelevant to the process? The question of a candidate’s ability to choose a supervisor is arguably embedded in the kinds of . The highest numbers of postgraduate degree upgrades at Wits come from the Science Faculty. Two of the Science candidates in this study had taken this route: Bill and Seamus. Bill is an energetic man in his mid twenties who is currently completing his PhD in Earth Sciences. During our interview, he did not focus on the reason for his upgrade – rather, he gave the impression that it was a logical step in his development as a researcher, and hence he only mentioned it in passing. He was much more specific, though, about the role that his principal supervisor, Professor Douglas Oxbridge, had played in his academic career: So, Doug was always my original supervisor… I’ve been working with him since my Honours and third year. I kind of decided that he was someone who I wanted to learn from, so I did my Honours project with him and then I decided to do a Master’s [with him]. 51 According to Becher (1994), the nature of knowledge differs between broad disciplines. For instance, in the ‘pure sciences’ (such as physics), knowledge is generally ‘Cumulative; atomistic…; concerned with universals, quantities, simplification; resulting in discovery/explanation’ (p. 154). By contrast, Humanities subjects such as history and ‘pure social sciences’ such as anthropology, understand the nature of knowledge to be ‘Reiterative; holistic…; concerned with particulars, qualities, complication; resulting in understanding/interpretation’ (p. 154). Marcus’ PhD conformed to this model in that it was largely interpretive and narrative in nature, and contained no discoveries or explanations of phenomena. 243 social and power relations that reside in their particular School. The role played by power will be unpacked in more depth in section 4.1 of the next chapter. Professor Oxbridge’s support for Bill extended beyond the immediate job description of a supervisor; while he had still been registered as a Master’s candidate, Bill found out about a conference that would be taking place in Australia, the substance of which was central to his research. However, this conference was ‘…specifically supposed to be for PhDs and post-docs’. Professor Oxbridge felt strongly that Bill would benefit from attending it, and so … He was actually quite proactive in getting a hold of the woman who was organizing it, whom he knows personally, and insisted that I was good and that I was going to do a PhD and that I should go. On numerous other occasions, Bill’s supervisor not only encouraged, but also supported financially, various forays to conferences, workshops and laboratory practices around the globe. As a result, Bill reflected that … I feel like I’m valued, you know? He lets me do a bit more of what I want to do and I think he’s got the balance right. In contrast to Marcus, Bill’s Master’s – PhD upgrade is in its fifth and final year. This relatively rapid progression arguably has more to do with the subject matter being researched than the student himself. Bill’s research is cutting edge. For it to remain cutting edge, though, he needed to have it completed and published within as short a period of time as possible, or face being upstaged by another researcher. Since Marcus’ research had a strong historical orientation, there was very little danger of his work being forestalled by others doing research in the same area. Therefore, from these two distinctly different cases, it would appear that the nature of the knowledge being developed in the doctoral research – be it historical (and hence unlikely to be revolutionary or groundbreaking) or scientific (with an emphasis on innovation, evolution and advancement) – prescribes in part, at least, the time-to-degree. For Marcus, the historical borders of his research were never going to change, and hence it was permissible for his research journey to have taken a decade to complete. Bill’s research, on the other hand, is (literally) groundbreaking, and hence there was a heightened sense of urgency in getting it completed and published. These two cases 244 represent, I believe, two ends of a continuum. They are artificial from the perspective that neither Marcus’ nor Bill’s time to completion may be considered to be characteristic of the norm of students who upgrade their degrees. Seamus, an intense young Chemical Science PhD candidate, seemed to occupy a middle ground between the two. Figure 6: Time-to-completion continuum (n=minimum number of years to completion) Bill’s experience with a supervisor who recognized the centrality of the student’s interests in the PhD research process fore-grounded another possible typology for attempting to understand the experiences of doctoral students, which I will call the x- centred approach, with x representing the key driver(s) of the research. In this approach, supervision could be categorized as being student-centred, research-centred or institutionally centred (see Figure 3, below). Bill and Marcus’ experiences fall within the student-centred approach – the research topic was chosen and directed by the student, with some input from the supervisor. Thus, the interest in the research topic lay intrinsically with the student. By contrast, a research-driven approach to supervision would be one characterized by the student being the passive recipient of a research topic – usually one generated by a larger project initiated and directed by the supervisor. In this scenario, the student’s interest in the topic may be more strategic than in the previous approach. In other words, the student’s primary goal is the degree, and whether she is interested in the topic or not is peripheral to this goal. Finally, the institutionally-centred approach to supervision puts the university’s needs at the top of Rapid n = 1 or 2 Protracted e.g. Bill e.g. Seamus e.g. Marcus 245 the agenda: it has been an overt practice for several years to recruit students to disciplines associated with the ‘scarce skills’52 http://www.skillsportal.co.za/download_files/professional/NSDS- Scarce_Skills_List_2007.doc sets identified by (amongst others) national government, industry and organizations like the National Research Foundation (NRF). According to the National Scarce Skills List 2007, these include engineers, scientists, medical and ICT professionals – located in the Faculty of Science at Wits (National Scarce Skills List 2007, retrieved 31 May 2010, from ). Figure 7: A typology of the x-centred approach to supervision. Note how all three possible typologies are contained by both the national and international contexts. 52 ‘Scarce’ or ‘critical’ skills are defined as: ‘an absolute or relative demand; current or in future; for skilled; qualified and experienced people to fill particular roles/professions, occupations or specializations in the labour market’ (Retrieved on 31 May 2010, from http://www.skillsportal.co.za/skills_guide/scarce-skills.htm). Research- centred approach Institutional- centred approach Student-centred approach National Internat - ional Science Humanities 246 In contrast to Bill’s rather laid-back persona, Seamus was an angst-ridden young man who had recently completed his upgraded PhD. He was very critical of the supervision that he received during the period of his research because it was ‘hands-off’. His supervisor, Dr ‘Monty” Montgomery, did not actively track Seamus’ progress (or more often, the lack thereof), but rather … ha[d] this bad habit of just dragging on and on and on, and [he was] having me do one thing and it wasn’t working and he just kept on: “Well, just try it again, just try it again, just try it again” as opposed to “Well, okay, let’s think about this and find a way around the problem”. In comparison to both Marcus and Bill, Seamus seemed never to have developed ownership over his work, but tended to rely on the advice of his supervisor before he took action. In other words, he would not – or could not – go forward in the research without the say-so of Dr Montgomery. He chose instead to wait for the supervisor to come to him first to provide direction, and when this did not happen, he became disillusioned. His dissatisfaction was evident throughout the interview, as he raged against the ‘system’ that he perceived to be preventing him from being supervised and developed as he clearly wanted to be. The quotation below is an edited compilation of the general tone of how he had experienced the supervision process: I felt that the supervision was lacking. I understand that during the PhD, they do push you and you’re supposed to be doing the research on your own, but I still think you’re paying for good supervision, and… in my opinion, a lot of the problem is that the supervisors are just bogged down with administration and so they never actually get into the lab. And another thing that I disagree with [is that] they … keep you there longer than you really should be there, and in my opinion, it’s a lack of project supervision. I was there for a year and a half longer than I should have been. And it was just because I was wandering around on my own trying to get the stuff to work – and I always resent them for this. Early on in my work, I spent months and months doing stuff I knew wasn’t going to work but they just said “Keep doing it, keep doing it”, and… they’re not guiding you to an end. And my opinion is when you are doing a PhD and you’re getting into the final year, they should be saying “Okay, what have you done? And let’s put down some goals on the table that you should finish.” They should be telling you “You need to finish this, this and that” in order to write up a document that’s of high enough quality to get your PhD, as opposed to just letting it drag on and on. 247 Evidently, Seamus was unable to reconcile his actual experience of supervision with his expectations of it. Furthermore, instead of saving time by upgrading his Master’s to a PhD, it seemed that the process took longer than he had originally anticipated – and yet he still managed to complete it. This suggests that the general context of the supervision in the department, as well as the specific practices of his supervisor, were sufficient to ensure that Seamus was awarded the doctorate even if he did not agree with the approach used. This served to disillusion him even more. A thought that occurred to me as I was transcribing his interview was: if Seamus had not upgraded his degree, would he have completed his Masters and then looked for a job in industry, or would he perhaps have registered for a PhD at another University? This is a significant point because it deals directly with the issue of student expectations versus actual experience, and what impact the correlation between these two factors has on postgraduate students. If – like Seamus – a PhD candidate53 To conclude this section of the chapter, it is useful to pull together the themes that have emerged from it. The first of these relates to the experiences of supervision that PhD candidates had had at the Master’s level: generally, if the experience had been problematic, then the student was unlikely to remain with the same supervisor for her doctorate. By comparison, students whose Master’s supervisors had challenged them to expects the ‘Rolls Royce’ treatment, consisting of extensive supervisor attention coupled with large doses of affirmation, (co)-publication and collaboration and does not receive it, then the consequences for supervisor, student and institution may be dire. These may include deregistration and non-completion – two situations that universities in South Africa are trying to avoid because, amongst other things, they represent a loss of government subsidy income. On the other hand, if the candidate possesses a strong sense of ownership over her research, the opposite scenario could be true. Thus, the expectations of doctoral students need to be explicitly dealt with from the start of their candidature in order to ensure a productive outcome. 7.3 Conclusion 53 This is arguably also true of Honours and Masters candidates, but to a lesser extent. Since the contribution to knowledge at these two levels is somewhat less than at PhD level, the ‘Rolls Royce’ treatment is not likely to be applied by supervisors in these contexts. 248 develop their thinking and problem-solving skills were more likely to continue with these supervisors during their doctoral research. They clearly valued the developmental approach taken by their supervisors, and wanted more of the same. There was a caveat attached to this, however: that a long-term relationship has the potential to become too comfortable for both parties, which in turn opens up spaces for unspoken (and hence untested) assumptions to be made; these could replace rigorous academic debate and feedback between supervisor and student. Another finding in this section related to the various models of supervision that existed within my interview sample. Aside from the ‘traditional’ Humanities and Science models, I noted that there was some ‘cross pollination’ or hybridization between the two, for instance, Humanities doctoral students who had more than the (customary) single supervisor, and Science candidates working with a single supervisor in non- laboratory settings. What was even more significant was evidence of an alternative model in which students proactively identified and made contact with experts in their field at an unofficial level – that is, individuals who were not recognized as formal, university-appointed sources of supervision for the doctoral students. In this model, the students went beyond the acknowledged expertise of their ‘official’ supervisors to find additional authorities in the field whose input was used to add depth and value to their research. What this suggests is that expertise in any discipline does not necessarily reside solely with individuals within higher education institutions, but may be sourced from other organizations too. Finally, the issue of upgrading degrees from Master’s to PhD level was discussed, and found to be both unusual and complex. In the following chapter, the remaining three windowpanes – process, relationships and activities/events - will be used to analyse my interview data. Read together, they will provide vivid insights into doctoral candidates’ experiences of supervision at Wits. 249 CHAPTER 8 ANALYSIS Through the looking glass: the journey, fellow travelers and milestones en route Chapter overview: The process, relationships, and activities and events windowpanes are developed in this chapter. Each reflects a particular aspect of the students’ lived experiences of their doctoral supervision. Taken together, they provide a panoramic view of these experiences, and offer the reader some important insights into the way that candidates make sense of their supervision. 8.1 Introduction This chapter presents an analysis of the data through the panes of process (how the research journey unfolds), relationships (the manner in which supervisor/s and students interact with each other during the course of the journey) and activities and events (opportunities for both parties to attend conferences together, co-publish and interact at a social level). Each of these panes offers insight into specific elements of doctoral supervision that can either inspire or demotivate the candidates, depending on how it is applied. As mentioned in the previous chapter, the boundaries of each of these panes are not absolute, and there are often ‘spillovers’ between them. In the pages that follow, this occurs on a number of occasions, highlighting how difficult it is to separate out the various levels of supervision. 8.2 The process For the purposes of this study, the term ‘process’ is used to describe the activities undertaken by student and supervisor during the research journey, as well as the artefacts (such as feedback) that emerged from these activities. I have excluded conference posters and research articles from this section – although they could be considered to be artefacts – because I felt that they fit better in the ‘events and activities’ window pane (discussed in Chapter 7). This is one instance in which the ‘leakage’ between panes has occurred. I decided not to include posters and articles in this section because I understand them to be measurable by-products emerging from the 250 findings of the research (although not always as a result of the engagement between student and supervisor) rather than tools used by both parties to advance the research . The research process is iterative in nature – that is, it contains ‘loops’ in which the student returns repeatedly to specific parts of her research to adapt, update or remove them as necessary (Badenhorst 2008: 50). Since the research path itself is rarely a straight one, these iterations may be anticipated. Their role is to provide the student (or novice researcher) with opportunities to fine-tune the data and its analysis. 8.2.1 Feedback The role of feedback is essential to the progress of PhD research (see for instance Brinko 1993; Hunt 2001; Hyatt 2005; Wisker 2005; Kumar & Stracke 2007; Wang & Li 2009). Whether this feedback is formal or informal, written or oral, constructive or not, it can nonetheless impede or facilitate the flow of the research. In this study, the majority of participants received a combination of written and oral feedback from their supervisors. The written comments were in all cases discussed and clarified with the students during formal supervision meetings. Zodwa’s experience of receiving feedback from her supervisor stood out for me: her supervisor, Professor Bellarosa, would often ..organise, just out of the blue, … her husband [also an expert in the discipline] and other friends, just people to sit and then I’d talk about my study… [Their feedback] made me realize [that] I’m going somewhere. Professor Bellarosa also cast around beyond this group of people to source additional experts who were willing to provide feedback on Zodwa’s work (always with the student’s permission). This had the double impact of exposing Zodwa to the core of her discipline’s community of practice, as well as getting her research out into this community. This practice of inducting doctoral student/s into the community of practice (see section 2.2.5) – for the purpose of eliciting feedback – represents a powerful opportunity for the student to access the widest range of experts who could, in turn, offer a depth and breadth of feedback that would be impossible for a single supervisor. 251 In only one of the cases – that of Alphons (African Literature) – was feedback solely oral. In this instance, the student in question was an independent, self-motivated individual who knew exactly what he wanted from his supervisors in terms of feedback, and oral communication was sufficient for him – there did not appear to be any desire on his part for written communication on his research work. His three supervisors would offer their feedback during the course of regular supervision sessions; nothing was ever given to him in writing. He enjoyed this verbal engagement with his supervisors, and described how he particularly valued the opportunity that these sessions (and the supervisors) gave him to be selective about what feedback he chose to use and what to discard. In some ways, this could be understood as an opportunity for him to negotiate the meaning of his data with his supervision team – whose meaning ultimately prevailed depending on his evaluation of the alternatives presented. He describes the process as follows: They would allow me to take what I want [from the feedback]. They’ll not impose anything on me. They’ll simply make suggestions … and at the same time, I’m figuring out ‘Okay, which one is the best that I want to take from here?’ So from there, I simply take hold of what it is that I want to take. However, neither the process of providing feedback to the students, nor the quality and timing of that feedback was identical in any of the cases in the study. In the case of the former, it took supervisors anywhere from a few hours to several weeks to return their students’ work. At one end of this continuum was Kenneth, a Paleo-Archeologist, who would literally receive feedback from one of his supervisors within hours of submitting it electronically. One particular example stood out in his mind: I normally come in very early, and Dr Lin was in early. [I] gave it [a chapter] to her at a quarter past seven; quarter past eight I had it back in my hands with comments and suggestions and it had been read properly. You could see she had read it – right down to a comma here, change the phrasing here etc – not only the content itself, but the grammar [too]. By contrast, Ace, a young Zimbabwean student registered for a PhD in Statistical Science had to ‘expect delays’. Since his supervisor was also the Academic Head of the Statistical Science department, her schedule was monstrously overloaded and thus for Ace, ‘... where you are expecting [feedback] tomorrow, it may actually come a month 252 after’. Situated more or less in the middle of this continuum - between an extremely rapid turnaround time of mere hours, to a ploddingly slow one of over a month - was Seamus, who was only months away from graduating with a Chemistry PhD when I interviewed him: I did my introduction, gave it to him, he got it back to me in a couple of days, and then I handed [in] the rest of the body of it, which was about three hundred plus pages, and he gave it back to me fully corrected in three days. This continuum is represented in Figure 4, below: Figure 8: Feedback response time continuum The quality of the feedback also seemed to be ranged along a continuum, from poor to superb. Seamus, for instance, did not feel that the feedback which came from his supervisor, Dr Montgomery, was particularly encouraging or motivating: I found that he only knows how to criticize – he can never say ‘Oh, well done! That’s good’. He said ‘Well done’ when I finished, but that’s obligatory. Dumisani also lamented what he considered to be poor quality feedback from his supervisor, Dr Mswati, although his conception of quality differed from Seamus’: You’ll give her the chapter, and instead of looking at the chapter … perhaps you forgot to check one word, she would just mark that word only and say ‘Go back and redo it’, without even checking the rest of the stuff. Kenneth Ace Rapid turnaround Delayed feedback time for feedback (more than a month) 253 Another example of feedback whose consistency was considered suspect by the student was related by Hannah, a fifty-something PhD candidate who had resigned from her corporate-sector job in order to work on her PhD full-time: He asked me to change things, and then three versions later, he asked me to change them back, and it depended on his whim and what he had for breakfast. Rebecca, on the other hand, described supervisor Professor Peterson’s feedback on her written work as ‘…the most pithy, extensive and subtle’ commentary that she had ever received. She reported that it was … always about pushing you to think in a more complex way about your own assumptions as you are operating in the field. And so, as difficult as it can be when you have to confront that, it pushes you to the next level. This comment echoed what Emily had had to say about her supervisor – that he challenged her to think for herself in such a way that she was able to develop these particular skills for herself over the course of her postgraduate studies. Both Emily and Rebecca seemed to relish the challenge of being ‘pushed to the next level’, arguably because they wanted to grow beyond just ‘knowing’ the content of their respective fields – they wanted to know how to think about their fields as well, so that they could continue to produce valuable research. This interest in self-development at a meta- cognitive level is arguably at the core of ‘good supervision’, although it may seem to be controversial. What these two particular experiences demonstrate is that the supervisors went beyond the immediate scope of the thesis to promote the development of researchers who could continue to contribute to their respective field after the completion of their PhD theses. The supervisors appear to recognize that the research endeavour does not end with a thesis54 54 This is an interesting point to ponder, given the government’s stated desire to play a more substantial role in the international knowledge economy. Should the National Department of Higher Education, as well as universities and bodies such as the NRF, not be training supervisors to look beyond the immediate windfall of the thesis being awarded, to the continuation of the research? Is this an opportunity to more actively promote post-doctoral research as a means of emphasizing the fact that productive research can continue after the thesis has been submitted? , but should rather continue far beyond it. Clearly, one way of ensuring this is by providing the doctoral candidate with the kinds of feedback that promote ongoing questioning, investigating and experimenting that continue long after the thesis has been examined and the degree awarded. This line of 254 reasoning echoes the observations made by Kumar and Stracke (2005), who suggest that in the supervisory relationship, .. feedback on written drafts is a form of communication, as it is through written feedback that the supervisor communicates and provides advanced academic training, particularly in writing, to the supervisee. It is through this feedback that the supervisee is able to understand that writing is a form of learning, as revising drafts after feedback can lead to a process of discovery, an integral part of PhD education. It is also the feedback that allows the supervisee to engage in critical thinking and writing; thus, communicate ideas. (p. 462’ italics added) Figure 5 shows the feedback quality continuum, which ranges from high quality and developmental at one end, to superficial on the other. Figure 9: Feedback quality continuum Despite this description of how feedback ought to be given and received, it does not take account of the issue of the power relations (to be discussed in section 8.4.1 of this chapter) that is often inherent in any supervisor-student interaction. For instance, it does not acknowledge that feedback could be used by both the supervisor and the student as a platform for dissent. Put another way, either party could potentially use feedback as a means of challenging the argument made or evidence presented by the other. Later on in the same article, however, Kumar and Stracke note that The supervisory relationship can propagate a power relationship in which one is the master and the other the learner. In such a circumstance, feedback is considered a form of teaching, where the traditional master-apprentice relationship is evoked (Kumar & Stracke 2007: 462). In summary, what was evident from the interviews was that feedback from supervisors (both the official and informal) was only useful when it contained substance and depth. Rebecca Dumisani High quality, developmental Superficial, inadequate 255 Although most of the candidates had received a combination of written and oral feedback, one type was not favoured over the other – as long as it led to progress, the students were satisfied. Overall, a composite of the ideal type of feedback seemed to consist of thorough, timely and constructive comments that students were not compelled to use if they did not agree with them. 8.3 Activities and events In this windowpane of analysis, the focus was on the specific enterprises that doctoral students engaged in along their research journey. While the majority of these revolved around the research itself – conference attendance, the (co)publication of articles, teaching/tutoring in the individual schools – I was also interested in any social events in which students and supervisors may have participated. As a number of the candidates noted, there was more to their lives than just their research. They shared experiences of social interactions that they had with their supervisors, such as attending braais55 The literature on postgraduate supervision practices places great emphasis on the centrality of exposing postgraduates to the culture of conferencing and publication as a fundamental part of their academic professional development (see, for instance, Sorcinelli & Austin 1992; Gaff & Lambert 1996; LaPidus 1997; Austin 2002 and Charlesworth et al 2007). However, this is more dominant in the pure Sciences rather than in the Humanities models of supervision. As the boundaries between the two models of supervision in these faculties begin to blur, however, so too do the range of practices regarding conferences and co-publication. These two activities are aimed at inducting the doctoral student into a specific research community of practice – and sports events together. While there is no evidence that these interactions altered the dynamic of the supervisory relationship in any significant way, the students who had spent time with their supervisors outside formal study hours reflected on how these events had cemented their relationship. In the sections follow, I will illustrate the impact that conference attendance, co-publication and social interactions had on the experiences of my interview subjects. 8.3.1 Conferences, co-publication and communities of practice 55 A braai is the commonly used Afrikaans word for a barbeque. 256 discussed in greater depth in Chapters 2 and 3. Essentially, doctoral students begin their candidature on the periphery of their research community of practice. Through their participation in conference and publications, and with some mediation by their supervisors (both official and unofficial) they begin to move closer to the ‘core’. Lave and Wenger (1991) remark on the inevitability of this movement, which they describe using the term ‘legitimate peripheral participation’: …the mastery of knowledge and skill requires newcomers to move toward full participation in the sociocultural practices of a community. ‘Legitimate peripheral participation’ provides a way to speak about the relations between newcomers and old-timers, and about activities, identities, artifacts, and communities of knowledge and practice (Lave & Wenger 2003: 29)56 56 For the purposes of this thesis, ‘old-timers’ are represented by the supervisors while the term ‘newcomers’ refers to the doctoral candidates. This is not unproblematic because it is a gross generalization. Nonetheless, I think that it is a useful framework on which to ‘hang’ the various supervisory practices that are discussed in this chapter. It is discussed in more depth in the literature review. Furthermore, these authors point out that ‘[p]eripherality suggests that there are multiple, varied, more- or less-engaged and –inclusive ways of being located in the fields of participation defined by a community’ (p. 36). My interviewees occupied a range of locations (or peripheralities) relative to their individual communities. For instance, Zodwa was located relatively close to the core of her community of (research) practice thanks to the active interventions of her supervisor in arranging get-togethers with renowned experts in her field. By contrast, Günther felt that he remained on the outer periphery because his supervisor, for example, never shared information about upcoming conferences with him. In an ideal supervisory relationship, the ‘old- timers’/supervisors would engage with their ‘newcomer’/students in the social learning contexts of conferences and opportunities for co-publication. This, in turn, would allow the student to become more engaged and included within their community of practice, adding to existing knowledge in it. As Lave and Wenger point out: [Peripherality] is also a dynamic concept. In this sense, … when it is enabled, suggests an opening, a way of gaining access to sources for understanding through growing involvement (Lave & Wenger 2003: 37) 257 In the data collected in this study, there was a clear absence of co-conferencing and co- publication in about a third of the cases. At this point, it is important to make a distinction between conferences which students attended and at which they presented papers on their own, on the one hand, and those at which they were accompanied by their supervisors, on the other. The former, while a useful professional experience for PhD candidates in terms of honing their presentation skills, nonetheless lacked the specific modeling behaviours that occur when students witness their supervisors – established researchers – in action. Thus, for instance, they do not get the benefit of observing how researchers network with one another, what kinds of questions they ask each other, how they handle disagreements about findings, and so on. Although they may witness the interactions between a range of researchers attending the conference, it is their supervisor57 We have co-authored papers together and he says to me ‘You are the first author, you are the owner of this paper, you must tell me what you want from who ‘unpacks’ for them the meta-processes associated with the specific skill of academic networking. Put another way, it is the supervisor who elaborates on the ‘art’ of making contacts instead of simply assuming that students already (inherently, perhaps) possess this skill. Therefore, it may be the case that it is not a good enough supervisory practice simply to fund conference attendance for one’s PhD candidates, because they may not gain the meta-level insights that will advance their academic careers. While both activities (co-conferencing and co-publication) are considered to be an integral part of the pure science model of postgraduate supervision, they were present amongst some Humanities candidates in my study, specifically those who belonged to a research collective within a Department or School. For instance, Ntombi, a Zimbabwean candidate in her early forties whose PhD topic was part of a collaborative research programme in the Department of Scientific Education, described her conference and publication experience with supervisor Anton Gould as follows: 57 This is admittedly a generalisation – not all supervisors possess these particular social skills. I would argue that promoting networking skills is a learned behaviour – that is, if the supervisor’s experience of being supervised had included being actively inducted into academic networks, then it is more likely that she will initiate such practices in her supervision of postgraduate students too. 258 this paper’… [and] he takes me to conferences. [H]e has actually sometimes introduced me to people and said ‘You know what? This one is the big authority in this, this, this that you’ve been talking about’, so to me, that says he admits that there are people who are bigger authorities in those areas than him, and he’s quite happy to introduce me. In fact … I’ve met almost all the people whose theoretical frameworks I’m using. This case was the exception, and not the norm in the Humanities. It represented an excellent example of an ‘old-timer’ initiating the (relative) ‘newcomer’ into the community of practice of scientific education and educators. By contrast, Nkosinathi - who was part of an inter-disciplinary research entity in the Humanities Faculty - seemed to shy away from attending conferences (either with or without his supervisors, who could be seen to be complicit in keeping him on the periphery of his community of practice). He rationalised: I have not gone to as many conferences as other people have gone to and I think that the reason is conferences are not so relevant to what I’m doing. And the other thing also … it’s a question of time because I can be so absorbed in my way of working [that] I don’t want to do other things. Interestingly, this contradicted a statement that he had made earlier in the interview, in which he described an ideal scenario that would suit his particular needs: Interpersonal interactions [at conferences] tend to be more beneficial because you get to talk about a problem there and then, and there’s that exchange, and I think that if you have something like that, say, twice a year, it would be very helpful because then it gives students a chance to, say, work on two chapters or work on one chapter which is going to be presented at the [conference]; in the next one, work on the other chapters, and exchange ideas, and [get] to know the level of research, where the research is, what are the new things that are coming into play? Because sometimes you lose out, sometimes you think you’ve researched enough only to find that there’s a body of literature that you don’t have access to, another body of thought that other people are dealing with which may be relevant to what you are doing. This ambiguity about how he constructed the role of conferences for his development as a professional researcher was interesting for a number of reasons, not least of which was that no mention was made of either of his supervisors joining him. Given that it was his nature to defer to authority (discussed in the previous chapter), I wondered 259 whether he felt that he was not ‘good enough’ to co-present a conference paper with either or both of his supervisors. If so, this did not reflect well on their supervisory practice since neither of them seemed to have disabused him of his self-constructed submissiveness. Furthermore, his seeming lack of a firm stance on conference attendance suggested that he was undecided about their potential value for his research. This was most evident in the contradiction between his asserting that conferences were not relevant to his work, on the one hand, and acknowledging that they were valuable platforms for presenting new knowledge in his discipline, on the other. Another element that stood out was how removed Nkosinathi appeared to be from the collective work of his research entity – in the intellectual, not physical, sense. While all six of his colleagues (consisting of both doctoral and post-doctoral candidates) in the entity were working on various facets of a core issue, there seemed to be little sharing of knowledge between them, and certainly no effort to collaborate on conference papers. This seemed to defeat one of the main purposes of the entity, which was to co- generate research. Thus, in this particular instance, it appeared that (potential) conference opportunities existed, but that Nkosinathi chose not to take these up, either on his own, with his peers in the research entity, or with his supervisors. Xolani, a Zimbabwean man in his mid-forties, was also a part of a research entity, although it was not as well established as Nkosinathi’s. His experience of conference attendance and co-publication with his supervisor, Professor Pieterse, was very different to Nkosinathi’s. Xolani described how the professor inducted him into the Brighton conference, an annual conference whose focus was in his area of education – thus unmistakably initiating him into this particular community of practice: I am now a member of the Brighton community because of his initiative. The first two Brighton conferences I attended with him and we made a joint presentation. Thereafter, he made sure that he facilitated my attendance to all the other Brighton conferences, every year… and at one point I became a committee member of the Brighton [conference] for two years. So, I have become networked with the Brighton people right up to now. Having co-presented a paper with Xolani at the first conference, Professor Pieterse gradually withdrew, possibly because he wanted his student to become more 260 independent, and possibly for the apprentice to become the master. The networks that Xolani accessed as a result of this initial joint presentation remained with him through his PhD and beyond. A variation on the pure Science model of working with one’s supervisor to both present and publish research was evident from my interviews: students who took the initiative to attend conferences and get their work published without much support from their supervisors. Hannah and Patricia were two such candidates, both of them in the Humanities. Hannah recalls I … made a point of writing a number of papers and making sure that I had papers out to journals and that kind of thing. But that’s my own experience – having been an academic – that it’s always important to have a few papers in the pipeline. So I did that, and it also helps you to get to grips with the literature you’re dealing with [but] it wasn’t something that particularly came from my supervisor. In fact, he was quite startled when I said ‘Oh, I’ve produced this paper and I’ve sent that paper off…’ She was one of the older, more independent PhD students amongst my cohort of interviewees, who had many years of work experience under her belt – including a number as an academic. Hence, it was not surprising that she had taken the initiative to get involved in publishing and presenting. What was especially remarkable about Hannah’s case was how differently her supervisor – like Xolani, she was supervised by Professor Pieterse – engaged with her regarding conference attendance and academic publication in comparison to Xolani. She was a part of the same research project as her peer, and was as competent an academic. Where the two diverged, however, was in their research areas of interest: Xolani’s topic revolved around the field of educational policy, while Hannah’s was focused on ‘models, pedagogies and student experiences’ of doctoral supervision practice. The most obvious conclusion that can be drawn from Professor Pieterse’s differing practices towards these two candidates is that his own research interest coincided more closely with Xolani’s than with Hannah’s. If this is the case, however, then the practice cannot go unchallenged: firstly, since both were Professor Pieterse’s students, it seems discriminatory that one student was favoured over the other in terms of their development in the spheres of conferences and publications. While it may be counter-argued that Hannah had come from an academic 261 background and hence knew the currency attached to publications, it is not a sufficient explanation – Xolani himself had worked at an educational non-governmental organization (NGO) and thus also knew the importance of publications. A second challenge to Professor Pieterse’s uneven treatment of these two candidates is that the role of the supervisor in, for instance, writing a paper with the student is not necessarily about the content of the paper, but rather about modeling effective research writing behaviours for the students.58 Further along the continuum of supervisor engagement in co-publication activities with the candidate is Patricia, a young mother who commutes between her home in Stellenbosch and Wits when she needs to confer with her supervisor, Professor Kerry Pinker. Their geographical separation is one of the reasons why the two do not collaborate on conference presentations or journal articles, but more significantly, Patricia’s research topic is outside Professor Pinker’s specific area of professional research expertise. While Patricia recognizes this divergence of interests, she Like Hannah, Xolani was not new to the contexts of publishing and presenting, and had a firm grasp on how to approach both. For these reasons, the seeming favouring of one candidate over the other bothered me. The source of my concern lay in the impact that it could have had on the two students, both in terms of their relationship with each other (and since they were both a part of the same project, this needed to be considered), and with Professor Pieterse as their supervisor. Since I did not mention to either of the two candidates that I was also interviewing one of their peers, I could not ask them how they experienced this situation, and neither raised the issue on their own. However, I would assert that publishing and attending conferences with one and not the other – whatever the specific background of each might be – is poor supervisory practice since it suggests that the work of one is more important or more valued than that of the other. This, in turn, could have a potentially damaging impact on the latter, with all the negative consequences that this brings with it. There is also the possibility that such unevenness could affect the future academic career of this candidate. 58 This was evident in the descriptions of co-publication of those who had written papers with their supervisors: in all cases, the students put their own name as first author, with the supervisor’s coming afterwards (or not at all, if this omission was warranted). 262 nonetheless feels torn about their lack of collaboration, as the following quote illustrates: We did a conference presentation. I submitted the abstract and then I submitted the paper with [both our names], and she took off [her name] and I presented it and I only realized it when I got to the conference and it was in the abstract book… and she had taken her name off it because it isn’t her project, and I was in two minds about it. I thought ‘Doesn’t she want her name on my paper? Or is it because she feels… like it hasn’t been hers?’ The dilemma that Patricia alluded to in the final sentence of this passage encapsulates the insecurities that many PhD candidates experience when it comes to publishing with their supervisors. In this particular case, however, it was offset by the fact that Professor Pinker attended several other conferences and workshops with Patricia – her presence here indicating to her student that she was invested in her research interests even if they did not necessarily merge with hers. What I also found interesting about Patricia’s responses to my questions about conferences and publications was that she was quite satisfied to have some attention from Professor Pinker in these areas, and was unperturbed that more support from the supervisor had not been forthcoming. It seemed, therefore, that the ‘Goldilocks and the three bears’ formula applied here – that she got neither too much nor too little attention from her supervisor regarding conferences and writing, but rather just the right amount. From the other interviews, I would argue that this formula – of what was considered to be ‘just right’ – would differ from student to student, depending in factors such as personality, motivation and willingness to embrace the challenges of writing and presenting. Figure 10: Levels of supervisor engagement with the doctoral student Prof Pieterse (with Xolani) Prof Pieterse (with Hannah) Very engaged Disengaged 263 In the context of the Humanities, a number of interviewees revealed that their research topic was not the same area of expertise of their supervisor’s, hence the absence of conference attendance and co-publication with their supervisors. Despite this acknowledged lack of common ground between them, these supervisors had been chosen by the students for other, possibly strategic, reasons. Rebecca, for instance, describes her experience as follows: My supervisor is in another division [and] has nothing to do, generally speaking, with the content area that I’m looking at. However, he is probably the most astute and subtle thinker in the School … and has a very good profile internationally for engaging with contemporary theory and philosophy… and I was fortunate that he had agreed to supervise me on that basis. Rebecca has, as a result, attended conferences and written articles and book chapters independently of her supervisor, but with his full support. This arrangement suited them both, and Rebecca has published book chapters and conference proceedings on her own. She has also attended a number of international conferences and Summer School programmes. Other students in the Humanities, however, have not had the opportunity to publish with their supervisors for a range of other reasons. Mojalefa’s experience is a prime example of this. He is several years older than most of the other students in this study, and had worked at a national research think-tank for four years before registering for his PhD in the Faculty of Humanities. His research topic straddled two countries, South African and Belgium, and he had a supervisor in each. His South African supervisor, Professor Liebowitz, had been intellectually remote from the beginning of their relationship, and had not provided a platform for the two of them to collaborate. When it came to conferences, she was even more remote – during the period of his PhD registration, Mojalefa had attended (and presented) at five international conferences, all of them funded by his European co-supervisor. At all five, he presented and published alone – no input from Prof Liebowitz was forthcoming. He argued that this was because he never actively sought her input, or invited her to work with him. His blunt statement ‘I did not write anything with her...’ seemed to express no regret about this state of affairs. Even institutional, non-research co-publications – such as the review of 264 a research entity to which they both belonged – never materialized. Mojalefa’s first realization that the review had been written and published by Professor Liebowitz on her own came when a copy was delivered to his office. In the Science Faculty, co-publication, poster presentation and conference attendance were generally a more common occurrence than in the Humanities. The biggest area of potential dispute in this Faculty revolved around the authorship of posters and papers. However, this minefield seemed to have been successfully navigated by all of my interviewees and their supervisors. Below are some of the students’ descriptions of their collaborative experiences with their supervisors. What is most notable is the similarity between the descriptions, even though they were describing different supervisors: I will write it [the paper] and then he’ll give me feedback. I kind of add him to everything that I do unless he doesn’t contribute much, I have to admit. [A]nd when he really hasn’t contributed, he says that he really doesn’t think that he should be on the paper. … I think that he would prefer to feel that if his name’s on the paper, that he’s contributed in some way. So sometimes, all he’s really contributed is a few ideas over a cup of coffee, but I usually put him on because … he’s provided the space for me to learn stuff. Usually, his ideas are really good – the few that he contributes [laughs] – and then sometimes, he puts a whole lot more in. But either way, I always put his name on. But he never wants to be first because I always do most of the work (Lee, Ecological Science PhD candidate; italics added) [T]hat’s generally the case always in geosciences – I’ve noticed always that the student is the first author and the supervisors are second and third authors unless it’s a case where the student has handed in a Master’s or something and then that’s the end of it, and then if the supervisor has to write up a paper afterwards, then they’ll put themselves as first author. But if the student writes the paper, the student is the first author (Bill, Geological Science PhD candidate; italics added) [A]ll his students, they are first authors. Even on the posters, we are always first authors, and he puts his name on [but] he never takes [full credit] even though he does so much. (Emily, Biological Sciences PhD candidate) Karabo, an Advanced Chemical Sciences doctoral candidate, had a supervisor who took this issue of intellectual property a step further: he helped her to patent an 265 invention that had evolved out of her research. She seemed puzzled when I described how certain supervisors would put their names first on research papers, even when they had not contributed much to them: I wonder why? Because, I mean, basically you are the one who is doing some work. I mean, the student is the one who is working… For Bruno, the issue of anyone but him being first author was a non-starter because …it’s my problem, I’m doing it … doing all the work. I’ve only really actually written two papers… and it’s my name on the articles. It’s not co-authored and Prof Stone and Prof Shriva haven’t had a problem with that because they realize … it’s my work. I mean, I’ve acknowledged them at the end of the article… It was interesting to note how jealously Bruno guarded his work – his use of the pronoun ‘my’ to describe the problem, the name on the article and the work itself attests to that. That he did not even consider including either of his supervisors on the articles – in any position of authorship – served to highlight his complete ownership of the work. The acknowledgement of the supervisors ‘at the end’ suggested that this was merely an afterthought on his part. Thus, conferencing and publishing with one’s supervisor could be equal part minefield, equal part godsend, depending on the supervisor. On the one hand, working with one’s supervisor in these contexts could provide the doctoral candidate with a model of the behaviours that are appropriate in each – a ‘watch and learn’ model, so to speak. On the other hand, in situations dominated by the supervisor, these contexts also open themselves up to intellectual abuse – although, fortunately, none of my interviewees experienced this. Furthermore, not all of the interviewees felt that the supervisor was needed at all when it came to conferencing and publishing – these students successfully asserted both their right and ability to undertake both activities independently, with or without encouragement from their supervisors. While not large in number (six out of thirty, or twenty percent), the candidates who did this nonetheless represent a significant trend in the research. 266 8.3.2 Social interactions While I did not raise the issue of social (rather than purely academic) relationships between the doctoral candidates and their supervisors in the interviews, it was nonetheless something that a number of them raised by themselves. For some, such as Karabo, Emily, Xolani and Thabang, social interactions played an important part in cementing the supervisor-student relationship. Others, like Günther and Mojalefa, felt aggrieved by the absence of a social element to their relationship with their supervisors. In their own way, each of the students expressed a need to engage with their supervisors at a personal level because it seemed to somewhat reconfigure any negative balance of power that might exist in the academic relationship. For Günther and Mojalefa, there was a clear desire for a more holistic human relationship with their respective supervisors, in which there were opportunities for relating to one another in a different context and with a different purpose than that expected from a supervisor- student relationship. Thabang’s experience epitomizes this holistic supervisor student relationship – he describes his social contact with Professor Baldwin as follows: I feel like he cares about [me], and I’ve been to his home, and I know his children and I play with them sometimes. They come and we play tennis here with his kids, and they attend football games at Wits when Wits is playing… Thabang subsequently described these interactions as like being part of a ‘family’. Professor Baldwin had also met Thabang’s mother at his Master’s graduation. Both student and supervisor clearly valued these connections, but the former also realized where the boundary lay: ‘When it comes to work, he’s my boss’. Emily echoed this limit: Like, we are friendly and we go out for drinks sometimes – like, the whole lab – so we do have a friendly interaction, but at the same time, I know he’s my supervisor… and I’ll never disrespect or cross that line. Karabo described a similar, laboratory-wide social interaction, and, like Thabang, compared it with being a member of a family: Because most of the time we have activities – we go out as a team with our supervisors, we go out and have fun. [W]e go to one of my supervisor’s place, have a braai, play together, so it’s like we are a family. 267 This use of the family metaphor was interesting because of what it implies about the supervision relationship. Metaphors, according to Deshler (1985), are ‘… one way that human beings have of sorting out perceptions, making evaluations, expressing feelings and reflecting on purposes. It is the stuff with which we make sense of our world’ (1985: 22). In the postgraduate supervision environment, the use of these family metaphors could be interpreted to mean that some doctoral candidates perceive their relationship with their supervisors as an emotional and protective connection – much like those that they experience with their biological family. This corresponds to the German term for supervisor – ‘Doktorvater’ or ‘Doktormutter’ (doctor father or doctor mother), which Aiken (2009: 1) describes as ‘…part teacher, part parent over the course of time’ which encapsulates the idea of the supervision relationship as one akin to that of a family, and the unconditional love and support that this implies. However, I am cautious about presenting the family metaphor of supervision as an unconditionally positive representation of the relationship. In some families, for instance, there exists an unhealthy dependency between child and parent, in which the former will only undertake a task with the complete support of the latter. In other words, the child will do nothing without the approval of the parent. Furthermore, the child may perceive that the parent’s validation of their work eclipses their own sense of the work’s worth. The application of this interpretation of the family metaphor to the supervision relationship highlights the potential for doctoral candidates to become intellectually dependent on the supervisor, possibly even paralysed from making progress in the absence of the supervisor’s (fatherly or motherly) guidance or approval. In the cases mentioned on the previous page, the negative connotation of this metaphor does not seem to apply; however, that is not to say that the dependency danger is not lurking at the periphery of the relationship. By contrast, Marcus’ relationship with Professor McNeil represented a healthier approach to supervision, one in which dependency and emotional connections were subordinate to the elements of challenge and inquiry. A more mature social relationship also existed between Xolani and Professor Pieterse, arguably because they belonged to a similar age group, and they had also worked together in a professional setting: 268 We would go for lunch together. We would go for tea together. For instance, he would just call me and tell me ‘You know what? I am tired and I just want to take a break and go for lunch. Would you like to come with me?’. And we would go and have lunch or tea together, and he did that right from the beginning… He [also got] to know my family. For instance, he would invite me sometimes to his house and I’d go there. When my wife visited me [from Zimbabwe], I would go and introduce her to him, so he actually got to know my family [and] I got to know his family and, you know, those kinds of things do help a lot .. in bridging the [power] gap between the student and the supervisor. For Kristelle and Mojalefa, by contrast, the absence of social contact with their supervisors made them feel alienated and isolated from these individuals. As Mojalefa describes it: Well, it is something that has, actually, bothered me throughout doing my PhD with her – I have always felt that our interaction was just too… impersonal, it’s just academic, you know? We talk about my work, I leave the office, but that’s it – there’s no interaction and I feel that this has been the biggest weakness in our interaction… I like being independent and being able to think alone, but … it’s also good to have a kind of cosy relationship with your supervisor. I mean, we have never even had tea together, or a drink together. I don’t know her house. I don’t know anything about her. I’m not saying we should be close friends or whatever, but at least there should be a level beyond merely being a student and supervisor. Mojalefa’s relationship with his Belgian supervisor in Minsk had been the opposite of this, which is perhaps where his expectation of a social relationship began. For him, it seemed that the supervisory relationship was about more than intellectual engagement. It required a social aspect in order to add value to the experience. Kristelle, an energetic young Mathematical Sciences PhD candidate who had recently put her degree into abeyance, never had the option of social interactions with her supervisor, Professor Maxwell, because he only held a fifty percent post at the University. In other words, he was only required to spend two and a half days per week at Wits. Most of this time was spent in undergraduate teaching. For the rest, he was engaged in his consultancy business. Therefore, there were no informal chats in the tearoom for them – ‘This was difficult [because] he’s not really in our tea room because he’s here and then there…’. 269 Overall, it seemed that while PhD candidates did not want to be their supervisor’s best friends and had identified clear boundaries, many nonetheless felt encouraged by having the opportunity to interact with the supervisor on a social level. Certainly, in the cases quoted in this section, it seemed that spending time with one another in a non- academic setting offered both supervisors and students the opportunity to get to know each other at a more personal, less hierarchical, level. Put another way, seeing one’s supervisor in a different (social) context effectively puts one on a different footing with her. This is important because it takes what could be a high-stakes relationship and balances it out somewhat so that the students, in particular, have the chance to experience the supervisor as someone other than a detached academic, intent on the research but not necessarily on the researcher. 8.4 Relationships and social structure 8.4.1 Power The presence of some form of power relation was, perhaps not surprisingly, a key element in each of the interview narratives. While I raised the issue in many of the interviews, in others it was the doctoral students themselves who beat me to it. This is significant because it signals a consciousness on their parts about both the presence and the role of power in the supervision process. Power, by definition can be positively or negatively applied by supervisors, and similarly positively or negatively experienced by students (for more on this, see Chapters 2 and 3). During the course of the interviews I undertook, I noted that there seemed to be a continuum along which the power relationship between supervisor and student could be plotted (see Figure 6 below). On the one end lay the all-powerful supervisor, and the inversely disempowered student; on the other, were the students who exerted strong agency over their work, and for whom the supervisor played a largely administrative role. In the following section, I will discuss PhD candidates who represent this wide range of power ‘experiences’ to demonstrate, amongst other things, that this element is not equally positioned in every doctoral supervision experience, even though it does exist to some degree. In addition, it was clear in many of the cases that the balance of power between supervisor and student shifted during the course of the relationship, with the 270 students becoming more empowered as they developed knowledge and expertise in their areas of research. Figure 11: The power continuum. Where candidates situate themselves often depends at what stage of the process they are. At the one end of the continuum was Nkosinathi, a diffident, dreadlocked Humanities candidate undertaking research in an inter-disciplinary field of study. He described himself as an ‘apprentice’ in relation to his two supervisors because … one’s experience is not so much as one’s supervisor’s, but at the same time, there’s always room for conversation. In other words, Nkosinathi was suggesting that he (willingly and knowingly) deferred to his supervisors because they had been in academia longer than he had, and were thus more conversant with what qualifies as ‘good’ research. In addition, they had (initially, in any case) a broader view of the field in which he was studying, and therefore were in a better position to give him the intellectual guidance that he needed. Nkosinathi seemed unperturbed by this imbalance – although I doubt whether he would construct it as an imbalance. Rather, the supervisor as master to the student’s apprentice seemed to Nkosinathi to be the natural and correct state of affairs in the relationship. Arguably, what helped in this particular context was the fact that neither of his supervisors were tyrannical or unreasonable: their commitment was to his development as a researcher, and not to their own elevation. Nkosinathi acknowledged this later on in the interview: I think it goes in stages. For instance, when I started, I would take the models that I’m given so that I’m on the straight. But then at some point, my supervisor would say ‘Now it’s yours to take, to do whatever you want to do’. All-powerful All-powerful supervisor candidate 271 Even though, towards the latter stages of his thesis, his supervisors had effectively turned over full control of the work to him, Nkosinathi did not appear to have embraced this repositioning of power. His mindset remained that of an apprentice. When I prodded him about this, he insisted that ‘…the reason is simply that one’s experience is not so much as ones’ supervisors’ experiences’. Gunther, a lanky Geosciences doctoral candidate from Sweden, seemed to situate the power relationship between his supervisor and himself in the middle of the continuum, although his description of it was somewhat contradictory: One reason that I went to [Professor van Onselen] is because I know what’s in her head and I want it… but she’s also the person who pays me, so in authority, yes, she had the upper hand, but we talk with each other as equals and I consider her a friend as well. At another point in the interview, he expressed his sense of powerlessness when his supervisor failed to give him the time and attention that he felt that he needed. He described this as follows: It’s hard to pin her down, that’s the problem with my professor. She’s very focused [on her own research]. If she’s very focused, it means she’s not paying attention if you’re … in the room. She should know that when I’m there, the attention should be got to me. Clearly, in the latter context, the representation of the two of them as ‘friends’ or ‘equals’ would not be applicable. What I found particularly interesting about this quotation was the almost childish need for attention expressed by the student - similar to a child wanting to be at the centre of its mother’s awareness before it can feel secure. This led me to ponder the connection between security and power in the supervision relationship. One needs the latter in order to achieve the former: security is the sense that one gets when there is (relative) freedom from risk or danger. In order to reach this state of being, one needs power – the capability and authority to influence someone or something else. For Günther, there appeared to be very little security around his research project because the power – understood in this case predominantly in terms of intellectual capacity and access to funding – resided with his supervisor, and he had been unable to access it to his satisfaction. This in turn seemed to lead to his sense of 272 disempowerment (or powerlessness), and from there into a downward spiral of insecurity and self-doubt. Gender – specifically his supervisor’s – could have contributed to the development and maintenance of this downward spiral. Professor van Onselen is a highly respected (female) researcher in the field of Geosciences – itself a generally male-dominated one. That she ‘… had the upper hand’ in the relationship suggested that the gender stereotype of a dominant male supervising a submissive (for want of a better term) female student had been reversed in this particular case. This ‘role reversal’, in turn, appeared to have caught Günther off balance from the start of his doctoral candidature. Megan, a middle-aged PhD candidate in the chemical sciences, also raised the imbalance of power between herself and her supervisor, but in contrast to Günther’s experience, she was not particularly worried or threatened by it: He [the supervisor] does enforce – or try to enforce – a power relationship. I think he’s letting go more and more as time goes by… [H]e’s a definite, authoritative person – that’s his personality. I knew [that] going into it, and I knew it was going to be something I was going to have to deal with. Since Megan had previously worked in industry, and had become accustomed to the absence of ranking boundaries between colleagues in the laboratory there, the ‘hierarchical relationship’ (Megan’s term) that her supervisor imposed on their relationship was initially an uncomfortable fit. However, since she was inherently an independent worker, she managed to take the enforced power imbalance in her stride, without making it into an issue. If anything, she radiated a sense of self-confidence based on her own personal satisfaction with the progress of the research that she was undertaking. In other words, she actively chose not make the power relations between herself and her supervisor a central focus of her PhD. Her supervisor, Professor Zakinsky, seemed to have accepted her independent orientation – she made no mention of being ‘punished’ for it. Indira is a serious young PhD candidate, also in the chemical sciences at Wits. She expressed a similar attitude to Megan’s towards power in her relationship with her 273 supervisor: she recognized its presence, but did not allow it to over-shadow the focus of her research: [W]hen I speak about the research, I feel like he knows what he’s talking about. I also think that with the co[-supervisor], the power differences are more pronounced relative to the actual supervisor, and I attribute that to confidence in one’s knowledge – because my main [supervisor] is confident in his knowledge, he doesn’t have to prove anything. She added the following reflection to the copy of the transcript that I had sent her for approval: If there [are] indeed any power struggles with my main [supervisor], then he does it very elegantly and imperceptibly – I don’t see it. The idea of power being exercised ‘elegantly’ is an interesting one, because it suggests a certain confidence on the part of this supervisor. He did not wield his power – in the form of his vast knowledge and authority - as a blunt instrument, but rather chose a more subtle approach that did not undermine the student. In common with Megan, Indira was a self-assured, focused student whose concentration was firmly on her research and thus, she did not allow the emotion and angst associated with a power play to prevail in the supervisory relationship. The fact that the supervisor facilitated this by moderating any overt displays of power was significant. That this appeared to be his general modus operandi and not just something that he deliberately did for Indira, was even more so. It suggested an innate recognition of the potentially damaging role that power – when used injudiciously – could have on the supervisory relationship, but specifically on the doctoral candidate. At the same time, however, it is possible that the supervisor had taken his cue from Indira: that is, recognizing her independence and intrinsic motivation, he realized that the ‘big stick’ approach to supervision was neither necessary nor appropriate in this particular case. Also at the ‘student empowered’ end of the continuum was Lee, a biological scientist in her late twenties. She dismissed the concept of power having any role whatsoever in her relationship with PhD supervisor Dr Alex Rathbourne. Lee attributed the absence of ‘some silly power dynamic’ to the fact that she recognized the capacity inherent in everyone to learn from other people. In her words: 274 I think human beings are human beings, and everyone’s got something to offer, and the two supervisors that I’ve had have had a huge amount of things to offer me and they’ve been helpful. I knew practically nothing when I first started – I think you can acknowledge all of that stuff without some strange power play. Lee exhibited clear signs of a student who had taken control over her research, and was not inclined to allow power (which she spoke of disdainfully) to get in the way of her research. She admitted that this attitude might have played out in ways that construed her as being disrespectful to more senior academics and practitioners in the field. Nonetheless, she was determined not to be disempowered by the process (or the supervisor, for that matter). The issue of respect as a form of power play did hover at the periphery of her thinking when it came to her engagement with experts in her discipline, though: I actually asked [my supervisor, Dr Rathbourne] the other day – we were at some conference and I started to wonder whether I am respectful enough because I see the way that other people interact with hot-shots in the field, and I realized that I don’t tend to do any of that kow-towing stuff. [S]o I asked his opinion, and he said ja, he didn’t think I was particularly respectful, but he didn’t think that I was disrespectful. Thus, it seemed that respect – specifically, showing respect – was more important in this supervisory relationship than power. The foregrounding of respect by Lee got me wondering: is the ability of a supervisor to command respect – even implicitly – not a form of power itself? Is respect not a signifier (and amplifier) of power? And does the manner in which the supervisor chooses to apply his or her power define the levels of respect within it? In this particular instance, Lee clearly held her supervisor in high regard, mainly because he was a renowned authority in the field and wielded substantial influence in it. She responded to his depth and breadth of knowledge in the discipline, rather than any potential power plays that he might have used (but did not) – for instance, trying to force the direction of her research into an area in which he was interested, insisting that his name went first on their publications, and so on. Ultimately, I realised that power becomes a non-issue in relationships where the student is able to assert her own authority over the research, while at the same time, the supervisor has the wisdom to both recognize and encourage this authority, hence contributing to the development of a confident, independent researcher. 275 Like Lee, Rebecca is also an empowered PhD candidate, although, unlike Lee, she did acknowledge the existence of some power interaction in her relationship with her supervisor, Professor Peterson, which she unpacked as follows: I do have a sense that he recognizes that I am an intellectual, and our values around thinking and intellectual labour and practice are in alignment… so it’s not like there isn’t any kind of power, but for example, when you go and see him, it’s like a therapist’s office – you never sit across the table. He forces you to sit like this [indicates two chairs facing each other with no table in between] and have a conversation like this [indicates face to face with her hands]. Implicit in this description is that even though the two are facing each other and are seated at the same level (suggesting an active attempt on the part of the supervisor to diffuse authority), Professor Peterson is still the ‘therapist’, the individual with the knowledge that can ‘unlock’ Rebecca’s contribution to knowledge. Hence, the power clearly resides with him even if it is mediated by his deliberate physical positioning during the supervisory meetings (which always occur in his office). In keeping with this idea of a ‘therapist’, it is always Rebecca who scheduled meetings with him, and not the other way round – just as a patient, who has decided that she needs professional help, would take the initiative to seek out the support of a therapist. What this suggests about this particular supervisory relationship is that the student is at its centre, and hence it is up to the student to define when supervisory inputs are required. It is not the supervisor’s role to enforce this, therefore implying that any power that may exist in the relationship resides with Rebecca, since it is she who decides when she needs to see him, and she who arranges these meetings. The supervisor, it appears, is simply a facilitator. The period of time over which the PhD is written might also have an impact on how power is wielded by the supervisor. It could be speculated that the equation between these two might logically be: the longer the PhD takes, the more the power of the supervisor diminishes. This is because (arguably) it is the student who has defined the length of time taken to completion, and not the supervisor. While the latter may well refuse to continue supervising the student beyond a certain point, the prevailing practice seems to be for the supervisor to withdraw support incrementally (see, for instance, Delamont, Parry & Atkinson 1998; Sambrook, Stewart & Roberts 2008). For 276 this reason, I was intrigued at the how the length of the supervisory relationship between Marcus and Professor McNeil played out in terms of power and its role, even though the two work together at the university. A decade is a long period of time to be engaged in such an intensive research process – in fact, it is a long period of time to be engaged in anything – and yet the two managed to stay the course without any upheavals in the power relationship between them (such as ultimatums for completion issued by Professor McNeil). The result was a PhD thesis that won numerous international and local awards for its intellectual contribution to the field. Marcus attributed this largely to the complete absence of any power differential between himself and Professor McNeil: [W]hen he’d say ‘This is crap’ or whatever, I’d say ‘No, it isn’t’, and we’d go from there. On reflection, he was generally right on these things, and I find it kind of creepy that a lot of things that I’d get interested in later, I’d go and find that he had written the definitive article ten years before, so in a way he influenced me but without dominating me all the time. Admittedly, personality would have played a vital role in the relationship – that is, how one or both parties perceived a) the power in the relationship; b) the intrinsic value of the research and the process leading up to it; and c) the balance between institutional demands (for more, and timely, PhD completions) and the individual’s need to produce a high-quality result. This could be represented graphically in a pie chart (see Figure 7 below), in which each ‘slice’ corresponds to one of these elements. Although I have represented each slice as being equal in size in this diagram, I recognize that in certain supervisory relationships, some may be larger than others. 277 Figure 12: The different drivers that may impact on the supervisory relationship. Note that the lines separating each are not solid, suggesting their variable nature and magnitude. 8.4.2 Agency To be an agent means to be capable of exerting some degree of control over the social relations in which one is enmeshed, which in turn implies the ability to transform those social relations to some degree (Sewell 1992: 20). The concept of agency is one that I have frequently referred to in this thesis because it has revealed itself in many of the actions of the doctoral candidates that I interviewed. Before I demonstrate exactly where and how, it is important to expound on the concept for the sake of clarity. As the name suggests, agency can be understood as the capacity to act in order to effect change. Cohen, Manion and Morrison contend that humans ‘… do not behave simply, deterministically like puppets’, but rather have ‘individual choice and intention’ (2002: 19). Bandura (2006) concurs: ‘[People] are not simply onlookers of their behavior. They are contributors to their life circumstances, not just products of them’ (2006: 164). Thus, the concept of agency implies that humans can competently and actively plot the direction that they want their lives to follow. Giddens (1984) makes a further contribution to the understanding of agency in his seminal work The Constitution of Society: Perceptions of the Intrinsic value power relationship of the research Institutional demands Individual needs Supervision relationship 278 Agency refers not to the intentions people have in doing things but to their capability of doing those things in the first place (which is why agency implies power: cf. the Oxford English Dictionary definition of an agent, as ‘one who exerts power or produces and effect’). Agency concerns events of which the individual is the perpetrator, in the sense that the individual could, at any phase in a given sequence of conduct, have acted differently (Giddens 1984: 9). Thus, he draws a clear distinction between the intentions of the agent, on the one hand, and her ability to act on these, on the other. In the context of this research, a number of the doctoral candidates that I interviewed exhibited qualities associated with a strong sense of agency – that is, they knew exactly what they wanted to research as well as what direction they wanted this work to take (intention). Furthermore, they were sufficiently confident in their research abilities to assert authority over the process, although they were responsive to supervisory input. Rebecca is a case in point. During our interview, she told me that ‘… by the time I registered, I decided for myself that it really needed to be a process that I had agency over and within, and that I would determine what I needed more of and when’. She subsequently explained: What I need from [Prof Peterson] is not someone to hold my hand and monitor everything I’m doing . I think I’m competent at that level, more than competent, but I do need someone who can sometimes help me troubleshoot because they’ve got a lot of experience around the politics of publishing or how to approach this… and I definitely appreciate having someone who is able to challenge my intellectual horizons, and that is the most important thing that I am looking for from a supervisor at this point because I really do think that I can manage the other things because I know where to go to… Rebecca’s point was essentially that the role of her supervisor was to help her through the peripheral issues related to her PhD, while she directed the substance of the research. Bill asserted a similar authoritative agency over his PhD: I chose my project quite specifically and I have what I feel is complete control over what I’m doing , so I know exactly what I’m doing and I know exactly what I want out of it, and that’s the kind of person I am. So, my supervisors, I use them to guide me but I don’t expect them to give me too much. I feel it’s up to me to do what I want to do, because this is for me. Bruno exhibited a unique version of agency over his doctoral research. He was matter- of-fact about his reasons for undertaking a PhD in the first place: ‘Registering for a 279 PhD was a prerequisite for getting employment at Wits’. Since he wanted a career as a Wits academic and since this entailed undertaking doctoral research, he decided to assert some control over the process by choosing a topic in which he (and not just his supervisors) was interested. It had emerged from a problem that had cropped up in his Master’s research, and he was convinced that it would make an original contribution to knowledge. He was emphatic about his agency over the research: I don’t want to do someone else’s research – I want to do my own. It was my problem originally… I knew the problem I was going to solve, I knew what I wanted to do, I knew it would work… Agency in this context seemed to be more about possession and possessing the research than with any innate excitement about the topic, or about becoming a PhD graduate. The degree was a means to an end – and that end was a permanent academic post at the university. At the other end of the continuum were doctoral candidates who had relied on their supervisor, not only for a topic but also for direction. Seamus was one such candidate. It was obvious from the start of our interview that he felt no real affinity to the work (which was part of a larger project overseen by his supervisor), nor did he really know or seem to care what direction it took. In response to my first question ‘Tell me the story of your supervision’, he described the following scenario: You arrive there, you get told, ‘well, this is the project and this is what we are going to get done and you’ll get a literature pack’ (which is fairly comprehensive). And when it comes down to that, you’re basically on your own and … my supervisor didn’t really approach me to see how the project was going. If I had problems, it was really left up to me to go and bug him about the project This disenchantment was the golden thread that ran throughout the interview. Not once did Seamus express an interest in his research or its findings, nor did he give any indication of having made decisions about the direction in which he wanted to go with the research. On the contrary, he expressed his alienation from the project in utterances such as: Chemistry has this bad habit of just dragging on and on. [T]hey were having me do one thing and it wasn’t working and they just kept on: ‘Well, just try it again, 280 just try it again, just try it again’ and it was obviously just my problem and my fault in the lab, and I was doing something wrong [said in an angry tone]. I would also like the sort of supervision where, when you’re getting to the end, they’ve laid out targets and goals. I’d like a supervisor who laid goals within a time frame: ‘Okay, first six months: at the end of these six months, we59 I just felt that I was hard done by, where it’s just a case of … we started the project, and I kind of felt that the supervisor kind of lost interest, and I must just wander on, just do it myself. It wasn’t a nice experience working there. It really wasn’t. have to have this and this done, and if this is not done, or it hasn’t worked, then we have to go to Plan B’ What was particularly interesting about Seamus’ case was that he had upgraded his Master’s degree to a PhD, and it appeared that his disillusionment had begun during the earlier degree. This is telling because it suggests that his alienation with the School and his research pre-dated his PhD – and yet he chose to continue with both. This made me wonder: is agency over one’s research a necessary and sufficient precondition for completion? For Seamus it clearly was not, but does that make him the exception rather than the rule? Between the two extremes – the candidates’ exercise of full agency over the project on one hand, or their utter detachment from it, on the other – are those who exercised it at strategic points during the research process. Patricia, for instance, was a case in point. She chose not to slot in to one of her supervisor’s existing research projects, but rather to pursue a topic in which she had an intrinsic interest: ‘ I came with an idea of what it was that I wanted to study, and it came from very deep personal experiences…’. Thus, at the start of the process, she asserted her agency over what she wanted to research. Thereafter, however, she allowed herself to be guided by her supervisor: I come to her with my ideas and she uses her very breadth of knowledge in various fields to then suggest a direction to me, and then it’s up to me to take that direction and develop it and come back and then she refines it a bit more… I would suggest that in this case, there is a shared agency over the research: while Prof Pinker, the supervisor, offered directions based on a broad knowledge of the field and how she thought Patricia’s work could best contribute to it, it was Patricia who 59 Note the use of the pronoun ‘we’ – it suggests that the research is not his own, but ‘belongs’ to his supervisor. Thus, it would appear that Seamus experienced no agency at all over his research. 281 ultimately decided which path to follow. This is clearly represented in the following quote: The image that comes to me is like an hourglass, where at the start of the process, she guided me in terms of very, very broad fields and broad, broad reading where she would say ‘Well, perhaps look at sociology’. Now, that took me a year, it took me an entire year to get to grips with sociology, to come back to her to say ‘I don’t believe sociology has a place in what I want to do’ and she would say ‘Absolutely right, I agree with you. Now, you need to move into anthropology’ and then it took me another year to do that, and then I came back to her and I said ‘Anthropology in a way, but I think I want to be more critical than that’ and she said ‘Absolutely right, this is who you need to read’. And so she has steered me from a very broad direction, to then channel me and narrow it down… Thus, the agency over the work seemed to alternate between Prof Pinker (who suggested directions for the research, based on her own knowledge and experience) and Patricia, who explored the suggestions and then selected the ones that she felt were the best ‘fit’ for her research. I have attempted to reflect these different levels of agency in Figure 7 – the continuum ranges from full student control over the research direction at one extreme, and disconnectedness from it at the other. In between, there is a middle ground in which the supervisor and the doctoral candidate negotiate the orientation of the process. As was the case with Patricia, this context is characterized by give and take, discussion and debate. The supervisor and candidate seem to be equally invested in the research. Figure 13: The agency continuum – from student-led research to student detachment Student agency Joint student/supervisor agency Student Detachment 282 8.5 What was missing from the candidates’ supervision narratives? Declarations and insinuations of authoritarianism, paternalism, sexism, ageism or racism were prominent by their absence from the interview data. This posed a dilemma for me, since I had assumed at the start of my research that at least some of these ‘isms’ would crop up during the course of the interviews. After all, the literature on postgraduate supervision is rife with tales of abuse and torment, so I had a reasonable expectation of similar encounters from among my interview cohort. I was therefore intrigued when no unambiguous accusations of supervisory misconduct were forthcoming. Certainly, there were some minor gripes, such as benign neglect by the supervisor (Professor Pieterse in relation to Hannah), continuous and deliberate neglect (of Seamus by Dr Montgomery), disinterest on the part of the supervisor (Dumisani’s impression of Dr Mswati) and a mismatch in expectations (between Kristelle and her supervisor, Professor Maxwell, for instance). However, these appeared to be idiosyncratic; that is, each was characteristic of only a single relationship. Thus, for instance, Xolani did not experience the same benign neglect that Hannah did from their supervisor, Professor Pieterse. Quite the contrary, Xolani and the professor often met for tea or lunch, taught together and attended the same conferences. So where were the tales of damage and despair? Why were they missing? And shouldn’t I feel happy that I had not stumbled across abuses at my institution? Didn’t that mean that Wits was (is?) upholding high standards of ethical and constructive postgraduate supervision? These questions sent me back to my original interview transcripts. I reasoned that perhaps I had missed something in the process of converting the transcripts to narrative profile, or even glossed over a case of authoritarianism or racism. Yet that was not the case. Nothing in the transcripts indicated that there has been a lapse in professional behaviour on the parts of the supervisors. I also considered the possibility that the candidates might not have told me their whole story, perhaps out of fear that their supervisors would somehow access the transcripts and subsequently punish them for any criticisms (real or perceived). The likelihood of this, though, was remote. Prior to each interview, I had given each candidate an undertaking that the transcripts would be kept on a disc at my home; hardcopies would also be stored there. In addition, none of the transcripts would contain the real names of the candidates, their supervisors or their 283 academic disciplines. All identifying details would be removed or re-labeled in order to maintain anonymity. Thus, I was able to dismiss the fear of victimization as a potential reason for the lack of data on supervisory abuses. However, this still left me reflecting on why, out of a relatively large sample, I could find no evidence of significant abuse when the literature was adamant that the ‘secret garden’ style of supervision was a breeding ground for this kind of behaviour. The conclusion that I drew was that the informal networks used by my interview candidates during the course of their research – friends, peers and unofficial communities of practice in their disciplines – had significantly diluted the potential for abuse in the one-on-one student-supervisor relationship. In other words, the fact that individuals other than the university-appointed supervisors were contributing to the progress of the doctoral candidates meant that supervisors were no longer the sole authority to whom their students could turn for advice and support. By accessing these broader networks, the doctoral candidates in my study had effectively deconstructed the ‘secret garden’ and replaced it with a more balanced mode of supervision. Furthermore, the context in which the supervision relationship occurs (cf. Figure 12) could also temper the tendency of supervisors to resort to ‘secret garden practices’. Thus, for instance, if the doctoral candidate possesses a strong intrinsic desire to earn a PhD, she will make the effort to engage with as many individuals in her specific community of practice as possible. In doing so, she will effectively tear down the ‘garden walls’ of supervision, thereby making the process more transparent for all involved. However, I suspect that HEIs in general and Wits in particular are a long way from achieving this state of affairs. 8.6 Conclusion The data analysed and interpreted in this chapter has shown a wide array of doctoral students’ experiences of their supervision. By examining the process through the lens of four separate, but often interlinked windowpanes – context/setting, process, relationships, and activities and events – I have attempted to unpack the practices that have both enabled and obstructed progress for my study’s interviewees. From the windowpane of context/setting, I discovered that even though both the Faculty of 284 Science and the Faculty of Humanities practice ‘traditional’ models of supervision, a third model appears to be emerging – one that I call ‘informal co-supervision’. This model combines the inputs of formal, university-appointed supervisors with contributions made to the research by unofficial co-supervisors who are often sought out by the doctoral candidates. The second windowpane – process – served to highlight the contribution that constructive, efficient, and relatively rapid feedback made to the students’ writing and thinking skills. My interviewees generally received both written and spoken feedback, which supervisors unpacked with them in supervision meetings. Where feedback was slow to come, or where it was perceived by the student to be superficial or over-critical in nature, it served little purpose but to frustrate the student. Feedback with substance and depth, and that was given within a reasonable amount of time, was highly prized, as was the opportunity afforded to students to disregard it if they deemed it to be irrelevant. The activities and events that occur within supervision – the third pane of my analysis – are fundamental to the supervision process. This proved to be the case even when activities and publications were not undertaken by a student in conjunction with her supervisor. What this windowpane exposed was how pursuits such as conference attendance, writing articles, or even just ‘hanging out’ with one’s supervisor helped to cement a more productive supervisory relationship. Furthermore, co-conferencing and co-publication also played a role in developing the doctoral candidate into a more confident researcher. My final windowpane – relationships and social structure – focused on the issue of power, and how this was played out in the supervisory relationship. I asserted that power was a key element in the relationships that my interviewees had with their supervisors, but it assumed different forms in each. In some, the supervisor was perceived to be its sole source while in others, the interviewees discounted its presence, indicating that it was irrelevant to both the process of becoming a PhD and to the supervisory relationship. In the former, it was the PhD student who constructed this hierarchy of the all-powerful supervisor; in the latter, the doctoral candidates were 285 sufficiently assertive and confident to keep any potential power plays on the part of their supervisors in check. Taking a step back and looking at all four windowpanes as they come together in one big picture window, what is most striking is that there is not one single model or practice or experience that can be universally applied to each of my interviewees’ experience of becoming a PhD. Thus, while each supervision experience encompassed all four windowpanes, the dimension of these panes differed in each case. For some students, one or two of the windowpanes dominated the others; for others, all four seemed to come together in a harmonious whole. Thus, each doctoral candidate had had a unique experience of supervision from the others – even when they had shared a supervisor. This suggests that the supervisors – certainly of this cohort – are sufficiently experienced to both identify what model of supervision would be most appropriate during the research process, as well as what specific practices best support this. Alternatively, it is possible that these supervisors have a unique supervision style that they can (and do) modify as and when the need arises. Where the ‘unique experience’ was not a positive experience, the problem seemed to lie with unmet expectations of the student – perhaps because she had never voiced these from the start of the process. What the picture window therefore suggests, via each of its constituent window panes, is that there is no single ‘correct’ way of supervising, nor is there a single ‘correct’ way of ‘being’ for students; rather, there are a range of practices, contained within a vast array of complex relationships, all of which are influenced by wider institutional and national imperatives. 286 CHAPTER 9 DISCUSSION AND CONCLUSION Chapter overview The final chapter of this thesis will focus on the key findings of the research, as well as on the conclusions that can be drawn from it. Using the window panes developed in Chapters 7 and 8, it will demonstrate the importance of the research to the practice of doctoral supervision. 9.1 Introduction In this chapter, I pull together the threads that have been woven throughout the thesis in order to answer the ‘so what?’ question – that is, why is this research important? At the most superficial of levels, the answer is (I suspect) that I wanted to know if I was alone in my experiences of supervision, from my Honours and Master’s level, right up to my encounters with my initial doctoral supervisor. Was all supervision negative and demotivating, or was I the problem? At a deeper level, I wanted to uncover the range of paths that other doctoral candidates had taken in their academic journeys. The purpose was to identify at least some of the practices (particularly the experience thereof) that seem to favour a positive and productive doctoral experience, as well as those that seem to get in the way of it – as defined by doctoral candidates. Herein lies the value of this research: the fact that it presents the voice of the student which, up until this point, has largely been veiled not so much in secrecy as in peripheral significance. Furthermore, I was interested in comparing the experiences of my interview candidates with the models of supervision described in the literature on postgraduate research supervision. Did the candidates’ experiences conform to these models, or had others emerged? Were communities of practice a significant feature of the students’ engagement with their research, both in respect of the attaining the PhD degree and of post-doctoral study? Pulling all these threads together, my ultimate goal was to consider how the range of candidates’ experiences of the process had (and would) impact on any future research career. In the remainder of this chapter, I will address these issues. 287 9.2 Formula-ting supervision From my research, it was clear that there is no ‘one-size-fits-all’ formula to, or model of postgraduate supervision. Even though the literature identifies a traditional ‘Science model’ and a traditional ‘Humanities model’ of supervision, each with distinctive characteristics and practices, I suggest that these no longer exist in their original configurations. Rather, they have evolved, largely in response to demands made on higher education institutions (and by implication, their academic staff and students) by the international knowledge economy. Thus, there has been a blurring in the formerly undisputed boundaries between faculty models. For instance, my study has shown how Humanities doctoral candidates are increasingly being supervised by more than one academic; historically, such co-supervision was considered to be a feature of the Science model. Furthermore, there has been a steady escalation in the number of academic papers that are co-authored by supervisors and their students in the Humanities – a significant break from the past, when scholarly articles were typically solo efforts. From the perspective of the Sciences, there has been a significant move towards supervision by a single academic, as opposed to co-supervision or supervision by committee. Recognizing that a one-size-fits-all approach to supervision does not exist is vital to future scholarship and practice in the area of postgraduate pedagogy because it challenges supervisors to reflect on and subsequently apply practices that are likely to lead to a better experience of the process for both them and their students. I have constructed the discussions of my findings – and the conclusions that can be drawn from these – according to the analytical window panes used in the previous two chapters, namely contexts, processes, relationships, and events and activities. It made sense to present them within these panes because they provided a useful organizing framework in which to discuss the thesis’ findings. 9.3 The presence of pedagogy An element of pedagogy (cf. section 2.3) was evident in the experiences of each of the doctoral candidates that I interviewed for this thesis. To restate my understanding of the term: it is a set of practices performed by both the supervisor and the student with a view to advancing learning. It consists of a social interaction in either (or both) a 288 formal or informal setting. It may also involve a critical element; that is, through these practices and interventions, it can effect change where inequalities of power exist. My interview candidates related numerous practices – both theirs and their supervisors – that contributed to their growth as scholars and researchers. For instance, some supervisors actively supported their students’ engagement in the intellectual life of the department in which they were registered. Prof Baldwin was a case in point: he offered Thabang opportunities to both lecture and tutor in the International Politics Department, recognizing the importance of these experiences to the development of Thabang’s identity as an academic researcher. Since Thabang had expressed the ambition of eventually being hired as a full-time member of staff in the department, these opportunities stood him in good stead. Xolani’s supervisor, Prof Pieterse, also exposed him to lecturing opportunities in their home Department. The supervisor co- taught with Xolani as an academic equal, not as his supervisor per se. Seamus had a somewhat different experience of being included in the intellectual pursuits of the Chemical Sciences Department: cheap laboratory labour. According to him, the research laboratories had a distinct hierarchy, with the PhD candidates to all intents and purposes supervising the MSc students, who in turn supervised the Honours cohort. Furthermore, this range of practices had a predictable impact on the students: Thabang and Xolani flourished, while Seamus became embittered. Another pedagogic practice drawn on by the supervisors was that of co-publication with their students – either journal articles, conference papers or, in the case of the Science Faculty, posters. Karabo, for instance, published several papers and prepared two posters with her supervisor, Prof Hurry. On all of these, her name was listed as first author. Leigh’s supervisor, Dr Rathbourne, has a similar ‘policy’ of insisting that the student’s name went first on these artefacts. According to her ‘I think that he would prefer to feel that if his name’s on the paper that he’s contributed in some way’. In the Humanities, Ntombi’s supervisor – Dr Gould - frequently co-published and co- presented at conferences with her. In fact, when her father died and she felt that she needed time to grieve, it was Dr Gould who suggested that she put aside her doctorate for a while, and focus instead on preparing articles for publication. That way, she could stay in the habit of writing while giving herself sufficient time to mourn. Xolani and his 289 supervisor, Prof Pieterse, were another student-supervisor team who regularly wrote and presented together. In each of these cases, it was evident that it was not only the practice and experience of writing or co-presenting that was valuable for the students, it was also the process of learning how to put together an article or presentation. Put another way, these opportunities afforded the doctoral candidates a space to learn how to communicate and disseminate their research at this level of academia from their ‘master’ (in Wenger’s terms). Moreover, in most instances, it was the supervisor who provided conference funding for the doctoral candidates – for both local and international conferences. This is a significant pedagogical practice because it represents an explicit investment on the part of the supervisor in the development of the student as a researcher. In other words, it denotes an appreciation of the fact that young researchers – the next generation - need to be actively nurtured and developed in order to achieve their full research potential. Exposing them to conferences, both as presenters and as an audience, gives them enormous scope to engage with their peers and in so doing, to advance their own skills and knowledge. In addition, it exposed them to a community of practice (see section 9.5) in their discipline. In this study, several of the participants were strategic in their selection of conference, choosing the ones at which the ‘big names’ in their field would be keynote speakers. Ntombi was one such student – as a result of receiving significant amounts of funding from her supervisor, she attended a number of conferences, during the course of which she met virtually all the people whose theoretical frameworks she had been using in her research. Bruno, from the Science Faculty, was a bit more ambivalent about the role of being sponsored to attend conferences: on the one hand, they represented an opportunity to share his knowledge with his peers; on the other, he was apprehensive about challenging long-established work in the field in his presentations. It seemed that he was anxious about how his research would be received in his disciplinary community. In sum, however, the pedagogic practice of funding students’ conference attendance – while not universally employed – added value to the students’ experiences of doctoral study because it exposed them to current thinking and the current thinkers in their discipline. It also gave them experience in presenting their 290 work in the appropriate disciplinary discourse (and community) – something that they would need beyond their PhD and into their future research careers. However, there were also a number of student-supervisor partnerships that did not yield any published articles or posters. This did not necessarily detract from the pedagogic experience, though. Hannah, for instance, chose to publish and attend conferences on her own, without any input or guidance from her supervisor, Prof Pieterse. As a confident academic (she had a background in lecturing at tertiary level), she did not feel the need to collaborate with him. She always kept him informed of these activities, offering to let him read her papers but never expecting any input from him. Neither did she express any regret that they had not collaborated. Rebecca and Prof Peterson were another case where co-publication was absent from the relationship. In this case, the reason was that the supervisor’s area of expertise was not the same as Rebecca’s and therefore collaboration was not necessarily a natural progression in their relationship. Günther was one of the few cases in the Science Faculty where there was no collaboration – conferences, papers or posters – with his supervisor. If anything, his supervisor (Prof van Onselen) seemed to avoid working with him in these pedagogical contexts. He attributed this to the professor’s heavy workload, which included the supervision of numerous other Masters and doctoral candidates as well as her own research. In this specific case, I believe that the lack of joint work was a missed learning opportunity for both of them. Prof van Onselen had so much experience in the discipline, while Günther had done extensive fieldwork – combining the two might have produced some valuable work. It might also have gone a long way in helping Günther feel more at home in South Africa. However, employing pedagogic practices did not only happen at the initiative of the supervisors. Many of the students that I interviewed for this study cultivated some courses of action of their own as a means of maximizing their own learning a form of self-empowerment (see 9.6). Dumisani, for instance, sought assistance from professors at other universities (both local and overseas) when it became clear that his supervisor was unable to meet his intellectual needs. This proactive stance meant that his learning experience was heightened because he was getting the quality of input that he had identified was necessary in order to succeed He perceived this course of action to be 291 wholly appropriate to ensuring his development as a researcher. In another case, Leigh was offered – and took up – the opportunity to undertake research at Princeton University, an Ivy League institution in the United States. Here, she engaged with some of the leading scientists in her field, which ultimately bolstered her research trajectory, not to mention her passion for it. Having said that, she was quick to point out that doctoral candidates need to proactively seek out opportunities like this for themselves instead of waiting for their supervisors to do it for them. Bill also took the initiative in identifying opportunities which could be used to advance his knowledge of his research field. The specific example that he mentioned – coming across a prestigious EU-led course on Isotopes and, in conjunction with his supervisors, motivating for funding – required dedication and determination to achieve. However, because he had a ‘big picture’ vision of what learning he wanted get out of his PhD, he was able to commit himself to finding opportunities and making full use of them. In the Humanities, Rebecca stood out as the doctoral candidate who epitomized this kind of pro-active stance. She applied for (and was granted) numerous local and international endowments which allowed her to travel to conferences and a Summer School in South America to advance her learning. For each of these endowments, it was Rebecca who found them, applied for them and adhered to all the conditions that were invariably attached to them. These processes were in and of themselves pedagogic practices which taught her how to deal with the bureaucracy attached to getting funding. What these particular cases point to is the fact that pedagogic practices are not necessarily originated only by supervisors – certain doctoral candidates have the wherewithal (and a strong enough sense of self-empowerment) to identify some of their own and to follow through with them. Furthermore, the contexts in which the pedagogy of supervision took place ranged from the very formal (such as the set three-weekly meetings that Ntombi had with her supervisor), to the extremely informal (epitomized by Bruno’s ‘meetings’ with his supervisors as he passed them in the passage). The context seemed to be dependent on what worked best for both supervisor and student, although this was not always explicitly discussed between them. George, for instance, would disappear for months at 292 a time when his formal employment kept him from his research work. Fortunately, his supervisor, Prof Pieterse, seemed to understand the demands of George’s work and was happy to meet with him whenever he ‘resurfaced’. This arrangement was never openly discussed or agreed upon – it just was. While George has not made significant progress in his research, he has nonetheless remained committed to it because his supervisor has recognized the pedagogic environment that best suits his style of learning and has adapted to it accordingly. It was interesting to note that not one of the interview participants expressed any deep dissatisfaction with their interaction with their supervisors. While some, like Megan and Kristelle, might have grumbled that their supervisors were not always as readily available as they might have liked, the supervision sessions themselves were never problematic. From this, I deduce that it is what happens in the supervision session itself that matters most to student, not the peripheral matters such as length, frequency or location of the meeting. Finally, I found no significant evidence of critical pedagogy amongst my participants. Undoubtedly, the interview candidates wanted to push the barriers of their learning and make their contributions to knowledge. However, none of them expressed an interest in changing inequalities that they might have come across in the course of their research. For all of them, the purpose of getting a PhD was either because it was a job requirement or for their own personal development. In the case of the former, one of the key criteria for promotion at Wits is a doctorate. It seems as though the PhD as a driver of new knowledge in the international knowledge economy has been replaced (or at least matched) by the PhD as guarantor of employment in the middle and upper ranks of the tertiary and research sectors – amongst my cohort of interviewees, at any rate. 9.4 Discourses of supervision I found evidence of all four of Grant’s (2005) key discourses of supervision – psychological, traditional, techno-sci and neoliberal - within the interview transcripts, 293 although in varying degrees. Marginal discourses were also represented in the data as I will demonstrate below. Each had a specific set of pedagogic practices (cf. 9.3) associated with it. The psychological discourse of supervision posits that the doctoral supervisor is a nurturing expert whose job it is to guide the student towards becoming an independent researcher (Grant 2005: 340). The supervisor is constructed as a sensitive, motivating presence in the research life of the student. Several supervisors described in this thesis seemed to favour this discourse. For instance, Bridget’s supervisor, Dr Martins, went out of her way to support Bridget in her journey to becoming a fully-fledged researcher: from introducing her to a colleague who was highly skilled at accessing National Research Fund grant money, to allowing her to stay at her home while she wrote her data analysis chapter, to the ‘Enjoy life’ comments that every piece of written feedback contained. All of these pedagogic practices – alone and in combination with one another – served to not only develop Bridget as a researcher, but also to build up her self-confidence. Emily’s supervisor, Prof Lambson, also made use of the psychological discourse. During the course of their supervisory relationship, he blended academic research training with pastoral care. The result was a doctorate, completed within the minimum time period by a young researcher who was confident in her ability to make significant contributions to knowledge in her discipline. The traditional-academic discourse is characterized by the ‘intellectual apprenticeship’ of the doctoral student to her supervisor (2005: 341). The power lies with the supervisor on account of her advanced knowledge of the field. There was a limited number of supervisor-student relationships in this research that exemplified this discourse, which could be an indicator of how empowered this group of doctoral students were relative to their supervisors. Nkosinathi was the exception. He actively constructed himself as an apprentice to both of his supervisors (cf. 9.5) because he recognized that they had far greater knowledge of the discipline than he did. This construction did not seem to be shared by his supervisors, who scaffolded their levels of supervision: that is, they were very directive at the start of the research, but gradually ceded full control of the process to him. Nkosinathi contended that this master-apprentice relationship was wholly appropriate, since he was ‘only’ a researcher 294 in training and still had a great deal to learn from his supervisors. The traditional- academic discourse applied in a limited way to Günther’s experience as well. His supervisor, Prof van Onselen, had thirty years’ experience in the discipline and for this reason was considered to be a ‘guru’ by Günther. She certainly seemed to be indifferent to him, as was evidenced by her failure to either tell him about or invite him to a local conference whose theme was directly relevant to his research. However, Günther was not always the submissive student envisaged by the discourse: he related how he enjoyed the independence that came with research, even if this meant straying from the instructions of his supervisor. Thus, the discourse of his supervision could be described as semi-traditional-academic. It was very interesting that such a small proportion of the sample group fell into the traditional-academic category. This suggests that it has lost its relevance in a world in which collaboration between supervisors and their students are the sites at which knowledge is increasingly being advanced. The third discourse described by Grant is that of the techno-scientific. A product of positivism, it constructs the supervisor as the expert scientist and the student as the ‘inexperienced trainee’ (2005: 342). It is the role of the former to train the latter in the ‘correct’ methods of research, using amongst other things, ‘surveillance’ and ‘judgment’. There were no identifiable instances of this discourse amongst my Humanities interviewees, and only weak renderings of it amongst the experiences of the Science candidates. This is worthy of note because, as Grant points out (2005: 343), this is the discourse most favoured by national governments and higher education bodies for its perception of the postgraduate research process as a linear one. That is, it understands the process as unfolding in a predictable way, one step logically following the other, thus making it easier for these bodies to budget for students in the system. Despite governments’ use of this discourse – including the South African government – the reality on the ground does not seem to support it. Thus, while doctoral candidates such as Indira, Emily, Nomsa, Karabo and Shanti all received some training from their respective supervisors in the experimental techniques required for their research, this took place within a framework of developmental pedagogic practices rather than the critical judgment inherent in the techno-scientific discourse. In all five of these examples, the research process was iterative – none of these candidates’ work 295 progressed in a straight line, but was rather typified by loops and the repetition of experiments when samples were tainted or deficient in some way. The discourse (and by implication, the institutional bodies that subscribe to it) thus fails to account for the fact that experimental work unfolds according to its own timetable; experiments can yield results in months or years, depending on a multitude of factors, usually beyond the control of the student. Imposing strict timeframes is therefore inappropriate, and could lead to the curtailment of vital research if a set of experiments is forced to come to a premature halt in order to adhere to them. Ultimately, this discourse could lead to a lose-lose situation: students suffer financially if they do not submit their doctorates within a set period of time; the government, university and possibly even the economy suffers because potentially ground-breaking (and hence income-generating) research has been restricted by timeframes. The neoliberal discourse of supervision constructs universities as commercial enterprises, and by implication, doctoral students as consumers of what these entities have to offer: education (Grant 2005: 343). The doctoral candidate pays for a service – supervision – and expects value for money. Thus, the PhD degree becomes an article of trade in which both parties have a set of rights and responsibilities in respect of one another. Therefore, the process of supervision becomes a contractual agreement between consumer (student) and service provider (supervisor), which requires a fairly rigid set of pedagogic practices to work. It is up to the supervisor to ensure that the student completes the doctorate in time in order to achieve maximum subsidy from the government, often at the expense of the intellectual development of the student. This discourse was not an overt feature of any of interview data. About half of the participants in this study were funded by a University staff bursary. The remainder had accessed grants or scholarships from various funding agencies. Thus, none of them could be described as consumers in the sense of the word that this discourse intended, because their funding had come from a source external to them. Theoretically, they did not have grounds to demand value for money, because the money was not theirs in the 296 first place. This made it impossible to measure the applicability of this discourse to my research data. Some of what Grant terms ‘marginal discourses’ of supervision were present in the data that was provided by the interview participants. Predictably, given South Africa’s political history, a discourse around race was evident in some of the interview transcripts, although this related to previous experiences of supervision (Honours and Masters) rather than to the PhD. Nomsa, for instance, described how she had been de facto marginalized at her MSc institution, which had been historically-advantaged, and largely Afrikaans-speaking. Her (white) supervisor made no effort to correct this state of affairs, thus maintaining his power over her and her research. At departmental functions, few people engaged with her – rather, the tendency was to ‘clique’ into distinct racial and language groups. By contrast, her (white) supervisor at Wits was more concerned with her development as a researcher, and treated her accordingly. The distinction between the two contexts was not lost on her. George’s experience of supervision was also shaped to an extent by the issue of race: as a Coloured man who had entered tertiary education at a mature age because Apartheid policies had economically marginalized his family, he was acutely aware of the value of education, as well as how Bantu education had stunted his educational opportunities. He made frequent reference to this throughout the interview, often in the same breath as declaring himself a ‘bad student’. On reflection, it is clear that there are elements of the various discourses described by Grant among the candidates that I interviewed for this thesis. However, none of them were an absolute match to the discourses espoused by these candidates. That is to be expected – Grant’s discourses are, after all, broad generalizations and therefore not intended to fit every individual context. Nonetheless, I was intrigued by how a distinct discourse – perhaps Wits-specific – had emerged: while the PhD was perceived, by and large, to be a commodity by the candidates, it was also an object of desire for them. In other words, the title, the prestige associated with the qualification as well as the career doors that it could open for the candidates was what motivated them to undertake the degree. Institutional and national imperatives were peripheral, in their view. I struggled to formulate a name for this discourse: could it be an egocentric discourse of 297 supervision, since the process revolves around the candidate and where the qualification can take her? Alternatively, could it simply be called the pragmatic discourse, because the candidate recognizes the PhD as her passport to a research career, and so does what is necessary to get one? Perhaps it is not a discourse at all, but rather an attitude towards the process. Regardless of its nomenclature, it represents an interesting departure from those proposed by Grant, and could be the start of a future research study. 9.5 Communities of practice and beyond The construct of communities of practice – as developed by Wenger (2000) – was evident to varying degrees amongst my research cohort. In the cases of Zodwa and Nkosinathi, for instance, there was a clear master-apprentice bond between them and their respective supervisors. However, Zodwa’s supervisor, Prof Bellarosa, advanced the relationship beyond this unequal power dynamic by introducing the doctoral candidate to her wide network, giving her opportunities to engage with the academics in it. In addition, she would send chapters of Zodwa’s thesis to scholars outside South Africa in order to get additional feedback for her. By contrast, Nkosinathi constructed himself as unenlightened relative to his masters/supervisors. It seemed that in so doing, he had condemned himself to an academic future on the periphery of his discipline. There was also evidence of candidates who had moved from the periphery to the core of their discipline. Thabang, for instance, had advanced from a position of apprenticeship to Prof Baldwin60 60 Although I doubt that Thabang would articulate his initial position as one of ‘apprenticeship’, per se. to one that virtually matched his supervisor: through the enterprises of writing and publishing academic papers, conference presentations, and ultimately consulting work in the field, Thabang had engaged with the community of scholars in his discipline. In this process of what Wenger calls ‘mutual engagement’ (1998a: 73), not only did Thabang advance his own knowledge, but he also made a contribution to that of others. In so doing, he moved away from being on the periphery of knowledge-creation, to being closer to the core. By implication, his power over the research direction also increased. His recent employment as a full-time member of the academic staff of the Department of International Politics has further consolidated this. 298 Faith’s trajectory was similar to Thabang’s: from being a novice researcher, dependent on the intellectual endowment of her supervisor, Prof Nzimande, she advanced her position in the community of scholars engaged in colonial literature research by publishing amongst other things, two of the chapters of her thesis. Once she had submitted her doctorate for examination, she was hired as a full-time academic at Wits. Subsequently, she was head-hunted by another university in the country, where she took up a teaching position in 2010. What this chain of events demonstrates is how she moved from the periphery of her community of practice (as a student) to a position closer to the core (having submitted her doctorate and been identified by employers in the tertiary education sector as an important thinker in the field). The transition from the periphery to the core was greater in some cases than in others. Hannah moved from her position as a relative newcomer to educational research to a full participant in the development of new knowledge largely on her own. Her supervisor’s benign neglect and her innate motivation to engage in the field – which she did by attending conferences, presenting paper, publishing articles and so on – meant that her transition was a self-directed one. By contrast, Karabo was extensively supported by her supervisors; she is the candidate who could be seen to have moved closest to the core in her community because she developed a new technique, which her supervisors helped her to patent. Thus, she is a considered a key figure – almost an ‘old-timer’ – because of the substantial contribution that she has made to knowledge in the field. In stark contrast to both of these, Seamus is a candidate who has remained rooted to the periphery of his community. He relied on his supervisor, Dr Montgomery, for intellectual input. The politics of the lab in which he worked were such that there was little collaboration or discussion between the students who worked there, which effectively cut him off from this potential community. While he did attend some conferences, he did so without his supervisor and thus missed valuable opportunities to co-present and to be introduced to Dr Montgomery’s network. Furthermore, the candidates that I interviewed did not seem to share a common idea of what constituted a ‘community’. Zodwa understood it to mean the networks that she had accessed through her supervisor (i.e. academics outside Wits); by contrast, Günther understood it in a more limited way to include only his supervisor and (some) of his 299 peers in the Department. He never made any mention of other members – the people in the field to whom his supervisor introduced him barely got a mention, suggesting that they did not have had any significant impact on his learning, nor did he seem to consider himself to a part of their community. George went out and actively found his own community: statisticians from work, scholars he had met at conferences and even family members who worked in a similar field as him. While none of these individuals were strictly in his field, they all had contributed in small, but different, ways to his research and hence formed a kind of community of practice. Thus, the notion of a community of practice – and the trajectory that it implies for doctoral candidates – was unevenly represented among my interview sample. This could suggest that the theory is not as applicable to the postgraduate supervision context as it might seem at face value. Alternatively, this unevenness might simply reflect the degree to which doctoral candidates and their supervisors are committed to contributing to the advancement of knowledge in their field: where the desire to develop the student as an active researcher who has the potential to contribute even more to the discipline is strong, the transition from periphery to core is evident. Where this is absent or indifferent in both parties, the student seems to flounder – not quite on the periphery, but not necessarily moving in the direction of the core, either. 9.6 Power and the doctoral relationship One of the critiques of communities of practice as a theory applicable to postgraduate supervision relates to its vague rendering of the concept of power (cf. 2.6.3). While Lave and Wenger (2003: 42) note that a great deal of power in the master-apprentice model is held by the ‘old-timers’ and is vulnerable to abuse by them, they fail to develop on the concept further. This presented a problem for my research because it did not help to account for the numerous variations in power relations (between old- timer/supervisor and newcomer/student) that emerged from my interviews. Nor did it sufficiently evolve to recognize that at some point, the doctoral candidate’s knowledge of the work overtakes that of the supervisor. If knowledge implies power – as Foucault (1975) argues - then at this point, it is the student who dominates the supervisor. In the cases of Thabang, Marcus and Shanti, the supervisors acknowledged this shift 300 (although the students seemed to be less convinced by it). They even encouraged it. By contrast, Günther, Seamus and Patricia were never motivated to see themselves in the context of ‘the student becoming the teacher’ – with the reversal of power that this implies. Tellingly, the latter could be interpreted as instances where the supervisor was resistant to ceding power to the doctoral candidate; perhaps, the supervisor felt threatened by the advances made by this new generation of researchers and wanted to protect their own standing in the Department or as ‘old-timers’ in their specific community of practice. 9.7 Reflections on the research design of the study Looking back, I am confident that the qualitative methodology that I used to elicit data for this project was appropriate in achieving what I had set out to do, namely to investigate doctoral candidates’ experiences of being supervised at this level. The interviews that I had designed to prompt the candidates’ narratives were effective in that they not only provided a ‘story’, but they also offered me the opportunity to ask follow up questions when I felt that further elaboration was required. In other words, although all of the participants were asked the same questions, the design of the interview tool allowed me enough leeway to probe deeper into their responses when I felt it was necessary. These interviews have been further consolidated by informal interactions that I had with some of the candidates when I bumped into them on campus. Ultimately, the results of the qualitative interactions utilised in the data collection process have fused to produce a rich and nuanced set of data that offers insights into the doctoral supervision experience that have thus far been contained in a ‘secret garden’ of their own. Encouraging doctoral candidates to breach the walls of this garden by sharing their stories is the first step in a larger process of critically studying the supervision process. 9.8 Further research The research contained in this thesis by no means represents the final word on the subject of doctoral supervision. Far from it. As I addressed the questions that I set out 301 to explore, so new avenues for future research occurred to me. I clustered these into three main categories, according to their focus: institutions, supervisory pedagogy and doctoral students. However, these categories are fluid and I recognize that there could be overlap between some of them. Future research about the role of the institution in postgraduate supervision could include a comparative study of supervision practices between historically advantaged and disadvantaged universities. Sixteen years into democracy, it would be practical to identify whether or not there is an imbalance in practices (and by implication, in graduation rates) between the two, as well as to investigate how supervisory practices could be brought into greater alignment (if this is necessary). Similar comparative projects could be done for merged and unmerged institutions (see section 1.4.1, footnote 4), as well as public and private HEIs. Ultimately, these research projects could be coordinated into a national repository of data about supervisory practices at the range of South African institutions with which new supervisors could engage as they begin their work in this academic role. Even more established supervisors could benefit from access to such a repository. Secondly, there is great scope for future research into the pedagogic practices employed by supervisors. One potential avenue would be to investigate if (and how) academics have adjusted their supervisory practices in response to demands from the international knowledge economy. Such research could be undertaken using either qualitative interviews or by encouraging supervisors to write reflective pieces about their experiences in this context. Another avenue for future investigation would be to study specific supervisory practices that overtly develop and nurture students’ sense of agency over their work. In other words, what do supervisors do to encourage students’ ownership of their research? Or is this process driven by the student? This would be an important contribution to current research in the field of postgraduate studies because it could start a valuable discussion on the empowerment of postgraduate students vis-à- vis their research. Yet another line of future inquiry relates to the racial and ethnic diversity of the South African student body: should (or do) supervisors adapt their practices to the (perceived) cultural characteristics of their students (Section 1.4.1) Put another way, are supervisors practicing culturally sensitive pedagogy in a country that 302 has historically favoured one group over others? This question is worth pursuing if South Africa is to make any advances in both redressing the educational imbalances of the past and in increasing the quantity and quality of postgraduate students graduating. The final category for future research – that of doctoral candidates – offers numerous possibilities. The most obvious of these is about students taking agency over both the content and direction of their research: How is such agency developed by students? That is, under what conditions does the candidate take ownership of her project? Moreover, does it evolve over the course of the research alongside the student’s scholarly identity? The significance of this research avenue lies in how the insights it generates could support supervision by highlighting good practice in guiding students to take ownership of their intellectual work. Methodologically, the interview technique that I used to gather my data could be replicated in various other supervision contexts. For instance, candidates could be asked to keep a narrative journal of their supervision meetings (and other interactions with their supervisors); subsequently, interviews with these candidates could be initiated to find out what they perceived the major issues in these experiences to be. Alternatively, doctoral candidates could be asked write an account of their supervision experiences (much like the one in the Prologue of this work), and to reflect on how these impacted on their attitude towards the research. Thus, there is a wealth of potential research that could be pursued from my project. Certainly, the richness of the narrative data that has been presented in this thesis has not thoroughly been mined – one thesis was not enough to do this. Therefore, more discussion and analysis of this data is possible. 9.9 Final thoughts What became clear as I wrote this concluding section was that there was no single overarching experience of doctoral supervision. Each student brought particular experiences, attitudes and abilities to their PhD studies, which in turn responded to the experience, attitudes and abilities of their supervisors. However, this individual context also operates within the broader context of institutional, national and international 303 pressures for more and better quality research, and competition for funding. These add multiple dimensions to the doctoral candidates’ research experience – dimensions that they were not always aware of. The result was a doctoral experience unique to each of them, even those who had the same supervisors. This is a clear indication that thesis- writing manuals and highly technicist checklists are both inadequate for, and irrelevant to, the doctoral experience of supervision because they fail to address the core truth that one size does not fit all. Rather, each supervision relationship needs to be approached as a new adventure by both parties, and the research journey as an opportunity for unbounded discovery and growth. In the absence of these conditions, there can be no advancement of new knowledge. 304 REFERENCE LIST 2009 List of Developing Countries. Retrieved 5 May 2010, from http://icce2010.org/docs/developingCountries.pdf Adair-Toteff, C. (1995). Ferdinand Tönnies: Utopian Visionary. Sociological Theory, 13 (1), 58 – 65. Address by the Deputy President, Kgalema Motlanthe, at the Higher Education Summit, Bellville, Cape Town, 22 April 2010. Address by Minister of Higher Education and Training, Blade Nzimande, at the Higher Education Summit, Bellville, Cape Town, 22 April 2010. Adkins, B.A. (2009). PhD pedagogy and the changing knowledge landscapes of universities. Higher Education Research and Development Journal, 28 (2), 165 – 177. Agrawal, A. & Gibson, C. (1999). Enchantment and Disenchantment: The Role of Community in Natural Resource Conservation. World Development, 27 (4), 629 – 649. Aiken, L.S. (2009). Leona S. Aiken, professor of psychology. ASU Graduate College. Retrieved September 7, 2010, from http://graduate.asu.edu/node/1154. Akoojee, S. & Nkomo, M. (2007). Access and quality in South African higher education: the twin challenges of transformation. South African Journal of Higher Education, 21 (3), 385 – 399. Alexander, R. (2008). Pedagogy, Curriculum and Culture. In K. Hall, P. Murphy & J. Soler (Eds), Pedagogy and Practice: Culture and Identities (pp 3 – 27). London: Sage Publications. 305 Ary, D., Jacobs, L..C. & Razavieh, A. (1990). Introduction to Research in Education. Fort Worth: Holt, Rinehart and Winston Inc. Austin, A. E. (2002). Preparing the Next Generation of Faculty: Graduate School as Socialisation to the Academic Career. The Journal of Higher Education, 73 (1), 94 – 122. Babbie, E. (2005). The Basics of Social Research. California: Thomson Wadsworth. Badenhorst, C. (2007). Research writing: Breaking the barriers. Hatfield, Pretoria: Van Schaik Publishers. Badenhorst, C. (2008). Dissertation Writing: a Research Journey. Hatfield, Pretoria: Van Schaik. Bandura, A. (2006). Toward a Psychology of Human Agency. Perspectives on Psychological Science. 1(2), 164 – 180. Barron, D. and Zeegers, M. (2002). ‘O’ for Osmosis, ‘P’ for Pedagogy: Fixing the Postgraduate Wheel of Fortune. Retrieved 22 November 2004, from http://www.aare.edu.au/02pap/bar02605.htm. Bartlett, A. & Mercer, G. (Eds) (2001). Postgraduate Research Supervision: Transforming (R)Elations. New York: Peter Lang. Bauman, Z. (1994). Alone Again: Ethics After Certainty. London: Demos. Bawa, A. (2007). Doctoral education in South Africa: the challenge for development. In S. Powell & H. Green (Eds). The Doctorate Worldwide. Berkshire: Society into Research for Higher Education & Open University Press’ 306 Becher, T. (1994). The Significance of Disciplinary Differences. Studies in Higher Education, 19 (2), 151 – 161. Becher, T. & Trowler, P.R. (2001). Academic Tribes and Territories. Buckingham: The Society for Research into Higher Education & Open University Press. Behar-Horenstein, L.S. & Morgan, R.R. (1995). Narrative Research, Teaching, and Teacher Thinking: Perspectives and Possibilities. Peabody Journal of Education, 70 (2), 139 – 161. Bell, C. & Newby, H. (1973). Community Studies: An Introduction to the Sociology of the Local Community. New York: Praeger Publishers. Bell, C. & Newby, H. (1974). The Sociology of Community: A Selection of Readings. London: Frank Cass. Bogden, R.C. & Biklen, S.K. (2007). Qualitative Research for Education: An Introduction to Theory and Methods. Massachusetts: Allyn and Bacon. Booth, W.C., Colomb, G.G. & Williams, J.M. (1995). The Craft of Research. Chicago: University of Chicago Press. Boyer, E.L. (1990). Scholarship reconsidered: Priorities of the professoriate. Princeton, New Jersey: Carnegie Foundation for the Advancement of Teaching. Braa, D. & Callero, P. (2006). Critical Pedagogy and Classroom Praxis. Teaching Sociology, 34(4), 357 – 369. 307 Brennan, M., Kenway, J., Thomson, P. & Zipin, L. (2002). Uneasy Alliances: University, Workplace, Industry and Profession in the Education Doctorate. The Australian Educational Researcher, 29(3), 63 – 83. Brinko, K. (1993). The practice of giving feedback to improve teaching: what is effective? Journal of Higher Education, 64, 574 – 593. Bröder, A. (1998). Deception can be acceptable. American Psychologist, 53, 805 – 806. Brown, J.S. & Duguid, P. (1998). Organising knowledge. California Management Review, 40, 90 -111. Bryman, A. (1984). The Debate about Quantitative and Qualitative Research: A Question of Method or Epistemology? The British Journal of Sociology, 35 (1), 75 – 92. Burbules, N & Berk, R. (1999). Critical Thinking and Critical Pedagogy: Relations, Differences and Limits. In T. Popkewitz & L. Fendler (Eds). Critical Theories in Education: Changing Terrains of Knowledge and Politics. New York: Routledge. Burman, E. (1997). Minding the Gap: Positivism, Psychology, and the Politics of Qualitative Methods. Journal of Social Issues, 53 (4), 785 – 801. Casey, K. (1996). The New Narrative Research in Education. Review of Research in Education, 21, 211 – 253. Cassell, C. & Syman, G. (2004). Essential Guide to Qualitative Methods in Organizational Research. London: Sage. Chapman, V. & Sork, T.J. (2001). Confessing Regulation or telling secrets? Opening up the conversation on graduate supervision. Adult Education Quarterly, 51(2), 94-107. 308 Charlesworth, G., Grossman, E., Hadingham, J., Janks, H., Mycock, D. & Scholes, M. (2007). Strategies for successful supervision. University of the Witwatersrand, Johannesburg. CHE (2004). Improving Teaching & Learning Resource No. 7: Postgraduate Research and Supervision. Pretoria: Council on Higher Education. CHE (2004). Criteria for Institutional Audits. Pretoria: Council on Higher Education. CHE (2004). Criteria for Programme Accreditation. Pretoria: Council on Higher Education. CHE (2007). HEQC Institutional Audits Manual 2007. Pretoria: Council on Higher Education. CHE (2009). Postgraduate Studies in South Africa: A Statistical Profile. Pretoria: Centre for Research on Science and Technology and Council on Higher Education. CHE (2010): Higher Education Monitor: Access and Throughput in South African higher education – three case studies. Pretoria: Council on Higher Education. Chisholm, L. (2004). Introduction. In L. Chisholm (Ed), Changing Class: Education and Social Change in Post-Apartheid South Africa. Cape Town: HSRC Press. Christians, C.G. (2000). Ethics and Politics in Qualitative Research. In N.K. Denzin & Y.S. Lincoln (Eds). Handbook of Qualitative Research. London: Sage. 309 Christie, P. (1991). Right to Learn: The Struggle for Education in South Africa. Randburg: Ravan Press. Clandinin, D.J. & Connelly, F.M. (2000). Narrative Inquiry: Experience and Story in Qualitative Research. San Francisco: Jossey-Bass. Cloete, N. (Eds)(2002). Transformation in Higher Education: Global Pressures and Local Realities in South Africa. Cape Town: Juta & Co. Cohen, L., Manion, L. & Morrison, K. (2002). Research Methods in Education. London: Routledge/Falmer. Connell, R.W. (1985). How to supervise a PhD. Vestes, 2, 38 – 41. Connelly, F.M. & Clandinin, D.J. (1990). Stories of experience and narrative inquiry. Educational Researcher, 19 (5), 2 – 14. Conrad, L. & Phillips, E. M. (1995). From Isolation to Collaboration: A Positive Change for Postgraduate Women. Higher Education. 30 (3), 313 – 322. Contu, A. & Willmott, H. (2003). Re-Embedding Situatedness: The Importance of Power Relations in Learning Theory. Organization Science, 14 (3), 283 – 296. Cortazzi, M. (1993). Narrative Analysis. London: Falmer Press. Craswell, G. (1996, 18 – 19 April). This unfathomable thing called supervision: negotiating better working relationships with supervisors. Paper presented at the Quality in Postgraduate Research: Is It Really Happening? Conference. Adelaide, Australia. 310 Craswell, G. (2005). Writing for Academic Success: A Postgraduate Guide. London: Sage Publications. Creswell, J.W. & Miller, D.L. (2000). Determining Validity in Qualitative Inquiry. Theory in Practice. 39 (3), 124 – 130. Cryer, P. (1996). The Research Students Guide to Success. Buckingham: Open University Press. Delamont, S. & Atkinson, P. (1997). Critical mass and doctoral research: Reflections on the Harris Report. Studies in Higher Education. 22 (3), 319 – 332. Delamont, S., Atkinson, P. & Parry, O. (1997). Supervising the PhD: A Guide to Success. Buckingham: The Society for Research into Higher Education & Open University Press Delamont, S., Atkinson, P. and Parry, O. (1998). Creating a Delicate Balance: the doctoral supervisor’s dilemma. Teaching in Higher Education, 3 (2), 157 – 172. Denzin, N.K. & Lincoln, Y.S. (1994). Introduction: The Discipline and Practice of Qualitative Research. In Denzin, N.K. & Lincoln, Y.S. (Eds), Handbook of Qualitative Research (pp 1 – 32). California: Sage. Denzin, N.K. & Lincoln, Y.S. (1998). Strategies of Qualitative Inquiry. California: Sage. Deschler, D. (1985). Metaphors and values in higher education. Academe, 71 (6), 22 – 28. 311 Dietz, T. & Burns, T.R. (1992). Human Agency and the Evolutionary Dynamics of Culture. Acta Sociologica, 35 (30), 187 – 200. DoE (1997). Education White Paper 3 – A Programme for Higher Education Transformation. South African Government Gazette, no. 18207, Pretoria. DoE (2001). Draft National Plan for Higher Education. Government Gazette no. 22138, Notice Number 230, Pretoria. DoE (2004). Creating Comprehensive Universities in South Africa: A Concept Document. Pretoria: Department of Education. Dunleavy, P. (2003). Authoring a PhD. Basingstoke: Palgrave Macmillan. Edwards, B. (2002, 1 – 5 December). Postgraduate Supervision: Is having a PhD enough? Paper presented at the Australian Association for Research in Education Conference, Brisbane, Australia. Etzioni, A. (1996). The responsive community: A Communitarian perspective. American Sociological Review, 61 (1), 1 – 11. Evans, T. & Green, B. (1995, 26 – 30 November). Dancing at a Distance? Postgraduate Studies, ‘Supervision’, and Distance Education. Paper presented at the Graduate Studies in Education: Innovation in Postgraduate Research and Teaching Conference, Hobart, Australia. Fontana, A. & Frey, J.H. (2000): The Interview: From Structured Questions to Negotiated Text. In N. K. Denzin & Y.S. Lincoln (Eds) Handbook of Qualitative Research. New York: Sage. Foucault, M. (1979). Discipline and Punish: The Birth of the Prison. London: Penguin. 312 Fox, S. (2001). Communities of Practice, Foucault and Actor-Network Theory. Journal of Management Studies, 37 (6), 853 – 867. Freire, P. (2000). Pedagogy of the Oppressed. New York: Continuum. Fuller, A., Hodkinson, H., Hodkinson, P. & Unwin, L. (2005). Learning as peripheral participation in communities of practice: a reassessment of key concepts in workplace learning. British Educational Research Journal, 31 (1), 49 -68. Gaff, J.G. & Lambert, L.M. (1996). Socialising future faculty to the values of Undergraduate Education. Change, 28 (4), 38 – 46. Gee, J.P. (1985). The narrativization of experience in the oral style. Journal of Education, 167 (1), 9 – 35. Gee, J.P. (1986). Units in the production of narrative discourse. Discourse Processes, 9, 391 – 422. Gee, J.P. (1991). A linguistic approach to narrative. Journal of Narrative and Life History, 1 (1), 15 -31. Gee, J.P. (1992). The Social Mind: Language, Ideology and Social Practice. New York: Bergin & Harvey. Gee, J.P. (1996). Social Linguistics and Literacies: Ideology in Discourses. London: RoutledgeFalmer. Giddens, A. (1984). The Constitution of Society: Outline of the Theory of Structuration. Berkeley: University of California Press. Giltrow, J. (1995). Academic writing: Writing and reading across the disciplines. Ontario: Broadview Press. 313 Giroux, H. A. (2004). Critical Pedagogy and the Postmodern/Modern Divide: Towards a Pedagogy of Democratization. Teacher Education Quarterly, 31(1), 31 – 47. Glatthorn, A.A., & Joyner, R.L. (2005). Writing the Winning Thesis or Dissertation: A Step-by-Step Guide. Thousand Oaks, California: Corwin Press. Golafshani, N. (2003). Understanding Reliability and Validity in Qualitative Research. The Qualitative Report, 8 (4), 597 – 607. Golde, C.M., & Walker, G.E. (Eds) (2006). Envisioning the Future of Doctoral Education: Preparing Stewards of the Discipline. San Francisco: Jossey-Bass. Grant, B. (1999, 12 – 15 July). Walking on a rackety bridge: mapping supervision. Paper presented at the HERDSA International Conference. Melbourne, Australia. Grant, B. (2000). Guidelines for Negotiating Supervision. Retrieved 15 October 2004, from http://www2.auckland.ac.nz/mdr/NegPGS.html. Grant, B. (2005). Fighting for space in supervision: fantasies, fairy tales, fictions and fallacies. Retrieved 22 November 2004, from http://www.aut.ac.nz/conferences/innovation/papersthemetwo/grantpapertwo.pdf. Grant, B. (2008). Agonistic Struggle: Master slave dialogues in humanities supervision. Arts and Humanities in Higher Education, 7 (1), 9 -27. Grant, B. (2009). Uneasy translations: taking theories of supervision into teaching. London Review of Education, 7 (2), 125 – 134. Graves, N. & Verma, V. (1997). Working for a Doctorate: A Guide for the Humanities and Social Sciences. London: Routledge. 314 Green, P. (2002). Coping and completing: The challenge of the professional doctorate in the knowledge economy. Retrieved 22 November 2004, from www.ecu.edu.au/conferences/herdsa/main/papers/nonref/pdf/PamGreen.pdf. Greenhalgh, T., Russell, J. & Swinglehurst, D. (2005). Narrative methods in quality improvement research. BMJ Quality & Safety, 14, 443 – 449. Gruenewald, D.A. (1999). The Best of Both Worlds: A Critical Pedagogy of Place. Educational Researcher, 32(4), 3 – 12. Gullan, P.J. (1996, March 13). Supervision of Research Students in Biological Sciences. Paper presented at a workshop for thesis supervisors on Supervising Research Students. Centre for Educational Development and Academic Methods, Canberra. Gurevitch, Z. (2001). Dialectical dialogue: The struggle for speech, repressive silence, and the shift to multiplicity. British Journal of Sociology, 52 (1), 87 – 104. Guri-Rosenblit, S., Šebková, H. & Teichler, U. (2007). Massification and Diversification of Higher Education Systems: Interplay of Complex Dimensions. Higher Education Policy, 20, 373 – 389. Habib, A. & Morrow, S. (2006). Research, research productivity and the state in South Africa. Transformation: Critical Perspectives on Southern Africa. 62, 9- 29. Haggerty, K.D. (2004). Ethics Creep: Governing Social Science Research in the Name of Ethics. Qualitative Sociology, 27 (4), 391 – 414. Hall, M, Symes, A. & Luescher, T. (2004). The Governance of Merger in South African Higher Education. Pretoria: Council on Higher Education. 315 Handley, K., Sturdy, A., Fincham, R. & Clark, T. (2006). Within and Beyond Communities of Practice: Making Sense of Learning Through Participation, Identity and Practice. Journal of Management Studies, 43 (3), 641 – 653. Hart, L. (2001). Doing a Literature Search: A Comprehensive Guide for the Social Sciences. London: Sage Publications. Heathcott, J. (2005). Trained for nothing. Academe, 91 (6), 14 -18. Henning, E. (2004). Finding your way in qualitative research. Pretoria: van Schaik. Herman, C. (personal communication. June 9, 2010). Hirst, E., Henderson, R., Allan, M., Bode, J. & Kocatepe, M. (2004). Repositioning academic literacy: the emergence of a community of practice. Australian Journal of Language and Literacy, 27 (1), 66 – 80. History of Wits (2010). Retrieved 12 May 2010 from http://web.wits.ac.za/AboutWits/IntroducingWits/HistoryofWits.htm Hoepfl, M.C. (1997). Choosing Qualitative Research: A Primer for Technology Education Researchers. Journal of Technology Education, 9 (1), 47 – 63. Horsthemke, K. (2006). The idea of the African university in the twenty-first century: Some reflections on Afrocentrism and Afroscepticism. South African Journal of Higher Education, 20 (4), 449 – 465. 316 Hunt, C. (2001). Climbing out of the void: moving form chaos to concepts in the presentation of a thesis. Teaching in Higher Education, 6 (3), 351 – 367. Hyatt, D.F. (2005). ‘Yes, a very good point!’: a critical genre analysis of a corpus of feedback commentaries on Master of Education assignments. Teaching in Higher Education, 10 (3), 339 – 353. Hyland, K. (2000). Disciplinary Discourses: social interactions in Academic Writing. Michigan: University of Michigan Press. James, R. & Baldwin, G. (1999). Eleven Practices of Effective Postgraduate Supervisors. Melbourne: CSHE. Jansen, J.D. (2009). Knowledge in the Blood Confronting Race and the Apartheid Past. California: Stanford University Press. Johnson, S. (1995). Building a sense of community in a Research Master’s course. Studies in Higher Education. 20 (3), 279 – 291. Kamler, B. & Thomson, P. (2006). Helping Doctoral Students Write: Pedagogies for Supervision. London: Routledge. Kamler, B. (2008). Rethinking doctoral publication practices: writing from and beyond the thesis. Studies in Higher Education, 33(3), 283 – 294. Kelly, F. (2003, July). Skills for an unknown future: the end of the doctorate and beyond. Paper presented at the 26th Annual HERDSA Conference on Learning for an Unknown Future, Christchurch, New Zealand. 317 Kiley, M. (2003, 6 – 9 July). Developments in postgraduate education in Australia over the last decade. Paper presented at the 26th Annual HERDSA Conference on Learning for an Unknown Future, Christchurch, New Zealand. Kiley, M. & Mullins, G. (2003, 26 – 30 August).Why conceptions of research might be significant in postgraduate education. Paper presented at the Improving Learning, Fostering the Will to Learn Conference, Padova, Italy. Knowles, S.S. (2003). Murmurs from underground: uncovering invisible pedagogies in PhD supervision. Retrieved 22 November 2004, from http://www.aut.ac.nz/conferences/innovation/papersthemetwo/knowlespapertwo.pdf, Kohler Riessman, C. (1993). Narrative Analysis. California: Sage. Kumar, V. and Stracke, E. (2007). An analysis of written feedback on a PhD thesis. Teaching in Higher Education, 12 (4), 461 – 470. Kvale, S. (2006). Dominance Through Interviews and Dialogues. Qualitative Inquiry, 12 (3), 480 – 500. Kvale, S. & Brinkmann, S. (2009). Interviews: Learning the Craft of Qualitative Research Interviewing. Thousand Oaks, California: Sage. LaPidus, J. (1997). Doctoral Education: Preparing for the Future. CGS Communicator, 3 (10), 1 – 8. Lave, J. & Wenger, E. (1991). Situated Learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lawson, E. (2001). Informational and Relational Meanings of Deception: Implications for Deceptions Methods in Research. Ethics & Behavior, 11 (2), 115 – 130. 318 Lea, M.R. (2004). Academic literacies: a pedagogy for course design. Studies in Higher Education, 29 (6), 739 – 756. Lea, M.R. (2005). ‘Communities of Practice’ in Higher Education. In D. Barton & K. Tusting (Eds), Beyond Communities of Practice: Language, Power and Social Context. Cambridge: Cambridge University Press. Lea, M.R. & Street, B. (2006). The “Academic Literacies” Model: Theory and Applications. Theory Into Practice, 45(4), 368 – 377. Leder, G.L. (1995). Higher degree research supervision: a question of balance. Australian Universities Review, 2, 5 – 8. Leonard, D. (2001). A Woman’s Guide to Doctoral Studies. Buckingham, UK: Open University Press. Lincoln, Y.S. & Guba, E.G. (1985). Naturalistic inquiry. California: Sage. Lindkvist, L. (2005). Knowledge Communities and Knowledge Collectivities: A Typology of Knowledge Work in Groups. Journal of Management Studies, 42 (6), 1189 – 1210. Matos, F. (2006, July). PhD: What is it? An analysis of how different stakeholders perceive the doctoral degree. Match or mismatch? Paper presented at the Higher Education Close Up 3 International Research Conference. University of Lancaster. Macauley, P. (2000). Pedagogic continuity in doctoral supervision: passing on, or passing by, of information skills? In M. Kiley & G. Mullins (Eds). Quality in 319 Postgraduate Research: Making ends meet. Adelaide: Advisory centre for University Education, University of Adelaide. Makgoba, M.W. (1997). Mokoko: The Makgoba Affair. A reflection on transformation. Florida: Vivlia. Manathunga, C. (2005). Early warning signs in postgraduate research education: a different approach to ensuring timely completions. Teaching in Higher Education, 10 (2), 219 – 233. Marginson, S. (2008). GLOBAL: The knowledge economy and global public sphere. University World News, Issue 0057, 1 – 5. Maxwell. J.A. (1996). Qualitative Research Design: An Interactive Approach. California: Sage. McWilliam, E. (1996). Pedagogies, Technologies, Bodies. In E. McWilliam & P.G. Taylor (Eds), Pedagogy, Technology and the Body. New York: Peter Lang. McWilliam, E. & Singh, P. (2002). Towards a research training curriculum: What, why, how, who? The Australian Educational Researcher, 29(3), 3 – 18. Melin Emilsson, U. and Johnsson, E. (2007). Supervision of supervisors: on developing supervision in postgraduate education. Higher Education Research & Development 26(2), 163 – 179. Merriam, S. B. (1995). What Can You Tell From An N of 1? Issues of Validity and Reliability in Qualitative Research. PAACE Journal of Lifelong Learning, 4, 51 – 60. 320 Merriam, S.B. (2002). Qualitative Research in Practice: Examples for Discussion and Analysis. San Francisco: Jossey-Bass. Mischler, E.G. (1986) Research Interviewing: Context and Narrative. Cambridge: Harvard University Press. Modest revolt to save research from red tape (12 February 2009). Retrieved 6 May 2010, from http://www.timeshighereducation.co.uk/story.asp?sectioncode=26&storycode=405335 &c=2 Morrow, W. (2009). Bounds of Democracy: Epistemological Access in Higher Education. Cape Town: HSRC Press. Mouton, J. (2006). How to succeed in your master’s and doctoral studies: A South African guide and resource book. Pretoria: van Schaik. Muendane, N.M. (2006). I am an African: Embrace your identity, escape victimization. Pretoria: Soultalk Publishers. Murray, R. (2002). How to Write a Thesis. Buckingham: Open University Press. Myers, M. (2000). Qualitative Research and the Generalizability Question: Standing Firm with Proteus. The Qualitative Report, 4 (1&2), 1 – 8. National Scarce Skills List 2007. Retrieved 31 May 2010, from http://www.skillsportal.co.za/download_files/professional/NSDS- Scarce_Skills_List_2007.doc 321 Ndimande, B.S. (2009). “It Is a Catch 22 Situation”: The Challenge of Race in Post- Apartheid South African Desegregated Schools. International Critical Childhood Policy Studies, 2(1), 123 – 139. Nelson, H. (1996, March 13). A Short Thesis on Supervision. Paper presented at a workshop for thesis supervisors on Supervising Research Students. Centre for Educational Development and Academic Methods, Canberra. Neuman, R. (2002). Diversity, Doctoral Education and Policy. Higher Education Research & Development, 21 (2), 167 – 178). Neuman, R. (2007). Policy and practice in doctoral education. Studies in Higher Education. 32 (4), 459 – 473. Northedge, A. (2003). Rethinking Teaching in the Context of Diversity. Teaching in Higher Education, 8 (1), 17 – 32. Nunkoosing, K. (2005). The Problems with Interviews. Qualitative Health Research, 15 (5), 698 – 706. Oakeshott, M. (1962). Rationalism in Politics and Other Essays. London: Methuen. Odora Hoppers, C.A. (1999). Indigenous knowledge and the integration of indigenous knowledge systems: towards a conceptual and methodological framework. A discussion document for the Human Sciences Research Council Project on A Comparative Study on the Development of Knowledge Systems in the Third World. Pretoria: Human Science Research Council. Odora Hoppers (2005). Culture, indigenous knowledge and development: The role of the university. Johannesburg: Centre for Educational Policy Development (CEPD), Occasional Paper No. 5, 1 – 50. 322 Ortmann, A. & Hertwig, R. (1998). The question remains: Is deception acceptable? American Psychologist, 53, 806 – 807. Park, C. (2005). New Variant PhD: The changing nature of the doctorate in the UK. Journal of Higher Education Policy and Management, 27(2), 189 – 207. Park, C. (2007). Redefining the doctorate. Retrieved 1 March 2010 from http://learning.ox.ac.uk/rsv.php?page =326. Phillips, E. & Pugh, D. (1989). How to get a PhD. Milton Keynes: Open University Press. Philp, R. (2010). SA brains choose dosh over degrees. The Sunday Times, 8 August. Powell, S. & Green, H. (2007). The Doctorate Worldwide. Berkshire: Society for Research into Higher Education & Open University Press. Principles of the Nuremburg Tribunal (1950). Retrieved 2 September 2010 from http://deoxy.org/wc/wc-nurem.htm. Purpose of the doctorate. (2005). Retrieved 21 March 2010, from http://www.learning.ox.ac.uk/rsv.php?page=326. Reid, I. & Parker, L. (2002). Framing institutional policies on literacies. Australian Review of Applied Linguistics, 25 (2), 19 – 27. Remenyi, D. & Money, A. (2004). Research Supervision: For Supervisors and Their Students. Kidmore End: Academic Conferences Limited. 323 Roberts, J. (2006). Limits to Communities of Practice. Journal of Management Studies, 43 (3), 623 – 639. Rose, M. & McClafferty, K.A. (2001). A Call for the Teaching of Writing in Graduate Education. Educational researcher, 30 (2), 27 - 33. Rudd, E. (1985). A New Look at Postgraduate Failure. Surrey: The Society for Research into Higher Education & NFER-NELSON. Schwandt, T.A. (1997). Qualitative inquiry: a dictionary of terms. Thousand Oaks, California: Sage. Searle, R, McKenna, S. & Harrison, L. (2005). It’s About Me, Not Just the PhD – Women’s Narratives About Supervision. Paradigms, Issue 12, 11 – 15. Shannon, A. G. (1995). Research degree supervision: ‘more mentor than master’. Australian Universities Review, 2, 12 – 15. Sewell, W.H. (1992). A Theory of Structure: Duality, Agency, and Transformation. The American Journal of Sociology, 98 (1), 1 – 29. Silverman, D. (2000). Doing Qualitative Research: A Practical Handbook. London: Sage. Singh, Z. and Kulski, M.M. (2003). Student-focused postgraduate supervision: A learning intervention process. Retrieved 29 October 2007 from http://www.ecu.edu.au/conferences/tlf/2003/pub/pdf/07_Singh_ZoraKulski_Martijntje. pdf. 324 Singh, Z. and Martijntje, M.K. (2003). Student-focused postgraduate supervision: A learning intervention process. Retrieved 22 November 2004, from http://www.ecu.edu.au/conferences/tlf/2003/pub/pdf/07_Singh_ZoraKulski_Martinntje. pdf. Soble, A. (1978). Deception in Social Science Research: Is Informed Consent Possible? Hastings Center Report, 8 (5), 40 – 46. Sorcinelli, M.D. & Austin, A. E. (1992). Developing new and junior faculty. San Francisco: Jossey-Bass. Soudien, C. (2008). Report of the Ministerial Committee on Transformation and Social Cohesion and the Elimination of Discrimination in Public Higher Education Institutions. Pretoria: Department of Education. St George, E. (2007). Higher Education for the Knowledge Based Economy. Higher Education, 52 (4), 589 – 610. The Skills Portal (2007). Scare skills: defining ‘scarce’ and ‘critical’ skills. Retrieved 31 May 2010 from http://www.skillsportal.co.za/skills_guide/scarce-skills.htm. Thomson, P. & Walker, W. (Eds) (2010). The Routledge Doctoral Student’s Companion: Getting to grips with research in Education and the Social Sciences. London: Routledge. Trafford, V. & Leshem, S. (2008). Stepping Stones to Achieving your Doctorate. Berkshire: Open University Press. Travis, J. (2009). Is the (Blue) Sky Falling in the UK? ScienceInsider, February 12, 1. 325 University of the Witwatersrand (2006). Strategic Research Plan: 2007 – 2011. Retrieved 8 May 2010 from http://web.wits.ac.za/NR/rdonlyres/F250662B-B4BD- 49DF-9030-B910374585EC/0/StratPlan.doc. University of the Witwatersrand (2010). Facts and Figures. Retrieved 8 May 2010 from http://web.wits.ac.za/AboutWits/WitsToday/WitsFacts.htm. University of the Witwatersrand (2010): Postgraduate Prospectus. Retrieved 8 May 2010 from http://web.wits.ac.za/NR/rdonlyres/B24601EB-45C1-4590-9245- BD5F35FE65E7/0/PostgraduateProspectus.pdf. Usher, R. (2002). A Diversity of Doctorates: fitness for the knowledge economy? Higher Education Research & Development, 21 (2), 143 – 153. Van Den Heever, M. (personal communication. September 7 2010). Vartuli, S. (Ed) (1982). The Ph.D. Experience: A Woman’s Point of View. New York: Praeger. Walker, G.E., Golde, C.M., Jones, L., Conklin Bueschel, A. & Hutchings, P. (2008). The Formation of Scholars: Rethinking Doctoral Education for the Twenty-First Century. San Francisco: Jossey-Bass. Walker, M. & Thomson, P. (2010). The Routledge Doctoral Supervisor’s Companion: Supporting Effective Research in Education and the Social Sciences. London: Routledge. Wang, T. and Li, L. Y. (2009). International research students’ experiences of feedback. The Student Experience, Proceedings of the 32nd HERDSA Annual Conference, Darwin, 6 – 9 July 2009, 444 – 452. 326 Ward, M. & West, S. (2008). Blogging PhD Candidature: Revealing the Pedagogy. International Journal of Emerging Technologies and Society, 6 (1), 60 – 71. Weedon, F. (1991). Feminist Practice & Poststructuralist Theory. Cambridge, Massachusetts: Blackwell Publishing. Wellington, J. (2010). Making Supervision Work for You: A Student’s Guide. London: Sage. Wenger, E. (1998a). Communities of Practice: Learning, Meaning and Identity. Cambridge: Cambridge University Press. Wenger, E. (1998b). Communities of Practice: Learning as a Social System. Systems Thinker, June, 1 – 10. Wenger, E. (2000). Communities of Practice and Social Learning Systems. Organization, 7 (2), 225 – 246. Whittle, J. (1992). Research culture, supervision practices and postgraduate performance. In O. Zuber-Skerritt (Ed). Starting research – supervision and training. Brisbane: Tertiary Education Institute. Winter, G. (2000). A Comparative Discussion of the Notion of ‘Validity’ in Qualitative and Quantitative Research. The Qualitative Report, 4 (3&4), 1 – 112. Wirth, L. (1926). The Sociology of Ferdinand Tönnies. The American Journal of Sociology. 32 (3), 412 – 422. 327 Wisker, G. (2001). The Postgraduate Research Handbook: Succeed with your MA, MPhil, EdD and PhD. Basingstoke: Palgrave Macmillan. Wisker, G. (2005). The Good Supervisor: Supervising Postgraduate and Undergraduate Research for Doctoral Theses and Dissertations. Houndsmills, England: Palgrave Macmillan. WMA Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Subjects. Retrieved 2 September 2010 from http://www.wma.net/en/30publications/10policies/b3/index.html. Wolfson, N. (1976). Speech events and natural speech: some implications for sociolinguistic methodology. Language in Society, 5, 189 – 209. World Bank. (2002). Constructing Knowledge Societies: New challenges for tertiary education. Washington D.C: World Bank. Wright, S. (2004). Markets, corporations, consumers? New landscapes of higher education. Learning and Teaching in the Social Sciences, 1 (2), 71 – 93. Yeatman, A. (1995). Making supervision relationships accountable: Graduate student logs. Australian Universities Review, 2, 9 – 11. Ylijoki, O.-H. (2001). Master’s thesis writing from a narrative approach. Studies in Higher Education, 26 (1), 21 – 35. Zhao, F. (2003). Transforming Quality in Research Supervision: a knowledge management approach. Quality in Higher Education, 9 (2), 187 – 197.