Sage Journals: Discover world-class research

Abstract

In this article the authors report on an approach that they used to enhance their understanding of the complex nature of doctoral supervision by analyzing e-mail communication within a supervisory relationship. Although some scholars have discussed research supervision, empirical research on the subject is limited, and the authors found no published attempts to explore doctoral supervision through the analysis of e-mail communication. The authors analyze correspondence between one doctoral student and two supervisors using discourse analysis influenced by the Foucauldian notion of disciplinary power. The findings revealed the discourses of unity and detachment operating throughout the course of the doctoral relationship. The authors suggest that research students might be no less detached from their supervisors on completion of their studies than at the beginning of their relationship and argue that understanding the discourses of doctoral supervision can enhance the quality and successful outcome of the experience.

Keywords

doctoral supervision power unity detachment research student Foucault

Introduction

In this paper we report on a discourse analysis study that we undertook in relation to research supervision. Our aim was to explore the taken-for-granted meanings and practices within our own doctoral relationship to enhance our understanding of this complex pedagogy. The issue of research supervision has captured the attention of scholars globally, with literature on the subject emanating from the United Kingdom (Bradbury-Jones, 2007; Denicolo, 2004; Hockey, 1995, 1997; Jones, 1999; Sambrook, Stewart, & Roberts, in press), Australia (Bartlett & Mercer, 2000; Conrad, 2003; Cullen, Pearson, Saha, & Spear, 1994; Gatfield, 2005; Green, 2005), New Zealand (Grant, 2005; Grant & Graham, 1999; McMorland, Carroll, Copas, & Pringle, 2003), and Canada (Chapman & Sork, 2001; Sullivan & Ogloff, 1998). However, despite the apparent widespread interest in research supervision, it is rarely the site of analysis (Chapman & Sork, 2001). Within available literature, methodologies that underpin some studies include action research/peer partnership inquiry (McMorland et al. 2003), case study (Holligan, 2005), ethnography (Malfroy, 2005), autoethnography (Sambrook, Irvine, & Bradbury-Jones, 2006a, 2006b; Sambrook, Stewart, et al., in press), and personal narrative (Chapman & Sork, 2001). In terms of method, interviews are a popular means of exploring doctoral supervision (Cullen et al., 1994; Gatfield, 2005; Grant, 2005; Hockey, 1995, 1997), as are questionnaires (Conrad, 2003; Cullen et al., 1994; Denicolo, 2004; Grevholm, Persson, & Wall, 2005).

Some researchers have used critical discourse analysis and genealogy (Grant, 2003, 2005) and feminist methodologies (Bartlett & Mercer, 2000) to examine current dominant discourses in research supervision. In this paper we have drawn on the work of Foucault (1995), specifically his notion of disciplinary power, to inform our analysis. We do not claim to have undertaken a Foucauldian discourse analysis per se, however. In fact, to suggest that there is a method for Foucauldian research is somewhat antithetical; Foucault resisted giving such a method (Cheek, 2000), and therefore one has never existed (Hook, 2001). We prefer to describe our method as Foucauldian infused, implying that during our inquiry we were mindful of Foucault's notion of disciplinary power, and, as we will show later, we used this as a point of reference for our analysis. Some discourse analysis studies use a number of textual sources (Turner, Keyzer, & Rudge, 2007), but we share our experience of using a single source: e-mail communication between one research student (Caroline) and two research supervisors (Sally and Fiona). We are not aware of any other studies on doctoral supervision that use e-mail communication as data.

Background

The lack of empirical research on the issue of doctoral supervision is a surprising omission given that the quality of the relationship between student and supervisor is paramount (Bradbury-Jones, 2007; Chapman & Sork, 2001; Grant & Graham, 1999; Phillips & Pugh, 2000; Tinkler & Jackson, 2004). Individual supervisory relationships are complex and dynamic (Cullen et al. 1994; Sambrook et al., in press), and although supervision can be full of possibilities and pleasures, it is a deeply uncertain practice (Grant, 2005). In addition, it is poorly understood (Grant, 2003), often problematic (Grant, 2003; Grant & Graham, 1999), and, as some would say, impossible (McMorland et al., 2003).

This impossibility fueled our curiosity as researchers and stimulated our inquiry into the dynamics of doctoral supervision. It was an inquiry underpinned by desire to enhance our own relationship and add to the existing knowledge base on the subject. Previously we have attempted to augment the limited body of knowledge in relation to doctoral supervision by placing our supervisory relationship as the subject of inquiry (Sambrook, Irvine, et al., 2006a, 2006b). Akin to other researchers who have studied their own relationships, we are aware that some might criticize our approach as self-indulgent (Chapman & Sork, 2001) or self-congratulatory (Bartlett & Mercer, 2000). However, we argue that such an endeavor has benefits beyond our immediate relationship and that sharing the insight gained provides a platform for further discussion and debate among those interested in the issue of research supervision.

The article sits comfortably within the body of literature concerned with enhancing the quality of research supervision. It is written from a nursing perspective, but we foresee that the issues raised might have transference to an array of other disciplines.

Method

One productive way of doing social research is through a focus on language, using a form of discourse analysis (Fairclough, 2003). The term discourse, however, is rather nebulous (Lupton, 1992) and means different things depending on the context. Confusingly, it can be a noun or a verb (Hardin, 2001, 2003). Discourse analysis is the close, critical study of talk and texts (Crowe, 2005; Quested & Rudge, 2003; Taylor, 2001; Traynor, 2006), but texts are always situated in their social, cultural, political, and historical context (Hardin, 2003; Kotecha, 2002; Polit & Beck, 2004). Best understood as a field of research rather than a single practice (Taylor, 2001), discourse analysis is based on the assumption that language is an irreducible part of social life (Fairclough, 2003) and that examining it can help reveal layers beneath taken-for-granted meanings and practices (Crowe, 1998; Horsfall & Cleary, 2000; Traynor, 2006). In turn, the revealing of such taken-for-granted forms, many of which are buried within discourse, can display forms of control, persuasion, and manipulation in the meanings inherent in the discourse (Lupton, 1992).

Language is always embedded in discourse, and it is important to recognize the influence that dominant social and cultural discourses have on the articulation of individual accounts (Hardin, 2003). The originator of the text is seen as unconsciously directed by social norms (Hallett, Austin, Caress, & Luker, 2000). This is what contrasts discourse analysis to humanist approaches, where researchers hold the unspoken assumption that individuals are free acting agents (Hardin, 2001) and language is treated as if it originates with the participant (Allen & Hardin, 2001). In discourse analysis the humanist idea of the whole human actor or agent is not a theme (Mason, 2002). Instead, it requires reading between the language of personal narratives and the social, cultural, and historical discourses from which those narratives drew (Allen & Hardin, 2001).

Discourse analysis can be employed within different epistemological paradigms (Crowe, 2005; Flick, 2006) that have roots in linguistics, cognitive psychology, sociolinguistics, and poststructuralism (Potter, 2004). In the last, associated with Foucault (1980, 1995), power is more in focus than in other versions of discourse analysis (Flick, 2006). The complex relationship between power and knowledge was addressed extensively by Foucault, and one of his main ideas was that knowledge and power are intertwined (Björnsdottir, 2001; Cheek, 2000). Foucault's (1980) thesis was that power exists in capillary form, which means that rather than be distributed top-down, it “reaches into the very grain of individuals, touches their bodies and inserts itself into their actions and attitudes, their discourses, learning processes and everyday lives” (p. 39); it is “exercised rather than possessed” (Foucault, 1995, p. 26). Hence, individuals are not powerless because power is also held by those who are governed in some way (Cheek & Porter, 1997). Thus, within doctoral supervision, for example, although clearly the supervision relationship is one of unequal power and status, both students and supervisors are capable of action (Grant & Graham, 1999).

Another important focus of Foucault's (1995) work was disciplinary power, which, he argued, operates through three processes: hierarchical observation, normalizing judgment, and the examination. Hierarchical observation is about being watched and is where individuals are subject to a constant “gaze” (Gilbert, 1995); in many respects it is like being viewed through a telescope or microscope (Rose, 1998). Observation takes a capillary form, which means that the gaze is cast not only from the top down but also sideways and from the bottom up. For Foucault this observation (gaze) takes two main forms: indiscreet and discreet. It is indiscreet because people are usually aware that they are under some kind of gaze, but in its discreet form people are unaware of the gaze because “it functions permanently and largely in silence” (p. 177).

Normalizing judgment is about being judged and compared to particular norms (Gilbert, 1995; Hardin, 2001). This involves regulation and evaluation of things such as conduct and manners, with judgments made as to conformity or deviation from such norms (Rose, 1998). Societies organize themselves around discourses of the self and, in doing so, engage in self-monitoring, self-reflection, and self-analysis (Allen & Hardin, 2001). In this self-monitoring, we compare ourselves with discourses of what we should be like: we compare and adjust ourselves accordingly. Foucault's (1995) description of normalizing practices reveals how particular behaviors are learned and regulated through such comparisons and how normalizing truths have been constructed, for whom, and for what purpose (Hardin, 2001). In a similar vein to hierarchical observation, judgments are not only from the top down; observed and observer are also tied up in making normalizing judgments.

The examination is a combination of hierarchical observation and normalizing judgment and is a surveillance (normalizing gaze) “that makes it possible to qualify, to classify and to punish” (Foucault, 1995, p. 184). When judgments are made, there is a penalty for deviating from the norm (Foucault, 1995); punishment is the result of failing to measure up to the rule. It is the examination, normalizing judgment, and hierarchical observation that underpin our inquiry into the discourse of doctoral supervision.

In the discussion that follows, we will show how capillary power, hierarchical observation, normalizing judgment, and examination seem to operate in our doctoral relationship through analysis of our communications with each other. Turning to written text for the purpose of analysis might imply that we perceive of discourse as being reduced to language, narratives, and text alone, which, according to Hook (2001), would render our analysis “woefully limited” (p. 530). We are aware that power cannot be fixed within texts and can really be grasped only through extratextual analysis of tactical and material force (Hook, 2001). For this reason, we have considered both text and context, for example, by drawing on the work of scholars globally to gain a broader perspective of doctoral supervision.

Text collection

E-mail communication between the principal investigator and her two research supervisors were collected between October 2004 and March 2007. This 30-month period began with their initial contact and ended midway through the second year of their supervisory relationship. During this period all e-mail correspondence was printed and the hard copies archived. Caroline took responsibility for securely storing the texts until the point when they were retrieved in their entirety to commence analysis.

Text analysis

Analysis was conducted using the literal, interpretive, reflexive framework described by Mason (2002). Although Mason's focus is on interviews, we found that it was congruent with analysis of the e-mail communication in our study. Mason proposed that reading of data can take several forms: it can be literal, in which the interviewer might be interested in the form and sequence of dialogue; interpretive, where inferences are made “outside” the text; and reflexive, whereby the researcher reads something of his or her own role within the text. Analysis can be underpinned by one or all three levels of reading the text. Gilbert (2005) used Mason's framework in a discourse analysis study and produced literal and interpretive accounts. Similarly, although we used all three levels, our focus was interpretive in the sense that to explore the texts discursively, they were “interrogated to uncover the unspoken and unstated assumptions implicit within them” (Cheek, 2000 p. 43).

Issues of rigor

There is little agreement about how to achieve rigor in discourse analysis (Nixon & Power, 2007), but “good” discourse analysis can be marked by two features: Quotations of text are included in the analysis (Lupton, 1992; Traynor, 2006), and authors do more than just summarize themes (Traynor, 2006). In addition, clarity in terms of research question, type of discourse analysis used, epistemological and ontological positioning, and analytical strategies is important (Nixon & Power, 2007). Using these markers of quality, we have aimed at transparency. We have included extensive excerpts from our communication to facilitate validation by the reader and we engage in a critical analysis of the findings. The excerpts that we share in this paper were initially selected by Caroline and later agreed with Sally and Fiona as representative of the dominant discourses of our supervisory relationship.

Findings and discussion

Two discourses were dominant in our analysis: unity and detachment. In the discussion that follows, we examine how these are reflected through our communication with each other, from the beginning of our relationship to the point of drafting this article. We begin with communication from Sally and Fiona to Caroline during the first 6 months of our relationship, a time of Caroline's preparation for registering for a doctoral program:

09/11/2004

Dear Caroline

… I have now had a chance to speak with Fiona about your proposed study. We think it might be beneficial for all three of us to meet to discuss your proposed topic in a little more depth … The dates we are both free are (two options) … I look forward to meeting you …

Best wishes,

Sally

26/05/2005

Dear Caroline

Sally and I have reviewed your proposal and our comments are attached. We think that it would be good to meet now and discuss the proposal and our comments. We can offer you two dates … See you soon Fiona

The discourse of unity is reflected by the united position of Sally and Fiona as supervisors and the relative detachment of Caroline as the research student. Unity is reflected by Sally and Fiona's use of we as supervisors: They have met with each other and have discussed and agreed on comments on Caroline's research proposal, but in her absence. Simultaneously, the discourse of detachment is reflected by Sally and Fiona's availability; they have some dates that they are able to “offer” Caroline to meet. In addition, Caroline's communication to her united supervisors reinforces their position as experts:

16/06/2005

Dear Fiona and Sally

Please find the second draft of my proposal. I found your comments and suggested reading invaluable and hope that this is reflected in the proposal … there are bound to be inconsistencies that you will see, but that I can't! Of course I will welcome your comments … Caroline

This “supervisor lead,” “expert” style of supervision is not uncommon and has been reported by others (Cullen et al., 1994; Jones, 1999). New students tend to expect supervisors to tell them what to do (Cryer, 2000), and in the getting-started phase students require such expert guidance (Jack, 1999). Some authors have pointed to the powerful position of supervisors in relation to students (Grant & Graham, 1999; Jones, 1999; Phillips & Pugh, 2000; Sullivan & Ogloff, 1998), the results of which can be manipulation and exploitation (Sullivan & Ogloff, 1998) and abuse of power (Grant & Graham, 1999) on the part of supervisors. However, viewed from the perspective of capillary power, our analysis shows that students also have power to act (Grant, 2003), as will be illustrated later in the paper.

The discourses of unity and detachment create a conundrum in the early stages of a doctoral relationship, and students report a tension between getting enough support and being independent (Cullen et al., 1994). Supervisors expect their students to be independent (Phillips & Pugh, 2000). Although this seems like a reasonable expectation, meeting too frequently will not allow time for progress and can be dispiriting for student and supervisors (Thompson, Kirkman, Watson, & Stewart, 2005). Paradoxically, supervisors are responsible for ensuring that students develop independency (Grevholm et al., 2005), yet meeting too infrequently can lead to difficulties (Thompson et al., 2005).

As the candidature progressed, the texts reflect a more negotiated relationship, in the sense that in relation to supervision meetings, dates are no longer “offered” by Sally and Fiona based solely on their availability; Caroline is included in the negotiation of a mutually convenient date:

24/01/2006

Hi Caroline

… Perhaps we could meet up to discuss how we are going to approach this. Are you coming over in the next week or so? Speak soon Sally

Students need skills of negotiation to navigate their candidature (Cullen et al., 1994), and supervision should become a two-way dialogue (Cryer, 2000). This requires developing a common language with assumptions that are shared and agreed (Murray, 2002). The following text reflects an increasing unity between Caroline, Sally, and Fiona, and “we” now refers to a triad:

20/07/2006

Hi Caroline

… I can't remember what dates I gave you, but next week I'm free on the following so hopefully we can all meet up … Hope one of these is OK for us all.

See you soon

Sally

During the first year of the candidature, we began to coauthor papers for publication, and in the name of Caroline's eventually becoming a detached, independent scholar, she produced first drafts of papers, which were revised in agreement with Sally and Fiona. Authorship is often a contentious issue (Cullen et al., 1994; Sullivan & Ogloff, 1998; Thompson et al., 2005). Some students feel pressured to include their supervisors or to give them first authorship when this is unwarranted (Sullivan & Ogloff, 1998), although, as indicated above, this was not the case in our particular relationship. The following text refers to a draft paper, a productive element of the increasing unity reflected in the tripartite relationship. Paradoxically, although writing for publication was a collaborative venture, the text reveals a strong detachment discourse, whereby Caroline demonstrates increasing confidence in asserting her views while still requiring supervisory support:

04/07/2006

Hello both

I have attached a copy of the paper that I have had a go at amending … although I think it is much improved, I am beginning to risk losing focus (in fact, I'm beginning to risk losing the will to live!). I don't know what your thoughts are, but I think we should submit it otherwise we could end up making changes ad infinitum.

Look forward to hearing from you.

Thanks

Caroline

26/07/2006

Hi both

The comments are really helpful but I don't know whether to set about addressing the changes myself, or whether to wait until we get together sometime. Only problem is that it will be ages until we get together. What do you think?

Caroline

An increasingly autonomous, detached student voice is illuminated:

29/11/2006

Hello

I have attached the revised paper. I must say that although it has been revised and revised this slimmed down version is better … I am actually reasonably happy with this (near … hopefully?) end result. What about you?

Caroline

Detachment and autonomy are reflected in the above text, but so, too, is Caroline's ability to exert power through persuasiveness in an attempt to gain the acquiescence of her supervisors to the submission of an article. Similarly, in relation to written work of doctoral students, Grant (2003) observed that students can exercise power by requiring supervisors to read many drafts, which in terms of capillary power demonstrates how both supervisor and student hold the potential to exercise power.

In addition to heralding collaborative writing, progression of the relationship signaled increasing unity to the point of friendship. The lunchtime meeting below refers to meeting socially over lunch, and unity is reflected in the relaxed tone: the address is a mere “hi,” and Caroline uses her abbreviated name to sign off:

06/10/2006

… I am around quite a bit over the next few weeks, so some scope for a lunchtime meeting (?).

Have a great weekend.

Carrie

In relation to the discourse of unity, our developing supervisory relationship revealed a different kind of becoming: that of friends. Traditionally in the United Kingdom, most supervision has been based on the “secret garden” model (Park, 2006, 2007), one in which supervision occurs behind closed doors, away from the public gaze (Park, 2006). Graduate education is frequently portrayed as an intimate, private pedagogical relationship between a supervisor and a student (Aguinis, Nesler, Quigley, Lee, & Tedeschi, 1996; Grant & Graham, 1999), and several authors have commented on the potential of positive relationships (Boucher & Smyth, 2004; Cryer, 2000; Sullivan & Ogloff, 1998). Certainly, supervisory relationships often involve activities of a social nature (Hockey, 1995), and gender, age, ethnicity and professional status are factors likely to influence the nature of supervisory relationships (Jones, 1999). As alluded to earlier, as we were three females of similar age, ethnicity, and social backgrounds, our friendship is perhaps not surprising. Analysis of gender issues within doctoral supervision would take us beyond the scope of this paper, but we cannot ignore that within our female supervisory relationship, such issues contribute to our positions. This is particularly significant given that there is evidence that same-gender relationships are advantageous to students (Kurtz-Costes, Andrews Helmke, & Ülkü-Steiner, 2006; Schroeder & Myatt, 1999).

Although friendships are a delightful bonus (Boucher & Smyth, 2004), conflicts may arise. A potential problem relates to the issue of feedback on academic work, and although this has never been particularly problematic in our relationship, it is an issue discussed in the literature. The role of critic is central to supervisory practice (Hockey, 1997), and the supervisor plays the role of “critical friend” (Grevholm et al., 2005). Students should receive detailed criticism and advice from their supervisors (Dunleavy, 2003) and advice to ensure that work is of the “correct standard” (Jack, 1999, p. 24). Supervisors cast over the students “a normalizing gaze, a surveillance that makes it possible to qualify, to classify and to punish” (Foucault, 1995, p. 184). In this context “punishment” is the ability of the supervisor to provide either positive or negative feedback, and, as previous discussion regarding friendship demonstrates, this creates more difficulty on both sides when feedback is critical (Cryer, 2000).

Another potential problem of friendship is the negative influence on the supervisor's capacity to fulfill his or her intellectual responsibility (Hockey, 1995). There is a danger of supervisors spoon-feeding students (Thompson et al., 2005) or jeopardizing their obligation to remain objective with respect to the student's personal development (Sullivan & Ogloff, 1998). As our supervisory relationship progressed, the text reflected a discourse of unity around the issue of writing for publication. Here, the notion of becoming was one of Caroline's becoming an author; an area of becoming discussed by others (Dunleavy, 2003). The potential for tensions between the competing discourses of unity and detachment were evident, with tensions between friendship and collaborative writing, on one hand, and the need for Caroline to progress as an independent, detached scholar, on the other.

The rubric for award of a doctorate is that it is the student's own work, but the discourse of unity holds potential for supervisors to transgress formal regulations concerning the autonomy of the student's thesis and contribute too much to its production (Hockey, 1995, 1997). The detached student should develop his or her own ideas (Cryer, 2000) and work autonomously (Thompson et al., 2005). Mindful of this, we placed high emphasis on Caroline's producing initial drafts of papers independently of Sally and Fiona in a bid to meet the criteria for doctoral study. In terms of disciplinary power, supervisors are thus called on to supervise within acceptable norms and cast a normalizing gaze on themselves. In this context disciplinary power “constantly supervises the very individuals who are entrusted with the task of supervising” (Foucault, 1995, p. 177).

The texts reveal a careful management of boundaries associated with a developing friendship. Some authors highlight possible tensions of becoming too friendly and suggest a separation of the roles of friend and supervisor in the event of role conflict (Sambrook et al., in press). Students cannot risk alienating their supervisors, and multiple relationships between students and their supervisors are potentially harmful (Sullivan & Ogloff, 1998). An added dimension in our relationship is the nature of our colleague supervision, something that Denicolo (2004) referred to as a “special case” (p. 693); at the time of collection of texts for this study, we were all colleagues within the same university. These kinds of relationships can be particularly problematic (Boucher & Smyth, 2004) as they have inherent tensions brought about by the multiplicity of competing roles and agendas.

Setting boundaries is one of the ways of dealing with the complexity of the close relationships (Boucher & Smyth, 2004). Students build up a sense of what can and cannot be said and how it might be said (Dunleavy, 2003); from a Foucauldian perspective, they learn to watch themselves. This kind of self-regulation and “discipline” is evidenced below as Caroline disciplines herself in relation to acceptable boundaries:

29/07/2006

Sally and Fiona

… I'm not contacting you again this weekend. I will leave you in peace and re-submit the paper on Monday.

Thanks

Carrie

08/08/2006

… will contact you again when I have got something together and for now I will leave you in peace.

Thanks

Carrie

The texts suggest a transition stage marked by the complexity of a developing collegiality, friendship, and unity of Caroline, Sally, and Fiona and a corresponding detachment of Caroline as an increasingly independent scholar. Other researchers have discussed the notion of becoming, a suggestion of transformation over time to becoming something other than what one already is (Barnacle, 2005). Transition is marked by increasing independence and responsibility, with control shifting gradually from supervisor to student (Thompson et al., 2005). Some have referred to this gradual detachment as weaning (Cryer, 2000; Phillips & Pugh, 2000) or have used the analogy of parenting to describe the gradual separation of supervisor and student (Cullen et al., 1994). The outcome of supervision is for the student to become a researcher (Grant, 2003). In terms of our relationship, Caroline's increasing detachment from Sally and Fiona was, arguably, therefore, a necessary, positive, and productive aspect of supervision.

Looming large in any supervisory relationship is the terminus: the examination/viva. Some way into the second year of our relationship, that very issue was raised:

21/02/2007

Fiona

I would love to be done-and-dusted for Jan 2009, but don't want to jeopardise the quality of the study by rushing the last bit, just for the sake of a speedy submission. I DO want to try and be done for graduation in 2009, so what is the latest date I can submit to achieve that? What is the normal timeframe for submission, examination, amendments etc? Asking these questions makes me very nervous …

Carrie

21/02/2007

Hi Carrie

… I guess we can talk more about this tomorrow and check things out with registry.

The excerpt above reflects a nervous Caroline faced with the final hurdle of the doctoral examination. The examination marks the final detachment, and several authors have referred to the rite of passage, a gateway to the discipline (Cullen et al., 1994; Dunleavy, 2003; Leonard, 2001; Tinkler & Jackson, 2004). As the privileged form of institutionalized knowledge the thesis is meant to be big, formal, disciplined, and original (Grant, 2003); similarly, the entire doctoral process is meant to be a challenge. Although it is a PhD, not a Nobel Prize (Mullins & Kiley, 2002), it is the fact that undertaking a doctorate is such a challenge that justifies the award (Phillips & Pugh, 2000). Because the oral examination is usually held privately (Phillips & Pugh, 2000; Tinkler & Jackson, 2004), it is a process that is “shrouded in mystery” (Tinkler & Jackson, 2004, p. 2). The viva is a pretty daunting experience (Leonard, 2001) because the thesis will be judged (Cryer, 2000) and the examiners will be the judges (Thompson et al., 2005).

Personal horror stories (Leonard, 2001) and grapevine stories (Tinkler & Jackson, 2004) combine to encourage supervisors and students to unite in the face of the challenge. Therefore, although the doctoral examination marks the final hurdle and epitomizes autonomy and detachment of the scholar, still, the discourse of unity exists, marked by the collegiality and unity of supervisors and student in preparation for the challenge. The capillary form of power is evidenced yet again within the examination process, in that the gaze of the examiner in doctoral examination is not restricted to the learner (Lovat, Monfries, & Morrison, 2004); the examiner casts a gaze over student and supervisors as a “unit.” In addition, in a similar vein to the requirement of supervisors to read draft copies of students' work, in spite of heavy workloads most examiners spend several days examining a thesis (Mullins & Kiley, 2002).

The findings reveal that at the beginning of the doctoral relationship Sally and Fiona were united in relation to Caroline's relatively detached position as a novice research student. Supervision is imbued with an expectation of independence and autonomy (Grant & Graham, 1999). One of the aims is to transform the student into an independent researcher (Grant, 2003), one who is capable of continuing with research (Grevholm et al., 2005). This was reflected in our analysis, with Caroline's gradual detachment as an independent scholar. In some sense, the analysis of our doctoral relationship reveals that Caroline was always detached to a greater or lesser extent.

As the relationship developed, it became less hierarchical and more negotiated. Joint publications, developing friendships, and preparation for examination collectively reflected a strong discourse of unity. Thus, discourses of unity and detachment operated in tandem, albeit in different forms throughout the period of our analysis. Paradoxically, and confusingly, it appears that as Caroline becomes a detached student, she is simultaneously attached, indeed more attached than at the genesis of the relationship. We have infused our analysis with Foucault's (1980, 1995) notion of disciplinary power, and we have discussed how power operates in capillary form. The final texts allude to this paper, and we present them as ones that capture the essence of our analysis. First, Caroline's communication to Sally and Fiona:

10/02/2007

I have started on the email communication paper and it is fascinating! Will discuss all when I come up … be careful, what you say and how you say it in the meantime! (only joking!)

Carrie

Followed promptly by Sally's reply:

10/02/2007

Aha! How shall I respond to thee now? Oh well … it's probably too late to change things now … enjoy the rest of the weekend

Sal

(or should I say PhD Supervisor, Director of … etc???)

Study limitations

This article is written from a nursing perspective within the United Kingdom, and we are aware of the multiplicity of models of research supervision within different disciplines and different countries. Indeed, Sally and Fiona supervise other doctoral students in different disciplines and different countries, and many of the issues raised here are apparent and relevant in these different contexts. Furthermore, our supervisory relationship is a positive one, and the article might not adequately reflect the more negative aspects of doctoral supervision that other students and supervisors might encounter. However, we do not present this article as an authoritative version of doctoral supervision, but, rather, a platform for further discussion/analysis.

Conclusion

Although some scholars have tried to reveal the unspoken dimensions of supervision (McMorland et al., 2003; Sambrook, Irvine, et al., 2006a, 2006b), the supervisory relationship is seldom problematized publicly (Boucher & Smyth, 2004; Chapman & Sork, 2001). Our view is that rather than taking supervision for granted, it is useful to ask which elements embedded in the process are problematic, for whom, and why, and how things can be different (Grant & Graham, 1999). The unsettling of experiences and assumptions that this generates can be used to develop a keener understanding of the complexities of doctoral supervision (McMorland et al., 2003). Like others, our intention has been to make the familiar strange (Grant, 2005; Meadmore, Hatcher, & McWilliam, 2000).

We used text in the form of archived e-mail communication to analyze and unsettle our experiences. Unity and detachment were dominant discourses in our relationship, but we are eager to state that many other possibilities exist. On this basis, we recommend that other researchers consider using a similar approach to explore their own supervisory relationship. We suggest that the article has international relevance because although different models of research supervision exist globally, many issues transcend geographical boundaries. Within the discipline of nursing we still have much to learn in relation to supervision, as is no doubt the case in other disciplines: There is more to be said, more to think about regarding the subject (Green, 2005).

References

Allen

Hardin

(2001). Discourse analysis and the epidemiology of meaning. Nursing Philosophy, 2, 163–176.

Aguinis

Nesler

Quigley

Lee

S.-J.

Tedeschi

(1996) Power bases of faculty supervisors and educational outcomes for graduate students. Journal of Higher Education, 67(3), 267–297.

Barnacle

(2005). Research education ontologies: Exploring doctoral becoming. Higher Education Research & Development, 24(2), 179–188.

Bartlett

Mercer

(2000). Reconceptualising discourses of power in postgraduate pedagogies. Teaching in Higher Education, 5(2), 195–204.

Björnsdottir

(2001). Language, research and nursing practice. Journal of Advanced Nursing, 33(2), 159–166.

Boucher

Smyth

(2004). Up close and personal: Reflections on our experience of supervising research candidates who are using personal reflective techniques. Reflective Practice, 5(3), 345–356.

Bradbury-Jones

(2007). Exploring research supervision through Peshkin's I's: The yellow brick road. Journal of Advanced Nursing, 60(2), 220–228.

Chapman

V.-L.

Sork

(2001). Confessing regulation or telling secrets? Opening up the conversation on graduate supervision. Adult Education Quarterly, 51(2), 94–107.

Cheek

(2000). Postmodern and poststructural approaches to nursing research. Thousand Oaks: Sage.

10.

Cheek

Porter

(1997) Reviewing Foucault: Possibilities and problems for nursing and healthcare. Nursing Inquiry, 4, 108–119.

11.

Conrad

(2003). Five ways of enhancing the postgraduate community: Student perceptions of effective supervision and support. Retrieved November 11, 2006, from http://surveys.canterbury.ac.nz/herdsa03/pdfsref/Y1033.pdf.

12.

Crowe

(1998). The power of the word: Some post-structural considerations of qualitative approaches in nursing research. Journal of Advanced Nursing, 28(2), 339–344.

13.

Crowe

(2005). Discourse analysis: Towards an understanding of its place in nursing. Journal of Advanced Nursing, 51(1), 55–63.

14.

Cryer

(2000). The research student's guide to success. Berkshire, UK: Open University Press.

15.

Cullen

Pearson

Saha

L. J.

Spear

R. H.

(1994). Establishing effective PhD supervision. Canberra, Australia: AGPS.

16.

Denicolo

(2004). Doctoral supervision of colleagues: Peeling off the veneer of satisfaction and competence. Studies in Higher Education, 29(6), 693–707.

17.

Dunleavy

(2003). Authoring a PhD. Basingstoke, UK: Palgrave Macmillan.

18.

Fairclough

(2003). Analysing discourse. Oxon, UK: Routledge.

19.

Flick

(2006). An introduction to qualitative research (3rd ed.). London: Sage.

20.

Foucault

(1980) Power/knowledge: Selected interviews and other writings 1972–1977. ( Gordon

Ed.). New York: Pantheon.

21.

Foucault

(1995) Discipline and punish: The birth of the prison ( Sheridan

, Trans.). New York: Vintage.

22.

Gatfield

(2005). An investigation into PhD supervisory management styles: Development of a dynamic conceptual model and its managerial implications. Journal of Higher Education Policy and Management, 27(3), 311–325.

23.

Gilbert

(1995). Nursing: Empowerment and the problem of power. Journal of Advanced Nursing, 21, 865–871.

24.

Gilbert

(2005) Impersonal trust and professional authority: Exploring the dynamics. Journal of Advanced Nursing, 49(6), 568–577.

25.

Grant

(2003). Mapping the pleasures and risks of supervision. Discourse: Studies in the Cultural Politics of Education, 24(2), 175–190.

26.

Grant

(2005). Fighting for space in supervision: Fantasies, fairytales, fictions and fallacies. International Journal of Qualitative Studies in Education, 18(3), 337–354.

27.

Grant

Graham

(1999). Naming the game: Reconstructing graduate supervision. Teaching in Higher Education, 4(1), 77–89.

28.

Green

(2005). Unfinished business: Subjectivity and supervision. Higher Education Research & Development, 24(2), 151–163.

29.

Grevholm

Persson

L.-E.

Wall

(2005). A dynamic model for education of doctoral students and guidance of supervisors in research groups. Educational Studies in Mathematics, 60, 173–197.

30.

Hallett

C. E.

Austin

Caress

Luker

K. A.

(2000). Community nurses' perceptions of patient “compliance” in wound care: A discourse analysis. Journal of Advanced Nursing, 32(1), 115–123.

31.

Hardin

(2001) Theory and language: Locating agency between free will and discursive marionettes. Nursing Inquiry, 8, 11–18.

32.

Hardin

(2003). Constructing experience in individual interviews, autobiographies and on-line accounts: A poststructural approach. Journal of Advanced Nursing, 41(6), 536–544.

33.

Hockey

(1995). Getting too close: A problem and possible solution in social science PhD supervision. British Journal of Guidance and Counselling, 23(2), 199–210.

34.

Hockey

(1997). A complex craft: United Kingdom PhD supervision in the social sciences. Research in Post-Compulsory Education, 2(1), 45–70.

35.

Holligan

(2005). Fact and fiction: A case history of doctoral supervision. Educational Research, 47(3), 267–278.

36.

Hook

(2001). Discourse, knowledge, materiality, history: Foucault and discourse analysis. Theory & Psychology, 11(4), 521–547.

37.

Horsfall

Cleary

(2000). Discourse analysis of an “observation levels” nursing policy. Journal of Advanced Nursing, 32(5), 1291–1297.

38.

Jack

(1999). Staying the course: The supervisee's personal experience of the supervision relationship. Nurse Researcher, 6(4), 19–27.

39.

Jones

(1999). Significant relationships: Planning for effective research supervision. Nurse Researcher, 6(4), 5–17.

40.

Kotecha

(2002). Exploring nurse learner wastage/persistence using a discursive approach: Towards a theoretical understanding of the subject. Journal of Advanced Nursing, 40(2), 210–217.

41.

Kurtz-Costes

Andrews Helmke

Ülkü-Steiner

(2006). Gender and doctoral studies: The perceptions of PhD students in an American university. Gender and Education, 18(2), 137–155.

42.

Leonard

(2001). A woman's guide to doctoral studies. Buckingham, UK: Open University Press.

43.

Lovat

Monfries

Morrison

(2004). Ways of knowing and power discourse in doctoral examination. International Journal of Educational Research, 41, 163–177.

44.

Lupton

(1992). Discourse analysis: A new methodology for understanding the ideologies of health and illness. Australian Journal of Public Health, 16(2), 145–150.

45.

Malfroy

(2005). Doctoral supervision, workplace research and changing pedagogic practices. Higher Education Research & Development, 24(2), 165–178.

46.

Mason

(2002). Qualitative researching (2nd ed.). London: Sage.

47.

McMorland

Carroll

Copas

Pringle

(2003, May). Enhancing the practice of PhD supervisory relationships through first-and second-person action research/peer partnership inquiry. Forum Qualitative Sozialforschung/ Forum: Qualitative Social Research, 4(2). Retrieved November 15, 2006, from http://www.qualitative-research.net/fqs-texte/2–03/2–03mcmorlandetal-e.htm.

48.

Meadmore

Hatcher

McWilliam

(2000). Getting tense about genealogy. Qualitative Studies in Education, 13(5), 463–476.

49.

Mullins

Kiley

(2002). “It's a PhD, not a Nobel Prize”: how experienced examiners assess research theses. Studies in Higher Education, 27(4), 369–386.

50.

Murray

(2002). How to write a thesis. Berkshire, UK: Open University Press.

51.

Nixon

Power

(2007) Towards a framework for establishing rigour in a discourse analysis of midwifery professionalisation. Nursing Inquiry, 14(1), 71–79.

52.

Park

(2006). The end of the secret garden: Reframing postgraduate supervision. Retrieved March 2, 2007, from http://luvle.lancs.ac.uk/CELT/supervision.nsf/f/A4F36FBFB1508827802571F60064C0B8/File/ChrisPark.pdf.

53.

Park

(2007). Redefining the doctorate. York, UK: Higher Education Academy.

54.

Phillips

Pugh

(2000). How to get a PhD: A handbook for students and their supervisors (3rd ed.). Buckingham, UK: Open University Press.

55.

Polit

Beck

C. T.

(2004). Nursing research: Principles and method (2nd ed.). Philadelphia: Lippincott Williams & Wilkins.

56.

Potter

(2004). Discourse analysis as a way of analysing naturally occurring talk. In Silverman

(Ed.), Qualitative research: Theory, method and practice (2nd ed., pp. 200–221). London: Sage.

57.

Quested

Rudge

(2003). Nursing care of dead bodies: A discursive analysis of last offices. Journal of Advanced Nursing, 41(6), 553–560.

58.

Rose

(1998). Inventing our selves: Psychology, power and personhood. Cambridge, UK: Cambridge University Press.

59.

Sambrook

Irvine

Bradbury-Jones

(2006a, April). Exploring doctoral supervision from three perspectives: A collaborative autoethnography employing the critical incident technique. Paper presented at the 12th Qualitative Health Research Conference, Edmonton, Canada.

60.

Sambrook

Irvine

Bradbury-Jones

(2006b, July). Exploring doctoral supervision from above, amidst and below: A collaborative autoethnography. Paper presented at the 7th Advances in Qualitative Methods Conference, Surfers Paradise, Australia.

61.

Sambrook

Stewart

Roberts

(in press). Doctoral supervision: A view from above, below and the middle. Journal of Further and Higher Education.

62.

Schroeder

Myatt

(1999). Graduate students' relationships with their male and female major professors. Sex Roles, 40, 393–420.

63.

Sullivan

Ogloff

(1998). Appropriate supervisor-graduate student relationships. Ethics and Behavior, 8(3), 229–248.

64.

Taylor

(2001). Locating and conducting discourse analytical research. In Wetherell

Taylor

Yates

(Eds.), Discourse as data: A guide for analysis (pp. 5–48). London: Sage/Open University Press.

65.

Thompson

D. R.

Kirkman

Watson

Stewart

(2005). Improving research supervision in nursing. Nurse Education Today, 25, 283–290.

66.

Tinkler

Jackson

(2004). The doctoral examination process: A handbook for students, examiners and supervisors. Berkshire, UK: Open University Press.

67.

Traynor

(2006). Discourse analysis: Theoretical and historical overview and review of papers in the Journal of Advanced Nursing 1996–2004. Journal of Advanced Nursing, 54, 62–72.

68.

Turner

Keyzer

Rudge

(2007). Spheres of influence or autonomy? A discourse analysis of the introduction of nurse practitioners in rural and remote Australia. Journal of Advanced Nursing, 59, 38–46.

Unity and Detachment: A Discourse Analysis of Doctoral Supervision

Abstract

Keywords

Introduction

Background

Method

Text collection

Text analysis

Issues of rigor

Findings and discussion

Study limitations

Conclusion

References