Doctoral supervision: sharpening the focus of the practice lens

ABSTRACT

The literature on doctoral supervision is heavily informed by a focus on the individual and dyad, together with a self-help ethic of supervisory improvement. On the margins is a disparate literature taking a ‘practice’ perspective on doctoral supervision. But this literature is disconnected and lacking in some important features. This article’s intention is to sharpen the focus and so to enhance the utility of a social practice theory lens. It refutes the idea that the practice perspective is inherently conservative, showing how a sharply focused lens can challenge assimilationism. The article first elaborates on the individualist approach and highlights differences with the practice approaches as already developed. It then sets out a social practice approach which builds on that existing literature. Finally, it applies this to the issue of colonialist assimilationism in doctoral supervision. The article is not directly based on empirical research, but instead uses existing literature to indicate ways in which a social practice approach might more effectively address the enhancement of doctoral supervision.

KEYWORDS:

• Social practice theory
• doctoral supervision
• assimilationism
• individualism

Introduction

This article argues that, while there has been a strand in the literature on doctoral supervision involving social practice theory, to date this work has been disconnected and partial. Meanwhile, there is a wealth of literature which adopts an individualistic perspective, concentrating on the supervisor, supervisee and this dyad’s interactions, experiences and behaviours. Taken together and developed, these two perspectives can provide an elaborated perspective which sees these interactions through a practice lens. As Hopwood (2018, p. 16) says ‘ … there remains untapped potential in researching and enacting doctoral education in ways informed by a theorisation of these doings, sayings and relatings as practices’. But that lens also needs to embrace ‘ … additional factors and settings’ (Cumming, 2010, p. 37). McAlpine and Norton (2006, p. 3) agree that understanding doctoral work ‘has not been conceived and implemented with sufficient breadth to integrate factors influencing [doctoral] … outcomes’.

Like Kelly (2017), I argue that an additional move is required, beyond adopting a practice perspective and being able to zoom out to broader influences. That is ‘panning up’ to apprehend the influence of national policies, incentives and sanctions and to consider the power of the ‘global policy ensemble’ (Ball, 2008). But beyond the policy domain, panning up also involves appreciating the significance of the ideational realm, the reservoirs of educational ideologies, symbolic and assumptive structures drawn on by candidates and supervisors as they perform practices. Recognising the significance of these and addressing their effects is the key to changing practices; this lies in the hands of policy-makers, institutional leaders and practitioners themselves.

This argument is elaborated below and then applied to the issue of the colonial assimilationism characteristic of much supervisory work, that is the absorption by doctoral researchers from the global South of colonialist forms of saying, knowing and practising (Manathunga, 2017). The article demonstrates how a practice lens has value in unfurling the nature of what is going on in this area, and how ‘what is going on’ might be changed in ways that go beyond pleas to behave better.

My approach is unusual in that it is not directly based on empirical research, but instead illuminates a path to a fruitful practice approach which aims to enhance doctoral supervision. To do this, the article uses findings and arguments from the literature on doctoral supervision, and from social practice theory to develop conceptual clarification and a reframed perspective.

Individualist accounts of doctoral activities

At the centre of doctoral supervision are just two people: the supervisor and the supervisee, at least in some disciplines and countries. The individualism of focusing just on them means that enhancement initiatives, for example, to increase rates of timely completion, widen access or improve doctoral students’ experiences, should focus on individual supervisors’ skills and attitudes (Jara, 2020). From this perspective, supervision appears to be infertile ground for analysis through a social practice lens and for attempts to enhance practices based on such analysis.

The literature on supervision is replete with such individualist accounts, asking questions that require only descriptive answers. Examples are: how do students experience the supervision process? (Hutt et al., 1983); how does their identity change? (Mawson & Abbott, 2017; Shultze, 2014); how is ‘doctorateness’ developed? (Boberg & Devine, 2016); what characteristics of supervision do students value? (Evans & Stevenson, 2011).

The same is true in relation to research on staff. Research questions usually take this form: what different approaches do supervisors take? (Carr et al., 2010); how are supervisors trained and developed? (Nichol et al., 2018); how do supervisors foster rapport with their students? (Rowan & Grootenboer, 2016); what attitudes do supervisors have about the purposes of the doctorate? (Åkerlind & McAlpine, 2017); how can supervisors improve their mindfulness traits? (Buirski, 2021) . Finally, in considering both students and staff together, examples based on an individualist frame include: what are the differences in perceptions and experiences of supervisors and supervisees? (Lee, 2009); how is power and emotion expressed and experienced within the relationship? (Doloriert et al., 2012); what are good online doctoral supervisor-supervisee processes? (Gray & Crosta, 2018).

This individualised perspective means that attempts to offer recommendations for the enhancement of supervision practices are frequently normative and aspirational without a pathway to achievement other than self-help. Usually, these are orientated to adopting the kinds of attitudes, behaviours and choices said to be characteristic of ‘ideal’ supervisors or doctoral researchers. Gray and Crosta’s (2018) systematic review of the literature in this area is replete with such appeals for supervisors and supervisees to be better people. There are appeals to be more creative, more flexible, better communicators, more culturally aware. Similar idealised descriptive versions of individuals are found in some chapters of Rowan and Grootenboer (2016).

This framing offers an attenuated version of people and contexts, seeing through a ‘de-contextualised, psychological lens’ (Bastalich, 2017, p. 1145). Real people in real places behave not only in consistently rational ways as they face difficult situations; they are influenced moment-by-moment by previous experiences and emotional responses not wholly under their control (Haidt, 2012). Moreover, the individualism of this framing sees agents’ attitudes and choices as the prime generative power in the social world. By contrast, applying practice theory, particularly the concept of teaching and learning regimes (Trowler, 2020; Trowler & Cooper, 2001), uncovers the forces shaping recurrent practices in this field. I move to this next.

The practice perspective on doctoral activities

A social practice perspective sees interpersonal interactions of doctoral candidate and their supervisor as the centre of a nexus of influences. ‘Social practices’ involve recurrent and organised constellations of multiple people’s activities aimed at accomplishing a set of (not necessarily mutual) goals (Reckwitz, 2002; Schatzki, 1996). In doctoral supervision, examples are: supervisory meetings; giving and receiving feedback on writing; talking about research; engaging with the research literature. Doctoral supervision, therefore, includes multiple activity types. Its components are underpinned by forms of knowledge, symbolic structures, ideologies, discourses, assumptions, values, expectations and subjectivities which are often in tension and are always multiple.

Valuable foundations for a social practice approach to doctoral activities, including supervision, have already been laid. Cumming (2007) describes a matrix which encapsulates the many different forms of practices involved (p. 116). Rönnerman and Kemmis (2016, p. 105) also disaggregate the different practices involved to demonstrate how an institutional practice architecture shapes, limits and holds them in place. Together with other authors (Hopwood, 2018; Lee & Boud, 2008; Manidis & Goldsmith, 2018), Cumming conceptualises supervisory and other practices as bundled with other forms of practice, each of which is an open, flexible system (Cumming, 2007, p. 87) and so influences the others (Cumming, 2007, 2008, 2010). They are ‘interconnected – mutually entangled and co-implicated such that changes in one have implications for the others’ (Hopwood, 2018, p. 16).

As well as being bundled horizontally in this way, some authors in the practice tradition have recognised the vertically nested character of doctoral practices which is ‘inclusive of the participants, the Academy and the community’ (Cumming, 2007, p. 5). McAlpine and Norton (2006) encapsulate this in an ‘integrative framework’, a concept taken up by other authors (Cumming, 2007, 2008, 2010; Hopwood & McAlpine, 2007; Rönnerman & Kemmis, 2016). Again, as Hopwood and McAlpine note (2007, p. 3), an integrative framework stresses not only that

[w]ork in universities … is situated in multiple different contexts, but that these overlap and interact. Importantly this perspective encourages us to combine an inward looking, specific analysis of particular programs or practices, with an outward looking, more general attention to broader structures. However, operationalising this remains a challenging task … 

Lee and Boud, adopting a very broad perspective on ‘practices’, also note (2008, p. 10) that ‘there is often a lack of articulation of concerns across levels’.

Individualistic perspectives are also capable of taking different levels of analysis into account in that way, and at the same time practice theory is capable of taking into account individualistic characteristics, but rarely does. These include ‘engagement with personal trajectories, identity, notions of academic values … issues around emotional investiture of doing a doctorate … [and] in the analysis of power’ (Hopwood & McAlpine, 2007, p. 8). Barnacle and Mewburn (2010) provide an indicative route to exploring the significance of artefacts and materiality in doctoral work, another gap in our understanding of doctoral supervision which sociocultural theories applied elsewhere consider significant.

These gaps are not insurmountable: a practice approach has the potential to do real work for our understanding of doctoral practices and how to enhance them. As Lee and Boud suggest:

[The social practice] frame allows us to attend to the increasingly complex array of sites of activity governing and framing the doctorate. Far from being the essentially private activity of individuals within the enclosed spheres of disciplines and departments, the doctorate is now the much more complex and public site of practices of governance, regulation and planning, as well as of research itself, the educative work of supervision and teaching and the activities and experiences of candidature. (Lee & Boud, 2008, p. 22)

Refocusing the practice lens

The challenge of applying a practice perspective to supervision has been initiated by the authors described above, but several lacunae remain. Socio-materiality is as yet under-developed. Issues around subjectivities and their trajectories, questions of values and affect as well as the operation of power are still in need of development.

While the bundled and nested character of supervision practices has been recognised, the elaboration and exemplification of the consequences of this has yet to be undertaken in depth. Moreover, the consideration of the hierarchically nested nature of practices has focused on policies, structures and institutional frameworks but has yet to fully take into account ideational influences. Structured sets of educational ideologies, frames of reference and symbolic structures operate ‘above’ supervisory interactions and structure the behavioural, affective and cognitive dispositions of the individuals involved. These constitute some of the building blocks of social practices, shaping and framing them, operating as proto-practice reservoirs, structured ideational resources, drawn on by social agents and expressed in practice repertoires.

The literature on the topic of educational ideologies, for example, consistently identifies four present in universities: traditionalism; progressivism; enterprise and social reconstructionism (Trowler, 1998). Bøgelund’s (2015) empirical work shows the potential of acknowledging their significance. He finds doctoral supervisors expressing three of these, which he names academic (traditionalism), market (enterprise) and society-changing (social reconstructionism). His respondents deploy one or a combination of these at different points, and he indicates the implications of these positions for their practices and subject positioning within doctoral supervision. Grant (2005), likewise, identifies ideologically rooted discourses of different sorts in both supervisor and supervisee: Psy; Trad; Techno; Com; Rad.

The ideational characteristics drawn on in the accomplishment of supervisory practices are dynamically realised locally and are often productive of contestation as well as consensus. In addition, contextual factors condition local practice repertoires and unique performances. These include the operation of power of different sorts (for example between supervisor and supervisee), as well as the material nature of the context of practice: the disposition of office spaces and software characteristics. Together these ideational and situational factors shape, but do not wholly determine, supervisory behavioural dispositions (Kemmis et al., 2012; Kemmis & Mahon, 2017). In this sense, practices are ‘situated’ and structurally conditioned but also co-constituted by reflexive agents.

The picture that emerges is not a bubble with just two or a few people in it but a web of connections, the threads of which transmit capillary forces which shape the object of interest: supervisory practices. There is an arterial topology to their accomplishment. As Nicolini (2012) and other practice theorists show, it is most valuable to zoom in on local practices of interest, and then to zoom out to their relationships in time and space. Without zooming out from the particular, one cannot make sense of how local practices are shaped: ‘when such a connection is severed or not developed, it is likely that we end up with an advanced version of something else’ (Nicolini, 2012, p. 122).

Sikanadar’s work (2018) offers an example of the operation of broader institutional practice frameworks in her critical discourse analysis study of language interaction shaped by (or free from) institutional discursive practices in two cases of postgraduate supervision. One case saw doctoral discursive interactions heavily shaped by institutional structure and control, while supervision in the other case was flexible and agentically shaped by participants.

Zooming out like this is valuable, but appreciating the operation of such practice ecologies is not enough. ‘Panning up’ shifts the lens upwards to see the connections and forces operating ‘from above’ on supervisory practices, forces that are wide-ranging in their effects, including ideational forces. Panning up and down involves unpicking nested forces that shape practices. Zooming out cannot do this at a general level, though it is valuable for showing how context shapes supervision practices and how practices locally are bundled and can affect each other in specific locales. For Schatzki (2016) there is nothing more than practices on the horizonal plane; clusters of relations within the ‘plenum of practices’, and no micro–macro relations. He acknowledges the value of zooming out, but not panning up. By contrast, Apple (2013) offers an empirical example in the same field as Schatzki (2016) but from the perspective of the oppressed, exploited and structurally disadvantaged. He demonstrates the invisibility of these oppressive relations to those who are advantaged by them, but highlights their significance for the lives of those made invisible by macro forces.

Nicolini describes zooming out as involving ‘making a connection between what is local and what is global’ (2012, pp. 221–122). Apple does this, but he does more. He pans up to the ideology and power of capitalism, and to the occlusion of exploitation. He looks up to the intangible but significant realm of power relations, positioned subjectivities, discursive repertoires, manufactured codes of signification. He also looks upwards at the very tangible ways in which the material world is restructured by capitalist relations. I return to this radical perspective later in the article.

Filling the gaps that remain in practice theory and integrating them conceptually can be partly accomplished through the application of the concept of ‘moments’; the disaggregated elements which constitute a social practice or family of social practices such as teaching and learning regimes (Trowler, 2020). These offer an organising framework of concepts which provide a foundation for an operationalised theory of social practices in doctoral supervision, one which is both explanatory and indicative of action for enhancement. They are what one sees when zooming in to social practices and through them, zooming further to their composite elements.

Eleven moments, set out in Table 1, flow through the social world of higher education and are situated at workgroup, departmental, institutional, national, global levels and in proto-practice reservoirs. In the doctoral relationship both supervisor and supervisee act as reflexive carriers of these moments. Understanding how moments operate is the first step to unravelling the forces shaping the accomplishment of their practices and, potentially, enhancing processes and outcomes. Table 1 indicates in the first column the structural forces which condition the nature of the moments, imbuing regularity to them in specific contexts while column 2 exemplifies them.

Table 1. Questions to ask about the moments flowing into the supervisory context.

Moments	Questions and examples
Proto-practice reservoirs
Implicit theories of supervision	What theory/ies underpin the approach to learning to become a proficient doctoral researcher, and are they shared? Boud and Lee (2008) describe the steady ‘pedagogisation’ of doctoral work which, among other things, involves peer learning and the directed acquisition of sets of research skills. This contrasts with an approach to supervision which is individualistic, personalised and prioritises the development of new knowledge over personal skill-sets. Grant (2005) sets out alternative theories of supervision.
Conventions of appropriateness	What appropriately happens in a ‘supervisory session’, and are there significant differences between participants’ expectations? What ‘rules’ are codified and which remain tacit? Zhang (2016) shows how Chinese doctoral students can struggle to come to terms with US conventions around who speaks and expectations of expressing criticality in public contributions to debate. She argues that learned ‘high risk-avoidance’ can make these students struggle to consider these to be appropriate behaviours.
Tacit assumptions	What are participants’ assumptions about what ‘doctorateness’ is, and do they differ? Henderson (2018) explores how assumptions about supervisory approaches develop, and highlights the consequences of them not being surfaced. This results in lack of mutuality.
Codes of signification	What does the word ‘research’ signify to those involved? Stubb et al. (2014) show how doctoral students may enter the supervisory relationship with one or more overlapping conceptions of it. Brew (2001) identifies four orientations held by academics, with similar orientations identified by Kiley and Mullins (2005) and Åkerlind (2008). Naturally, supervisor and supervisee may not share the same orientations, and this again can be problematic.
Discursive repertoires	To what extent are ways of saying and meaning shared, and how far is there intersubjectivity (mutual understandings)? How are assumptions and theories encoded in written and other texts, what do they concern and what are their effects? Holligan (2005, p. 267) argues that ‘the conflicting array of ideological discourses exercising authority over the university sector may undermine the concept of scholarly originality and the underpinning academic skills and dispositions’.
Local characteristics
Power relations	Who makes decisions, how are they made, and what do they concern? What effects do they have? Who is empowered to be agentic and creative, who is not, and how is this enacted? Ismail et al. (2013) offer a case study in which the power plays occurring between two supervisors disrupted the candidate’s progress.
Backstories in process	What histories do those involved bring to the supervisory relationship, and how do they affect other moments, processes and outcomes? Evans and Stevenson (2011) show how the negative impact of moving from being a high-status professional to undertake a PhD, and occupying the student role, can have significant effects on self-esteem and sense of self-worth.
Regimes in interaction	How do other families of practices, both horizontal (bundled) and vertical (nested) impinge on the accomplishment of doctoral supervision, and vice versa? What different contextual concerns are in play for those involved in the supervisory relationship? Eisenbach (2013) shows how practices associated with and expected of motherhood bundled with those of doctoral research involve pressures and stress for women. Rönnerman and Kemmis (2016) demonstrate how supervisory practices are enabled and constrained by the practice architectures of the international doctoral programme within which they are nested.
Materiality in interaction	In what ways do physical layout, hardware, software and artefacts configure the accomplishment of supervisory practices, and how are they re-shaped by it? How far does materiality constrain or scaffold desirable outcomes for doctoral research? Hunt et al. (2012) highlight the significance of office space, and in particular shared offices for PhD students, in facilitating networking, peer support and a culture of collaboration. Kelly (2017) likewise stresses the significance of ‘spaces of doctoral research’.
Subjectivities in interaction	How do the subjectivities involved configure each other, and what is the effect of this on supervisory practices and outcomes? What forces operate on supervisory practices to structure the positionality of the subjectivities involved and their identity trajectories (McAlpine et al., 2014)? Hutt et al. (1983) show how subjectivities structured around an axis of criticism, anxiety, resentment of authority and powerlessness are disruptive for progress.
Descriptive
Recurrent practices	What is recurrently said and done in supervisory practices to accomplish them, and what are the effects of this on outcomes? How and to what extent are newcomers socialised into this new family of practices? Skakni (2018) shows how routinised, taken-for-granted practices in a department or research field can impede students’ progress.

The palette of 11 moments can enable researchers and change agents to better understand, depict and discuss the practices in operation in their context, and to identify room for enhancement. Combined with the concept of panning up, they can operate to identify and challenge structured advantage and disadvantage and the operation of power. In the next sections I move on to discuss and illustrate this.

Understanding and addressing assimilationism

Social practice theory has been criticised for having an only-descriptive rather than explanatory approach to change and for not being critical of current inequalities (Archer, 2010; Ashwin, 2009). The argument runs that practice theory explains stasis well, but cannot explain change, only describe it. In this sense, it provides a model of change rather than an explanatory theory. This model involves describing changes as emergent from the past, as incremental and as heavily influenced and constrained by present practices and structural forces. Archer (2010, p. 229) talks about the ‘chronic recursiveness’ that social practice theory describes. She argues that it provides no analytical grip on which forces for change are likely to prevail or under what circumstances. It is a theory of social reproduction, in this view, seeing people as social dopes whose behaviour is predetermined.

However, as Apple has shown, the ability not only to zoom out but to pan up to broader power structures and ideologies can give explanatory power, challenge the status quo and provide a path for change. This is as true for supervisory practices as for any other. I want to interrogate these propositions in relation to the issue of assimilationism in supervision.

Manathunga has led the critique of assimilationist supervision pedagogies which bracket out non-hegemonic, non-colonialist ways of thinking and practising (Manathunga, 2017). I would include in the term the absorption of local repertoires of practices, of site ontologies (Schatzki, 2003, 2005), in doctoral education by increasingly hegemonic ways of thinking and practising in that field. Manathunga shows how these occlude and deprecate overseas doctoral researchers’ own geographies and cultural knowledge, considering them irrelevant. Most of the literature on overseas doctoral researchers in global North universities adopts an individual deficit model, reporting how they have difficulties with the English language, in fitting in socially, in adjusting. By contrast, in discussing intercultural doctoral supervision, the postcolonial perspective proposes a transcultural approach, which seeks to situate place, time and diverse cultural knowledge at the heart of supervision (Manathunga, 2017, p. 119). She writes:

… postcolonial and Indigenous understandings of intercultural supervision pedagogies argue that mutual respect, dialogic approaches to supervision and the recognition of the intellectual resources diverse students bring with them represent the core principles of empowering and effective intercultural supervision. (Manathunga, 2017, p. 115)

The post-colonial sensibility when applied to supervision stresses, like the application of practice theory proposed here, the significance of the backstories, the emergent, historically contingent nature of today’s practices, the centrality of power, ideologies and regimes in interaction.

Pinto and Araújo e Sá (2020, p. 8) agree that the supervision process is an intercultural encounter that usually involves a clash of the powerful and the weak to achieve progress. However, an enlightened approach which apprehends the different ideational structures at work and seeks to harness them is possible. This puts

… an emphasis on an ethno-relative perspective focused on developing mutual and transformative learning which arises from the encounter of different knowledge, skills and values within a ‘dialogic space’ (Robinson-Pant, 2009) where interaction between supervisors and students’ cultures provides distinctive opportunities for both to learn from each other. (Pinto & Araújo e Sá, 2020, p. 9)

Manathunga (2017, p. 122) concurs, arguing that approaches to supervision should ‘situate time, place and cultural knowledge at the centre of doctoral pedagogies’. There is … 

the need for Western/Northern supervisors and students to learn more about respectful and effective intercultural communication from their Eastern/Indigenous/Southern colleagues and to develop … humility and awareness of ‘cross-cultural ignorance’ … . (Manathunga, 2011, p. 375)

But in addition to deliberative changes in approaches, the practice perspective suggests that ways to address the forces shaping behaviours and attitudes are necessary. Looking ‘above’ the practices themselves, panning up, to see and tackle the generative forces is necessary. These include university practice architectures (Kemmis & Mahon, 2017), often informed by neoliberal ideology, technical-rationalism and ontological individualism.

Sikanadar’s (2018) findings about the different discourses at work in two universities, described above, exemplifies the discursive dimension of these processes. Isike’s work illustrates another significance of this. He attempts to ‘highlight how structure and culture [at the University of Zululand] combine to impact on his supervision work at the institution’ (Isike, 2018, p. 112):

UniZulu’s postgraduate policies, guidelines and principles constitute an important structure which impacts on the quality of postgraduate students admitted and thus the trajectory of postgraduate supervision. For example, the policy stipulates that Heads of Department (HoDs) are responsible for postgraduate admissions. According to the university’s Policy on Postgraduate Admission and Registration … doctoral candidates are to approach the HoDs of their prospective departments with mini proposals. The HoDs are then required to use the mini proposals to determine the availability of supervisors and, based on such availability, admit the students. … This has left the selection and admission of candidates at the hands of HoDs, some of whom have actually abused the process to admit low quality students who did not meet the required standards for admission. (Isike, 2018, p. 120)

Here the social-political arrangements, especially the question of who holds the power to make decisions relevant to supervision, are highlighted. Practice architectures shape the present situation, foreclosing and occluding alternative possibilities.

Practice architectures partly shape local ‘site ontologies’ – the specific practices on the ground, ‘below’ them. These architectures consist

… of what practices are composed of, and how practices are shaped by, and shape, the cultural-discursive, material-economic, and social-political arrangements … with which they are enmeshed in sites … of practice. (Kemmis & Mahon, 2017, p. 109)

Universities’ discourse and policies project doctoral pedagogies in terms of efficient project management (Krejsler, 2006). This creates a generic map for supervision which occludes alternative knowledges and backstories (Grant & McKinley, 2011, p. 384). Power relations, resources and material dispositions reinforce this occlusion.

However, the institutional level analysis is only a partial aspect of panning up. Elliot and Kobayashi (2019) encourage us to pan further upwards to national cultural contexts of intercultural supervision practices. Starting with Hofstede’s broad typologies, they analyse cultural characteristics in Denmark, as articulated in social practices such as the prevalence of self-regulatory codes, tolerance of disagreement but a preference for decisions based on consensus, high trust and the importance of a sound work-life balance. They say:

Nordic education, from primary school onwards, rests on the belief that individuals can develop and that society influences the development of the individual. (1984, p. 916)

The corollary of this developmental perspective is resistance to the foreclosing effects of summative assessment and an evaluative culture, in contrast to the Anglo-American educational system with its performance goal orientation (Dolin, 2016). Consequently, Danish supervisors tend to resist judgements of performance or of the person. For doctoral researchers from Japan, Bangladesh, India, China, Iran and Kenya this often resulted in annoyance with time-wasting ‘interactive’ discussions, frustration at a lack of judgemental feedback, resistance to taking an autonomous rather than directed approach and problems in moving from respect for the authority of the written word to critiquing it (Elliot & Kobayashi, 2019).

While this study could easily be critiqued for adopting stereotypical understandings of cultures, for agglomerating very different student cultures and for being over-appreciative of Danish culture, the points being made are familiar. Expressed in terms of social practice moments, they highlight the significance for doctoral supervision of backstories, especially in relation to: power relations; implicit theories of teaching and learning; conventions of appropriateness; tacit assumptions; subjectivities in interaction and recurrent practices. There is, in short, considerable friction built into diversity in the host and the guest regimes because of displaced practices (Diamond, 2013). This can be exacerbated when the student’s backstory and new situation involves a sudden change in status where, for example, they come from a high status, high-powered post in their own country and are funded for three years to gain a PhD abroad. The subjectivities in interaction are suddenly very different, and this too can create potential friction.

Conclusion

Individual attitudes, behaviours and choices form only part of the supervision story, and only part of an approach to enhancement. Where enhancement efforts focus on improving personal relationships for more productive processes and outcomes these are important. But many supervision issues involve wider social practices, as Cumming (2007) and others show us. Enhancement aimed at institutional-level outcomes such as timely completion rates as well as wider agendas such as de-colonising research and knowledge involves addressing recurrent behaviours and underpinning attitudes and assumptions which flow from ideologies and symbolic structures through the social world. While recognising the significance of subjectivities and backstories is important here, so is foregrounding the shared, contested and often unrecognised assumptions, theories and power plays at work. Understanding the characteristics of the moments which comprise supervision practices in specific locales and designing interventions aimed at them rather than just a personal improvement provides the route to robust, sustainable enhancement initiatives. Not to do so risks tackling epiphenomenal characteristics rather than the root causes of behaviours, which are structural in nature and are located in proto-practice reservoirs and local conditioning factors.

So, this article has argued the need to integrate individualistic, voluntaristic approaches to doctoral supervision with a social practice approach. Currently, however, the latter has not been integrated into a conceptual framework which can be deployed for enhancement purposes. The article has indicated areas where further development could yield heuristic power and shown how zooming up to the proto-practice reservoirs which shape supervision practices on the ground and the bundles of other practices around them yields greater explanatory power. Recognising how organizational practices and policies in areas outside doctoral research influence, shape and limit it, for example, helps explain its characteristics in specific locales and why some changes are easier to make, others more difficult. Illuminating the ideational, discursive and symbolic structures that shape practices from ‘above’ is a first step to enhancing supervision practices where this is deemed necessary.

Disclosure statement

No potential conflict of interest was reported by the author(s).