Abstract
In this paper, I explore the relationship of democracy to educational leadership; more specifically, to the notion of distributed leadership as it unfolds within policy-mandated multi-site school collaboratives, with particular reference to practices in Malta. Under the policy framework ‘For All Children To Succeed’ introduced in Malta in 2005, Maltese primary and secondary state schools embarked on the process of being organized into networks, legally termed ‘colleges’. I explore leadership distribution among the leaders constituting the college and the subsequent inherent tensions within this educational scenario. The notion of distributed leadership as perceived by the leaders is examined, and especially the leaders’ reception of its presentation in the policy document as the leadership discourse; and its eventual (non-)enactment at both school and college level. A Foucauldian theoretical framework, specifically Foucault’s concepts of power relations, governmentality, discourse, and subjectification, is used to carry out a case study of a Maltese college, collecting data via semi-structured, in-depth interviews, participant observation and documentary analysis. Narrative is both the phenomenon under study and the method of analysis. The policy discourse does not unfold in a participatory democratic manner in practice, resulting in an organizational paradox where leadership enactment in a Maltese college is ‘directed’ from above, rather than ‘distributed’. These findings may be significant for educational practice, policy and theory in terms of the generation of problematization which may lead to further research on this contested topic.
Keywords
Introduction
It is evident that in many educational systems worldwide there has been a partial dissolution of the traditional single school model, with a resultant move towards more flexible modes of organizational link-up taking the form of increased collaboration among schools. Networking and collaborative school improvement programmes have emerged in various cultural contexts, from Australasia to North America and Asia, as well as in Europe (Chrispeels and Harris, 2006). Frankham (2006) and Ozga (2009) questioned the deployment of the ‘network’ metaphor in the field of education, thus illuminating the tensions embedded in its promise of democratic, decentralized governance, and its associations with trust, flexibility and responsiveness. Frankham problematized the discourse used in association with networking (partnership, collaboration, community, connection, flow) that she claims ‘all suggest unproblematic notions of equality within groups, obscuring the inevitable power/relations that exist’ (Frankham, 2006: 672–673). The network’s promise of a stronger democracy through distributed leadership and ‘shared’ decision-making may serve as a technology to mask top-down power and centralized accountability. This notion of distributed leadership (in relation to democratic procedures) is problematized on various criteria. Woods (2004) argued how distributed leadership appears to incorporate democratic procedures, insisting that democratic leadership and distributed leadership are not synonymous. Distributed leaders are appointed rather than elected (Hartley, 2007); power differentials do exist (Harter et al., 2006); and distributed leadership may give rise to autocracy (Spillane, 2005).
It is within this leadership narrative that I set out to explore the relationship of democracy to educational leadership; more specifically, to the notion of distributed leadership as it unfolds within policy-mandated multi-site school collaboratives. In Malta, under the policy framework ‘For All Children To Succeed’ (Ministry of Education, Youth and Employment, 2005) (hence forth referred to as FACT), Maltese primary and secondary state schools embarked on the process of being organized into networks, legally termed ‘colleges’. I explore leadership distribution among the leaders constituting the college and the subsequent inherent tensions, within this educational scenario, and examine the notion of distributed leadership as perceived by the leaders, and especially their reception of its presentation in the policy document as the leadership discourse (hence my focus on one principal document – FACT) and its eventual (non-)enactment at both school and college level.
The paper is structured as follows. The first section presents the local policy context, with a particular emphasis on the role of distributed leadership in school reform. The second section presents a literature review in which I problematize the concept of distributed leadership in relation to democratic procedures, anchoring it in the deployment of the network metaphor in education, with its promise of a stronger democracy through distributed leadership and shared decision-making. The third section outlines my methodological approach, which is based on narrative. This is followed by a presentation of my theoretical framework, explaining how I draw on Foucault’s concepts of power relations, governmentality, discourse and subjectification in order to explore leadership distribution and the subsequent inherent tensions. Finally there is a presentation and discussion of the findings and conclusions are offered regarding the distributed leadership dilemma, together with consideration of the implications and points for reflection.
The local policy context
The Maltese educational system has been undergoing a structured, gradual but steady change in terms of decentralization and increased school autonomy, with the main aim being renewal – modernizing the system in line with global policy development.
The reform process that Maltese education has been going through since 1964 has reached an extremely significant stage in its journey in the last decade, predominantly as a result of the publication of the policy document FACT in 2005 which mandated collegiality in the state school system. Under this reform, Maltese state primary and secondary schools were organized into colleges according to their geographical location, in order to ensure a smooth transition for students across levels through internal exams, control and accountability. This major reform necessitated the introduction of new roles and new responsibilities, amongst which was the deployment of the College Principal, designated to be the educational leader of the college as a whole. Besides ensuring ‘an effective and efficient dialogue with all Heads of School and stakeholders’, the Principal is also ‘expected generally to execute and implement efficiently the policies of the College’ (FACT, 2005: 73). The Education Act (2006) compels the Principal to hold a monthly meeting for all the Heads of School in the college, which is legally known as the ‘Council of Heads’(CoH), to enable all the leaders to build and maintain open channels of communication within and beyond the school community. The Head of School, who, according to the policy document FACT, is expected to lead and manage, is explicitly required …to collaborate with other Heads of College Schools…in a manner that maximizes networking under the leadership of the Principal and according to the direction and guidelines established by… other competent authorities. (FACT, 2005: 74, emphasis added)
The policy document fosters a strong belief in ‘shared or co-leadership’, which is important for the distribution of the leadership function across more than one school location, thus offering the potential of generating ‘healthy dialogue and debate’ while fostering a ‘satisfying and fruitful team spirit’. While advocating distributed leadership, FACT justifies the need for senior leaders – ‘Distributed leadership only thrives where there is effective senior leadership’ (FACT, 2005: 39).
Article 57 of the Education Act (2006) makes provision for a certain degree of autonomy for the colleges, stating that, The Minister and the Directorates shall promote the application of the principle of subsidiarity in the management and administration of the Colleges, within a framework of decentralization and autonomy of the educational operation and services given by the Colleges and their schools according to the priorities, targets and national strategies adopted by the Government. (Education Act, 2006, Article 57: 1) Autonomy is not to be confused with complete deregulation…autonomy and decentralization predicate a grasp by the Education Ministry and the central education entities. (FACT, 2005: 29)
At the same time, Bezzina (1999: 60) notes that ‘Over the years, the meaning of leadership has not quite been understood in Malta’ – that is, despite being strengthened by the advent of decentralization, the leadership aspect of the Head was not given its due importance. The Maltese education system operates within a context that has bred a dependency culture, supporting isolation and Balkanization (Bezzina, 2007): it has therefore been attempting, albeit in different ways and mostly spearheaded by ministerial reforms, to move away from the shackles of a highly centralized system. Various studies have been carried out, exploring distinct areas of leadership from various angles. Whilst it is not within the scope of this paper to give an overview of studies exploring school leadership in Malta, I want to make readers aware of the interest spurred by recent reforms, especially that of networking. A study exploring the leadership styles of Maltese secondary school Heads (Mifsud, 2008), revealed distribution as delegation, both within schools and the education system at large, supported by the tension between decentralization and centralization (strongly in favour of the latter). The fundamental role of both College Principals and Heads of School in the successful establishment and function of colleges as network leaders and collaborative leaders respectively, was explored by Fabri and Bezzina (2010). Furthermore, effective school leadership in the new college system was examined by Fabri through reflections on various leadership aspects, such as structures, qualities, approaches, centralization and decentralization, delegation and distribution, competition and collaboration, in addition to internal and external demands – all framed within a context of change (Fabri, 2012).
I now explore the literature dealing with distributed leadership, further problematizing this notion in relation to democracy. This section outlined the policy context, with a particular reference to the prominence given to the exercise of distributed leadership within the colleges and the subsequent interplay between autonomy and regulation. This was followed by a very brief overview of leadership studies within the Maltese education reform context which my own work will contribute to.
Exploring the literature
Distributed leadership in education
Lumby argues that the concept of distributed leadership has dominated educational leadership theory and practice, thus becoming ‘the theory of choice for many’ (Lumby, 2013: 581), undergoing a transformation ‘from a tool to facilitate the comprehension of leadership ecology to a widely stipulated praxis’ (Lumby, 2013: 581). She expands on this by adding that distributed leadership theory has moved on into practice. Bush (2013) further states that this leadership model has been given ‘normative preference’, with Gronn noting the ‘accelerating amount of scholarly and practitioner attention’ given to this model in an undivided manner (Gronn, 2010: 70). Education practitioners and policy-makers have, for various reasons, been lured by the attractive notion of leadership distribution. Mayrowetz (2008) explores the value of distributed leadership in a pragmatic sense, pointing out four benefits that can be generated through its practice. It may be regarded as a theoretical lens for exploring the leadership dynamics; as a means for furthering democracy – despite distributed leadership and democracy being very distinct, as outlined by Woods (2004); as a way of improving organizational efficiency and effectiveness; and, finally, as leading towards human capacity-building. The last benefit relates to Gronn’s notion of ‘concertive action’ (Gronn, 2002: 3, which results in an outcome which is greater than the sum of individual actions, illustrating leadership as an emergent property of a network and resulting in a product of conjoint activity.
Hadfield states that ‘the increasing popularity of network-based approaches…is generating new and emergent leadership challenges, which in their turn are likely to create new leadership approaches’ (Hadfield, 2007: 259). Jopling and Spender argue that there is ‘no simple, single solution to leading networks’ (Jopling and Spender, 2006: 5), with Hadfield describing how ‘…the very nature of a network makes it difficult to define who its leaders are…’ (Hadfield, 2007: 260). Jopling and Crandall (2006) assert that networks of schools do require some form of leadership despite their shift towards more plural, distributed and adaptive forms. They explain that network leadership is qualitatively distinct from traditional notions of hierarchical school leadership because it is facilitative rather than directive, it is about leadership emerging from interactions and relationships between people, rather than charismatic individuals – it seems to work best when it is distributed, being more responsive to context. Harris describes the state of networking: ‘we are witnessing a proliferation of collaborative possibilities…schools are frenetically pursuing the goal of networks and networking’ (Harris, 2005: 5). Consequently, Harris foresees the generation of leadership challenges in terms of the necessity of more innovative and dynamic approaches resulting in lateral and vertical forms of leadership practice. Distributed leadership is thus regarded as a complementary mode of thought about leadership practice which meets the shifting demands of multi-site collaboration as a result of the conception of this leadership theory being based on the metaphor of the network with its ideas of interaction, undefined boundaries, wide distributions, and power flows.
Definitions of distributed leadership abound in literature. Harris describes it as ‘collective leadership responsibility rather than top-down authority’ (Harris, 2005: 1), subscribing to Spillane’s (2005) ‘leader plus’ perspective, moving us from a ‘person solo’ to a ‘person plus’ perspective, suggesting multiple leaders at multiple levels. This is premised on a collective approach to capacity building in schools (Harris and Lambert, 2003) through a recognition that leadership practice is constructed by means of shared action and interaction. In his emphasis on ‘leaders, followers, and their situation’, Spillane (2005) implies an interdependent influence between followers and leaders, who emerge as leadership co-producers. The follower dimension incorporates followers as influencers in the determination and shaping of leadership practice. Bennett et al. (2003) identify three characteristics of distributed leadership: as an emergent property of a network of interacting individuals; as operating within undefined boundaries that can only vary along a continuum between wide and restricted; and having widely distributed expertise and leadership opportunities. Zepke (2007) builds on Gronn’s definition of distributed leadership as ‘structurally conjoint agency’ (Gronn, 2002: 543), describing it as a ‘community for action’ (Zepke, 2007: 305) where power flows from leader to leader as new leadership roles emerge and are nurtured. Harris critiques the positioning of distributed leadership as the antithesis of top-down, hierarchical leadership; she regards it instead as a form of co-leadership involving ‘both formal and informal leaders, it is not an “either/or”’ (Harris, 2013: 548).
Various reasons have been given for the motivation for distribution. Those which feature the most prominently are what Fullan (2005) proposes as the recognition of the limitations of the ‘charismatic hero’ which, according to Bush (2012: 649), have been ‘supplemented but not supplanted’ by concepts of shared leadership, mirroring the trend towards self-management, together with the pragmatic popularity of distributed leadership to ease the burden of principals and senior leaders who have become overloaded, as evidenced by Hartley (2010). However, Bush (2013) argues that distributed leadership oversteps this instrumental motive to an acknowledgement of the conjoint expertise of organization members. Jopling and Crandall (2006) suggest that distributed leadership is viewed as an important practice in school networks because it offers a new way of talking about leadership in which all the voices of relatively ‘unrecognized leaders’ are legitimized and the language of leadership is extended. Relatively new, emergent leadership roles are given the space to expand across the network. Distributed leadership helps leaders in challenging expectations, adding new perspectives and making lateral and latent leadership practice more visible: as Bennett et al. state, it is ‘an important analytical tool for thinking about leadership and re-orientating thinking about its nature’ (Bennett et al., 2003: 7).
Distributed leadership already features in policy frameworks in a number of countries and is being actively advocated. This is the case in the Maltese state education system, where, similar to practice in England, distributed leadership underpins the new models of schooling, as researched by Chapman et al. (2010). Published work by both Harris and Jones (2010) and Harris (2011) describes how, in the Welsh case, distributed leadership is a vital part of system-wide reform, manifesting itself through a national infrastructure of professional learning communities within, between and across schools. Hartley (2007) explores the emergence of distributed leadership in education through political and cultural considerations; besides providing a response to the focus of recent policy shifts on the merging or networking of work-based activities, distributed leadership resonates with Bauman’s (2000) concept of ‘liquid modernity’ in a contemporary culture that favours the trend ‘from organized social structure to network culture’ (Page, 2006). Dispersion of leadership within and across organizations resonates with the flexible ‘liquid modern’ view of time and space. Moreover, Hartley (2010) claims that it fits in well with the contemporary trend of organizational learning within a so-called knowledge economy.
Problematizing distributed leadership in relation to democratic procedures
Hartley (2007) suggests that one of the main reasons behind the emergence of distributed leadership in schools, at least at the level of policy rhetoric, is it being regarded as a response to policy shifts. The notion of distributed leadership might just be used as a mask by policy producers and government officials to ease in their agenda as a normalizing discourse in schools. Hartley (2010) elaborates further on this notion of distributed leadership by stating that it ‘is a means to an end whose purpose is organizational…development…It is mainly about accomplishing the organizational goals which comprise the instrumental tasks and targets set by officialdom’ (Hartley, 2010: 281, original emphasis). Hall et al. recognize distributed leadership as the ‘officially sanctioned model of good practice’ (Hall et al., 2011: 32) advocated by government departments, yet suggest that discussions about this notion reflect normative narratives and are merely part of policy rhetoric designed to claim that power and autonomy are being shared with schools, whereas reality points to centralization and managerialism. Hartley regards it as ‘yet another sign of an institutional isomorphism’ (Hartley, 2007: 211), whereby the public sector purports to legitimate its policies by appeals to the new organizational forms within the private sector (Alvesson and Thomson, 2005: 488). Gunter and Forrester (2008) detect a control imperative in the relationship between policy and practice in the professional practice of ‘policy entrepreneurs’ (a term coined by Kingdon, 2003), thus concluding that the primacy of the single person (that is, the ‘solo’ leader) remains, with distribution coming downward and used as a form of sophisticated delegation. Youngs (2009) states that education reforms by neo-liberal policy agendas have privileged the economic purpose of education, leading to a state of opposition between self-management and mandated external accountability. Youngs points to the lack of critique on the policy environment and queries the extent to which distributed leadership practices will be shaped or hindered by ‘official and delegated leadership practices informed by neo-liberal ideals’ as opposed to the ‘more educative and democratic ideals’ (Youngs, 2009: 382) informing the professionals’ performative environment. I explore the extent of the influence of the exercise of distributed leadership as a policy mandate on the leadership practices within the individual schools and the college, within this narrative of policy rhetoric, self-management and external accountability.
Are distributed leadership and democratic leadership synonymous?
Some critics of the distributed leadership notion ask whether such a concept offers a genuine alternative to other forms of leadership, or whether it merely serves as ‘the emperor’s new clothes’ (Bolden, 2011: 254) or as a pragmatic response to society’s demand for equity and purpose. Lumby accuses distributed leadership of offering ‘yet another nuanced rebranding, anchoring the nebulous concept of leadership to a seemingly fresh and inclusive activity’ (Lumby, 2013: 589). She argues that in this way distributed leadership subjects educators to the ever-increasing limitations posed by centralized curriculum control, surveillance and marketization. This arises because distributed leadership appears to loosen the bonds in encouraging staff to embrace it as a ‘technology of the self’ (Foucault, 2000b: 177), thus leading to the successful enactment of policy discourse. Hartley suggests that it is ‘little more than an emancipatory rhetoric’ (Hartley, 2010: 279), building on the accusation by Hargreaves and Fink of distributed leadership being a ‘more subtle and clever way to deliver standardized packages of government reforms and performance targets’ (Hargreaves and Fink, 2008: 238–239). On a similar note, Spillane et al. argue that distributed leadership has ‘effortlessly entered the conversation about school leadership and management…often with simplistic and unwarranted mantras such as everyone is a leader or the more leaders the better’ (Spillane et al., 2011: 159).
According to Woods (2004: 22), as with other discourses of legitimation, such as ‘empowerment’ and ‘ownership’, the notion of distributed leadership appears to incorporate democratic procedures. Woods addresses the danger of the notions of democratic leadership becoming colonized by distributed leadership discourses, stating that the two notions cannot be regarded as synonymous. Distributed leadership incorporates a degree of control and autonomy, within which there is the scope for dispersed initiative and the boundaries of participation. Despite being distributed, leadership does not imply the lack of a hierarchical pyramid because it varies along the continuum between control and the autonomy participants are allowed to exercise. This positioning of distributed leadership across the control/autonomy continuum determines whether autonomy and empowerment are extended or if constraint and control are exercised in novel ways. Democratic leadership runs the risk of being regarded as another way of denoting distributed leadership; however, the concept of democratic leadership simultaneously draws open and goes beyond that of distributed leadership. Whereas distributed leadership may obscure the deeper questions inherent in democratic leadership through its widening of leadership boundaries, democratic leadership attempts to make these deeper questions visible. Woods (2004) thus argues that distributed leadership promises a ‘hollow’ democracy, further entailing a ‘democratic deficit’, which, according to Woods and Gronn (2009: 430), leads to its critical interrogation. Is it the leadership of one or the leadership of many?
Hartley (2007) argues that distributed leaders do not arrive at their position as the result of an election but, rather, of an appointment, with a ‘presumed harmony and consensus’ about the whole affair. Hatcher (2005) and Storey (2004) are of the view that ‘distribution’ tends to underestimate the micro-political aspects of leadership practice, while completely ignoring the ‘distribution’ of wealth beyond the school and its causal relationship to attainment. However, leadership cannot survive without difference: this seems to be echoed by Harter et al. (2006), who argue that leadership dynamics are ‘unequal’ in one way or another – power differentials exist despite the notion of distributed leadership. Woods et al. (2004) propagate the existence of distributed leadership and strong senior leadership which allows for strong partnerships with the simultaneous power disparities between the partners. Watson (2013) draws upon McMurry’s (1958: 82) notion of ‘benevolent autocracy’ to show its particular relevance to school leadership, thus implying that distributed leadership practices are ‘only’ possible ‘where an autocratic leader at the top ruthlessly ensures that participatory ideals are…adhered to by those lower down the hierarchy’ (McMurry, 1958: 259).
Leadership can be stretched over leaders in a school, but is not necessarily democratic: Spillane (2005) argues that a distributed leadership perspective may also give rise to autocracy. Is there a single distributor who ‘distributes’ leadership? One has to question what is being distributed, how, and to whom, as well as the way in which this distribution occurs. Distributed leadership does not guarantee better performance; it is not a panacea for success, it does not possess any innate good or bad qualities, it is not friend or foe. Much depends on how leadership is distributed and the intentions behind it. (Harris, 2013: 552, emphasis added)
My theoretical framework
It is within this discursive narrative of multi-site collaboration, ‘shared’ decisions-making, ‘promised’ democracy, self-management and external accountability that I explore leadership distribution among the college leaders, and the subsequent inherent tensions. These issues are analysed by means of a Foucauldian theoretical framework, using Foucault’s theories of power (1979, 1980), discourse (2002c), subjectification (2002d) and governmentality (2002a).
Conscious of the fact that Foucault was keen to avoid being seen as offering a ‘general system, an overarching theoretical framework or worldview’ (Foucault, 2001: 240), I take a ‘piecemeal approach to his work’ (Allen, 2012), by viewing it as a ‘tool-box’ (Megill, 1987). Foucault warns that, ‘What I’ve written is never prescriptive either for me or for others – at most it’s instrumental and tentative’ (Foucault, 2002b: 240). Gillies (2013) demonstrates the value of Foucault’s trident of scepticism, critique and problematization to operate within educational discourse, the purpose of which is ‘to question, probe, and identify weaknesses, contradictions, assumptions, and problems’ (Gillies, 2013: 19).
One central thread running through Foucault’s writings, which helps my exploration of the modes by which educational leaders in a network are positioned by the interweaving and often battling discourses, is that of relations of power. According to Foucault, power is ‘the moving substrate of force relations which, by virtue of their inequality, constantly engender states of power…[that] are always local and unstable’ (Foucault, 1998: 93). Foucault is very critical of what he terms the ‘repressive hypothesis’: ‘If power was never anything but repressive, if it never did anything but say no, do you really believe that we should manage to obey it?’ (Foucault, 1979: 36). He moves away from a negative conception of power, extolling instead its productive nature. Foucauldian power is an ‘exercised’ strategy existing within relationships – it is therefore ubiquitous, anonymous and comprehensive, exercised unconsciously with its effects often being repressed. Power, in this sense, is both coercive and enabling, in that it is not imposed from ‘outside’ or ‘above’, but circulates within institutions and social bodies, producing subjects who exert a ‘mutual “hold”’ on one another. This is termed by Foucault as ‘a mutual and indefinite “blackmail”’, which binds superiors and subordinates in ‘a relationship of mutual support and conditioning’ (Foucault, 1980). Thus, despite institutions having a system of power with a ‘pyramidal form’, power does not derive from the summit which stands in a symbiotic relationship with the lower elements of the hierarchy. Freedom is a condition necessary for the exercise of power – ‘at the same time its precondition…and also its permanent support’ (Foucault, 2002d: 342). It is a form of power ‘that subjugates and makes subject to’ (Foucault, 2002d: 331), functioning in what Foucault terms a ‘political “double-bind”…the simultaneous individualization and totalization of modern power structures’ (Foucault, 2002d: 336).
Foucault conceives of power dynamically, by proposing a model in which power relations dissipate through all relational structures of the society. He insists that ‘relations of power are not something bad in themselves…I don’t believe there can be a society without relations of power’ (Foucault,1980: 96). He maintained strongly that ‘power is employed through a net-like organization…individuals are the vehicles of power, not its points of application…The individual is an effect of power and the element of its articulation’ (Foucault, 1980: 98). This way of understanding power exhibits two key features: power as a network of relations; and individuals as the locus where power and resistance to it are exerted (Mills, 2003: 35). Foucault’s (1998: 93–94) concept is that ‘power is exercised from innumerable points…[it] comes from below’; we get a picture of dispersed power, not present at specific locations, but always at issue, being ‘produced from one moment to the next’. According to Foucault, power is not possessed by a dominant agent, nor located in that agent’s relations to those dominated, but is instead distributed throughout complex social networks.
Foucault describes ‘discourses’ as ‘practices that systematically form the object of which they speak’ (Foucault, 2002c: 49). Within a Foucauldian approach, discourses are inextricably linked to institutions and to the disciplines that regularize and normalize the conduct of those brought within the ambit of those institutions. As Foucault (1991a: 217) stated it, the individual is ‘fabricated’ into the social order – people are woven into, and woven out of, discourse. Foucault implies that the subject is produced ‘as an effect’ through and within discourse, within specific discursive formations, and has no existence, as well as no transcendental continuity or identity from one subject position to another. MacLure translates this as: ‘The individual achieves agency as an active subject by being subject-ed to the disciplinary machineries of discourse’ (MacLure, 2003:176). Foucault also argues that ‘discourse is not a place into which the subjectivity irrupts; it is a space of differentiated subject-positions and subject-functions’ (Foucault, 1991b: 58). This notion of the subject is important because Foucault is referring to the idea that subjects are not only shaped by social structures, but also actively take up their own discourses through which they are shaped and by which they shape themselves (Blackmore, 1997). Notions of agency and structure are always present when considering educational leaders, because these individuals are expected to formulate visions and enable change while simultaneously being constrained and normalized by bureaucratic processes and mechanisms (Niesche, 2011). Discourses structure our sense of reality and our notion of our own identity, both of which are not fixed but, rather, constantly being made and remade. Foucault emphasizes the productive capacity of discourse. For Foucault, ‘discourse transmits and produces power; it reinforces it, but also undermines and exposes it, renders it fragile and makes it possible to thwart it’ (Foucault,1998: 101). These ‘regimes of truth’ enable an exploration of how the subject is produced ‘as an effect’ through and within discourse and within specific discursive formations – how they are positioned by the leadership policy discourse, and how they, in turn, position themselves according to their distributed leadership performance.
Foucault’s concept of ‘subjectification’ deals with the ‘way a human being turns him- or herself into a subject’ (Foucault, 2002a: 208), with a specific focus on those processes of self-formation in which the person is active. To be a subject is to be subjected. Individuals are in the unknowing grip of an insidious power operating through invisible strategies of ‘normalization’, even when they are under a ‘misconception’ of a state of total freedom. Foucault’s subjectification is formed through multiple ‘practices of the self’, as he himself remarks: …the subject is constituted through practices of subjection, or … through practices of liberation, of freedom…starting of course from a certain number of rules, styles and conventions that are found in the culture. (Foucault, 2002f: 331)
This concept helps me explore the ways in which educational leaders are ‘subjectified’ in a college, in the changes that occur in their leadership conduct due to the creation of new roles as a direct effect of the policy document.
In his analysis of the concept of governmentality, consisting of methods of shaping others’ behaviour, Foucault (2002a) stresses that institutions are fragile and have a great potential for change. Thus he implies that power is subject to negotiation, each individual having their place in the hierarchy, no matter what its degree of flexibility. Gillies maintains that governmentality retains ‘theoretical potency’ as a concept, a ‘portmanteau neologism’ (Gillies, 2008: 415) that combines government and rationality, while proceeding to outline the ways in which Foucault deals with this concept. Foucault understood the term ‘government’ as ‘the conduct of conduct’ (Foucault, 2002e: 341), in both a wide and narrow sense, encompassing forms of activity to affect the conduct of others, as well as the relation between self and self. Governmentality encompasses both political rationality (dealing with mentalities, conceptions and discourse), and technologies of government (dealing with the ways in which government is exercised) (Olssen, 2003: 197). Foucault reveals a preoccupation with the ‘art of government’ (Foucault, 2002a: 201), when he enquires ‘How to govern oneself, how to be governed, how to govern others, by whom the people will accept being governed, how to become the best possible governor’ (Foucault, 2002a: 202).
According to Foucault, modern governmental rationality is simultaneously individualizing and totalizing, in its attempt to explore what it is for an individual, and for a number of individuals to be governed. In his essay ‘Governmentality’ (Foucault, 2002a: 205), he alludes to the ‘multifarious’ practices of government ‘concern[ing] many kinds of people’, further describing how the art of government involves establishing a continuity, in both an upwards and a downwards direction, learning to govern both oneself and others. In a political context where there is a double movement of state centralization and dispersion, Foucault identifies a ‘problematic of government’ that emerges through ‘how to be ruled, how strictly, by whom, to what end, by what methods, and so on’ (Foucault, 2002a: 202). Gordon (2002) further suggests that the ‘problematic of government’ provides Foucault with a more practical way to address the power-freedom association, as power only functions in the presence of freedom. Thereby, certain discourses (that is, rationalities of government) are ‘transactional realities’, tools for negotiation, which may eventually lead to dissenting ‘counter-conducts’.
Gillies explores the use of a Foucauldian governmentality perspective in relation to the discourse of educational leadership: this allows one to examine the rationality of its exercise, the justification of its own activity, and the way it comprehends its own function (Gillies, 2013: 66). Foucault’s concept of governmentality is a very useful analytical tool in my exploration of the leadership behaviour at college level because it facilitates the study of how leadership is justified, and how its exercise is to be understood – leadership being one of the main issues that emerges in the FACT policy document. It allows me to explore the extent to which the leaders’ behaviour is shaped by FACT.
Methods and methodology
The data presented in this article derive from a case study of a Maltese college, collected via in-depth, semi-structured interviews with the Heads and the Principal, and participant observation of their Council of Heads meeting, all subjected to narrative analysis after being transcribed and translated by me, in addition to documentary analysis of the FACT policy document. This multi-method approach involves an engagement with crystallization, what Richardson regards as ‘a postmodernist deconstruction of triangulation’ (Richardson, 2000: 934). While the main purpose of the interview questions was to incite narrative using only a minimum of framing when spontaneous coverage was fully exhausted, participant observation helped to consolidate and/or contradict interviews, as well as to discover things participants might not talk about freely, by observing the ‘effects’ of the policy in action. Documentary analysis of the policy enabled me to explore the leaders’ narrative about and performance of distributed leadership as the leadership discourse explicitly set out in the policy narrative.
Narrative is both the phenomenon under exploration and the methodological approach adopted for analysis. My approach to narrative analysis roughly follows Mishler’s framework for understanding the different approaches to narrative analysis (Mishler, 1995), based on meaning, structure and interactional context. A focus is placed on the content of the narrative, on its structure and on its performance, for a disclosure of the interactional and institutional contexts in which narratives are produced, recounted and consumed, while analysis is based on meaning, structure and interactional context. In a ‘story analyst’ (Smith and Sparkes, 2008) stance, I identify the main themes in the individual ‘analysis of narratives’ (Polkinghorne, 1995) – these narratives referring to the transcriptions of my data. Narrative is the mode used to explore ‘policy reception’ (Gowlett et al., 2015: 149) by the various leaders: in the footsteps of Bansel (2015), I use qualitative narrative experience as an alternative mode of policy analysis through ‘accounterability’ (Kamuf, 2007, quoted in Bansel, 2015: 10) which allows for pluralism and multiplicity. According to Bansel, ‘this vivifies the day-to-day, moment-to-moment practices through which the subject of policy is constituted and regulated…It also articulates the ways in which multiple and often competing narratives, are managed, mediated, revised and resisted’ (Bansel, 2015: 10). Close attention is paid to the policy performance of these educational actors as subjects of policy in their mundane leadership practices amidst their multiple iterations and relations.
DemoDistributa College, the establishment studied, is made up of a number of primary and secondary schools. 1 Despite the fact that since undertaking the fieldwork both the Principal and the network composition have changed, I cannot reveal any additional specific information about the individual schools and their leaders due to the sensitive nature of the data involved and the bounded nature of the Maltese educational community. The participants are all given the female gender and referred to numerically (those starting with ‘P’ indicating primary school Heads, while those starting with ‘S’ refer to those at secondary level), except for the Principal who has to be identified (as ‘CP’) because of the purposes of this present research. The leaders’ quotes, which are henceforth used as testimonials for the themes that emerge, emanate from both the interviews and the observation. Underlining in the quotes extracted from the policy document FACT indicates my added emphasis.
I am well aware of the fact that my case study constitutes what Damianakis and Woodford (2012) have identified as a ‘small connected community’ where ‘unintentional identity disclosure’ may occur due to tensions regarding the issues of anonymity, privacy, confidentiality, and betrayal. The participants were informed of the limits of confidentiality and the possible risks involved; they all opted to continue with their participation. Transparency about my personal and theoretical attitudes and research purposes enabled me to maintain good relationships with the participants. Furthermore, engaging in reflexivity throughout the research process helped me focus my attention and awareness on ethical nuances which, according to Guillemin and Gillam (2004), might arise during the research process beyond initial perceptions.
In my research, I attempt to adopt the position where ‘Space is left for others to speak, for tension and differences to be acknowledged, celebrated, rather than buried alive’ (Koro-Ljungberg, 2010: 706). As an ethically responsible researcher, I deal with dilemmas, uncertainties and indecision by facing what Koro-Ljungberg regards as ‘the ultimate responsibility: the unknown’ (Koro-Ljungberg, 2010: 605) and acknowledging that ‘what we see depends on our angle of repose’ (Richardson, 2005: 963). I highlight my vulnerabilities and positionality, embracing Richardson’s claim that, ‘There is no such thing as “getting it right”, only “getting it differently contoured and nuanced”’ (Richardson & St. Pierre, 2005: 962). Through my research, I do not assume to have arrived at any ‘final truths’; rather, interpretations that are multiple, partial, contingent, relative and relational.
Presentation and discussion of findings
FACT, distributed leadership and the leaders of DemoDistributa College
The notion of distributed leadership is perceived and performed by the Heads and the Principal in very distinct, diverse, particular and idiosyncratic ways, both within DemoDistributa College and the individual schools. The discourse of distributed leadership thus emerges as both the ‘object’ and the ‘site’ for struggle, as the thing ‘for which’ and ‘by which’ there is struggle (Foucault, 1981: 52–53).
FACT is presented to the educational leaders as an attractive reform to be embraced and discursively practised as the norm in schools, leading Heads into believing that a contingent application of the policy is possible and desirable, when the discursive framework proves otherwise, because ‘leadership’ is directed ‘from above’. FACT, while avowing to address ‘the issue of the The start up of school networks is an exciting opportunity for our schools to develop local solutions. It will provide us with the capacity to I didn’t have to be told to practise distributed leadership because it is my style. (P1) There are certain things in life which do not have to be dictated by anyone, they don’t have to be issued by the Directorate. They’re almost natural. The policy document just put in writing all my practices! (P2) …shared or co-leadership … distributes the leadership function across more than one school location within a given network. Shared leadership creates a capacity for We do take initiatives together. For example, my school organizes a day trip to Rome for Year 6 pupils. Before, we used to take our pupils only, now it is open to all the Year 6 students of the college. We even give them a t-shirt with the logo DemoDistributa College, in order for them to be identified. Another example of sharing is ‘Merit Day’. (P1)
The Principal’s view of leadership distribution practices among the Heads contrasts very sharply with accounts of their own practices: School Heads are gradually learning how to distribute leadership. I’ve come to enjoy observing Heads working a lot more closely with their Assistant Heads and not simply delegating tasks to be carried out. Distributed leadership is the leadership of everybody…of everybody…leadership that is felt by everybody…Leadership is not the prerogative of a particular position, but it is the prerogative of everybody. Leadership is coordinated by the Head who ensures that it will lead to results. The Heads will surely tell you about it. (CP) One of the things that I insist on is that I have to be informed about every single thing. I need to know about everything! It is a distributed type of leadership but the ‘knot’ is in my hands. (P2) Distributed leadership is leadership which includes everybody, that invites you to contribute and to be a protagonist. Our schools need leaders who distribute leadership. (CP) Distributed leadership has its pros and cons – it can lead to isolation with the lack of a common vision as all the Heads lead in isolation without communication. (P3) I agree with the fact that you distribute and you delegate, but you have problems. Not every Assistant Head carries her own weight. If you are a perfectionist and others are not, then you’ve had it! (S2) How distributed is it going to be? Who is going to shoulder the responsibility? How are the tasks going to be divided? The FACT policy does not mention any of this. (P2)
FACT provides the following answer: This does not mean that no role exists for senior leaders in networks … Distributed leadership only thrives where there is We have to accept certain ‘mistakes’ of others. That goes against my beliefs, but I cannot be perfect! In some way or another, I have to accept certain ‘imperfections’ in others. (P1) I am very reluctant to delegate as I would like to do certain things my own way. (S3) Distributed leadership is a very positive aspect as no leader can work on her own. This is a model that we should reflect about more in schools, that we should try to understand much better. Unfortunately, this is not a practice that is being performed so widely in many of our schools. (CP)
The aim of FACT is clearly set out: ‘autonomy and decentralization of State schools’(FACT, 2005). Is the discourse of distributed leadership intended to be used as a technology to generate more autonomy and decentralization – defined by FACT as ‘a greater say by schools in determining their own management’? Despite the construction of distributed leadership as the dominant discourse, it fails to be enacted as such and is instead overshadowed by the ‘hidden discourse’ of centralization. This emerges both in the leadership practices of the individual schools, where P2, P3, S2 and S3 speak about a distributed leadership that positions them at the centre and, more so, with DemoDistributa College, where a ‘centralized’ model of distribution seems to be enacted because it comes from the top. Indeed, centralized control is still heavily present in DemoDistributa College. A case in point seems to be the exercise of School Development Planning, in place since 1996, which witnesses schools creating opportunities for internal dialogue within and across schools. It is supposedly drawn up by the Head of each individual school together with the other School Management Team (SMT) members, in dialogue with the members of staff, according to the specific needs of their particular school. According to P3: The schools are not autonomous; I would not want to say that they are never autonomous! When it comes to the SDP, I’m supposed to cater for the individual needs of my school.
2
But if the Principal imposes the inclusion of healthy eating, of eco-school, of literacy, of AfL
3
…how many items can be included? I then end up not catering for my school’s basic needs! (P3)
Some Heads (S2, P1 and S3) practise distributed leadership as ‘informed delegation’, with S3 even using the verb ‘delegate’ rather than ‘distribute’, because they must have knowledge of everything for the retention of leadership power in their schools. Practice of distributed leadership therefore leads to the lowering of the ‘threshold of describable individuality’, using these practices as ‘a means of control and a method of domination […] it is a new modality of power’ (Foucault, 1975: 189). This mode of distributed leadership may be considered as a ‘disciplinary technology’ – on a par with the examination – that aims at achieving both objectification and subjectification of the individual, implying more leadership transparency and accountability for the various leadership levels in the college hierarchy as a result of networking. Distributed leadership may be interpreted as a form of panopticism – imposition of a particular conduct on a particular human multiplicity (Foucault, 1991a); the conduct of accountability. I draw myself in a flat structure, positioning myself in the centre of the circle immediately surrounded by my SMT members, reaching out to everybody while leading the school in a transparent manner…I do not believe in hierarchical leadership. In the college I think I occupy the same position. In the sense that the Principal is in the centre and we all surround her. (P2)
There is a certain degree of control present in the distributed leadership narratives recorded here. How is this leadership distributed? It turns out to be an ambiguous concept, because the term itself contains a directional element of power – distributed leadership is done to the others. Do the ‘distributees’ always accept the ‘distribution/s’ of the ‘distributor’? One can detect a battle within leadership discourse in the narratives of some Heads who narrate themselves as distributors who retain control – distributed leadership for some, but not for those at the lower levels (in this case, the Assistant Heads). A paradoxical leadership identity emerges through their exercise of Foucault’s concept of gouvernmentalité (Foucault, 2002a) – where their discourse acts as a form of masked power, albeit unconsciously. The voices of these Heads’ undermine what the Principal said, about ‘Heads working closely with Assistant Heads and not simply delegating tasks’.
The subsequent discussion the Heads (P3, S2, P2, P1, and S3) have regarding problems in the enactment of distributed leadership reveals an ‘intradiscursive battle’ within distributed leadership practice. This ‘intradiscursive battle’ reveals indirectly how the Heads have been subjectified by the discourse of efficiency, accountability and marketization – they want ‘perfection’, ‘my intentions’, and ‘my standards. However, in this way they are taking up the positioning as ‘effective senior leaders’ laid out by FACT. But is this senior leadership practised to such an extent as to be detrimental to the practice of distributed leadership?
The Principal’s discourse regarding distribution is expressed in terms of spatial dimensions: I think that I distribute a lot. I work very closely with the Heads – I have my own space and they have their own space – spaces that come together. But I do not invade their space, and they do not impinge on mine – the spaces are very close to each other but they are not the same. (CP) An effective leader never works on her own. I don’t think that being a leader means being superior to anyone else. It doesn’t mean that if you are a leader, you are at the top and someone else is at the bottom. As Principal, I am not of the opinion that a good idea doesn’t materialize if it doesn’t emanate from me – the last word on the matter shouldn’t always be mine. I think that I do not only preach this reality, but I live it, as well. (CP) Autonomy? I still have to see it being born! I cannot even organize Prize Day the way I like – it has to conform to the Principal’s standards and be similar to that of the other college schools. (P1) The intensification of networking based on communication is the way learning communities can Each school still requires its
In FACT, leaders and leadership are discursively situated as the solution to everyday organizational and professional problems (Gunter, 2001). The performance of leadership is expected to take place within a setting in which there is a specific public purpose and a deliberate casting to ensure agreed outcomes are met through careful adherence to the policy script. Politicians find educational leadership an attractive concept because it places responsibility for success or failure largely in the hands of professionals (Humes, 2000). Heads are promised a sole focus on the provision of ‘professional educational leadership’ (Humes, 2000: 42), with other issues to be handled by appropriate and qualified personnel. Qualifying leadership in these ways may serve to make it more palatable to the potential leaders and deflect attention from its directive aspects.
Power and control remain centralized – a fact that had been previously noted in literature (Strain, 2009 and Smyth, 2011). In addition, the tactics of decentralization have co-opted leaders to act for and on behalf of the State – government-driven reform that requires leaders to enact policies at school level is no less than the downward delegation of the managerialist project, where leaders become the deliverers of reform (Gunter, 2001, 2012).
Conclusions and implications
The distributed leadership dilemma – autocracy and/or democracy?
My findings reveal the appeal of the ‘chameleon-like quality’ (Harris, 2007: 315) of distributed leadership to Maltese policymakers and leaders, each with their own ‘variabilities and dualities’ (Woods et al., 2004: 439) of the term. One can detect its various adaptations to describe an infinite variety of leadership approaches with a specific focus on highly structured forms of limited, or ‘directed’, delegation, rather than genuine distribution, especially at school level.
A dilemma of distributed leadership that emerges in the Heads’ discourse is the selective nature of this co-leadership practice – distributed leadership for some, but not for all – in this case, the Assistant Heads positioned lower in the leadership hierarchy. This dilemma points to the inequality in leadership dynamics (Harter et al., 2006), throwing into question the claim by Katz et al. (2009: 5) of leadership being ‘defined by activity other than formal position’, while simultaneously contradicting Simkins (2005) on the claim of informal influence being more relevant than formal power.
Distributed leadership is specifically set out in FACT as the leadership style to be practised by all the leaders in the college. Leadership practices in DemoDistributa College transpire otherwise. Despite the leaders’ claims, distributed leadership does not unfold as such a prominent praxis. The discursive framework of FACT itself has inherent contradictions with regard to the involvement of the Heads as policy implementers in the enactment of networking, implying the assumption that they will simply be accommodated within the pre-set discursive positionings and roles, with further expectations encompassing involvement. The policy discourse, however, does not unfold in this participatory democratic manner in practice, because not all the Heads are allowed to make their voice/s heard, thus contradicting Jopling and Crandall’s (2006) view of distributed leadership which legitimizes all the leaders’ voices. This underlines the tension between ‘empowerment and direction’ as identified by Smith and Lewis (2011: 384), resulting in an organizational paradox where leadership enactment in this particular Maltese college is ‘directed’ rather than ‘distributed’ from above. 5
Hierarchical forms of accountability are still inherent in the system, because what is to be ‘distributed’ remains within the strategic parameters set by the government (Hartley, 2007), thus creating a tension between collaboration and accountability (Elkins and Haydn, 2004). A control imperative in the relationship between policy and practice (as explored by Gunter and Forrester, 2008) can be detected in DemoDistributa College, with distribution coming downward and used as a form of sophisticated delegation. This points to the ‘constraining’ nature of distributed leadership (Woods et al., 2004), which leads to the ‘dysfunctional dynamics of control-collaboration tensions’ (Sundaramurthy and Lewis, 2003: 399). Despite the hegemony of distributed leadership discourse as constructed and positioned in FACT, the primacy of the ‘solo leader’ remains in the figure of the Principal. The narrative of distributed leadership in DemoDistributa College unfolds as part of the policy rhetoric to claim that power and autonomy are shared with schools and within and across schools, whereas the mundane leadership practices point to centralization and managerialism (when analysed in relation to research in distributed leadership carried out by Hall et al., 2011). 6
Both the Principal and the Heads make claims to leadership distribution according to a set of very specific criteria, which is tantamount to ‘imposed’ distribution. At times, distributed leadership also serves as ‘a smokescreen for the more authoritarian practices’ of the leaders (Crawford, 2012), in which case, the notion of distributed leadership only ‘appears’ to incorporate democratic procedures (Woods, 2004: 22). A distributed leadership perspective may also give rise to autocracy (Spillane, 2005) because leadership can be stretched over leaders without being democratic – evident in the college under study. The discourse of the Principal exposes the ambiguous nature of the term ‘distributed leadership’ – this contains a directional element of power, implying that distributed leadership is done to others. This confirms Youngs’ (2009) findings about the inequality in the leadership hierarchy between the ‘distributors’ and the ‘distributees’, and the subsequent positioning of the locus of power in educational contexts.
The findings presented in this paper may thus be significant for educational practice, policy and theory in terms of the generation of problematization which may lead to further research on this somewhat contested topic. These questions arise: Can distributed leadership and hierarchical leadership co-exist? Is there a single distributor who distributes leadership? Do the distributees ever question the role of their leader as leadership distributor, how s/he distributes, and the intentions behind the distribution? Distribution: Autocracy: Democracy: Direction… It is problematization that responds to these difficulties […] This development of a given into a question, this transformation of a group of obstacles and difficulties into problems to which the diverse solutions will attempt to produce a response, this is what constitutes the point of problematization and the specific work of thought. (Foucault, 2000a: 118)
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.