Distributed Leadership and bureaucracy

No brave new world

The field of educational leadership has been charged with failing to build empirically based knowledge on a progressive basis (Heck, 2015). Instead, there is a tendency towards ‘reliance on new panaceas’, noted as early as half a century ago (Stephens, 1967: 9). The latest idea to improve schools appears and is hawked about for a period of ascendancy, but the fundamental inequalities in outcomes that the latest theory is argued to address persist (European Union, 2015). The latest iteration of this habitual behaviour is the widespread adoption of distributed leadership, promoted as a means to respond to the zeitgeist of a super-complex world, and to replace perceived solo leadership and its alleged inevitable accompaniment: hidebound and pernicious bureaucracy. Repeatedly, distributed leadership is conceived as dismantling ‘the tyranny of bureaucracy’ (Fitzgerald, 2009: 51), enabling all to ‘work and learn beyond bureaucratic enclosures’ (Hairon and Goh, 2015: 694) to the benefit of learners.

This article considers distributed leadership, placing it as part of the extensive literature on post-bureaucratic organisations. It also considers bureaucracy as one of the adaptations and developments of organisation reported in this body of literature. Some studies adopt a thesis that experimentation with new forms of organisation stems from the ‘imminent demise of bureaucratic organisations and their irrefutable replacement’ (Johnson et al., 2009: 38). New post-bureaucratic theories of organisation, including distributed leadership, are argued to be characterised by factors that are the antithesis of the building blocks of bureaucracy (Grey and Garsten, 2001). A different narrative runs in parallel. This insists that bureaucracy is the most enduring of organisational forms, that it is embedded in the ‘grammar’ of all institutions (Hartley, 2010: 282) and that the reason for its persistence is its utility as support for their positive functioning. This article argues that both bureaucracy and distributed leadership are ideal types, and that both are potentially related to abuses of power that contribute to the stasis of educational organisations. It further argues that, of the two, bureaucracy has engaged more stringently with issues of power, is more prevalent across the world and offers a necessary means of engaging with current leadership challenges.

Bureaucracy

Many people have an understanding of the form of organisation that Weber (1947) described as bureaucracy, as in Greenwood and Lawrence’s (2005: 497) summary: ‘a fixed division of labour (horizontal differentiation), a hierarchy of authority-based positions (vertical differentiation), written documents and general rules (standardization and formalization), and the use of expert personnel (specialization)’.

Weber’s original intention was to take an historical perspective on how forms of organisation had evolved in response to the changing environment and to explore the place of human values. However, his depiction of bureaucracy has repeatedly been simplistically caricatured and attacked as destructive (Fitzgerald, 2009), losing sight of his complex and nuanced engagement with the relationship of organisational form and human capacity (Samier, 2002). Bureaucracy has, at times, become a convenient whipping post used to excoriate all that is ill in current leadership: the ‘bacillus of bureaucracy’ (Courpasson and Clegg, 2006: 323). It is often depicted as a key tool in subjugating workers to management and/or the state, and linked to concepts such as neoliberalism (Hyslop-Margison and Leonard, 2012; Gobby, 2013), managerialism (Meyer, 2002) and a performative culture (Burnard and White, 2008). In Du Gay’s (2000: 1) laconic summary, ‘These are not the best of days for bureaucracy’. Yet the challenges remain that Weber, and others since, set out to address. How can organisations both coordinate human effort and nurture innovation and initiative (Courpasson, 2000)? How can a range of human motivation and resultant behaviour be accommodated within a productive framework (Blau, 1963)? How can organisations, driven by imperfect human beings, be made to work?

Weber insisted that ‘every sphere of social action is profoundly influenced by structures of dominancy’ (Weber, 1968: 941). Given widespread agreement that resulting hierarchies are ubiquitous in organisations and in social relations (Magee and Galinsky, 2008), all theories of organisation and of leadership are fundamentally shaped, explicitly or implicitly, by notions of ‘the problems of control and power’ (Crozier, 1964: 148). The capability to achieve desired outcomes – that is, power – is sometimes confused with the sources of that capability – role, gender, expertise, control of resources and so on – but they need to be distinguished. Power itself, irrespective of its source, is conceived in many different ways (Lumby, 2013). Giddens’ (1984: 257) general notion of ‘the capacity to achieve outcomes’ needs to be supplemented with an explanation of the kind of social relations envisaged to achieve outcomes. Concepts of power range from the ability of one individual to impose his/her will on another to Orwellian-type ideas of influencing the thinking of others so that no compulsion is necessary to achieve a desired change; people think and as a result act as intended by those in power. Foucault (1979) proposed a notion of ongoing fluid negotiation of power relations where power is a changing, ever-present shaper of social relations and organisations. This kind of understanding of power is emerging in analyses of distributed leadership (Woods, 2016). From this perspective, control of others is then inevitably partial and fluctuating, so that any form of leadership that assumes perfect control of others is untenable. Equally, a form of leadership that is predicated on the deliberate distribution of power to others is unrealistic.

Leadership is intimately bound up with such complexity and, just as ‘conceptual pluralism’ (Bolman and Deal, 1997: 11) is needed to view leadership from varying perspectives, a similar plurality of concepts is needed to unravel the use of power in leading. An assessment of the utility of leadership theory in supporting leaders, and in particular in addressing the inequities embedded in education, requires scrutiny of how power is understood and embedded in action. Crozier (1964: 145) suggested that power is the ‘central problem of the theory of organisation’ and that power, in all its complexity, has remained a key concept in understanding how organisations, including bureaucracies, function.

Bureaucracy and power

Weber originally conceived of bureaucracy as an organisational form that potentially enabled human beings to achieve the greatest capacity when working together: ‘the most rational known means of carrying out imperative control over human beings. It is superior to any other form in precision, in stability, in the stringency of its discipline, and in its reliability’ (Weber, 1947: 337). Others have noted the positive contribution of bureaucracy (Dahlström et al., 2011; Greenwood and Lawrence, 2005; Gronn and Woods, 2009) and feared that over-hasty moves to new forms of organisation might be ‘bought at the expense of guarantees of honesty and fair dealing and of security and resilience’ (Hood, 1991: 16). However, discussion on the positive aspects of bureaucracy has been overwhelmed by charges of ‘offences ranging from the relatively banal – procrastination, obfuscation, circumlocution and other typical products of a ‘red tape’ mentality – to the truly heinous – genocide, totalitarianism, despotism’ (Courpasson, 2000: 1).

In the field of educational leadership, the negatives have generally been highlighted.

The range of critiques of bureaucracy and the reasons for the approach of each critic would take a full article to explore, but a spectrum of thrusts is apparent from even a cursory glance at the educational leadership and management literature. Wenger argues that communities of practice are impeded by bureaucracy because ‘the formality of the bureaucracy can come in the way of open knowledge sharing’ (Wenger, 2011: 4). Apple (2013: xv) suggests that bureaucracy binds schools to the mindset of commercialism and ‘metaphors of markets, profits and the accountant’s bottom line’. Chubb and Moe (2011) analyse the problem differently: it is not bureaucracy in schools, but the degree of bureaucracy in schools that is the problem. Also, they believe that bureaucrats tend to work to bolster their own power rather than to support professionalism and innovation. It would appear that each author has a particular and very different angle on why schools are not as they would have them be, and bureaucracy is cited as a cause in each case. Bureaucracy appears to offer a useful chameleon-like nature that is adaptable to criticism of education from many different angles. As such, it is a most helpful tool for those looking to identify what to blame. Though there is occasional acknowledgement that bureaucracy – or at least a key element of bureaucracy: hierarchy – persists, at best this tends to be couched in rather concessionary terms (Harris, 2013) and, at worst, excites excoriating criticism (Fitzgerald, 2009).

This is not to argue that the ideal type of bureaucracy presents no issues. Despite his statement that bureaucracy offers the potential for the most efficient form of human organisation, Weber (1947) recognised its inherent dangers. Others have not been slow to follow in cataloguing the problems. The poor fit between the perfect rationality of the bureaucracy ideal type and messy, irrational, emotional human beings was exposed from the inception of the concept: ‘Research has demonstrated that the ideal type of bureaucracy is far from being completely efficient’ (Crozier, 1964: 177). Any assumption that workers would unquestioningly follow the instructions of their superiors was exposed by Crozier’s (1964) empirical research on French bureaucracies. His research showed that workers play a power game, whatever level of the hierarchy they inhabit. More recent research into supposedly hyper-bureaucratic organisations, the universities of the UK, has uncovered similar power play as fundamental to leadership (Lumby, 2015). The critique of bureaucracy, that too much power is placed in the hands of those in leadership roles in the hierarchy, underplays the complex, contested and fluid nature of power.

Ongoing analyses have continued to deepen our understanding of both bureaucracy and the flows of power and resistance that are integral to its nature (Du Gay, 2000). The idea that bureaucracy is a means to embody the domination of many by one or a few leaders ignores decades of developments in thinking based on research on bureaucracies (Diamond, 2013). However, this research has been conducted in disciplines other than educational leadership, which has generally rejected bureaucracy as a relevant and positive conceptual frame to research leadership in schools. Other fields offer broader and more varied conceptualisations; for example, research of simultaneously formal and informal hierarchies (Diefenbach and Sillince, 2011) or variations in leaders’ control, accountability and empowerment in post-bureaucratic organisations (McKenna et al., 2010).

Setting aside crude criticisms of bureaucracy as mere domination, a more nuanced discussion recognises that bureaucracy embeds a sophisticated framing of power. Weber did not shirk from articulating the necessity for organisations to secure the obedience of employees. Such terminology is likely to alienate current educators, schooled in the language of empowerment and professional autonomy. However, clearly, some degree of obedience is required. This does not equate to simplistic notions of domination. Employees are not automata, to obey every instruction, yet there must be some kind of framework that does not demand critical consideration and agreement on every action. Teachers and lecturers must turn up each morning, must prepare their classes and teach well. Given such a foundational framework, analysts of bureaucracy have nevertheless moved far beyond the idea that undertaking the tasks necessary to secure effectiveness involves robot-like conformity to the requirements of authority. Rather, Blau identified the necessity for ‘voluntary compliance with legitimate commands and suspension of judgement in advance of command’ (Blau, 1963: 28). Staff therefore act neither with complete autonomy, nor with complete obedience.

Neither is there any shrinking from acknowledging that part of the system is pressure to comply: what Du Gay refers to as ‘soft coercion’ (Du Gay, 2000: 154). There are disincentives and penalties for those who move too far outside what is expected; for example, those who discipline children inappropriately or fail to produce assessments in a timely way. The ‘reflexivity of actors’ (Du Gay, 2000: 154) makes judgements about the legitimacy of those in authority roles, in part because personal reputation rests on a degree of conformity, and in part in response to the ethical stance of the leader. Bureaucracy has come to be detached from any notion of ethics in current thinking in the field of educational leadership, yet its founding principles were very much about the shaping and containment of power. Du Gay (2000: 3) acknowledges that most see bureaucracy ‘as inherently unethical’, but argues that this is a distortion, a kind of unthinking propaganda against bureaucracy:

The ethical attributes of the ‘good’ bureaucrat – adherence to procedure, acceptance of sub- and superordination, commitment to the purposes of the office and so forth – do not therefore represent an incompetent subtraction from a complete all-round conception of personhood. Rather, they should be regarded as a positive moral and ethical achievement in their own right. (Du Gay, 2000: 4)

(Bolden et al., 2012: 14) that educators employ, based on reflexivity. However, this is within a bureaucratic framework of rules and requirements that are generally obeyed because of the perceived legitimacy of the rules themselves (Lumby, 2015). Bureaucracy assumes a degree of obedience, but also enables individual judgement. Critical consideration of every action by all actors would lead to paralysis, not least because staff often do not agree on what should be done. Consensus is a fine thing, but even a cursory knowledge about staff meetings and corridor conversations brings an understanding that empowerment and autonomy cannot be assumed to lead to agreed action.

Contrary to popular views of bureaucracy as the rigid imposition of control, the concept of power most evident here is closer to that of Arendt, who insisted that: ‘When we say of somebody that he (sic) is in power we actually refer to his being empowered by a certain number of people’ (Arendt, 1970: 44); that is, ‘a concept of power and law whose essence did not rely on the command–obedience relationship’ (Arendt, 1970: 237). Bureaucracy is assumed to rely on a ‘command–obedience relationship’, but a bureaucratic structure, with its insistence on appointment by objective assessment of expertise and experience, and action that stays within defined ethical parameters, endows leaders with legitimacy. Such legitimacy shapes the organisation so that educators are not making decisions about every action every day. Rather, they can rest on legitimate practices which, nevertheless, they are able to judge and influence. Bureaucracy does not deny the presence of required compliance, which would be both unrealistic and dishonest. Rather, it provides limits to its exercise; it flavours compliance with legitimacy. Delving further, an undertow of micro-politics is reached. As Crozier learned in his study of French bureaucracy, ‘a human being…is free to decide and to play his (sic) own game’ (Crozier, 1964: 149). A constantly changing flow of power relations results, as described by Foucault. One example provided is the power that relates to expertise. Crozier (1964) pointed out that such power, unchecked, would be as potentially detrimental as any power, whatever its basis, which is not confined or balanced in some way. The benefit of the interplay of expert-based power and role-based power within a bureaucracy is indicated. He further suggests that, as the new knowledge and skills introduced by those who have expertise become widely known, the basis of the power of the initiator declines. In consequence, power that derives from control of knowledge is self-limiting. The authority invested in formal roles is in tension with the flows of power that derive from expertise. In summary, in Crozier’s (1964) thesis, the power plays of resistance, considered by some to be the dark side of bureaucracy, are symbiotic with its functional aspects and at the heart of how bureaucracy operates: the negatives are intrinsic to the positive functioning of the organisation.

Two conclusions emerge from this brief exploration of power in bureaucracies. First, the concept of bureaucracy acknowledges the existence of power as a vital element of both organisational functioning and leadership. Indeed, its basis is to shape power productively and control it within ethical parameters. Second, over time, analyses of bureaucracy have involved a complex understanding of power flows and power plays that are very far from the simplistic notions of the rigid management control and hyper-standardisation of popular thought. The ideal type of bureaucracy is a heuristic stimulating reflection, not a rigid formula preventing thought.

Distributed leadership

Distributed leadership is not underpinned by a clear definition in the way that bureaucracy is. Its definition is problematic, in part because of the bifurcating path of its history. In its earliest incarnation in the field of educational leadership and management, distributed leadership was adopted as a research framework through which to understand the totality of leadership practice in an organisation, including both formal and informal, planned and emergent activity (Gronn, 2000; Spillane et al., 2004). It rapidly took another route and became a particular practice of leadership and, beyond that, the most promoted form of leadership practice in the first decades of the twenty-first century (Parker, 2015), including ‘romanticized and idealized accounts of distributed leadership in practice’ (Harris and DeFlamis, 2016: 143). Despite its rapid spread and enthusiastic adoption by many, the nature of distributed leadership remains opaque. Following their meta-analysis of hundreds of articles, Tian et al. (2016) conclude that there is no clear definition of distributed leadership.

Nevertheless, a variety of generally rather broad definitions are in use. Jones et al. (2014: 604) conclude that ‘the common ingredient in each of these theories is the acknowledgement of the role of leadership at multiple levels, both formal and informal, and the need for collaborative networks to engage within complex systems’. Leadership at multiple levels could equally apply to hierarchical systems, and many discussions of distributed leadership acknowledge the persistence of hierarchy. The distinctive nature of collaborative networks is more difficult to define exactly. Jones et al. attempt it:

a more holistic perspective of organisational work and a focus on emergent approaches. This enables the complexity of interactions that occur between various subjects, stakeholders and other interested parties, viewed in context with the artefacts and instruments (rules, community and division of labour) that affect these interactions, to be identified as an activity system. (Jones, et al., 2014: 606)

Despite its popularity and rapid spread, distributed leadership has attracted much critical comment. A particular focus of critique has been the relationship of distributed leadership and power. If theories of leadership are fundamentally about power relations within organisations, distributed leadership has generally been found wanting. Hatcher (2005: 256) points out that theories in the field of educational leadership have detached distributed leadership from the original underpinning within other disciplines, with the result that they ‘do not have an adequate theory of power’. Within the distributed leadership literature, power is often referred to as being distributed and there are references to overarching concepts, such as democracy or empowerment, which imply or connote changed power relations (Woods and Gronn, 2009). The preference of some to remain in a control system in which instructions are given is set aside. The more critical literature expresses further doubts. Bolden et al. (2009) suspect that, within higher education, distributed leadership is actively used to disguise underlying power dynamics. How, then, is power conceptualised within distributed leadership? What mechanisms are in use? How far are the putative power differentials of hierarchies undermined or not?

Distributed leadership and power

A characteristic claim of distributed leadership is that it diminishes the power of the solo leader, enabling others also to take a leadership role. For example, Murphy et al. (2009: 182) suggest that leaders promoting distributed leadership are ‘rethinking conceptions of power’. In a longitudinal study of six schools, Murphy et al. accuse current structures of failing to meet the needs of learners and of inhibiting change, in general, and shared leadership, in particular. They argue that new structures are needed. One example given of such changed structures to accommodate shared leadership is an adjustment in timetabling that enabled staff to come together to discuss and plan work. However, the resulting meetings appeared to be rather ineffective until ‘the principal tapped someone to act as the informal leader for the various groups’ (Murphy et al., 2009: 189). The principal also provided protocols and a booklet in which to record the activity of the meetings. ‘Halfway through the year, she actually collected those notebooks and took a look at them…[to] make sure people were doing the work’ (Murphy et al., 2009: 189). This looks remarkably like the principal in a formal authority role continuing to select leaders, delegating work and shaping the parameters within which they function. Power remains largely unaltered.

This is but one of various accounts of a school or university asserting that it is using distributed leadership, when the evidence suggests that power relations remain undisturbed (Bolden et al., 2009; Harris and DeFlamis, 2016; Woods, 2016). Hairon and Goh discovered the same limitations in changes in power in Singapore. Their study started with the premise that distributed leadership requires ‘the relinquishing of decision-making power’ (Hairon and Goh, 2015: 698). They go on to assert that, in the distributed leadership system, influence can emanate from anyone and that it does not equate to delegation, where the principal imposes or achieves agreement for another to take on a leadership role. However, their study exposes that, within this putative distributed leadership system, ‘empowerment is distinctly bounded…. In other words, teachers’ autonomous decisions are not without the superiors’ knowledge and approval, even if it may be silent’ (Hairon and Goh, 2015: 708). This accords with the conclusion to be drawn from Murphy et al.’s (2009) case study: that the notion of power being distributed to a much wider range of people who are able independently to take decisions and initiatives is not supported by evidence.

Despite all the objections and assertions to the contrary, what appears to be happening resembles much more closely formal and informal delegation within a bureaucratic system. One might argue that this points to a failure of a distributed leadership system to be such, but the problem is that the researchers are arguing that this is indeed distributed leadership in action. From their meta-analysis, Tian et al. (2016: 157) agree with other researchers that ‘distributed leadership entails a deliberate organisational redesign by the principal and purposeful engagement by the other school staff’. However, they conclude that, ‘Nonetheless, so far, researchers have been unable to clearly describe how different agents use their initiatives to influence leadership work’. While there is scant evidence that power, if it is defined as the ability to take decisions and implement them, moves from those in formal authority roles and in particular the principal, there is considerable evidence that power as it is analysed within bureaucracy remains; that is, those in formal authority roles retain power.

Nevertheless, power relations can and do change persistently. Authority is buoyed and resisted, shaped by micropolitical activity and the kind of power flows envisaged by Foucault (1979). Theoretically, and apparently in practice in a large-scale US project (Supovitz and Riggan, 2012), a greater capacity to take decisions can be developed in staff. A key concern is how realistic it is to imagine that this happens, or could happen, reliably. Cryss Brunner explored the aspiration of specifically female school superintendents and principals in the US to espouse ‘power with/to’ (Brunner, 2000: 134) or ‘coactive power’ (Brunner, 2000: 137), rather than ‘power over’ (Brunner, 2000: 138). Her research suggests how difficult it is for a principal to abandon ‘power over’.

Cryss Brunner’s work with female leaders in schools also uncovered a paradox. In apparently giving away power to others, you increase your own. One of the women professionals explained that ‘by enabling your people and improving your people you really give yourself a power’ (Brunner, 2000: 145). Equally, some believed that the explicit use of power as domination served to rob one of power. Some women in the study felt intensely uncomfortable discussing power, preferring to refer to leadership. These findings are evident in other education sectors, such as universities, and are true of men as well as women. As a registrar in a study of UK higher education put it: ‘We don’t like using the word power’ (Lumby, 2015: 58). Apparent rejection of power as domination is a preferred stance. Another respondent believed: ‘If you have to exercise power in its naked form you’re perhaps doing something not quite right’ (Lumby, 2015: 59). Respondents suggested that middle leaders are comfortable acknowledging power with, or power to, or power for, which makes theories such as distributed leadership attractive, but also that solo uses of power were necessary, especially as a leader of large numbers of staff.

In an education environment, cultural pressures demand that power remains hidden, or ‘unobtrusive’ (Pfeffer, 1981: 137). Nevertheless, it is a resource that leaders are often aware of and wish to retain or build, in part, paradoxically, through endowing others with power. The concept here is very different to the idea of power as a zero-sum game (Parsons, 1963), as a commodity which can be given away and that you therefore no longer own. Rather, an individual who apparently gives away power thereby increases his or her power. Those who employ power to impel become less powerful.

The overt concept of power in much writing about distributed leadership is of power as single dimensional, a zero-sum game, a commodity released by the principal or other senior leader to staff who were previously considered more junior within a hierarchy (Harris and DeFlamis, 2016). The explicit concept is one where the apparent sharing or endowment of power decreases the power of the solo leader in favour of others. An alternative analysis of the relevant literature suggests that leaders both augment their own power by empowering others and simultaneously retain a hold over what happens through their authority role.

Murphy et al. suggest that another structural inhibitor of shared leadership by which the mechanisms of power can be discerned is who is actually able to become a leader. They provide an example in their case study of ‘principal favouritism that hindered the spread of leadership beyond the leadership team’ (Murphy et al., 2009: 187). The principal appeared to control the process of access to leadership roles, ‘appointing or anointing teacher leaders’ (Murphy et al., 2009: 187) from those in their inner circle. Favouritism was perceived by staff. The workings of power catapult some into becoming persons of influence and others not; in effect, elites are created. Crozier (1964: 149) stated that ‘the elite is made up of those who have merited good marks in the competition of life, or drawn winning numbers in the lottery of “social” existence’. Belief in the equality and empowerment brought about by distributed leadership can only survive if blind to the considerable literature from sociology, psychology and anthropology that details ‘the interrelated topics of stereotyping, prejudice, intergroup relations, gender, race, and class discrimination’ (Sidanius and Pratto, 2001: 3) and the equally extensive literature on power.

It is not that human societies and organisations cannot embed equality, but that the overwhelming evidence suggests that they do not. In the face of this evidence, the pursuit through distributed leadership of ‘symmetrical power, where every member in the organization has equal opportunity to assert influence over another regardless of hierarchy’ (Hairon and Goh, 2015: 710) appears wilfully disingenuous. In studies of distributed leadership, acknowledgement of the persisting exclusion of women and black and minority ethnic people from leadership is becoming more frequent (Diamond and Spillane, 2016; Woods, 2016), but, as yet, there is far less emphasis on researching whether the emergence of leaders in a system claimed to be distributed reflects the same levels of discrimination and differential power as are evident in formal appointment systems. Rather, under pressure from ongoing criticism of distributed leadership theory, undisturbed power relations in terms of who accesses leadership are beginning to be acknowledged.

The nature of the workforce

Distributed leadership as a research framework for understanding the complexity of the leadership of an organisation may remain a useful lens, though early proponents such as Gronn (2016) have more recently questioned even this. The second incarnation of distributed leadership as a specific way of leading is predicated on a particular set of assumptions. It views leaders as a key factor in establishing effective, efficient and humane systems, and assumes their wish to do so. It also assumes that education staff, if given the opportunity, will contribute to the best of their ability to the benefit of learners and the organisation more broadly. It is an almost perfect embodiment of McGregor’s Theory Y, that:

The motivation, the potential for development, the capacity for assuming responsibility, the readiness to direct behaviour towards organizational goals are all present in people. Management does not put them there. It is a responsibility of management to make it possible for people to recognize and develop these characteristics for themselves. (McGregor, 1960: 169)

While Theory Y is demonstrated by many dedicated education staff worldwide, who work hard to forward the interests of learners, there is considerable evidence that many educators do not exhibit solely either Theory X or Theory Y, but, persistently, a mix of the two (Watt et al., 2012). International evidence that has persisted over time suggests that assumptions of the wholehearted commitment of teachers above all else to the benefit of learners idealises and misrepresents the nature of the school workplace environment. At the very least, it suggests that the promotion of distributed leadership by organisations such as the Organisation for Economic Co-operation and Development across a range of cultures and geographic regions, irrespective of the likelihood of emergent or willing leadership by many, is insensitive to the different economic and cultural conditions prevailing in schools, colleges and universities across the world. Leaders in formal positions, too, are not always competent, benign or culturally aligned to the demands of distributed leadership (Eacott, 2011; Hatcher, 2005; Wright 2001). Harsh approaches to leadership are often cast as the outcome of increasing bureaucracy (Schlechty, 2011) but, as argued earlier in this article, such practice is an abuse of power: a distortion of bureaucracy, not its embodiment. Nevertheless, many endorse distributed leadership as an antidote to such behaviour and it has been globally promoted on this basis (Firestone and Martinez, 2007; Woods et al., 2004).

Bureaucracy revived?

The discovery of persistent hierarchy in sites where distributed leadership has been researched suggests that at least elements of bureaucracy are a near-universal approach to leadership. While Murphy et al. (2009: 186) may suggest that, to accommodate distributed leadership, ‘schools will need to be restructured in significant ways’, there is little compelling evidence that such shifts have taken place. On the contrary, not only in schools but throughout public organisations, ‘bureaucracy, far from being superseded, is rejuvenating’ (Courpasson and Clegg, 2006: 319).

The interaction between human behaviour and organisational forms creates an imperative to seek those forms that are most likely to result in behaviour with positive outcomes for learners. It is not argued that bureaucracy is necessarily that form. Such absolutes are untenable. However, when Weber (1946) first conceived of the notion of bureaucracy, he believed that any idea of replacing it was utopian, and so it has proved. Bureaucracy arguably remains a universal within education organisations, manifest not just in systems close to the ideal type described by Weber (1947) but in a variety of hybrids that have since developed. It potentially embodies a range of positives, including sophisticated balances, checks and protections. These advantages do not always appear: ‘Real organizations will typically embody a mix of enabling, coercive, and ceremonial forms of bureaucracy, and this mix would leave employees ambivalent about the overall phenomenon’ (Adler, 2012: 248).

Unfortunately, ambivalence has been replaced by a hostility reflecting stereotyping. Just as stereotyping people results in facile, negative assumptions that may seriously underestimate their abilities and limit their potential, so bureaucracy has become a kind of organisational stereotype. Such prejudicial dismissal of an enduring organisational form and its proposed replacement by distributed leadership is another manifestation of a sequence of reliance on the theory of the moment in educational leadership that was noted at the start of this article.

As a consequence, despite the near universality of bureaucracy, it is rarely the subject of study while research on distributed leadership continues to burgeon and claims about rigorous research underpinning distributed leadership’s positive impact grow (DeFlaminis, 2013). There are problems with such research. First, though student test results may be one side of statistical analysis, the variables on the other side are generally garnered by staff self-report: what staff believe to be the case. What staff think and feel is an indicator, but it has the same weaknesses as any self-reported qualitative data, even when subjected to statistical analysis. If staff have bought into the distributed leadership approach, their perception is likely to be biased towards this concept. Second, when researchers or consultants work with schools to support the development of distributed leadership and point to improved outcomes, there is no counterfactual (Levačić, 2005). Other schools, which have not received an equivalent level of attention, may not improve to the same degree, but the Hawthorne effect would ensure those receiving attention from researchers are likely to improve whatever concept was in use. If there were a comparison with other schools in which research teams were attempting to work within a different conceptual frame to effect improvement – say transformational leadership, for example – the distinctive claims of distributed leadership would be more convincing.

However, such is the hold of distributed leadership that a paradigm shift would be needed to effect sufficient change in research. If the field were to fully acknowledge conceptual and empirical weaknesses and shake off the emotional hold of the concept, it would then be in a position to take account of the much wider literature beyond educational leadership that relates to organisational development and leadership in contexts that have evolved since bureaucracy was conceived. Research could focus, at least in part, on the more enduring framework – bureaucracy – and research the accommodations of power, decisions and actions made within authority frameworks and, in turn, the effect of these on learners’ experience and outcomes. More account of cultural differences and of the enduring discrimination that has limited the potentialities of staff could be factored in. This would not shut the door on recognising emergent or spontaneous leadership, but would place it more accurately within a power system, acknowledging that its existence is ultimately facilitated by explicit or implicit allowance of those in authority roles, who gain power through enabling initiative. If bureaucracy is the more enduring framework, then a multiplicity of approaches to leadership within such a framework could provide a rich resource in understanding power flows and their impact. More sophisticated research, using conceptual pluralism rather than the monotheistic concept of distributed leadership, offers a more productive future.

This article is but one of a series of articles over time that have critiqued the rigour and progress of the field of educational leadership and management from a variety of perspectives (Hallinger and Heck, 1996, 2005; Tooley and Darby, 1998). Criticisms are generally met with a robust defence of the status quo. Nevertheless, researchers and practitioners of leadership continue to have the opportunity to reject comfortable, but ultimately limiting, notions of leadership to grapple with the tough problems of authority and power, which are at the core of leading. This article has only had space to indicate in broad brushstrokes the nature of the wider literature on organisational development and its relationship to leadership. Much work remains to answer the question initially posed by Weber as to how human beings can work together most productively in organisations. Specifically, how do formal and informal hierarchies interact and impact on learners? How do formal and informal processes of decision-making interlock and to what effect? How do staff find their place in formal and informal hierarchies? There is little research on formal appointment processes, which remain something of a black box. There is even less on how ‘appointing or anointing’ (Murphy et al., 2009: 187) to informal roles happens. To imagine either formal appointment or informal adoption of leadership role rests purely on objective criteria or self-choice, irrespective of bureaucratic and power structures, sets aside a considerable literature suggesting the contrary. We need research studies of conceptual and methodological complexity if the field is to overturn a history of inequity in educational organisations, both in achieving a leadership role, whether formal or informal, and the processes and outcomes that learners experience. These are just some of the questions, the answers to which might carry the field forward to the benefit of learners and allow us to offer a complexity of analysis of bureaucratic and post-bureaucratic organisations comparable to other disciplines.