New perspectives on middle leadership in schools in England – Persistent tensions and emerging possibilities

To afford school middle leaders meaningful opportunities to initiate change, we must provide them with the space and flexibility to engage with agentic and creative responses to policy and practice. Whilst we argue that the tensions identified in Bennett’s seminal reviews persist, there may, nonetheless, be opportunities for school middle leaders to creatively influence educational agendas. Through engaging in a critical interpretative synthesis of school middle leadership literature, we consider how the subjectivities of such leaders are discursively constructed. We argue that a culture of performativity has diminished opportunities for middle leaders in English schools to develop a strong sense of agency, educational ideology and authentic professional responsibility. However, a current governmental focus on subject knowledge may have opened spaces for a collegial agency, despite the prevailing neo-conservative policy discourse. We thus identify, the potential for movement beyond a discursive position to one where school middle leaders take greater responsibility for developing practice to align more closely with their educational values. Utilising a dialogic theoretical perspective we examine how middle leadership in English schools is currently practiced and mediated in relation to the changing political landscape, and suggest that seemingly contradictory positions provide a fruitful site for new research.


Introduction
The purpose of this paper is to overlay key trends in the Conservative government's educational policy onto the canon of scholarship relating to middle leadership in schools to re-explore the field, and to identify fruitful possibilities for future research.The field of school middle leadership research in England shapes and is shaped by educational policy and practice.This interconnectivity between research, policy and practice means that a timely examination of the field is called for to recognise key 'policy, actors, discourses, connections, agendas and solutions' (Ball, 2021: 3).Additionally, Francis (2015: 440) has stressed how enmeshed we are as researchers in this complex milieu, and that we 'are all involved in the production and evaluation of these discourses.'It is these temporal entanglements that have developed over the last fifteen years that we want to unpick in this paper, and in doing so, revisit the issues raised in Bennett et al.'s seminal reviews of 2003 and2007 to identify enduring tensions and potential possibilities that have emerged from the literature.Our paper identifies three inter-related strands to contribute to future thinking about middle leadership: (1) The field is constructed from competing and polarised perspectives.Usually, research is carried out from either an instrumental/scientific or critical/emancipatory position (Gunter, 2016).Although the contradictions and tensions for practice are recognised, this in itself creates 'discursive site[s] of contestation' (Francis, 2015: 449).While we acknowledge the necessity of these positions for knowledge production in the field, we suggest an alternative approach that allows an exploration of these contestations facilitating the exploration and understanding of how middle leaders practice and mediate the tensions identified.
(2) While the relationship between policy and practice is not straightforward (Ball et al., 2012), and it is outside the remit of this paper to fully theorise this, we have suggested that recent policy developments may be stretching the key tension of controlled decontrol (du Gay, 1996) to a point that it has unintentionally allowed agentic spaces to open up.Conservative educational policy continues to discursively construct the subjectivities of middle leaders through neoliberalism and a culture of performativity.This suggests limited opportunities for middle leaders to develop a strong sense of agency, educational ideology and authentic professional responsibility (Ball, 2021).Within this paper, we want to explore a seemingly contradictory position: how constraining neoliberal policy developments may be affording spaces for an agency for those middle leaders who are tasked with leading learning either as heads of department or subject leaders and who we will frame as practicing pedagogical leadership (Male and Palaiologou, 2017).
(3) We propose that future research using a dialogic theoretical framework (Davies and Goodley, 2021;Francis, 2015;Holland et al., 1998;Braathe and Solomon, 2015) could prove a fruitful way to explore how middle leaders have responded to this new policy landscape, and to develop new conceptualisations of middle leaders and their practices.We argue that this approach exposes possibilities for research that move middle leaders beyond discursive positioning to one that acknowledges their complex relationship with policy discourse, and its important implications for mediating agency within their practices.

Our approach
This paper offers a critical interpretive synthesis of the literature that has allowed us to develop new perspectives and insights around the construction of middle leadership discourse since 2007 through critical review and analysis (Dixon-Woods et al., 2006).It builds upon our recent literature review exploring middle leadership which was commissioned by Ambition School Leadership to inform the training and development needs of Middle Leaders with a focus on Subject Leaders (Hammersley-Fletcher et al., 2018).This work constituted a comprehensive literature review of the school middle leadership literature and summarized current evidence in relation to the development needs of subject leaders.In this paper, we consider the nearly fifteen years of literature production on the activity of middle leadership since |Bennett et al. (2003,2007) so that we could provide an analysis of how the field has evolved since this liminal stage.Significantly, we found that the research from an English school perspective was somewhat limited; therefore, our original review and this paper also draw upon relevant international studies.In contrast to a recent bibliometric review by Harris et al. (2019), this paper adopts a more qualitative stance and considers how Conservative educational policy developments have affected constructions of middle leadership.
Our earlier review (Hammersley-Fletcher et al., 2018) adopted a systematic and positivist approach to identify relevant sources using a broad range of pre-defined search terms, followed by snowballing to mine further citations.The studies were then collated and grouped according to the themes covered within them and used to provide a summary of the literature relating to middle leadership in schools in England.A comparative analysis was also conducted, drawing upon the literature, to identify the development needs of middle leaders in charge of specific subjects.While this previous more instrumental approach focused on identifying the skills, knowledge and abilities required by middle leaders, the current paper re-explores the literature including relevant articles published since our review from a poststructuralist perspective in order to identify the meta-narratives and dominant discourses that have shaped this body of work (Dixon-Woods et al., 2006), and to then suggest future research approaches.We are thus able to problematise and challenge constructions of middle leadership in relation to the policy context.In doing so we acknowledge research as social practice (Gunter, 2016) and that knowledge is assembled and (re)produced within changing policy and practice environments (Thompson, 2017).
Whereas our earlier systematic review sought to collate ideas and findings from across the middle leadership literature and present them in the form of an accessible summary to inform Ambition's development programmes, the approach used here moves beyond 'aggregation' of the ideas and findings within the middle leadership literature towards a discursive analysis of the field (Dixon-Woods et al., 2006: 2;Francis, 2015).In other words, we returned to the literature base and the themes identified within the original conventional review and used them to develop a 'synthesising argument' (Dixon-Woods et al., 2006: 5) in relation to policy developments.This process allowed us to critically explore how the changes in the socio-political educational context have altered the way in which knowledge about middle leadership is constructed, and consequently how middle leadership is practiced and understood.While the initial thematic analysis conducted for the original review, led to a descriptive organisation of ideas from the literature into themes (e.g.responsibilities of the middle leader) and sub-themes (e.g.middle leader as administrator, middle leader as an innovator), the analysis conducted for this paper sought to synthesise these themes into a deeper conceptual narrative.In practice, this involved an iterative process of organising and reorganising the original set of themes into a structure which supported articulation of the relationships between the various ideas, policies and actors (Braun and Clarke, 2006).
In the section that follows, we begin with an overview of the changing policy context and the way that it acts to construct middle leadership in particular ways.We then explore the three strands, posited in the introduction and make suggestions for future research.

Middle leadership as constructed by the policy landscape
While it is hard to define (De Nobile, 2018), the term middle leadership tends to be applied to those within school structures who inhabit roles in the layer between senior leadership teams and teachers.It is also noted, that as school governance systems become more complex and diverse (e.g. in the wake of academisation), middle leaders might be considered more broadly to be anyone who is 'sandwiched between' different hierarchical levels (e.g. a headteacher working under a CEO) (Hammersley-Fletcher et al., 2018).
The linking of education with national economic imperatives creates powerful policy discourses in educational contexts that discursively shape and construct middle leadership practice and identities.Yet, the concept of middle leadership is contested (Hammersley-Fletcher and Strain, 2011) and complex (Forde, 2011), resulting in ambiguity when exploring notions of middle leadership, especially from a functional perspective (Fluckiger et al., 2015).Indeed, the complexity of the middle leader role is compounded by the fact that they are both teachers and leaders (Bennett, 2007) and may need to draw upon differing sets of practices.We argue some recent policy developments are shaping middle leaders as pedagogical leaders whereby there is a congruence of the leader/teacher role (Male and Palaiologou, 2017), especially relevant for those middle leaders who lead subject-specific departments or areas.These are usually referred in the literature as either department heads or subject leaders.
Notions of middle leadership first came to prominence as New Labour education policy emphasised the role of leadership as a means to bring about school improvement (Gunter and Forrester, 2008).Despite their rhetoric of welfarism and social justice, neoliberal discourses of marketisation and competition dominated educational policies that set about reforming and improving the performance of schools (Ball, 2021).The need to demonstrate improvement and effectiveness resulted in a preoccupation with measurement (Apple, 2018;Biesta, 2009) and this intensified the pressure on schools through extreme accountability and performative regimes.That said, New Labour invested heavily in education and wanted to tackle inequality through policies to 'close the gap' that aimed at lifting aspirations as well as raising attainment.Thus, the pivotal role that schools could play in providing services that 'wrapped around' a child became a key feature of the welfarist arm of New Labour educational policy and the notion of an integrated and holistic approach to education emerged (Whitty and Anders, 2017) that demonstrated this wider view that education was not just about teaching and learning (Ball, 2021).
Through developing leaders at middle levels within schools, responsibility for meeting reform and improvement imperatives was spread across multiple staff.This was argued to be important in a system where growing complexity made it impossible to hold responsibility alone (Hartley, 2007).Hence, leadership could be devolved or distributed from senior to middle leaders, who could, in turn, raise the expectations placed upon other staff to meet educational policy requirements.This focus on middle leaders as the change agents to bring about rapid educational reform perpetuated this discursive and semantic turn.The building of 'leadership density and capability' (Bush, 2015: 651) was a consequence of 'the [perceived] recognition that distributing leadership and building leadership skills early creates more effective schools' (Hammersley-Fletcher et al., 2018).The growing importance of middle leadership resulted in a plethora of research that reflected this policy push.Consequently, a literature review was commissioned in 2003 by the then National College of School Leadership (NCSL), to consider the role of the middle leader, a role that reflected these growing neoliberal policy imperatives.Bennett et al.'s (2003) review was mainly focused on secondary education where these roles already existed in the form of subject leaders or heads of departments.In primary schools, middle leadership was being introduced, leading to the then 'curriculum co-ordinators' becoming re-cast as subject leaders (Hammersley-Fletcher, 2002).Indeed, Bennett et al. found that constructions of middle leadership were never presented in research as a homogenous practice across schools and it incorporated an extensive array of roles and responsibilities such as key stage leaders, phase leaders, departmental leads and subject leaders (Hammersley-Fletcher et al., 2018).Bennett et al. (2007) brought together their findings from the 2003 commissioned review with a re-commissioned perspective, again for the NCSL, conducted in 2005.They concluded that middle leadership was enacted within a context that created tensions and contradictions for practice.The main tensions highlighted were that middle leaders were caught between competing strategic and departmental expectations, and that middle leadership practices continued to be enacted within hierarchical managerialist structures that focussed on performative and prescriptive practices that did not allow for leading in an agentic sense.That is, their role orientated around responding to directions from others but did not seem to include activity around creating imaginative, new educational avenues for their work.That is not to say that there were no middle leaders acting imaginatively, but that this was not the generally reported experience.This approach to middle leadership was thus hampering the more collegial and collaborative potentials of the role and restricting the development of middle leaders' professionalism (Bennett et al., 2007).
Since the Bennett reviews, the Coalition government and then successive Conservative governments have intensified the restructure of the education system through neoliberal and marketised approaches.We have identified two key policy trends that are pertinent to our later discussions about middle leaders, especially for subject leaders.Firstly, this involved a wholesale dismantling of the more holistic elements of New Labour policy, resulting in a narrowed view of how schools can contribute to social equality.According to Michael Gove, this would allow schools 'the freedom to concentrate on what matters' (Ball, 2021: 103).What mattered to the Conservative government was developing an education system that would be competitive in the 'knowledge economy'.The notion of the 'knowledge economy' emphasised the economic imperative to reform and improve education but it also began to influence what pupils should be taught.The 2007 National Curriculum was replaced because according to the government it was 'based on a series of general aptitudes with insufficient subject based content' (Gibb, 2021, online).This focus away from skills to knowledge was emphasised during a speech given by former schools minister, Nick Gibb, where he referenced Young's (2015) concept of powerful knowledge and eulogised that 'Many schools across the nation have risen to the challenge of putting a knowledge-rich curriculum at the core of what they do' (Gibb, 2021, online).This discourse of powerful knowledge has become a 'curriculum principle' (Carlgren, 2020: 323) and according to Hordern (2021: 1), the government is 'using it as a basis to justify curriculum interventions in schools.' Alongside this curriculum shift, the other main policy trend involved a rapid and far-reaching expansion of the academies schools programme deviating from its original intentions.This 'new education landscape' (Gilbert et al., 2013) involved establishing networked governance mechanisms to organise and deliver new autonomous school systems through the establishment of multiple Multi-Academy Trusts (MATs) (West, 2015;Ehren and Perryman, 2018).Indeed, the momentum does not appear to be abating and the government intends that all schools in England become academies and part of a MAT by 2022 (Department for Education, 2016).Despite the programme being 'highly contested' (Dennis, 2018: 50) and 'irreversible' (Francis, 2015: 438), this rapid wholesale restructuring has created complexity and complications as schools try to make sense of their new positions within these new structures.Indeed, Ehren and Perryman (2018: 947) have identified the 'problematic nature of accountability of networks' and how this has led to stretching of the controlled decontrol tension to such as an extent that the Academies Commission has identified that the 'feasibility of the DfE [to play] an active role in academy governance is remote' (Francis, 2015: 438).
It is within this complex and emerging context that middle leadership identities and practices are being re-constructed.We now consider how positions within the literature around the construction of middle leadership tend to be polarised towards either an instrumental focus on how middle leaders can be developed to be as effective as possible or an emancipatory focus, where the aim is to critique the way that middle leaders are constrained by the policy landscape.

Effectiveness versus emancipation
The diversity of studies conducted into middle leadership over the last fifteen years reflects the growing complexity of the middle leadership role and much of the literature is concerned with how to develop the capacity and capabilities of middle leaders so that they can be effective.The challenges faced by middle leaders vary according to position, for example, subject versus year group leader (De Nobile, 2018; Thorpe and Bennett-Powell, 2014); subject or area of responsibility (Spillane, 2005b;Spillane and Hopkins, 2013); the age group or phase of schooling (Bickmore, 2011); and the locality of the school, for example, rural versus urban settings (Thomson, 2009;Tuck, 2009;Woodhouse and Pedder, 2017).Middle leaders' needs, therefore, vary crosssectionally involving different categories of middle leadership and also vary temporally in response to the dynamic nature of the contexts within which middle leaders enact their roles.Consequently, it is important to recognise that middle leadership does not readily reduce to a neat list of desirable attributes to be acquired; rather middle leadership is developed through a situated contingent process of professional learning which is a 'context specific, time consuming, messy and fluid' (Stephenson, 2010:155).That said, the middle leadership literature is still framed by a discourse that constructs middle leadership development from an instrumental and prescriptive perspective.It thus focuses on codifying the competencies and skills required to be an 'effective' or 'good' middle leader to meet the needs of the educational market.In other words, middle leaders were recruited to aid what Sachs (2003) and Ball (2007), amongst others, discussed as the greater levels of accountability demanded of school systems and the rise of international competitiveness in education.Attempts to identify a 'what works best' approach to conceptualising middle leadership development persists.
Instrumental positions identify characteristics of middle leaders aimed at improving practice (Flemming, 2014;Harris et al., 2019;Toop, 2017).For example, within De Nobile's (2018) theoretical model of middle leadership in schools, the competencies of middle leadership are collated under five broad headings representing key behaviours which middle leaders enact to fulfil their various roles: leading teams, managing relationships, managing time, communicating effectively and managing self.Other attributes include emotional intelligence (Wong et al., 2010;Held and McKimm, 2012), honesty and respectfulness (Duignan, 2012), and having the necessary subject knowledge (Gurr and Drysdale, 2013;Heng and Marsh, 2009).However, such atomistic approaches to defining what makes effective middle leadership are problematic if viewed from a critical perspective for being reductive and aligning with the neoliberal agenda: pejoratively described as 'the recipe books about how to be an effective manager or leader' (Gunter, 2013: 4).This provides a site of discursive contestation between the two positions: those that see the value in developing instrumental approaches to support middle leaders in understanding their role, and those that dismiss it as reductionist and lacking any critical intent.However, within the complexity of the middle leadership space, while critical perspectives are nevertheless important, middle leaders' interpretations and enactments of these approaches may not always be merely about conformity.It seems that instrumental approaches could provide the social and cultural resources that help middle leaders construct understandings of themselves in their rolea particularly challenging and complex space.Further research is needed to explore how these middle leaders make sense of the social and cultural assets available to them, and how these might even be mediated to produce agentic responses, especially in light of policy shifts discussed (Holland et al., 1998).
From a critical perspective, neoliberal entanglements are always present.Critical research in relation to middle leadership aims to make the production of this discursive positioning visible.Morrison (2013) emphasised that schools remain rigidly hierarchical as a consequence of performativity and accountability.This has been argued to hinder genuine leadership practices in favour of managerial approaches that focused on more direct and explicit monitoring and assessment behaviours (Ghamrawi, 2010) within a framework of responsibilisation through regulation and competitive enterprise (Hammersley-Fletcher et al., 2020).The onus on middle leaders having responsibility for student outcomes (Thorpe and Bennett-Powell, 2014) has meant that they are frequently cited as an essential mediator in the drive to meet market-led school improvement targets, which authors such as Gunter (2016) have argued was the underlying rationale for developing the role.In other words, middle leadership is a political mechanism to develop compliant schools.So on the one hand middle leaders are assigned the role of innovator, charged with leading creative practice within their teams (Boylan, 2018: 94); on the other, they are described as 'technicians' (Forde, 2011;Hammersley-Fletcher and Qualter, 2010) forced to follow initiatives which 'specify in some detail what is expected of them and require them to transmit and deliver externally defined agendas for change' (Hammersley-Fletcher and Strain, 2011: 881).So much middle leadership research has explored the quandary of whether the aim of middle leadership is to cultivate efficient managers who work with their teams to instantiate the 'vision' from above, or whether this is a role to empower multiple leaders to make innovative strategic decisions rooted in their professional values and practice.Cranston (2013) speaks to this question, arguing that the orthodoxy around how leadership is defined and evaluated needs to be challenged, making way for other understandings of leaders as being responsible for much more than raising narrow measures of academic attainment, instead orientating around a wider notion of professional responsibility.This is echoed elsewhere in the field and the development of a strong sense of collegiality continues to be viewed as an emancipatory alternative for middle leaders (Elmore, 2008;Lambert, 2009;Stephenson, 2010).Collegiality and its relationship to a 'cultural ethos of professional collaboration' (Ghamrawi, 2010: 314), are viewed as crucial to facilitate the improvement of teaching and learning (Forde, 2011;Harris and Jones, 2010;Leask and Terrell, 2014), allowing the creation of professional communities of practice (Hargreaves, 2003;Rönnerman et al., 2015;Wenger, 2000), where middle leaders encourage staff to innovate in their own classrooms and share with others (Forde, 2011;McIntyre and Hobson, 2016).As part of the general emphasis on participatory approaches to leadership within schools, collaboration and collegiality are preferred in the creation of learning communities (Freidman, 2011) and it is argued that this should be achieved through a democratic discussion of professional practices, managed through consensus, rather than through imposition (Jarvis, 2012).Moreover, the creation of participatory learning communities promotes the development of teachers' professional identity and sense of agency within the team (Sachs, 2001), whilst also improving their motivation and self-efficacy (Friedman, 2011).
This vision is consonant with the teacher identity described by Boylan (2013) which is constructed around participation and collaboration within the school community, promoting the development of relationships and activism based on moral and social purposes.As Boylan notes, teachers are outspoken about the moral purpose of their role, but consideration of ethical educational drivers is often missing in discussions around leadership.A participative emancipatory approach to leadership in combination with a shift in focus at the broader political level from accountability to professional trust and responsibility might allow a more settled, focused identity and vision for middle leaders, where they are positioned as 'proactive reflexive leadership professionals, not reactive managers' (Cranston, 2013: 139).In the meantime, while leaders continue to work within an 'increasingly authoritarian policy context' (Hammersley-Fletcher et al., 2017: 591), practitioner research invoking a kind of agonistic democracy (Wenman, 2013) has been put forward as a way for 'teacher-researcher-leaders' to innovate their practice, providing a 'spur to ethical aspiration' (Hammersley-Fletcher et al., 2017: 600).
The 'leader' part of the middle leader title would suggest that those in these positions can construct understandings of their professional selves that align with their own educational values, and that, crucially, they would have agency to shape their role and practice.A context of collegiality supports the shared endeavours involved when undertaking change or innovations in practice (Busher et al., 2007) and should allow for agentic practice to flourish.A desire for collegiality was a significant theme within the Bennett reviews, which reported that subject (middle) leaders did not feel capable or comfortable with exercising influence at a whole school level, seeing themselves as mainly subject administrators.Our recent review revealed that the tension between a desire to adopt a collegial approach and the emphasis on performativity endured and remained problematic (Cranston, 2013;Ghamrawi, 2010) and that this was limiting the agentic possibilities for middle leaders.As Hall (2013) has explained, It points to key tensions and contradictions underlying this discursive intervention.Through this lens it highlights the strangeness of this development by contrasting the rhetorical invitation to teacher and school leader agency as part of the shift to DL [distributed leadership] with the restrictive managerial intent of the NPM [new performance management].(Hall, 2013) In an effort to emphasise the collegial possibilities of this move, middle leadership in England was aligned with notions of 'distributed leadership' where teachers at all levels of the organisation became responsible for some aspect of the work of schools that went beyond the classroom, this being argued to be helpful (i.e.Harris, 2004;Spillane, 2005a;Timperley, 2005) or problematic (i.e.Gunter et al., 2013;Hall, 2018;Hatcher, 2004).This highlights how the field has become split between the competing discourses of those prescribing the efficacy of leadership practices for schools from scientific and instrumental positions, and those from more critical positions who align current leadership practice as a conduit of neoliberal education policy and then offer emancipatory alternatives 'that often lead nowhere' (Savage, 2021: 282).Both of these positions are important but there is a need for research grounded in the realities of middle leaders' work which explores how middle leaders might successfully bridge the chasm between these two camps.In other words, further exploration is needed into potential agentic practices which might allow middle leaders to simultaneously be 'effective' middle leaders in the neoliberal sense (successfully implementing government directives) while also finding spaces for autonomy and authentic innovation.In the section that follows, we will argue that the two key shifts towards a 'knowledge economy' and academisation provide examples of how the policy context, while constraining the work of middle leaders in particular ways, may also open up new opportunities for subject leaders to develop a greater sense of agency through curriculum innovation and collaboration.

Opportunities for agency within the contemporary landscape
At face value, our review of the literature suggests that the tensions highlighted by Bennett et al., continue to be endured by middle leaders in contemporary school settings.Critical perspectives demonstrate the constraining influence of performativity on work of subject leaders, positioning them as conduits of compliance rather than innovators and interrogators of practice.According to Erss (2018: 242): … policies that encourage teacher autonomy do not tend to favour absence of control.The control is merely transformed from that of being exercised through external authorities, to that operated by the school, teaching profession and/or each individual teacher.
As a result of this control, middle leaders' work has focused largely around narrow conceptions of results-based improvement, falling under the scrutiny of the schools inspectorate, parents and indeed the media (De Nobile, 2018;Forde, 2011;Hammersley-Fletcher and Strain, 2011;Hatcher, 2012).It is therefore, unsurprising that middle leaders have expressed concerns about taking risks due to the possibility that unorthodox practice might negatively impact on student achievement and overall school performance and reputation (Cranston, 2013).In this way, high levels of surveillance and the omnipresent threat of inspection arguably stymie, rather than foster, innovation (Hall, 2013).
However, the new National Curriculum, underpinned by the discourse of powerful knowledge, has clear implications for subject leaders, which warrant further exploration.We suggest that this change may in fact offer potential opportunities for subject leaders to engage collegially and more agentically in their practices as pedagogical leaders who consider 'pedagogy as an epistem' for framing their leadership practices (Male and Palaiologou, 2017).Indeed, policy makers are demonstrating a renewed interest in the development of subject leaders (vis-à-vis our earlier review) and this in turn has reinvigorated middle leadership research (De Nobile, 2018;Harris et al., 2019), again highlighting the relationship between policy, research, and practice.For example, the field of curriculum studies has responded to this shift, theorising epistemological approaches and discussing the repercussions to curriculum design and pedagogy (Deng, 2021;Muller and Hoadley, 2021).This opening up of philosophical and educational professional dialogue about what constitutes subject knowledge may be running counter to the critical discourse concerning overly prescriptive and narrow curriculum constructions.These 'discursive[s] site of contestation' (Francis, 2015: 449) require further investigation, for example, exploration of how subject leaders mediate and construct their own understandings of the relationship between powerful knowledge and knowledgeable practice (Hordern, 2021).
Additionally, the often-cited requirement for middle leaders to hold 'strong pedagogical expertise' requires further deconstruction within this knowledge-driven context.As noted by Southworth (2011: 78), education is a 'practical, value-laden, hands-on practice' and so the identification of what 'strong pedagogical expertise' might look like is far from straightforward.While Irvine and Brundrett (2017) explained that teachers need to acquire tacit 'craft' knowledge from more experienced teachers in order to develop their practice, Southworth (2011) suggested that middle leaders play an important role in bringing tacit knowledge to the surface so that it may be shared within their teaching teams.Further, research is needed which theorises the different kinds of knowledge which middle leaders might usefully acquire and which facilitates the construction of what we might call (following Shulman, 1986: 9) a 'pedagogical content knowledge' for leaders.Given Southworth's (2011) proposition, that dialogue is central to the articulation, reformulation and analysis of such expertise, it is better to consider such 'practical, actionable knowledge' (p.78) as something to be constructed within teaching teams, rather than as an attribute for middle leaders to develop and then transmit.This highlights the important role that middle leaders play in leading professional development and this is a prominent feature in more recent research.The middle leader's role in developing the pedagogical expertise of others (Barber and Mourshed, 2007) provides them with potential opportunities to operate at a more critical level.The move away from a skills-based curriculum to a knowledge-based curriculum, and the focus on 'curriculum intent' from Ofsted (2019) has reinvigorated discussions around what the essential concepts, knowledge, skills and principles are within a given subject, how to sequence them, and how to teach in a way which promotes recall and retrieval (e.g.Priestley, 2019).While there is significant controversy around the framing and content of both the National Curriculum and the Ofsted Inspection Framework, the onus on schools to generate and articulate a carefully constructed curriculum approach for each subject, presents opportunities for subject leaders to reconnect with the essence of their role as subject experts and motivates vigorous debate within communities of practice.
The increase in networked governance also presents further possibilities for middle leaders to engage in collegial practice as they are increasingly engaged in delivering training and development across complex networked systems such as school academy chains and multi-school groupings.This has been highlighted by Boylan who demonstrated that professional development is now led across systems and through networks (Boylan, 2018) and importantly this has involved some reconfiguring of their role as 'boundary manager', responsible for smoothing the relationships between teachers, senior leaders and the policy environment (Somech and Naamneh, 2017: 57).So middle leaders are now engaged with practices that 'moved back and forth across the team boundary', reporting a significant association between boundary management activities and team learning and between team learning, innovation and organisational citizenship behaviour.In other words, there seemed to be some potential here for middle leaders to play a crucial role in connecting strategic and operational activities both in one school and across school networks.Working flexibility and drawing upon collective professional knowledge and supporting each other in navigating context-specific challenges, affords middle leaders some possibilities to develop agency especially given the issues raised with networked governance and its stretched relationship with accountability (Ehren and Perryman, 2018;Francis, 2015).

Towards a dialogic approach
The field of middle leadership is still dominated by competing discourses that construct our understandings of middle leadership.Accordingly, Ozga and Lingard (2007: 78) caution that educational research cannot 'be reduced to totally instrumental activity' but neither can it just be about critique even if neoliberal educational policy agendas offer a fertile ground in which to do so (Savage, 2021).While we acknowledge the importance of research from both critical and instrumental stances, we recognise that a binary of knowledge positions has emerged.Additionally, the research tends to be dichotomous based on these opposing conceptualisations and offers either emancipatory (critical) or neoliberal (instrumental) solutions.This is not conducive to developing knowledge in the field that engages with the complexity and messiness of the realities of middle leadership practice.Indeed, reframing the debate from a dialogic perspective may move us from these overly discursive constructions to one that allows for the recognition that for 'individuals and organisations … engagement with and mobilisation of, policy rhetoric's, in ways that lend support to both dominant and oppositional discourses' (Francis, 2015: 439) reflects the realities of middle leadership practices.A position that may at first seem contradictory but in actuality offers the space for agency.As Holland et al., assert, 'Human agency may be frail, especially among those with little power, but it happens daily and mundanely, and it deserves our attention' (1998: 5).
We thus need to understand the ways in which these tensions manifest themselves and how they are mediated in practice.A dialogic theoretical lens provides us with the thinking tools to conceptualise the practices of middle leaders by exploring the relationships between them and their sociocultural context (Holland et al., 1998).Understanding how school middle leaders constitute themselves in practice, respond to the structural systems within which they are entangled, and exploring the possibilities of agency rather than offering emancipatory alternatives is important.These 'micro critical practices' (Ball, 2016) deserve our attention, and the work of Holland et al. provides a conceptual framework to explore these moments of resistance and also the more subtle practices that are more dialogic and seemingly contradictory.It is imperative that researchers working from critical stances also explore and amplify middle leadership practices that are not just about rejection or compliance but are often about being in both spaces at the same time.As researchers, we are interested in exploring these dialogic spaces in order to contribute to this 'complex milieu in which we are enmeshed'.

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.