The phantom national? Assembling national teaching standards in Australia’s federal system

Abstract

In this paper, we use the development of the Australian Professional Standards for Teachers (APST) as an illustrative case to examine how national schooling reforms are assembled in Australia’s federal system. Drawing upon an emerging body of research on ‘policy assemblage’ within the fields of policy sociology, anthropology and critical geography, we focus on interactions between three dominant ‘component parts’ in the development of the APST: the Australian federal government; New South Wales state government agencies; and the Organisation for Economic Cooperation and Development. While policies like the APST claim to be national in form and scope, our analysis suggests ‘the national’ is much more disjunctive and nebulous, constituted by a heterogeneous and emergent assemblage of policy ideas, practices, actors and organisations, which often reflect transnational traits and impulses. We thus see national reforms such as the APST as having a phantom-like nature, which poses challenges for researchers seeking to understand the making of national policies in federal systems.

Keywords:

• Actors institutions
• sociology

Introduction

The past decade has seen unprecedented efforts in Australia towards achieving greater national consistency in a variety of schooling policy areas. One of the central policy mechanisms deployed towards this aim has been the introduction of standards-based reforms in areas that include curriculum, teaching standards, testing and reporting. These reforms have been promoted as policy solutions for putative policy problems in Australia’s federal system, including overlap and duplication across states and territories, and inequalities in provision and performance nationally. Standards-based reforms have also been advocated on economic grounds, with schooling framed as an investment site to be harnessed towards making Australia more competitive in the global economy (Rizvi and Lingard 2010). National reforms are driving rapid changes within state and territory systems. Although state and territory governments maintain responsibility for schooling, major policy agendas and reforms must now be navigated at the national level and through complex intergovernmental channels that are strongly influenced by the federal government and transnational policy influences (Savage 2016).

In this paper, we use the development of the Australian Professional Standards for Teachers (APST) as an illustrative case to examine how national reforms are assembled in Australia’s federal system. In doing so, we build on emerging research in the fields of policy sociology, anthropology and critical geography, which uses the concept of assemblage to analyse policy. We begin the paper by elucidating our understanding of an assemblage approach to policy analysis, arguing that an assemblage analytic is particularly generative for understanding policy development processes in federal systems. We then analyse development processes leading to the APST, a national set of teaching standards that were established in 2011 with the aim of enhancing and assessing teacher quality by outlining what Australian teachers ‘should know and be able to do’ across four career stages: Graduate, Proficient, Highly Accomplished and Lead. With a focus on reform processes from the 1990s to the early 2010s, we examine interactions between three dominant component parts of the national schooling policy assemblage that have been central to shaping the APST: (1) the Australian federal government, which played significant roles in promoting and incentivising the reform; (2) New South Wales (NSW) state government agencies, which played significant roles in shaping the national standards; and (3) the Organisation for Economic Cooperation and Development (OECD), which worked transnationally to promote teaching standards in ways that strongly influenced Australian reforms. We conclude by considering the implications of an assemblage analytic for conceptualising national reforms in federal systems. We argue that while policies like the APST claim to be national in form and scope, our analysis suggests ‘the national’ is much more disjunctive and nebulous, constituted by a heterogeneous and emergent assemblage of policy ideas, practices, actors and organisations, which often reflect transnational traits and impulses. We thus see national reforms such as the APST as having a phantom-like nature, which poses challenges for researchers seeking to understand national policies in federal systems.

Assembling policy in federal systems

The past decade has seen a proliferation of work in the fields of policy sociology, anthropology and critical geography, which has sought to understand how policy ideas and practices are generated and enacted in ways that overcome the limits of established theories in policy research, including rational-technical, institutionalist and state-centric accounts (e.g. Clarke et al. 2015; McCann and Ward 2012; McFarlane and Anderson 2011; Peck and Theodore 2015; Rizvi and Lingard 2011; Shore, Wright, and Pero 2011; Ureta 2015). A central aim (and challenge) of this work has been to develop a more nuanced set of concepts and methods for making sense of how policies move, mutate and manifest in an era of intense transnational flows of ideas and practices, in which systems of governance are increasingly polycentric, with complex networks of government and non-government actors and organisations working to develop and enact policy. Many ideas and concepts have been articulated, re-articulated and tested in this work. In this paper, we engage with one fast-emerging concept in these fields: policy assemblage.

The term ‘policy assemblage’ emerges from a broader field of assemblage theory with roots in the philosophical works of Gilles Deleuze and Felix Guattari, who used the concept of assemblage (agencement in French) to understand the heterogeneous composition of complex social and non-social formations. Deleuze and Guattari framed assemblages as ‘complex constellations of objects, bodies, expressions, qualities, and territories that come together for varying periods of time to ideally create new ways of functioning’ (Livesey 2010, 18). An assemblage ‘transpires’, therefore, ‘as a set of forces coalesces together’: a process that can be applied to ‘all structures, from the behaviour patterns of an individual, the organisation of institutions, an arrangement of spaces, to the functioning of ecologies’ (ibid.). Despite being central progenitors of assemblage theory, Deleuze and Guattari wrote relatively little about assemblage (DeLanda 2006), leaving the term open to a variety of translations, interpretations and uses.¹ The term assemblage also bears conceptual similarity to other terms designed to capture the complexities of multi-faceted formations, including dispositif, apparatus and regime (Li 2007). This variety of theoretical and empirical uses means some careful framing is needed to clarify how we interpret and use the term ‘policy assemblage’ in this analysis.

In what follows, we draw attention to two core conceptual foundations that underpin our approach to using the concept of assemblage to analyse policy: first, a distinct conceptualisation of the relationship between parts and wholes; and second, the contingent and emergent nature of policy assemblages. In drawing attention to these foundations, we are not suggesting these are the only (or even the definitive) features of an assemblage approach.² Instead, these are core conceptual elements that we have adopted (and in some cases adapted) from the broader field of assemblage research to inform our analysis of national schooling reforms in federal systems. In outlining our approach, we also suggest the concept is generative in two distinct senses for understanding policy development in federal systems. First, assemblage provides a powerful lens for understanding how specific policies are developed and enacted in federal systems. The focus here is on tracing the development of the policy itself as an assemblage, akin to the way Ureta (2015) analyses the Transantiago public transport system in Chile, or how Peck and Theodore (2015) analyse the movement and development of conditional cash transfers (CCTs) and participatory budgeting (PB) models within and across multiple systems and contexts. Second, we see an assemblage approach as powerful for understanding the broader nature of federalism, in terms of how the component parts of Australian federalism operate to produce broader trends and features – or, as DeLanda (2006) puts it, the wider ‘properties that emerge from interactions between parts’ (10). Here, the emphasis is broader, insofar as we understand federalism as an assemblage of diverse, heterogeneous and emergent components, rather than something that has a specific essence.³ In both senses, assemblage provides a counterpoint to conventional political science and public policy literature, which tends to downplay the ‘fluid and processual’ (McCann and Ward 2012, 326) nature of policy processes, meaning it cannot capture the development, movement, mutation and enactment of policy ideas and practices. Following Clarke et al. (2015), therefore, we see the concept of assemblage as ‘unsettling policy and its study’ (9), and challenging an established ‘policy studies orthodoxy’ (12).

1. Against essences and totalities: conceptualising parts and wholes

An assemblage approach is particularly useful for understanding relationships between parts and wholes in the making of complex social formations like Australian federalism or national schooling policies. An assemblage analytic rejects the idea that any social formation has an inherent nature or logic, or can be understood as a coherent whole. This is because the concept of assemblage presents a distinctive understanding of the relationship between micro and macro levels of social reality that is against both essences and totalities (DeLanda 2006). Instead, assemblages are framed as synthetic, emergent and constituted by a heterogeneous set of interacting components (e.g. policy actors, governments, departments and agencies, organisations, unions, professionals, ideas, practices and more), with these components having a contingent rather than essential relationship to each other (DeLanda 2006). As Ureta (2015) argues, ‘assemblages are never fully stable and well-bounded entities’ and ‘don’t have an essence’, but instead ‘exist in a state of continual transformation and emergence’ (11–12, italics original). At the same time, assemblages do reflect certain broader properties (i.e. features, traits) that emerge because of the specific nature of interactions between component parts. These properties, however, are not the same as an essence, but are also shifting and emergent.

DeLanda’s (2006) ‘reconstructed theory of assemblages’ (4) is instructive for unpacking this relationship between parts and wholes. Building on Deleuze and Guattari, DeLanda elaborates his take on assemblage by analysing a range of social assemblages, including networks, organisations, governments, cities and nation-states. DeLanda argues that assemblage helps us move beyond ‘an ontological commitment to the existence of essences’ (4), which he sees as still dominating and hampering the social and political sciences. DeLanda critiques what he calls ‘the organismic metaphor’, whereby social scientists treat social institutions as akin to biological organisms that have internal logics and order, and in ways that mistakenly position component parts as ‘bodily organs’ (8). Such perspectives, he argues, produce a skewed understanding of the relationship between parts and wholes – or, more precisely, ‘the link between the micro- and the macro-levels of social reality’ (4).

DeLanda argues that this ‘micro-macro problem’ (4) is plagued by two contrasting forms of reductionism. First, there is a tendency towards micro-reductionism, whereby wholes are framed as ‘a mere aggregate’ (4) of individual component parts. For example, a researcher might understand federalism as simply an aggregate of the many policy actors, institutions, laws and other phenomena that ‘make up’ a federal system, much like organs within the body of federalism. If we adopt DeLanda’s arguments, however, then a major problem with understanding the whole as simply the sum of its parts is that we obscure the ‘irreducible properties’ (10) of the whole: that is, properties that exist only as a result of complex interactions taking place between parts (i.e. properties that would be different if the components were organised differently):

… the reason why the properties of a whole cannot be reduced to those of its parts is that they are the result not of an aggregation of the components’ own properties but of the actual exercise of their capacities. These capacities do depend on a component’s properties but cannot be reduced to them since they involve reference to the properties of other interacting entities. (11)

When viewed as an assemblage, policies or political systems reflect distinct properties that exist only because of contingent interactions between components at specific moments in time.⁴ Different interactions and different components thus provide different ‘conditions of possibility’ for what the core features of policies or systems are capable of being. Again, understanding the nature of interactions between components is of central importance here. Simply recognising that individual components exist or focussing on what individual components do is thus useful, but ultimately insufficient. For example, as we show in our analysis to follow, changes in the way one component (e.g. the federal government) operates in relation to schooling or, indeed, the development of a new policy (e.g. the APST) can have significant impacts on how a wide range of other components interact.

Understood in this way, we see that policies and systems do not have a particular essence or nature, but instead can be understood as ‘a complex assemblage of historically situated, contingent and emergent structures, forces and relationships’ (Savage 2016, 848) – reflecting distinct properties that are products of interaction between shifting components. To research and understand policies or systems as such, we must attend to interactions between components (thus adopting ‘bottom-up’ approach to analysis) to ascertain which component parts exert which kind of influence over the greater whole. In this paper, we do not begin by assuming a coherent ‘national’ exists, but instead start by analysing how various component parts interact to produce policy ideas and practices that claim to be national in form and scope. Only once we have considered such interactions between components can we come close to understanding what ‘the national’ might be and the conditions of possibility it might produce.

The second tendency DeLanda critiques is macro-reductionism, which, in contrast to micro-reductionism, frames individual components of an assemblage as being ‘mere products’ (4) of the whole. Here, the reverse problem is created. For example, a researcher might understand one key component of federalism (e.g. a federal department) as having certain features or operating in certain ways based on views or assumptions about the broader nature of federalism itself. In political science literature, for example, there is a strong tendency for researchers to evoke ‘ideal types’ to understand relations between governments in federal systems, describing different federalisms as ‘collaborative’, ‘pragmatic’, ‘coordinate’, ‘concurrent’, ‘coercive’, ‘corporate’, ‘contractual’ and so on. Even if we acknowledge Max Weber’s arguments that ideal types do not imply essences, there is still a clear problem of macro-reductionism that emerges when seeking to understand federalism through recourse to ideal types. For instance, if we adopt a position that describes Australian federalism as ‘coercive’, we are inclined to look for examples of the federal government or other key components that reflect such politics of coercion, even if this invariably obscures properties of federalism that fall outside the lens of coercion. This is not to suggest that researchers do not ‘induce’ from empirical realities an ideal type and use it as a heuristic device. However, even this process of induction is fraught, because as the process towards formulating an ideal type unfolds, the researcher must begin to narrow the lens and choose certain empirical examples over others to inform the ideal type being proposed, and it is impossible to formulate an ideal type without this process of narrowing. Such recourse to ideal types thus rarely reflects the complexity and fragmented nature of interactions between components, especially in Australia’s federal system where there is evidence of ‘collaborative’, ‘competitive’, ‘coordinate’, ‘concurrent’ and many other ideal type tendencies simultaneously at work.

By rejecting macro-reductionism, an assemblage approach is also useful for understanding the development and enactment of specific policies, as well as the work of specific policy actors and organisations, especially in contemporary contexts marked by intensified transnational flows of policy ideas and practices. As we show in our analysis, the APST might ostensibly be a national and Australian policy, but it is misguided to ‘explain’ this policy as simply a ‘product’ of Australian schooling policy. This is, in large part, due to the strong transnational influences that have guided its development. The same applies to understanding the work of specific policy actors, where a policy actor might move from working with one policy organisation to another, and, in the process, operate differently (and with different impacts) in each. Imagine, for example, that a policy actor works for several years within a state education department in New South Wales, then takes a position for two years working at the OECD in Paris, then returns to Australia to work at the federal department in Canberra. As a component part of Australian schooling, this policy actor cannot be ‘explained’ as simply a product of Australian schooling, nor as a product of any specific organisation. Put differently, the actor’s properties are not simply constituted by their relationship to the broader ‘whole’ of Australian schooling, nor the ‘whole’ of any specific organisation. Instead, we see that the actor/component can be ‘unplugged’ from one assemblage and ‘re-plugged’ into another, a process that has the capacity to influence other components within the assemblages through which the actor moves. The way the policy actor interacts with and seeks to influence policy through working with a state education department, therefore, could be quite different from how that actor interacts and works with a global policy organisation (and with different impacts). The movement of policy actors is just one example of component parts that move in and out of different policy assemblages. Indeed, policy-making is increasingly defined by the complex movement of people, technologies, ideas and practices, which are translated and assembled differently in different contexts (see Clarke et al. 2015).

2. Re/dis/assembly: the contingent and emergent nature of policy assemblages

The second foundation that guides our approach is a framing of policy assemblages as both contingent and emergent. Policy assemblages are never static, but rather are always undergoing processes of assemblage, disassemblage and reassemblage, often simultaneously (Youdell and McGimpsey 2015). For this reason we draw attention to processes of re/dis/assembly in the making of the APST and the broader making of Australian federalism. This allows us to capture not only how assemblages are assembled, but also how existing assemblages are disrupted, dismantled and reassembled. To borrow from Clarke et al. (2015), our analysis is premised on the view that policy ‘is never a singular entity’ (9). Or, as Ureta (2015) puts it, policies are not ‘solid or stable’, but are ‘temporary concatenations of heterogeneous entities, always on the verge of becoming something completely different’ (12). Just as natural ecosystems are assemblages of diverse plant and animal species, we see policies and political systems as also the result of shifting relations between diverse component parts.

In emphasising processes of re/dis/assembly in our analysis, we also have in mind Deleuze and Guattari’s use of the terms territorialisation, deterritorialisation and reterritorialisation. Deleuze and Guattari used these terms to describe how assemblages come together (territorialisation) and come undone (deterritorialisation), or cases where existing or disrupted assemblages are reassembled (reterritorialisation), in relation to a variety of social and non-social formations (e.g. from political spaces through to ecosystems) (see Deleuze and Guattari 1987). Conceptualising policy in terms of re/dis/assembly processes is useful for understanding policy change, especially when change involves the development of new policies. For example, in a federal system like Australia, shifts towards new national reforms can significantly disrupt existing sub-national (state and territory) policy assemblages, subjecting these to forms of deterritorialisation (as existing formations are dismantled) and reterritorialisation (as dismantled formations are reassembled in relation to new components). Again, these processes often occur simultaneously. As Ureta (2015) argues,

the act of assembling a certain entity always goes hand in hand with disassembling other/s, or, using the terms developed by Deleuze and Guattari, the territorialization of a certain entity always goes hand in hand with the deterritorialization of other/s. (Ureta 2015, 12)

Assemblages always include, therefore, elements involved in an ‘interplay of territorialization and deterritorialization’ (12), with some components working to stabilise the assemblage and others working to transform it. Or, as McCann and Ward (2012) put it, ‘[a]n assemblage is always in the process of coming together … just as it is always also potentially pulling apart’ (328). This notion of simultaneous re/dis/assembly is thus an important feature of assemblage thinking, as it emphasises that assemblages are composites, with different parts of the assemblage potentially undergoing different processes at the same time. To reject this proposition is essentially to argue that policy objects and systems can be, or are, static. Of course, emphasising simultaneity does not mean policies like the APST are never ‘formed’ (i.e. the standards do exist as a written policy text). However, as we outline below, emerging processes around the policy text reflect highly diverse interpretations and enactments of the standards amongst different actors and organisations. In time, it is likely that diverse translations of the APST could lead to contestations and perhaps even a revised set of standards at the national level, or within different state or territory systems. It is also possible that the standards could fail and be dissolved entirely, as so many ambitious reforms have done so in the past. So, while policy texts might give an illusion of stability, a wider view ultimately reveals constant motion and new beginnings. As Deleuze argues, ‘the current is not what we are but rather what we are in the process of becoming’ (Deleuze 1992, 164; in Allan and Youdell 2017, 71).

Understanding policy assemblages in this way necessitates both a relational and material approach to policy analysis, which pays close attention to how policy ideas and practices are assembled in specific contexts and at specific moments in time. It also requires attention to the mutually constitutive nature of such ideas and practices (i.e. interrelationships and forms of coevolution), as well as tensions and contradictions. In this sense, a central concern of an assemblage approach is how certain ideas and practices are ‘made to cohere’ (assembled) or, alternatively, are subject to forms of disruption and change (disassembled or reassembled). As Li (2007) argues, an assemblage approach invites attention to ‘the hard work required to draw heterogeneous elements together, forge connections between them and sustain these connections in the face of tension’ (264). The concept of assemblage thus positions researchers to examine how heterogeneous formations hold together ‘without actually ceasing to be heterogeneous’ (Allen 2011, 154; in Ureta 2015, 12). This perspective is reflected strongly in Ureta’s (2015) conceptualisation of policy assemblages, which itself borrows from Law’s (1994) notion of ‘modes of ordering’ to emphasise the dynamic processes by which heterogeneous elements come together to transform existing assemblages into something new. Modes of ordering, Ureta (2015) argues, ‘are heterogeneous and variable but always include the search for strategic effects, the aim to transform an existing situation in a certain predetermined way through the establishment of particular sets of relations between new and existing entities’ (12). In some cases, governments and formal policy organisations dominate modes of ordering, whereas in other cases a broader range of actors is involved. Attention to modes of ordering, and how ideas and practices ‘cohere’, is highly generative for understanding how specific policies are developed and enacted, particularly in complex federal systems like Australia. As we argue below, the APST has been assembled through the efforts of multiple policy actors and organisations, working within and across multiple policy spaces. Over time, various ideas and practices have been proposed, contested, rejected, rearticulated and so on, to produce what is now considered, on paper at least, to be a national policy. However, as we also show, the ‘holding together’ of the assemblage becomes increasingly difficult once a national policy like the APST is enacted, as multiple interpretations and enactments of the policy emerge. As Shore, Wright, and Pero (2011) argue, ‘a key quality of policies is that, once created, they often migrate into new contexts and settings, and acquire a life of their own that has consequences that go beyond the original intentions’ (3), meaning that policies are both ‘productive’ and ‘continually contested’ (1).

In positioning researchers to consider how policy ideas and practices cohere or otherwise, an assemblage approach invites attention (either implicitly or explicitly) to questions of power, politics and agency. Indeed, these elements are of crucial importance if we wish to understand why some ideas and practices are brought together and gain traction (as well as how, and with what effects), but also why other ideas and practices do not. Such an approach positions us to analyse the role policy actors and organisations play in highly contested processes of putting together and pulling apart policy ideas and practices. A central practice that needs to be addressed in this regard is that of problematisation; that is, ‘how problems come to be defined as problems in relation to particular schemes of thought, diagnoses of deficiency and promises of improvement’ (Li 2007, 264). In this way, an assemblage approach speaks productively to other work on problematisation that uses Foucauldian perspectives (e.g. Bacchi 2012; Webb 2014). We would add to this focus on problematisation an interest in why ‘other’ problems do not gain prominence, and what such processes of either bringing into being (or obscuring) problems says about the conditions of possibility in given policy spaces. For example, as our analysis shows, ‘teacher quality’ needed to be constructed as a policy problem, and made amenable to reform, for the APST to be developed. This process of problem construction involved the success of specific actors and organisations gaining traction and shaping the terms of debate through constructing the problem of teacher quality in a specific way, and the offering of policy solutions that could generate national support. However, this invariably comes at the expense of other less successful or powerful actors, organisations and ideas. Moreover, the ability to shape agendas is not simply an effect of actors and organisations proposing certain ideas and practices, but is also influenced by the overarching conditions of possibility that a given policy space provides. In other words, certain policy contexts provide conditions amenable to certain policy ideas and practices, but not others, and the agency of policy actors and organisations, and the fidelity of policy proposals, is thus always ‘situated’ and context dependent (Bevir and Rhodes 2010).

While drawing attention to the messiness and simultaneity of policy processes can make the job of a policy researcher quite complicated, it arguably offers a more nuanced account of policy development and enactment than what is offered by traditional ‘policy cycle’ approaches, or studies of ‘policy implementation’ and ‘evaluation’, which tend to obscure such complexity (see Colebatch 2006). Indeed, an assemblage approach is at odds with ‘the overemphasis in the literature on policies as forms of order and planning (Ureta 2015, 13). Moreover, as Clarke et al. (2015) argue, in contrast to approaches that position policy as a static object that moves between different political systems and exerts impacts upon entry, an assemblage approach frames policy as involving ‘practices of translation as policies are interpreted, enacted, and assembled’ (9). An assemblage approach thus pays attention to ‘what policies become’ (25) and how policies are ‘re-made or re-shaped’ (25), due to changes within or between different social and political contexts. Understood in this way, policy (and federalism) is ‘always ‘unfinished’, and always open to having its intentions, its meanings and its substance bent, or re-appropriated’ (30).

Re/dis/assembling the APST

Having outlined our understanding of an assemblage approach to policy analysis, and how it can contribute to better understanding the dynamic, contingent and emergent processes of policy development, we now deploy it to analyse the development of the APST in Australia’s federal system. With a focus on reform processes from the 1990s to the early 2010s, we examine three dominant component parts in the development of the national teaching standards: the Australian federal government; NSW state government agencies; and the OECD. We stress the focus on ‘dominant’ components here, as these are clearly not ‘the only’ components.⁵ Instead, we see these components as central to shaping the APST, to the extent that the standards would arguably have looked very different (or may not have been developed at all) if not for the roles these components played. Moreover, as we outline below, focussing together on these three parts demonstrates how national reform has emerged through complex interactions between actors and organisations located within and across multiple policy spaces. In making this argument, however, we want to stress that our analysis does not presume that these spaces are in some way prefigured or scalar in nature. In other words, we do not see our choice to focus on federal, state and transnational spaces as implying that the reform is best understood as a product of interactions between different ‘scales’. Indeed, we strongly urge against seeing these components as forming any sort of ‘Russian Doll’-style nested hierarchy. Instead, our approach stresses both assemblage and topological analytics, which focus on understanding interactions that cannot be captured neatly (or even feasibly) using a scalar lens (see Allen 2011). As McCann and Ward (2012) note, ‘policy regimes of various sorts are relationally interconnected’ (327), meaning that, methodologically, we need to avoid assumptions that fixed scales and territories exist, and instead focus on the specific interactions that make policy: that is, to ‘follow the policy’ (see also, Gulson et al. 2017; Peck and Theodore 2012). This approach does not preclude an attempt to use ‘the national’ as a heuristic to explore how policies that claim to be national are made. It does, however, allow us to analyse national reform in a way that does not assume the national exists in an a priori manner. Instead, an assemblage analytic allows us to see the national as something that is actively made through processes that bring various ideas, practices, actors and organisations together into strategic relations, and with specific effects.

Constituting ‘the national’ through the federal government

The issue of teaching standards has long been a major feature of Australian political and policy debate, with many federal and state reviews conducted into teacher training, preparation and quality over the past four decades. While the federal government has played a steadily increasing role in schooling policy since the 1960s (Lingard 2000), its efforts to influence teaching and teacher education at the national level significantly intensified from the 1980s and 1990s onwards (Bourke 2011). While there is often a tendency to conflate ‘federal’ and ‘national’ policy initiatives, such a conflation is highly problematic in the Australian federation. As we now argue, the federal government has strategically worked to promote and incentivise national reform in line with federal agendas and objectives, but in doing so has been just one (albeit powerful) component part in assembling the national. Moreover, the dominance of the federal government has arisen out of (and been contingent upon) interactions with other leading component parts. For this reason, we focus not only on what the federal government did, but on the particular nature of interactions between federal agencies and other actors and organisations.

While the federal government reflected a growing interest in teaching standards during the 1990s, a major milestone was in 1998 under former Liberal Party Prime Minister, John Howard, when the Australian Senate released A Class Act: Inquiry into the Status of the Teaching Profession (Australian Government 1998). A Class Act was significant in establishing teaching standards as a putatively ‘national’ issue and in a much stronger way than earlier reports. It was also the first time the federal government officially supported and promoted the establishment of common national teaching standards, and the creation of a national body to create, oversee and administer these teaching standards. A Class Act focused largely on addressing the perceived decline in the status of the teaching profession and did so largely through arguing the benefits of professional standards. Six of the report’s nineteen recommendations focused on either teacher professional standards, teacher professional development or the establishment of a single national body to oversee reforms. Recommendation 1, the most significant in detail, called for a ‘national professional teaching standards and registration body to have the responsibility, authority and resources to develop and maintain standards of professional practice’ (x). While to be overseen (and facilitated) by the federal government, this body would be ‘national’ insofar as it would collaborate with state and territory governments, as well as peak professional organisations and teacher unions across the nation. As such, the federal government was not positioned as ‘taking over’ but instead was to be a central force in driving a more ‘national’ approach to education policy. It was envisaged, however, that the new national body would make recommendations to the federal Education Minister, and in this way A Class Act represented an important repositioning of the federal government as both steering and enabling national reform. In other words, the federal government would play a key role in bringing together other component parts to form the beginnings of a new national policy assemblage, but would also seek to maintain control over these processes.

A Class Act set in motion further reviews and reform shifts in the early 2000s that eventually led, in June 2004, to the federal government establishing the National Institute for Quality Teaching and School Leadership (NIQTSL). NIQTSL was designed as a national professional body for teachers and was given the responsibility of overseeing the development of national teaching standards. Mirroring the vision of A Class Act, the federal government exerted a dominant influence by guiding its development and dedicating $10 million to fund its operations. Even though NIQTSL was ostensibly ‘national’, it was owned by the federal government, with no formal ownership by states or territories. Instead, its Board (which was appointed by the then federal Minister for Education, Science and Training) comprised leading academics and representatives from professional associations representing teachers and school leaders. By creating a body to develop standards to be implemented across states/territories, but without formal state and territory involvement, we see here a significant evolution and enhancement of the federal government’s role in seeking to constitute a national policy space and agenda in education. This is a clear example of the federal government beginning, in these early stages at least, to assemble the national by seeking to bring together previously diverse actors and organisations (but a select group only) into a new national policy assemblage.

Of course, the development of NIQTSL (and the willingness of states and territories to let the federal government lead the process) should not be seen simply as ‘federal imposition’ or ‘state acquiescence’, but instead needs to be situated in relation to several national agreements made between the late 1980s and 2000s, which formed the backdrop to the developments canvassed thus far. Central in this regard was the Hobart Declaration (1989) and the Adelaide Declaration (1999), both which were signed by all Australian education ministers (federal, state and territory). These declarations promoted commitments to achieve greater national consistency in core areas of schooling policy, and provided strong justifications for the kinds of federal interventions seen in A Class Act and the subsequent establishment of NIQTSL. So, while state and territory governments were not directly steering NIQTSL, the organisation’s agenda formed part of an existing national assemblage in schooling, which had a strong trajectory towards strengthening national consistency in a wide range of schooling policy areas. This said, and even though such nominally ‘national’ declarations were in place, the presence of the federal government in steering the shape of these reform agendas was still dominant. For example, successive federal governments (from both sides of the political divide) played important coordinating roles in bringing the states and territories together around such national education goals, which ultimately played an important role in creating the conditions of possibility for a single ‘national’ set of teaching standards to emerge. From the beginning, therefore, we can see that the emerging national assemblage was not ‘even’ or ‘smooth’ in terms of power relations (and relative contributions) between different component parts. It was an emerging ‘national’, therefore, but a national with traits that strongly reflected federal interests and agendas.

As the 2000s ensued, the combined influence of the federal government and the reform agenda set out in the subsequent intergovernmental Melbourne Declaration (2008) provided the conditions of possibility for the current APST to emerge. While NIQTSL had begun work towards national standards, the organisation was soon reconstituted and renamed, in late 2005, as ‘Teaching Australia: Australian Institute for Teaching and School Leadership’ (Teaching Australia), which continued the standards agenda. However, by the time the Melbourne Declaration was signed in December 2008, a national set of standards had still not been agreed to by all governments. The Declaration included an explicit commitment to ‘supporting quality teaching and school leadership’ (MCEETYA 2008, 11), and was soon followed by the federal Labor government’s National Partnership Agreement on Improving Teacher Quality (‘the Agreement’). In addition to framing the issue of standards squarely around individual teacher quality, rather than the previously more professionally oriented focus on teaching quality,⁶ the Agreement was also significant in that it included an explicit system of federal funding incentives that encouraged states and territories to adopt national teaching standards.⁷ In other words, states and territories agreed to national teaching standards, but with the caveat that they would receive more federal money. This coercive strategy, however, was accompanied by a language of collaboration centred on working together in matters of the national interest. For example, the Agreement noted that all layers of Australian government (federal, state and territory) have ‘a mutual interest and shared responsibility in improving educational outcomes in the areas of principal, teacher and school leader quality, and supporting reforms to achieve these outcomes [was needed]’ (COAG 2008, 4). However, once more it was the federal government that would drive this reform and financially incentivise states and territories into action. This complex assemblage of coercive and collaborative traits was reflected in a broader suite of reforms pursued by the Labor government following its election in late 2007, including national reforms in other areas of schooling like curriculum, testing, reporting and funding (see Savage 2016).

To operationalise this refreshed national agenda, the Labor government replaced Teaching Australia with the Australian Institute for Teaching and School Leadership (AITSL) on 1 January 2010. AITSL’s primary task was to continue developing national standards, but through collaboration with states and territories, school sectors and the profession. Importantly, however, AITSL was established as a public company, owned exclusively by the federal government, with the federal minister for education its sole member and allocated the power to directly appoint members to the governing board. While the inaugural board included a more representative selection of individuals from across the nation (encompassing schooling authorities, educators and the Australian Education Union), recent years has seen an intentional shift towards an ‘expert board’ structure, which has seen union/teacher representation removed, uneven state and territory representation emerge, and a host of new actors come on board. Ultimately, therefore, the more representative structure of AITSL’s board was short-lived, with the pendulum eventually swinging back to a tight-knit group (but with far less representation of ‘the profession’ than NIQTSL) that does not equally represent states and territories. This positioning of the federal government and minister contrasts with the ‘joint ownership’ model between the federal, state and territory governments that underpins the other major organisation responsible for national schooling reform, the Australian Curriculum, Assessment and Reporting Authority (ACARA). Despite this federal dominance, however, the combination of funding incentives and a willingness on the part of state and territory ministers to cooperate eventually led to the successful establishment of the APST, which were endorsed by all ministers in 2011.⁸

The thirteen years of reform spanning from the release of A Class Act to the introduction of the APST illustrates the strategic ways that the federal government has worked to problematise the teaching profession, cement standards-based reform as a common-sense policy solution, and bring together diverse policy actors and organisations into a national policy assemblage. These shifts have seen the federal government achieve unprecedented influence in setting national agendas, which have had a powerful influence over what happens in schools and, importantly, a major impact on initial teacher education and the accreditation of tertiary education courses. This is because the ‘Graduate’ level standards of the APST have been tied to graduation requirements from teaching degrees across the nation, as well as teacher registration in state and territory schooling systems. These influences have become particularly prominent since the 2014 release of the federal Teacher Education Ministerial Advisory Group (TEMAG) report, entitled Action Now: Classroom Ready Teachers (Australian Government 2014). The report argued that the APST were being ‘weakly applied’ and that ‘implementation timeframes are too slow’ (xi). In response, the report recommended more explicit alignment between teacher education programmes and the linking of registration requirements to the Graduate standards. It also recommended that AITSL should ‘be reconstituted to undertake a stronger role to ensure high standards of initial teacher education in Australia’ (xiv). The federal government subsequently increased AITSL’s budget by $16.9 million in the 2015–2016 federal budget to implement recommendations from TEMAG,⁹ giving it the capacity to ‘overhaul and manage the accreditation of initial teacher education programs’ across the nation (xiv).

The national policy assemblage that has emerged around the APST is illustrative of DeLanda’s (2006) point that ‘the whole’ (i.e. the national) ‘emerge[s] from interactions between parts’ (10), but cannot be reduced to any one of those individual parts. While the federal government has clearly served as a dominant component part, gaining significant power and traction over the interactions between other component parts (especially interactions between state and territory governments), its influence has always been (and continues to be) contingent upon its interactions with these other parts. Indeed, to understand the national, and the traits it reflects, attention is needed such the nature of such interactions. To further illustrate these points, we now turn our attention to consider two other component parts that have also been central to making the national assemblage cohere.

Constituting ‘the national’ through New South Wales state departments and agencies

While the federal government has served as a dominant ‘component part’ in shaping the national assemblage of the APST, it is important not to discount the powerful role that states and territories (i.e. sub-national component parts) have played in determining the characteristics of the reform. While there is a strong argument to be made that the APST would not have been assembled in the first place if the federal government had not pushed the agenda over several decades, driven the development of AITSL, and provided powerful financial incentives to the states and territories, it is incorrect to see the national as solely constituted via the ‘top-down’ imposition of the federal government. Moreover, while a ‘national consensus’ around teaching standards can be traced from earlier collaborative declarations and agendas since the 1980s, we should not assume ‘the national’ has been assembled through even contributions, powers and relations between sub-national governments. Indeed, as we now argue, one state in particular, NSW, the most populous state within the Commonwealth, has played a leading role in the development of the APST, and to such a great extent that the eventual form of the APST strongly resembles teaching standards previously existing in NSW.

NSW has historically been a leading contributor to the debates around teaching standards, with the Correy Report (New South Wales Government 1980) being one of the first official inquiries into the reform of initial teacher education. This precipitated a series of influential reports commissioned by the NSW government that often pre-empted similar reports by the federal government. For instance, the watershed 1998 federal senate report, A Class Act, which largely set the agenda for a national approach to teaching standards, was preceded by multiple state-level publications in NSW, including Desirable Attributes of Beginning Teachers (1994), Raising the Standards of Teachers and Teaching (1997) and Vocational Education and Training in NSW Schools: Meeting the Need for Appropriately Skilled Teachers (1997). These all sought to address issues relating to initial teacher training and ongoing professional development, as well as defining the attributes teachers should possess. These reports positioned NSW not just as an ‘innovator’ amongst the states and territories, but also positioned it at the vanguard of developing teaching standards at the national policy level.

However, NSW did not just ‘contribute to’ national developments, but also ‘responded to’ these broader developments, often seeking to anticipate and stay a step ahead of national reforms. For example, the intergovernmental policy A National Framework for Professional Standards for Teaching (MCEETYA 2003) was endorsed by all federal, state and territory ministers of education in July 2003. Although this Framework did not outline specific standards per se (‘a framework of standards’), it nevertheless provided the parameters in which standards could be developed (‘a framework for standards’), implying that the individual states and territories would still retain the overall responsibility for developing and administering teaching standards. Anticipating this national development, the NSW Professional Teaching Standards (PTS) were released in draft form in June 2003, a month before the national framework. At the same time, moves were also undertaken to establish a state-level statutory body, the NSW Institute of Teachers, that would oversee the administration of these standards. This comprehensive approach drew on decidedly more piecemeal professional standards work being undertaken at the same time in other Australian states.¹⁰ Significantly, NSW also looked internationally to other schooling systems to inform the development of the PST, including professional teaching standards that had been developed in the USA, England and Wales, Scotland, Canada and New Zealand.

The central role of NSW in the initial work around the national teaching standards is also reflected in the fact that the NSW representative to the Board of AITSL from 2009 to 2014, Patrick Lee, was also, concurrently, the Chief Executive of the NSW Institute of Teachers, and had been centrally involved in the writing of the PTS in NSW. Lee was also on the Interim Advisory Council of NIQTSL. The APST were thus influenced not only by NSW and its PST, but also by a key state-level policy actor who contributed directly to influencing the shape of the national standards. A recent review into the activities of the NSW Board of Studies, Teaching and Educational Standards (BOSTES) – the organisation into which the NSW Institute of Teachers was eventually merged – speaks precisely to the leading role of NSW in the development of national teaching standards and, by association, a national policy space in which these standards could be implemented:

New South Wales, through the activity of the former Institute of Teachers, has been a national leader in the development of a consistent approach to high standards of teaching. As a result of this leadership, the Australian Professional Standards for Teachers, developed by the Australian Institute for Teaching and School Leadership (AITSL) and approved by Education Ministers in 2011, were heavily based on the NSW approach. The Standards are now in place in all states and territories. (Louden, Paul, and Lambert 2016, 30; emphasis added)

This quote reflects (accurately in our view) how NSW effectively laid important groundwork for the subsequent APST. To further illustrate this point, we have highlighted (see Table 1) the extensive similarities that eventually resulted between the NSW standards and the national APST.¹¹ This highlights the relationality inherent in contemporary policy-making processes and especially the development of a national policy space within a federal system, in which state-level policies are variously re/dis/assembled into a national approach – but in uneven ways, with one state able to exert much more influence over shaping the assemblage than others.

Table 1. Comparing the features of the NSW and Australian Professional Standards for Teachers.

Features of the standards	NSW Professional Teaching Standards (PST), 2003	Australian Professional Standards for Teachers (APST), 2011
Four Career Stages	Graduate Teacher	Graduate
	Professional Competence	Proficient
	Professional Accomplishment	Highly Accomplished
	Professional Leadership	Lead
Three Domains of Teaching	Professional Knowledge	Professional Knowledge
	Professional Practice	Professional Practice
	Professional Commitment	Professional Engagement
Seven Elements (PST) or Standards (APST)	1. Teachers know their subject/content and how to teach that content to their students	1. Know the content and how to teach it
	2. Teachers know their students and how students learn	2. Know students and how they learn
	3. Teachers plan, assess and report for effective learning	3. Assess, provide feedback and report on student learning
	4. Teachers communicate effectively with their students	4. Plan for and implement effective teaching and learning
	5. Teachers create and maintain safe and challenging learning environments through the use of classroom management skills	5. Create and maintain supportive and safe learning environments
	6. Teachers continually improve their professional knowledge and practice	6. Engage in professional learning
	7. Teachers are actively engaged members of their profession and the wider community	7. Engage professionally with colleagues, parents/carers and the community

Despite this overwhelming similarity, the release of the APST by AITSL in February 2011, and the subsequent endorsement by all federal, state and territory ministers of education in October 2011, meant that the national standards would effectively supersede the PST that had been in place in NSW for eight years. This process was further emphasised by the passage of the Australian Education Act 2013, which legislated that the provision of federal funding was contingent on the states and territories adopting the APST. However, the introduction of the APST in NSW went largely unremarked, as many NSW teachers and policy-makers saw the new AITSL standards as being analogous (or even identical) to the earlier state-based standards. As Clinton et al.’s (2015) evaluation of the implementation of the APST shows, the introduction of the APST was much more disruptive as a policy intervention in states that had not previously had teaching standards, or in those where existing standards differed considerably from the APST.

Despite the clear similarity between the NSW and Australian standards, and the dominant powers NSW had in shaping the APST, the development of a national approach created new levels of overlap and duplication between state and national levels (namely, between BOSTES and AITSL), even though NSW was instrumental in the initial conception and development of the standards. In recent years, this has led to new questions about the relevance and sustainability of state-level organisations like BOSTES. The recent review of BOSTES, for example, repeatedly questioned whether needless duplication was now taking place between AITSL and BOSTES. At the same time, the review suggested the APST is producing a sense of intrusion of ‘the national’ and ‘the federal’ into the constitutional responsibilities of a subnational jurisdiction:

… there is a concern about the potential for national intrusion in areas of state responsibility. While for others, there is a sense of duplication of effort, with the suggestion that BOSTES should defer to AITSL. Since the development of the national standards, AITSL has been responsible for developing tools and resources to support all jurisdictions. BOSTES too is producing resources to support the application of the standards in New South Wales. A number of stakeholders have found the BOSTES and AITSL teacher accreditation materials to be conflicting, leading to confusion for teachers and principals … [W]hile the volume of material supporting the teacher accreditation process is extensive, it doesn’t always appear to align. (Louden et al. 2016, 32)

This suggests an emerging tension between state and national standards and organisations. In this case, there appears to be an implication that unnecessary state-level policy work should be stopped and the lead should be given to the national – or, in the words of the Review Panel, that NSW should ‘adopt and adapt’ (Louden et al., 2016, 4) from the national to the state level. While it remains to be seen how these recommendations will manifest within policy and/or practice,¹² the review nevertheless seems to lay groundwork for facilitating a further increase of federal government power (via AITSL) in the constitution of the national, insofar as a leading state such as NSW is seemingly repositioning itself, vis-à-vis the teaching standards, from a ‘norm innovator’ to one that now ‘adopts and adapts’. If such trends continue, this would arguably see a lessening of state government influence relative to the federal in a national policy context in which the federal government can generate state acquiescence through funding deals.

We see this dynamic (re)positioning of NSW in relation to the national, and the re/dis/assemblage of the NSW PST into the national standards, in light of DeLanda’s (2006) point that assemblages are characterised by relations of exteriority, in which ‘a component part of an assemblage may be detached from it and plugged into a different assemblage in which its interactions are different’ (10). This insight is particularly useful for understanding policy assemblages, especially in terms of the ways that different individual actors (e.g. Patrick Lee), different organisations (e.g. NSW Institute of Teachers or AITSL) or even policies themselves (e.g. the NSW PST or national APST) can move in and out of different assemblages and, in doing so, forge new interactions and assume different roles. For instance, the interactions of the NSW Department of Education and its policy actors with the federal government will differ depending on the exact nature of the relation, such as whether they are contributing actively to developing the national standards or, alternatively, are negotiating the uptake of the NSW PST as a re-inscribed set of national standards, with attendant conditions of funding and compliance.

This speaks to the notion that assemblages are dynamic, with different parts of the assemblage undergoing different (and at times competing) processes, thereby enabling simultaneous processes of re/dis/assemblage. For instance, we can see this in the assembling of the NSW state-level PST, in response to national developments, while at the same time the beginnings of APST at the national level, building on the emerging PST. In turn, we then see the enactment of the APST by the NSW state government and educational authorities, but in ways that are best described in terms of ongoing translation and reassemblage, rather than a simple ‘policy transfer’ from the national to the state. As Clarke et al. (2015) note, policy is ‘never a singular entity: it is put together – or assembled – from a variety of elements that are always in the process of being reassembled in new, often surprising, ways’ (9). Again, this underlines the point that a national policy assemblage cannot be explained as simply the sum of its parts. Instead, to understand ‘the national’, we must focus on such processes of translation and interaction between components, and the broader conditions of possibility informing how component parts and policies can be re/dis/assembled. For instance, a ‘national’ organisation like AITSL could technically exist but might have a far less significant influence were it not for the force and authority provided to it by federal backing and, importantly, state and territory consensus. This has allowed AITSL to steer the interactions between the other component parts of the assemblage, and thus over the characteristics of the APST as a ‘whole’ assemblage at the national level.

In this sense, what arguably matters more is how AITSL (or any other such organisation) contributes to re/dis/assembling diverse components, and the ‘form/nature’ of the assemblage that is produced as a result. How AITSL is structured, therefore, plays a powerful role in determining how individual components can exercise agency. Moreover, we can see how the states and territories very much actively constitute the national in addition to but also in combination with the federal government. Invariably, this involves the disassembling of existing policy formations, especially at state and territory levels (as reflected in the rearticulation of the NSW PST into the AITSL APST), but at the same time, even with these new forms of re/assemblage at the national level, this does not create a national ‘policy monoculture’. Instead, we can see complex processes of re/dis/assemblage unfolding within and across subnational levels, as state and territory systems seek to respond to national reforms. It is for this reason that we would also caution against construing the development of national standards as producing homogenisation at either national or state levels within federal systems. While the notion of national standards-based reform might seemingly suggest standardisation as an end product, we feel that our analysis suggests something quite different, especially when one further considers the agentic and contingent nature of enactment.

Constituting ‘the national’ through the OECD

Up to this point, we have argued that understanding national reform in federal systems is difficult when using an analytic lens that treats ‘the national’ as a coherent (or a priori) whole. Acknowledging Ball and Junemann’s (2012) view that contemporary policy-making processes reflect a thoroughly re-spatialised ‘topology of policy’ (78), we see policy-making to represent processes that go beyond traditional notions (and borders) of nation-states, and indeed, of elected governments (Lingard and Lewis 2016). In light of this, we see any attempt to understand the national, especially in federal systems, as requiring us to consider not only the complex interactions between component parts of the assemblage within Australia, but also how more global policy ideas, practices and organisations interact with the federal and state-level governments to help constitute the national, and to produce what is ostensibly national change.

Given the increasing policy influence of the OECD, and its contribution to emergent modes, spaces and relations of global education governance (Lewis, Sellar, and Lingard 2016; Sellar and Lingard 2014), we turn our attention here to how the policy work of the OECD has contributed to the assemblage of the APST. Again, the benefit of our assemblage approach is that it enables us to emphasise that the OECD is not somehow directing the development of the APST from afar in a linear way. Instead, it is one component part (albeit dominant) of a broader policy assemblage that flows across national boundaries and spaces, but which, at the same time, plays a key role in assembling national spaces. Put differently, we see the OECD’s ideas and agendas being taken up and re/dis/assembled into specific national contexts. This underlines McCann and Ward’s (2012) argument that policies ‘are assemblages of parts of the near and far, of fixed and mobile pieces of expertise, regulation, institutional capacities, etc. that are brought together in particular ways and for particular interests and purposes’ (328).

Even though we could focus on the overwhelming influence of the OECD’s Programme for International Student Assessment (PISA) on the constitution of a broader national policy space in Australia (see Lingard, Thompson, and Sellar 2016), this is not our primary concern. Of arguably greater significance to the development of the APST was the OECD report Teachers Matter: Attracting, Developing and Retaining Effective Teachers - Final Report (‘Teachers Matter’), which was released in June 2005 and highlighted the key influence of quality teaching on student outcomes. Importantly, this report coincided with the establishment of the NIQTSL and played a central role in positioning teacher training, development and status as global concerns. Indeed, an analysis of Teachers Matter shows how influential it was within NIQTSL and Teaching Australia, insofar as it highlighted: (1) the importance of teachers to student learning outcomes; and (2) the benefits from involving teachers in education policy development.

These OECD arguments were taken up directly in Australian policy developments and used as justifications for pursuing national reforms. For example, and emphasising the relevance of the OECD to the early development of national standards, a Teaching Australia report opened with the following paraphrased excerpt from Teachers Matter:

[The] broad conclusion is that of those variables which are potentially open to policy influence, factors to do with teachers and teaching are the most important influences on student learning. In particular, the broad consensus is that ‘teacher quality’ is the single most important school variable influencing student achievement. (OECD 2005, 2; emphasis added)

Also of particular relevance is that Teachers Matter concluded that improvement to teacher quality, and hence student performance, could best be achieved by developing a national representative body for all teachers and principals that can give a single united ‘voice’ to the profession, and that governments and other partners should work together to develop and implement a comprehensive approach to teacher policies. Such policies, according to the OECD, would address the image and status of teaching, make teacher education more flexible and responsive to changing global conditions, strengthen teacher professional development, encourage more flexible employment conditions, reward effective teaching and engage teachers in policy development. Beyond seemingly vindicating the creation of NIQTSL, Teachers Matters also reflects the policy relevance of the OECD to matters of national schooling policy in Australia, since the intended audience of the OECD’s policy recommendations, and (for that matter) the substantive funders of the OECD’s policy work, are typically national governments. In this way, national policy initiatives are given precedence over individual state or territory reform processes, which, in the Australian context, was translated largely into increased federal involvement in promoting national reform initiatives.

It is also interesting to consider how the work of the OECD contributed to subsequent (post-Teachers Matter) developments around teaching/teacher standards that tended to marginalise the voices of teachers and professional/union organisations, both from policy development processes and, for that matter, from AITSL itself. These developments were accompanied by a reframing of teachers as ‘the problem’ that needed to be ‘fixed’. Earlier reports from the late 1990s and early 2000s recommended creating a national body that would be driven ‘by the profession and for the profession’. For example, A Class Act had a putative focus on improving the morale and community standing of the teaching profession, and saw the locus of change being with the profession itself, which would be enabled by government. However, the past decade has seen a dramatic shift away from these initial hopes, with the AITSL board shifting from a more ‘representative’ board to an ‘expert’ board (see previous section), with agendas strongly set by the federal government. These shifts reflect a broader reframing of the ‘policy problem’, which has articulated ‘teacher quality’ (and specifically a lack thereof) as the central problem that needs fixing, a policy narrative that the OECD itself has been central in promoting. Governments, in this narrative, are reframed as needing to intervene in the profession (rather than enable it to manage itself) in order to tackle the teacher quality problem, and AITSL’s ‘expert’ board is thus supposed to comprise individuals with the expertise to solve this problem.

The OECD’s PISA has been a central lever in constructing the ‘teacher as problem’ discourse as a global policy concern. In Australia, debates have been infused with a strong sense of political panic about the quality of the teaching profession in response to declining student achievement on PISA. At the same time, PISA and other OECD data have formed central reference points in a converging body of international research that supports the OECD in arguing that teacher quality is the most important factor in determining student achievement (e.g. Darling-Hammond et al. 2005; Hattie 2009). This research cannot be abstracted from the development of the APST, as it is the former that has strongly shaped the logic and form of the latter. Recent PISA 2015 results show that Australia’s performance has declined, in both absolute terms and relative to other countries, since PISA began in 2000. Over the past decade in particular, this decline has significantly amplified the debate about teacher quality and has seen successive federal governments (from both sides of the political divide) consistently argue that national teaching standards and reforms in teacher education are necessary to address Australia’s declining PISA performance. Such trends are reflected internationally, particularly in the USA, where governments have taken similar steps to reform teacher preparation and the profession. President Obama’s federal funding initiative Race to the Top, for example, introduced a range of new accountabilities for teachers and teacher education, and these have had significant consequences for the profession and for how ‘teacher quality’ is understood. In short, PISA performance is framed as a pressing policy problem that requires some form of immediate intervention from governments, and tackling the problem of teacher quality is increasingly a dominant response.

This transnational field of debate has centrally informed the APST, which have been introduced in Australia as a policy mechanism intended to both enhance and assess teacher quality, with the view that doing so will ultimately lead to improved student achievement on standardised tests. Choosing ‘standards-based reform’ as the preferred policy solution, however, is also a development strongly influenced by the OECD, which has argued that standards-based reform is a logical policy solution to fix the problem of student underachievement. Indeed, standards-based reforms have been strongly promoted by the OECD through its policy advice and advocacy work, PISA and ‘Education at a Glance’ reports, and by a variety of think tanks, research institutes and transnational policy entrepreneurs, whose work is informed by a converging body of evidence about ‘what works’ as policy solutions in schooling (Ball 2012; Lewis 2017; Thompson, Savage, and Lingard 2016).

In summary, therefore, we see the OECD playing a particularly powerful role informing what concepts like ‘student achievement’, ‘teacher quality’ and ‘effectiveness’ mean and look like, both in policy and practice. By framing the links between these in particular ways, and by advocating specific policy problems and solutions, the OECD has been a central player in shaping national level processes in Australia. We are careful not to suggest, however, that the OECD has had a direct ‘top down’ influence over the assemblage of the APST. Instead, we see the OECD as simply one (albeit dominant) component part that forms part of a broader transnational policy assemblage, within which Australia is located. We also see the OECD has having particularly strong ties to the Australian federal government, relative to states, underlying the unevenness of assemblage processes. Moreover, ideas and practices promoted by the OECD are not simply ‘taken up’ or ‘transferred into’ Australia in a linear kind of way, but are instead translated and re/dis/assembled into the Australian context.

Seeking (but not finding) the national?

Through our focus on the development of the APST, we have not only sought to demonstrate the utility of an assemblage approach to analysing policy developments, but also the analytical difficulty inherent to researching national policies in federal systems. As we have argued, understanding ‘the national’ requires attention to be paid to complex assemblages of policy ideas, practices, actors and organisations, which work within and across a variety of policy spaces. An assemblage analytic, we argue, appears well suited to understanding both the development of specific policies and the broader operations of federalism.

One of the key insights to emerge from our analysis is that when we go ‘looking’ for the national, it is difficult to find. In other words, when we break down what is claimed to be a national policy like the APST into its component parts, we see that there is no inherent essence to the policy at the national level, beyond the written document itself. This is not only because the APST have been assembled from multiple parts and influences, but also because as Australia now moves into the enactment phase of the reform, distinct versions of the standards are being adopted and put into practice across different jurisdictions (Clinton et al. 2015). This diversity is further exacerbated by differential uptake within Australia’s three main schooling sectors (government, independent and Catholic), as well as significant differences between how the APST are being understood and enacted between schools and individual educators (Clinton et al. 2015). Both policy development and enactment processes, therefore, are marked by diverse translations and re/dis/assemblages.

The relational understanding of policy provided by an assemblage approach also allows us to understand space (a national space, in this case) no longer in terms of a priori coordinates – a fixed stage upon which events occur – but instead as something formed by relations between heterogeneous parts, where ‘points … that are distant can also be proximal (categorically as well as spatially and temporally)’ (Michael and Rosengarten 2012, 93). In this sense, space (and time) are decidedly more a posteriori in nature; that is, emergent, contingent and dynamic (Lash 2012), and emerging out of material relations. This relational, or ‘topological’, understanding of space and time is powerful for theorising how power is constructed and exercised across political boundaries, including ‘imagined communities’ (Anderson 1991) such as the nation-state, as well as revealing how these relations help to actively constitute modes of governance (Lewis and Hardy 2016). For example, even while collaborative moments – such as the Hobart, Adelaide and Melbourne Declarations – help to imagine and bring into being an ‘ideal’ national space, we argue that it is the relations between actors and agencies (which are themselves enabled by such processes of imagining) that are key to understanding what ‘national’ policies are.

Thus, and to borrow from Walter Lippmann’s (1925) argument that ‘the public’ is ultimately a ‘phantom’, we see ‘the national’ in similar terms. Put differently, we suggest it is extremely difficult (if not impossible) to ever find or clearly identify the national, per se, because at best one will find complex assemblages of policy ideas and practices – all of which coalesce to form what is ‘claimed’ as being national like the APST. In this way, the national only appears coherent if we consider it in terms of written policy documents like the APST, or in terms of ‘consensus policies’, like national agreements (e.g. the Hobart, Adelaide and Melbourne Declarations). But, of course, if we follow Ball (1993) and consider ‘policy’ to be more than the mere written text, then finding the national becomes even further complicated. What we have, therefore, is ‘the phantom national’, insofar as it has no definable essence. In other words, the national is an inchoate and messy assemblage of components that is always unfolding because of changing relations of power and politics, and will manifest differently at different times and in different policy areas. The national, therefore, is somewhat illusory. Much like phantom pain, it is ‘real’ insofar as it is experienced and thus ‘exists’, but it cannot be easily found if one goes looking for it. What ‘the national’ looks like in relation to national teaching standards is not only unique to that specific policy area, but is effectively impossible to ‘pin down’ analytically in any sort of ‘complete’ way. This is because, as an assemblage, ‘it exists in a state of process whereby it continually passes into something else’ (Message 2010, 280). The national, therefore, can be considered as something akin to Deleuze and Guattari’s notion of ‘territory’, as manifesting in ‘a series of constantly changing heterogeneous elements and circumstances that come together for various reasons at particular times’ (Message 2010, 281). This is not to suggest that the teaching standards do not have real effects in systems, sectors and in the classrooms of teachers across the country. Nor are we suggesting that seeking to analyse national policy processes is a fruitless endeavour. Instead, we are again stressing that any analysis that seeks to understand such processes needs to start from a position that does not assume ‘the national’ to exist in an a priori manner. Instead, we should begin by focussing on existing ideas, practices, actors and organisations, and the interconnections between these parts (as well as the resulting impacts).

All of this arguably poses a much bigger question: ‘What are the APST?’ National in name and in text, perhaps, but certainly something quite different in the policy worlds beyond that. Put differently, rather than it being possible to define an authoritative or essentialised version of the national standards, the nature of policy development and enactment means that the APST are only ‘made real’ when translated and assembled into material practices, and these acts of translation and assemblage depend largely on the context in which the Standards are being enacted (e.g. different jurisdictions, sectors, schools and classrooms). This insight will no doubt please scholars and activists who fear that the rise of standards-based reform will drive the creation of rigid national monocultures. Instead, we see a clear and important difference between ‘standards’ and ‘standardisation’, whereby the former does not guarantee the latter. Standards-based reform, therefore, does not necessarily produce standardisation. Moreover, instead of viewing the development of national teaching standards as the ‘unfolding’ of a coherent reform agenda, an assemblage approach helps us see policy reform as a highly dynamic and messy process that is always subject to change and mutation. As Ureta (2015) puts it, assemblages are never fixed with an essence but are, rather, ‘alive and moving [and] continually evolving as new elements are added, removed, and/or transformed’ (13).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Glenn C. Savage is a senior lecturer in Public Policy and Education at the University of Western Australia. His current research examines the development of national schooling reforms and how policies in federal systems are mediated by transnational flows of policy ideas and practices. He currently holds an Australian Research Council ‘Discovery Early Career Researcher Award’ (DECRA) titled ‘National schooling reform and the reshaping of Australian federalism’ (2016–2019).

Steven Lewis completed his PhD in the School of Education at The University of Queensland, addressing the development and effects of the OECD’s PISA for Schools programme in terms of global educational governance. He has recently been appointed as the Alfred Deakin Postdoctoral Research Fellow at Deakin University, Australia, with his research focusing on how new global modes of standardised testing and data, and evidence around ‘what works’, help to reshape local schooling practices, teachers’ work and student learning in lower-SES communities. He has recently published in the British Journal of Sociology of Education, Comparative Education Review, Critical Studies in Education and Journal of Education Policy.

Funding

This work was supported by the Australian Research Council [grant number DE160100197].

Acknowledgements

This paper has been enriched by conversations and collaborations with a number of colleagues. We extend particular thanks to Bob Lingard, Sigrid Hartong, Laura Engel, Jennifer Wallner, Sam Sellar, and a number of other colleagues involved in a symposium titled ‘Towards a global standardisation of education? Exploring the Re-scaling of Policy and Governance in Multi-level Education Systems’ held in November 2016 at Schloss Herrenhausen, Hannover, Germany. We also thank Kate O’Connor for reading an early draft of this paper, as well as Helen Sullivan, Fazal Rizvi and Lyn Yates for contributing insights to our development of the assemblage analytic we use in this paper.

Notes

1. See Buchanan (2015) for a useful discussion of assemblage in Deleuze and Guattari’s work. Buchanan argues that the concept has ‘drifted a long way from its origins’ (382), as it has been taken up in different fields.

2. It is also worth highlighting that there is no ‘singular’ or homogenous approach to policy assemblage analysis, with the various authors we cite herein adopting distinct approaches. Unsurprisingly, therefore, the study of policy assemblage is itself an assemblage of heterogeneous ideas and approaches, but which reflect a family likeness.

3. We see these two broad but distinct ways of using assemblage reflected in broader theory and research. For example, there are several researchers who use assemblage to frame policy itself as the object of analysis and the phenomenon being assembled (e.g. Ureta 2015), whereas others treat policy as ‘just one element’ in seeking to understand broader forces and formations in areas like education (e.g. Youdell 2015). We see both as generative, but as serving distinct (yet related) purposes.

4. In this way, an assemblage approach walks an interesting line in relation to historical debates in the social sciences between methodological individualism and holism, and about structure and agency. In many ways it cuts through these binaries. This is because, on the one hand, an assemblage analytic stresses that the whole is more than the sum of its parts (an argument it also shares with complexity theory), but (as we outline in more detail below) it also resists a temptation to seek explanations at the level of the individual, instead framing individuals as component parts within webs of interaction that serve as mediating conditions of possibility.

5. A key challenge for researchers using an assemblage approach is deciding which components to focus on from a potentially very broad set of options, and which components to leave out. Again, an assemblage approach shares similarities in this regard with related concepts such as dispositif or apparatus, which also take a broad view of policy as constituted by multiple ideas, practices, people, objects, organisations and so on (see, for example, Bailey 2013). In our view, the choice about which components to focus on should ultimately be driven by a specific research question or problem guiding the analysis. As our key interest is in understanding the development of the APST, we have chosen to focus on those components that appear most central to shaping this reform.

6. The Agreement acknowledged that there were ‘significant challenges’ to ‘maintain’ the quality of the Australian teaching workforce (COAG 2008, 4), with the goals of the Agreement predicated upon ‘provid[ing] the platform for raising student performance’ (4). Here, we can see a clear discursive link between teacher quality and student performance, and a correlation between declining student performance and (presumably) low teacher quality.

7. This is possible in Australia due to a high level of ‘vertical fiscal imbalance’. For example, as a result of historical shifts in taxation resources from the states to the federal government, the federal government has gained much greater revenue raising capacity. However, the states have remained constitutionally responsible for the delivery of expensive social services such as education. This has created a dependent relationship between the states and the federal government and opened the way for greater federal involvement and intervention.

8. The Australian Professional Standard for Principal (note singular 'Standard' instead of plural ‘Standards’) was also endorsed by all ministers (federal, state and territory).

9. See: https://www.education.gov.au/australian-institute-teaching-and-school-leadership-aitsl.

10. At the time that the NSW PTS were being developed, Queensland had similarly established a standards framework for ‘graduate teachers’, Victoria was in the process of developing standards for ‘accomplished teachers’, and WA had developed ‘exemplary standards’ for ‘Level 3 classroom teachers’. However, NSW was distinguished by being the only state or territory to articulate ‘an integrated career long approach to the development of professional teaching standards’ (Hayes 2003, 34).

11. The 7 APST standards have been reordered to emphasise their alignment with the 7 NSW PST elements.

12. One major change to already occur as a result of the review is that BOSTES has been renamed the NSW Education Standards Authority (NESA).