Initial teacher education in the panopticon

The schools’ inspection regime in England has shifted in recent decades from a focus on external assessment of practice to a scrutiny of external data and schools’ self‐evaluation, culminating in a normative system based on self‐surveillance by school senior managers. This model of inspection (characteristic of the performative approach to public sector accountability) is now being extended to providers of initial teacher education, with providers using a standardised self‐evaluation template. A qualitative analysis of this template demonstrates its attempt to normalise and manage the development of initial teacher education programmes in order that they reflect political priorities rather than being based primarily on the professional knowledge and judgements. Further, the potential for a conflict of interests between the government agencies responsible for delivery and inspection is considered.


Introduction
The past two decades has seen a well-documented revolution in the management of public sector services, not just in the UK, but globally (Mahony and Hextall 2000), with two related developments emerging. The first of these has been the increased marketisation of service delivery, with the second being the growth of an 'audit culture' in which performance is rigorously measured against externally-determined targets. It is this second development that we examine here in the specific context of the work of teachers and teacher educators in England, by tracing the changing ways in which their performance is regulated and inspected. Since the early 1990s, inspection has been transformed from the relatively collegial regime of assessment/guidance of 'experts' and peers (Her Majesty's Inspectors/Local Authority advisory teams) via a relatively short period of intensive inspection (and sanctioning of 'underperforming' schools) by external state agencies (Ofsted, the English school inspectorate), into a model in which the 'inspectocracy' has withdrawn to the role of overseer whilst the profession polices itself. We examine the implications of this self-policing, which, despite being located in a discourse of 'self-evaluation' and constructive inter-professional dialogue, has replaced teachers' fear of the triennial Ofsted visitation with the permanent presence of the mechanisms of surveillance within the school itself (Troman 1997). This is a particularly significant discussion because it takes place as a new inspection methodology for initial teacher education (ITE) 1 is being introduced which largely replicates the self-surveillance methodology applied to schools. *Corresponding author. Email: caw11@leicester.ac.uk The value of data-driven auditing of the performance of professionals is frequently portrayed incontestable, the 'common sense voice of the people' in a 'consumer democracy' in which critical perspectives can be portrayed as backwards-looking and obstructive (Mahony and Hextall 1997). However, whilst being lauded by government ministers as increasing accountability and the quality of service delivery, this policy has caused growing concern in public sector professions because of a perceived 'coercive instrumentalism'. These concerns have been reinforced by research evidence (often gathered by government agencies themselves) about the negative impact of the audit culture. Examples from education include a narrowing of the curriculum (QCA 2001), decreasing pupil motivation (Harlen and Deakin Crick 2002) schools' 'malpractice' in respect of GCSE course work (QCA 2005) and a 'risk averse' culture of conformity developing in the teaching profession (Hayes 2001;Robinson 2005). This paper explores some factors that may contribute to these negative outcomes, and concludes by discussing the possible implications of the present drive to 'complete the circle' of education self-surveillance via the inspection of ITE.

Opening up the secret garden
The period from the 1950s to the 1970s saw the development in the UK of a strong teachers' professional culture and the active promotion of theories of learning and teaching set within a broad consensus about the purposes of school and the most effective ways of fulfilling these purposes. Schooling was increasingly portrayed as an empowering, egalitarian process, raising aspirations as well as attainment and contributing to a progressive social agenda, exemplified by the move to comprehensive schools at secondary level, the growth of 'child-centred' teaching strategies and the increased attention given to addressing issues of social justice within both school organisation and curriculum content. These developments coincided with the establishment of teaching as a graduate-only profession in the late 1960s and the growth of a powerful teachers' union movement.
The first signs of a significant challenge to the 'progressive consensus' came with the publication of the polemically neo-conservative Black Papers (Cox and Dyson 1969), but the key turning point was Prime Minister James Callaghan's Ruskin College speech in 1976, which opened up the 'secret garden' of the teaching profession. This brought the politics of schooling into the political spotlight, preparing the ground for a 'conservative renewal' in educational discourse following the election of the Conservative government in 1979. The development of comprehensive schools, child-centred teaching and the acknowledgement of social issues in the curriculum were portrayed as 'leftwing indoctrination' (Hillgate Group 1986;Palmer 1986) and a succession of neo-liberal and neo-conservative policies were put in place, designed to expose state schooling to the forces of the market (e.g. the devolution of aspects of management, including budgetary ones, to school level) and to rein in the autonomy of the teaching profession (Furlong 2005, 122). Teacher education was the focus of particular attention from the 'radical right' for its perceived left-wing bias (Hill 1990;Demaine 1995), paving the way for the confrontational relationship between governments and teacher educators throughout the 1990s.
This shift in political climate led eventually to the introduction of the National Curriculum in 1990; whilst this simply prescribed what should be taught, the literacy and numeracy strategies that followed towards the end of the decade increased the degree of prescription by specifying how it should be taught, a trend that has continued since then (for example, the active engagement of government ministers at the supremely micro level of how primary schools teach reading through synthetic phonics). With the school curriculum in place, a 'quasi-curriculum' for initial teacher education (ITE) could be introduced with the establishment of the Teacher Training Agency (TTA). Initial teacher education was re-branded as initial teacher training, reinforcing the shift in emphasis to learning in higher education to training in the classroom; in its early days the TTA appeared ideologically wedded to the notion that the balance of power should move away from universities and towards schools, a stance characterised by a faintly disguised distrust of 'the educational establishment' in university education departments (Gilroy 1998(Gilroy , 1999. This increased emphasis on training in the 'craft skills' of teaching has been a key driver in defining teacher identity as less a response to inculcation into a professional culture and values (Bourdieu's habitus) than to externally-regulated instrumentalism (Leaton Gray 2006). Once curriculum control in both schools and ITE was in place, the mechanism for inspecting them could follow, and in 1993, the Office for Standards in Education (Ofsted) was established to take over the role of inspection of schools and ITE.

The rise of performativity
The management and assessment of schools via the audit culture, together with their exposure to market forces, can be viewed as a characteristic example of 'performativity' (Turner-Bisset 2007). Ball defines performativity as a mode of regulation in which judgements about individuals'/organizations' are measured by 'displays of "quality", or "moments" of promotion or inspection" (Ball 2003, 216). Entwined with wider notions of market efficiency, performativity necessitates the use of multiple tiers of auditing, from individual to school and local authority level. This has not been the result of a single initiative; rather it has developed through a succession of policy reforms promoting a culture of hyper-accountability, with schools being increasingly driven by targets where success indicators are determined by political imperatives rather than professional judgements (Mansell 2007). These reforms have not only been wide-ranging, but ideologically paradoxical, comprising as they do of a blend of 'neoliberal commitments to the market and a supposedly weak state… [with] neo-conservative emphases on stronger control over curricula and values' (Apple 2005, 11).
The justification for these reforms has been that not only do they raise the quality of service 'outcomes' but that they lead to greater professional status and autonomy. However, some have argued that the neo-liberal project of public sector reforms led to what Du Gay (1996) calls 'controlled de-control', where teacher self-regulation becomes the norm. Inglis (2000) argues such self-regulation is a form of neo-Foucauldianism which: … has dominated the human sciences for a couple of decades because the great man can be enlisted in a theory of the totalitarianism of the modern state, achieved largely by its capacity so to order and generate fields of language that discourse itself coerces all conduct, and men and women everywhere in chains forged by linguistic legislation. (Inglis 2000, 423) Adopting the language of the market, efficiency and effectiveness have become central tenets of the new educational landscape, further reinforcing the importance of accountability. Poulson (1998) sees accountability in education as being ambiguous; it can be viewed as being self-regulatory, enshrined in teachers' responsibility to students, parents and colleagues, but this view has been largely downplayed in favour of externally imposed accountability through contract, testing and inspection, legitimised accountability as a legal necessity rather than a moral obligation (Inglis 2000). As Ball (2008) notes, this appeal to pragmatism is embedded into the 'third way' ideology of the Labour government of Tony Blair elected in 1997, 'the "what works" ideology of the third way… presented as "beyond" politics' (150). This technorationalist argument was signalled in advance of his election by Blair when, on the twentieth anniversary of James Callaghan's Ruskin speech, he stated that his education reforms would be 'practical not ideological' (Blair 1996).
Since 1997, the pace at which the audit culture has become embedded into education has continued, not just in the UK but globally (Moon 2003), with ever more sophisticated means of measuring the performance of schools and teachers put into place, at all times supplemented by increasing marketisation (e.g., the introduction of 'parental choice' in allocation of school places was set alongside the publication of pupil attainment league tables, with the clear expectation that the latter provided appropriate 'consumer information' to enable meaningful use of the former). In ITE, the exposure to market forces has been even more marked, with the active promotion of a 'mixed economy' in the production of the teaching workforce (Furlong 2005). This process began in the early 1990s when moves to develop models of partnership emphasising competency-based training rather than theoretically-driven education coincided with an increased emphasis on the role of mentorship by experienced classroom practitioners, particularly in the assessment of trainee teachers' competence. Whilst also firmly in the techno-rationalist discourse, this policy is underpinned by an implicit distrust of 'academics in their ivory towers' and an unproven assumption that expert teachers are necessarily better placed to train new teachers (Gilroy 1992). It also presupposes the teacher educators were naturally resistant to increasing involvement of schools in their work, whilst the reality is that they actually initiated the development of increasingly collaborative ITE partnerships (Furlong 1996). Whilst early initiatives in ITE partnerships were driven by a notion of balancing complimentary skills, this purportedly 'democratising/devolving' pressure can conversely be seen as a centralising measure designed to dilute the traditional independence of universities by introducing the values of performativity embedded within the 'new professionalism' of schools.
The TDA has long promoted the virtues of the mixed economy by encouraging the growth of schools' involvement in ITE. Initiatives such as the graduate teacher programme and school-centred initial teacher training (ITT) programmes have been established, although as these have only ever represented a small proportion of the 'market share', their role might be viewed as a statement of market ideology rather than to have a significant impact on teacher supply. More significantly perhaps is the ways in which the inspection process has steered higher education institutes (HEIs) to structure their ITE programmes and relationships with partnership schools, favouring practice-focused, school-based coaching/mentoring elements of programmes over university-based theoretical elements.
In the next section we outline the role that the inspectorate has played in the radical changes in the governance of state schooling in the last two decades, examining the ways in which, because the very nature of performativity requires that individuals and institutions become self-policing, the role of the inspectorate has become increasingly that of overseer rather than of assessor. This is done primarily to set the scene for an examination of recent developments in the inspection of ITE that largely replicates changes in the inspection of schools.

The development of school inspection methodology
As part of the 1992 Education Act, inspection of state schools and local education authorities transferred from Her Majesty's Inspectorate to the Office for Standards in Education (Ofsted). In the period from its inception to 2005, an emerging audit culture developed which increasingly ensured that educational work was carried out in the 'correct' and 'most efficient' way when set against pre-determined external criteria and targets. Central to this process was periodic inspection (generally every four to five years) in which teachers would be observed and graded to measure the quality of teaching and learning within the school, whilst documentation would be scrutinised and interviews held to ensure that the school was offering both a 'quality' service and delivering 'value for money'. Ofsted thus became the central agency for ensuring the practical application of the government mission of school improvement, and by extension part of the government machinery which ensures compliance through accountability.
Perryman (2006) characterises Ofsted as a state disciplinary mechanism through her discussion of the inspection experience of a school in special measures (a categorisation for schools which are deemed to be underperforming, and/or which have acute problems). Within her description of Ofsted's frequent visits to the school there is a keen sense of the creation of educational normalisation as a requirement to leave special measures. She argues that 'normalisation becomes an instrument of power', demonstrating that the quantitative and reductionist view taken by the inspectorate has to be accepted by schools if they are to be deemed successful (152). Fielding (2001) argues that 'accountability tends to be a largely negative instrument of social and political control' (699), and that although the inspection methodology is designed to appear objective 'it is high time the value laden nature of judgement within the Ofsted lexicon was properly acknowledged' (697).
Set against the notion of normalisation is a legitimate concern with the need for accountability within the public sector (Rowe 1994;MacBeath and McGlynn 2002). Accountability can be seen as an intrusion on professional legitimacy, but can be argued to aid understanding of quality and aspiration in educational establishments. This view leads to the recognition that there is a perceptual debate about how the structures of inspection should be understood. However, the view of Wrigley (2003) is persuasive, arguing that 'the mechanisms of accountability are everywhere, corroding educational processes, and affecting teachers and students alike' (45). This might be seen as leading to what Mahoney and Hextall (2000, 102) describe as a 'high surveillance, low trust' system.
Taking the view of Mahoney and Hextall, Ofsted, prior to the autumn of 2005, might therefore be considered as the medium through which governmental power could be realised, a panopticon where schools were under 'surveillance' to ensure conformity and a normalised approach to the work they do. As Foucault states: The success of disciplinary power derives no doubt from the use of simple instruments; hierarchical observation, normalizing judgement and their combination in a procedure that is specific to it, the examination… The exercise of discipline presupposes a mechanism that coerces by means of observation; the apparatus in which the techniques that make it possible to see induce effects of power, and in which, conversely, the means of coercion make those on whom they are applied clearly visible. (Foucault 1977, 170-1) In 2005, a radically revised inspection framework was established, with its purported guiding principle evident from the title of the document setting out the new methodology, A new relationship with schools: Improving performance through school self-evaluation (Ofsted 2004). This revision places self-evaluation at the centre the inspection process, so that 'a robust and professional SEF will virtually amount to the school's inspection report on itself' (Ofsted 2004, 12), with Ofsted making much shorter visits to schools to confirm (or otherwise) the conclusions of school managers. Under the revised framework, schools are provided with a standardised selfevaluation form (SEF) in which they are expected to summarise and substantiate their judgements. Although use of the SEF is not statutory, it is treated as such by an overwhelming majority of schools, who generally perceive the choosing of an alternative format to guarantee more intensive scrutiny by Ofsted. Guidance on the Ofsted website gives some credence to this view by explicitly conflating non-standard submissions with poor submissions; the guidance sets out what should happen 'where inspectors faced with a weak SEF to draw on or no SEF at all' (Ofsted 2008).
The development of self-evaluation can be seen as a positive move to greater school-level autonomy, and a far-reaching opportunity for the development of distributive leadership. However, the notion of distributive leadership as genuinely leading to a more collaborative, 'democratic' model of professionalism has been questioned (Woods 2004). Hatcher (2005) critiques the level to which genuine distributive leadership is applicable to the current accountability system, stating 'the contradiction between the proclaimed intention of greater freedom for teachers and the continuing, and in some cases even stricter, apparatus of centralised control, over them has been noted by a number of commentators' (255). This is especially true in relation to greater levels of external control. Hatcher therefore believes that 'There is therefore a tension -I would say a fundamental contradiction -between distributed leadership and government-driven head teacher managerialism' (255).
This ambiguity is apparent in a study by Bubb et al. (2007) who suggest that the SEF has been welcomed by many schools, but also note the ambiguity of its dual use, as a tool for self-evaluation and improvement planning whilst at the same time being a mechanism of self-inspection. MacBeath (2006) comments that SEFs must not only be well written, but must show both political and presentational skills, suggesting that the process has its ultimate utility defined by inspection. It is also interesting that Bubb et al. (2007) state that 'In some schools, the SEF has "shone a light into some dark corners" because the school has been asked to think about areas not previously considered or neglected' (35).
Whilst this view can be seen as a positive feature of the inspection methodology, in terms of increasing accountability and enhanced school improvement, it can also be viewed as a strongly 'normalising' device, by which an inherent value judgement determining the most important aspects of schools' performance, is externally enforced. This panoptic self-surveillance therefore ensures that internal monitoring and quality assurance systems focus purely on the facets of school life deemed significant by Ofsted and devalue others that may be considered more important to the school community.
The self-evaluation system, with its concomitant school improvement and quality assurance systems, appears designed to provide high quality SEF based data, with schools forced into a performative 'straightjacket', constantly tightened by themselves, whilst Ofsted micro-manages the system from a distance. As Ball (2003) states: 'the work of the manager, the new hero of educational reform, involves instilling the attitude and culture within which workers feel themselves accountable and at the same time committed or personally involved in the organisation' (Ball 2003, 219).
This focus on developing teachers' 'productive autonomy' through the device of performativity has led to a creeping instrumentalism in the classroom (Torrance 2007), whilst simultaneously allowing the government to present their educational policies as increasing professional autonomy, by 'outsourcing' inspection to senior managers in schools. School managers, fully aware of the high stakes nature of the inspection process, are faced with a choice between allowing school-level policies and practices to emerge through internal professional dialogue (with the consequent risk of a resultant mismatch with the particular priorities indicated by the Ofsted framework) or simply ensuring the compliance of their junior colleagues. Ofsted is able to micro-manage the system remotely by periodically shifting the focus of SEF scrutiny, determining which particular 'dark corners' are worthy of investigation.
The SEF, therefore, as the key interface between Ofsted and schools, is a crucial tool of performativity that ensures that the remote surveillance of schools (and the consequent homogenisation of approach by school managers) is maintained. In the 'self-managing' school the resident inspector (i.e., the head teacher) is the 'critical reality definer' (Troman 1997, 362-3).
The critique of an overbearing and surveillance-led system is not intended to demean the teachers who experience its perhaps unintended consequences. Galton and MacBeath (2008) demonstrate clearly a profession engaging and attempting to enhance the quality of student learning whilst reacting to external demands, which may even extend to envelope the professionalism of the inspectors themselves: Such a comment [a quote concerning the narrow view of inspection criteria] may do violence to inspectors who struggle to liberate teachers from the delivery straightjacket but it does reveal a widespread perception that teachers have no 'wiggle room', a perception reinforced when there are adverse experiences of inspection and a prevailing pressure not of teachers' own making. (Galton and MacBeath 2008, 55) This is a process that, for MacBeath, appears to impact on the senior leaders in schools in equal measure: While in rhetoric, and enshrined in the publications of the National College of School Leadership (NCSL), we have emerged from the dark era of managerialism into the enlightened age of leadership, Workload Reform seems to have driven schools backwards, at least in the secondary sector, to a business-like concern for efficiency rather than effectiveness, for delivery rather than growth, for executive decision-making rather than consultation. (Galton and MacBeath 2008, 56)

Role of inspection in ITE
Whilst the framework for inspection within ITE has followed a somewhat different trajectory to that in schools, the parallels are striking. Ofsted's brief was expanded to include ITE in the mid-1990s, using frameworks of assessment jointly developed with the TTA (later to become the Teacher Development Agency (TDA)), and initially at least adopted a similarly intensive, 'hands-on' scrutiny of practice as with schools. However, the coercive element of inspection is even more direct here than in schools, since the TDA has used inspection grades awarded to determine each provider's 'quality category', this being the key variable in determining the allocation of funding. The link between the outcome of the inspection and the viability and reputation of the provider is therefore direct, with evidence that unsatisfactory inspection reports have led to the closure of courses and, in some cases, whole institutions (Ghouri and Barnard 1998).
A key element of neo-liberal management of public sector provision is a 'quasidevolution' of different aspects of state functions, and the relationship between the government and the key agencies directing ITE demonstrates this. The TDA is a 'delivery' agency charged with delivering a high quality teaching workforce, whilst Ofsted are an 'auditing agency' responsible for measuring the quality of the product delivered by the TDA. Whilst the two are clearly inextricably linked, a façade of independence/objectivity is maintained; although the inspection methodology is jointly determined, the TDA argues that the inspection outcomes determining 'quality categories' are independent objective evidence. However, whilst Ofsted is formally accountable to parliament rather than government, its inspection methodology is clearly determined by the agenda of government.
From September 2008 the methodology for the inspection of ITE has been brought into line with that of schools, with the self-evaluation document (SED) the equivalent of the school SEF. This new framework has been driven by a 'pseudo-consensual' discourse; consultation over inspection criteria showed no significant dissatisfaction amongst providers, the key lesson of self-evaluation in schools is that it is the restricted, data-driven nature of evidence sought that is of most concern. As with the SEF, the use of the SED is not compulsory, but the response of providers, given the high stakes of inspection outcomes, is likely to be similar to that of schools (whilst data are hard to find, the proportion of schools not submitting the standard SEF is almost certainly vanishingly small). Furthermore, the rhetoric in support of 'selfmanagement' of schools is currently being recycled in ITE, despite the sustained critique that when applied to schools, 'self-management' simply creates an illusory freedom to mask the reality of 'self-policing' (Ball 1994;Grace 1996;Troman 1997).
Whilst the parallels between the two self-evaluation methodologies are striking, they differ in one crucial aspect. For schools, the hypotheses drawn from examination of the SEF are developed (and then tested) independently by Ofsted, but in ITE the process is politicised through the construction of an additional tier of the panopticon. Providers' annual submissions will be initially analysed not by Ofsted but by the TDA, in order to help set the context for inspection (this could involve determining the timing, focus and level of scrutiny that Ofsted will give to the provider). By giving the 'delivery agency' a direct influence over the content and 'tone' of individual inspections, the independence of the inspectorate appears to be compromised. Furthermore, the coercive nature of inspection is intensified by making the implicit explicit; not only will the TDA continue to use the outcome of individual inspections as the primary basis of resource allocation, it will now be in a position to directly influence the process and outcome of those inspections.
This conflict of interest calls into question the traditional, and fiercely defended, independence of the inspectorate. In the future they will not only be required to adopt a timetable and inspection tariff set by the TDA, but also take any initial TDA hypothesis regarding the quality of individual provision as their starting point for each inspection. Once on site, they are then required to adhere to an exhaustively prescriptive methodology which determines not only the scope and extent of their role, but sets out in minute detail exactly what types of evidence are admissible, how they should be presented by providers and how they should interpret them. Inspectors have little or no autonomy to use their professional judgement, but rely on the application of a mechanistic framework through a managerial model (with each inspection team including a 'managing inspector' responsible for the performance management of their team); the gaolers in the panopticon are thus under scrutiny themselves.
For senior managers in HEIs, the process of completing the SED is complicated by the need to predict not only the 'strategic delivery agenda' of the TDA but the 'quality auditing agenda' of Ofsted. MacBeath's point about the need for SEFs to demonstrate political and presentational skill is just as pertinent in respect of the SED, if not more so given the ways in which the inspection of ITE has been used to drive changes in the ways programmes are structured. The ITE inspection framework (specifically, the descriptors used to describe 'good practice') has always encouraged particular approaches in course design and content (despite the rhetoric about the quality of outcomes being of primary significance). An example of this is the promotion, through scrutiny of the 'effective allocation of resources within the partnership', of a particular model of partnership with schools which passes over ever-increasing responsibility for key elements of training (particularly assessment against Standards for QTS) to schools. As noted earlier, this can be seen as a move to shift training further away from the 'hard to reach' universities and closer to the 'performativised professional' domain of the school.
Just as with the SEF, the SED is the key interface between Ofsted and ITE providers, and performs the same function of facilitating remote surveillance and developing a normative model of 'good practice'.

Assessing the self-evaluation document
There are three central themes which can be drawn from the SED framework. Firstly, it is primarily driven by quantitative data analysis, this being clear from the fact that providers are supplied with a TDA 'data profile' which, according to Section 1.1, is intended to "assist their evaluation and review". Throughout the SED, providers are reminded to support their evaluations with data. This quantitative approach is most apparent where providers are asked to grade three aspects of their provision ('Selection', 'Training and assessment', 'Management and quality assurance', 'Overall effectiveness' and 'Capacity to improve'). It is these judgements that Ofsted will use to make decisions about the overall quality of provision and the degree to which they agree with internal assessments made by providers.
Giving such prominence to a reductionist model must ultimately coerce providers into adopting a simplistic, quantitative approach to self-evaluation (using Ofstedprescribed 'grade characteristics'). As with schools, the quality of dialogue with inspectors is reduced, since no opportunity for dialogue occurs until the 'remote inspection' process is well underway. Before inspectors meet with providers, the SED must be submitted for scrutiny by the TDA who, also using individual provider and sector-wide performance data, set the inspection tariff. The TDA will then inform Ofsted when and how to inspect, including identification of 'inspection trails' (the 'dark corners' worthy of scrutiny?) enabling the lead inspector to form initial hypotheses about the quality of provision. He/she will then pass this on to the inspection team to shape their initial views and 'lines of attack'. The arrival of inspectors on a Monday morning to meet with provider representatives for the first time is therefore not the beginning of the inspection; it is the beginning of the end. Performance data, and the extent to which they measure up against the TDA's strategic priorities, have become the lens through which remote decisions concerning the quality of provision are made.
This emphasis on quantitative evidence is apparent throughout the SED and makes it difficult for providers and inspectors to engage in any consideration of the complexity of the activity under scrutiny and with the contextual meaning of data. In section 6.5, the SED asks for a three-year summary of evaluations (of trainees and schoolbased partners) undertaken of training and assessment, highlighting evidence; these internal data are matched against external data from the TDA Performance Profile (which includes data gathered from the NQT Survey (a sector-wide questionnaire 'perception survey')). Whilst such data are undoubtedly valuable, they are only so when used to contextualise and inform professional judgements; without this, providers are forced to adopt a market ethos where the need for improved ratings in 'customer/stakeholder surveys' becomes paramount. The consequential distortion of professional judgement by this aspect of performativity has been well-documented in other areas of educational policy, such as the use of examination results and standardised national testing in schools (Mansell 2007).
The primacy of data is particularly apparent in the sections of the SED relating to 'training outcomes', where providers are required to give data about their grading of students and to evaluate 'trends over time' and 'the effectiveness of any action you have taken to address issues arising from the data' (sections 5.1, 5.2 and 5.3). Given the array of contextual variables that a meaningful 'trend analysis' would need to take into account, it is hard to see how this can provide a robust evidence base for inspection. However, the approach to evaluating 'effectiveness of action' is the most troubling aspect of the framework. Because of the reliance on trends in quantitative outcomes, a mechanistic discourse is encouraged, in which the provider is expected to claim that 'In Year 1 we took action x, in Year 2 achievement rose by y%, therefore x was the cause of the rise'. Most educational professionals, one would hope, would understand the danger of confusing correlation with causation, but are drawn into this flawed thinking by the inspection process.
The second theme within the SED is of normalising provision. Throughout the document, types and forms of evidence are 'suggested' rather than required, often hinting that the providers 'may wish to' go beyond the scope of currently available data (for instance, in the section 'Progression of trainees' it is suggested that 'Where data are available, the proportion still in teaching three years after completing the course… should be noted'; section 6.4). As with the 'suggested' use of the SED itself, the high stakes nature of the inspection process means these suggestions become de facto requirements. The need for providers to use Ofsted grade characteristics (in the section 'Training outcomes') is another demonstration of the normalising nature of the SED. Whilst providers are at liberty to use their own assessment mechanisms internally, when submitting their SED are required to 'translate' these to the Ofsted grade frameworks; they are therefore faced with either investing significant additional time in 'cross-matching' two assessment mechanisms, or simply compliantly adopting the Ofsted framework.
The SED therefore can be seen to be an effective tool for standardising provision, implicitly demanding that providers alter their provision to meet the needs of TDA/ Ofsted, not only in respect of their evaluation systems, but in the way the course is managed and delivered. This would be in keeping with the experience of schools since the introduction of the SEF. As such the SED provides a clear TDA/Ofsted view of how ITE should be structured and delivered, and may provide a barrier to providers with 'alternative' approaches.
The third theme of the SED is of managerialism. Threaded throughout the document is a focus on 'improvement', and, moreover, emphasises that self-evaluation of improvement should concentrate on measurable outcomes 'using statements of fact rather than writing evaluative comments' (section 3). In keeping with the first theme giving primacy to measurable, quantitative impact, improvement is to be simply judged through the simplistic correlation of 'action + outcome = impact'. As such, the focus on 'improvement' as demonstrated through data outcomes may have significant consequences for the management of ITE. The presumption that quantitative data in themselves lead to a clear assessment of the quality of provision in turn suggests a model of management in which the causes of 'underperformance' can be easily isolated and redressed, leading to definable improvement. This mechanistic model ignores the complexity of the educative processes involved, and may encourage shorttermist, target-driven management, replacing strategic leadership.

Conclusions
The inspection of schools has shifted from the direct monitoring of 'service delivery in action'; inspectors observing teachers in action and making a professional judgement about the quality of their work, to a data-driven monitoring of professionals' self-assessment. This shift has created an internalised self-surveillance culture in which senior managers have become 'the inspector within', managing the work of their colleagues, and ensuring that key performative targets are met. Central to creating and maintaining this self-policing culture is the SEF; it ensures that schools fall into line with whatever is currently judged (by the government) to be the model for an 'effective school' and for 'best practice', that these are adopted nationally, and that as government initiatives regarding 'best practice' change, so to does school practice.
This culture suggests a strong tendency towards normalisation, with the pressure for compliance that is inimical with the traditional values of critical engagement and the exercising of professional independence. The quantitative focus of the methodology makes resisting the normative tendency almost impossible, since this focus is more sophisticated than a simple positivist one of 'what matters is what can be measured'; the parameters are carefully set to ensure that only those foci deemed important by government are measured and reported on publicly. Within this developing framework, Ofsted has become the 'remote gaoler'; schools are constantly under the panoptic gaze, exposed by their examination/SATs results and the wider information published in their SEF. Senior managers become the 'on-site gaoler' who are responsible for directly coercing colleagues into a compliant approach, whilst the fact that the coercion comes from within the profession enables a façade to be maintained by which the profession is portrayed as having increased professional autonomy.
Performativity necessitates a compliant workforce philosophically responsive to the needs of the performative state, otherwise professional values and practices based on critical enquiry will challenge and may undermine the central functionalism of the mechanism. In England, the introduction of the various apparatus of performativity, in particular its punitive inspection regime, have created an ethos that encourages compliance rather than critical engagement. In the process, the notion of reflective practice has been diluted, and rather than a genuine engagement with critical issues in professional practice leading to the 'bottom up' development of new initiatives (or indeed the critique of policies/strategies) it has come to mean simply assessing the 'effectiveness' of new initiatives by use of simplistic measures of 'what produces better results'.
With a secure panopticon overseeing the work of schools, the final part of the performative framework of education is being constructed in ITE. As long ago as the 1980s, the then Conservative government was aware that to change the culture of schools it was necessary to control the training and education of teachers; if universities continued to produce inconveniently independent-minded newcomers to the profession, changes in school culture would be likely to meet with resistance from within. An explicit control mechanism is therefore required to ensure a compliant ITE sector. As the SEF has allowed remote surveillance and control of the school system, so the SED is capable of controlling the ITE sector. The SED is strongly suggestive of both a normative and quantitative process mimicking that which has developed within schools, and so it is likely that, over time, it will become a mechanism of internal policing in the same way as the SEF has done in schools. In fact, the direct relationship between the outcomes of Ofsted ITE inspections and the allocation of TDA funding to providers suggests the SED will become even more explicitly coercive. The additional context of the role of the TDA in undertaking the first scrutiny of completed SEDs, and in setting individual inspection agendas, means that the pressure on providers to adopt compliant approaches to course planning, content and structure will be immense. If providers themselves are unable or unwilling to shine light into the dark corners of their work, the TDA will do so for them (before handing the torch to Ofsted). It will be a brave manager in the high-stakes game of ITE inspection to resist the pressure to become the internal, ever-present gaoler, the 'inspector within' who maintains the panoptic surveillance of their colleagues to ensure that data are produced, targets are met, standards are maintained and initiatives are delivered.
When the performativity loop is truly closed, where the profession fully polices itself, dissent becomes increasingly ineffectual because it is increasingly viewed with suspicion and alarm from within. The consequences of challenging normalisation are potentially disastrous for institutions, and so educational management has increasingly become dominated by avoidance of risk. Management systems are at their most effective when they, as Lyotard puts it, 'make individuals "want" what the system needs to perform well ' (1984, 62). Whilst inspection by 'self-evaluation' is claimed by the government to give educators the professional autonomy that they have long fought for, we are left to speculate whether in fact this is the most secure prison of all.

Postscript
Our argument is not intended to foretell an inevitable drift towards a normalised delivery of initial teacher education, but to draw attention to the parallels with the self-evaluation model of inspection in schools and the evidence of impact this has had. Space limits our opportunity to explore the potential for professional values to assert themselves in this particular manifestation of the performative culture. Furlong et al. (2000), Sachs (2003) and others have considered ways in which a 'transformative' teacher professional identity might emerge through more collaborative, participatory approaches to professional discourse (engaging both within the profession and with the diverse communities they serve). Galton and MacBeath (2008) note with approval good teachers' capacity for subversion, their refusal 'to collude with the victim mentality which relinquishes initiative, self-belief and a sense of agency' (115). A similar resilience is required from teacher educators if we are to find ways of maintaining professional integrity and creativity under the gaze of the panopticon. Note 1. The preferred official identification of initial teacher education is 'Initial Teacher Training' (ITT). Within this paper ITT is used when official documents are quoted. We use 'Initial Teacher Education' (ITE) in all other cases as we are dedicated to the principle of a wider concept of education as opposed to a narrower 'instrumental' ideal suggestive of ITT.