Subjects of , or Subject to , Policy Reform ? A Foucauldian Discourse Analysis of Regulation and Resistance in UK Narratives of Educational Impacts of Welfare Cuts : The Case of the ‘ Bedroom Tax ’

This paper draws on material generated from a qualitative study of educational impacts of a British welfare reform affecting housing rent subsidy, size and location commonly known as ‘the bedroom tax’ (Bragg et al., 2015), which was partly taken as a topic for study specifically because of its iconic status as a controversial and unpopular welfare ‘reform’ (or cut). The analysis draws on Foucauldian understandings of subjectification or subject-formation as elaborated both from within and in relation to the social (Foucault, 1970; 1983; Ball, 1990; Olssen, 2006; Popkewitz & Brennan, 1998), read through new and newly available perspectives (Allen, 2015; Pêcheux, 2014). This approach is applied to discuss how those addressed by educational policy, and also as research participants, are both subject to prevailing political and practice-oriented discourses (of educational ‘problems’, and of the neoliberal frameworks by which poverty and welfare cuts are discussed), but also – at times – how they can become the subject of – in the sense of reformulating – these discourses in their accounts of everyday activities. After outlining our approach and the context for the study, we focus on four examples drawn from the narratives of the various stakeholders in the study – parents/carers, school staff and other community-based organisations as illustrations of how this discursive approach can provide rich readings of relevance to educational policy debates. From these we not only take further discussions of the production and regulation of subjectivities via social and educational policy practices, but also offer indicative glimpses of resistance to this as expressed by those who are its primary subjects, and where in one case such resistance brings our own research commitments under critical scrutiny. As such, the contribution of this article is both topic-related (concerning the educational impacts of policy) but, crucially, also conceptual and methodological, in motivating for a Foucauldian-influenced discursive approach that is sensitive to struggle and resistance.


Introduction
This research reported on in this paper mobilises a Foucauldian discursive approach to analyse processes of subjectification produced by policy practices, taking as the policy field a specific recent UK 'welfare reform', the 'bedroom tax' which although not specifically directed to educational change nevertheless, through both general and specific impacts on family income and relationships and alongside (and as part of) a wider programme of cuts in welfare provision, also has educational effects.The 'bedroom tax' is a UK housing-related welfare reform which was introduced in April 2013 (under the official name "removal of the spare room subsidy") as part of wider welfare and public spending cuts, of which it is one of the most controversial and unpopular.This new reform specifies room allocation criteria for working age tenants of social housing and mandates a reduction in housing benefits for those households deemed as having one or more spare rooms. 1  The examples discussed in this paper arise from a recent study addressing connections between welfare, wellbeing and education in relation to this new so-called 'tax' (which is in fact a welfare cut) 1 Housing benefits are reduced by 14% for one 'spare bedroom' and 25% for two or more 'spare bedrooms', meaning that tenants impacted by the tax currently lose on average £14.92 of their housing payments per week (Wilson & McInnes, 2014).The policy allocates one bedroom for-adult couples , single adults over 16 years old, two children of the same gender up to age 15, -two children of either gender up to age 9, an overnight carer (where required, but such 'requirements' are highly circumscribed).(Bragg et al., 2015), which was the first study specifically considering its links with schooling and education.
The suggestion that housing policy, even within the context of wider cuts to social services, can be analysed as part of educational policy discourse may seem odd at first glance, especially since public spending on schools has until very recently overtly been protected from recent UK Government spending cuts (Lupton & Thomson, 2015).Yet as Ball (2005, p.20) pointed out, 'In failing to take account of the ways in which education is embedded in a set of more general economic and political changes education policy, researchers close down the possibilities for interpretation and rip the actors who feature in the dramas of education out of their social totality and their multiple struggles.'Similarly, as Anyon (2005, p.66) shows '…job, wage, housing, tax, and transportation policies maintain minority poverty in urban neighborhoods, and thereby create environments that overwhelm the potential of educational policy to create systemic, sustained improvements in the schools.'As such, these policies should '…be part of the educational policy panoply as well, for these have consequences for urban education at least as profound as curriculum, pedagogy, and testing' (ibid).Following Anyon's analysis, the rationale for our discussion here is that education is not synonymous with schooling.By this statement we do not only mobilise well-known claims of how little education happens in schools, or alternatively how children learn how to fail or perhaps more accurately how they successfully learn the various 'hidden curricula' of social stratification (according to class, gender, or 'race', for example) via schooling.Rather, the approach informing this paper is that education and education policy need to be understood in their broadest senses (Fielding & Moss, 2011;Burman, 2016a;Greenstein, 2016), to include contexts of and for educational engagement that lie outside the schoolgates.This includes the range of discursive as well as material practices that policies inform and produce (taking the perspective here that the material and discursive are closely related), including how they address and engage with institutional contexts that lie outside, but closely interface with, schools and schooling, such as families, communities and workplaces (Wallace, 1961).Specifically, parents are often directly addressed by educational policies, in particular on early intervention or strategies to promote attainment, but as our discussion below will show, community-based organisations are also significant players in, or at least relayers of, the enactment of contemporary educational policy discourse.
A key contribution of this paper concerns how, whether and when speakers do more than reproduce the dominant discourse, questions which combine conceptual and methodological as well as policy concerns.That is, as we outline in more detail later, we draw on Foucauldian discourse analysis to attend to instances encountered within speakers' narratives where they narrate themselves as subjects of, and subject to, those policy discourse(s).From a Foucauldian perspective, as participants within specific historical and cultural-political contexts, it may be impossible not to be subject to (and so being rendered the object of) those dominant discourses.Nevertheless, as Ball (2005) points out, policy discourses are not entirely coherent and shift in specific contexts of practice, so that the analytic project is one of '…attend[ing] to the ways in which policies evolve, change and decay through time and space ' (p.17).In this paper we discuss instances where our participants' narratives reformulate, comment upon and even offer alternative framings of the dominant policy discourse, and in this sense they become subjects of discourse.That is, they are not merely spoken by discourse but they also demonstrate the capacity to speak of it.
We are therefore engaging with a more activist reading of Foucault, in particular as elaborated by both recent and contemporaneous theorists.Hence we will be suggesting that, even though policy may speak through us as subjects, sometimes we also speak back to it.Such a claim can be made without invoking notions of 'agency' that are incompatible with a poststructuralist understanding of the production of individual subjectivity, such as Foucault's strong constructionism (see e.g.Danziger, 1997).Rather, through close attention to particular accounts or narratives of the practice of policy, arising from and situated within particular contexts, the play and interplay of repetition/reproduction, reflection and re-formulation can be attended to, which -as we illustrate below -can sometimes offer more indications of resistance that predominating (governmentality) readings of Foucault allow.
It should be noted that the Foucauldian focus here arises not only as a tool to analyse participant accounts of the educational impacts of welfare reforms, but also as a way of interpreting the wider circulation and functioning of educational discourse.Various commentators have highlighted how educational discourses have become a key feature of current neoliberal political practice, in the sense of shifting from direct prescription or legislation to motivating for ways to better teach or guide the performance of good citizens.This pedagogical state (Jones et al., 2013), not only positions citizens as in need of learning the correct (political) lessons (of how to behave, work, live, etc.), but it also selectively promotes the regulation and scrutiny of some social sectors over others -in particular welfare recipients who form the focus of significant moral and scientific policy discourse.2From this perspective, then, not only does any policy affecting children's engagement in schooling count as 'educational' for our purposes, but we are also concerned with the ways policy now mobilises educational discourses (of teaching and learning) that produce particular forms of subject position (Davies & Harré, 1991) or subjectivity (Henriques et al., 1984(Henriques et al., /1998) ) or, in the terms we use here, subjectify (Patton, 1986;Davies, 2006) particular groups of citizens as teachers, students or learners.As we shall see, not only are children so addressed, but also parents, with particular conceptions of the social correspondingly enacted.
After outlining the discursive parameters of the policy context surrounding the 'bedroom tax' and our methodological approach to analysis, the rest of this paper focuses on the ways various stakeholders in the study -parents/carers, school staff and other community-based organisationsnarrated its educational impacts, attending to the subject positions assumed by the narrator as well as those narrated about.Four examples are discussed to address two key aims: firstly, the different positionings available to, and narrated by, these different stakeholders thus illuminating the educational impacts of current policy discourse in the UK.Second, we explore these as methodological illustrations of different ways a Foucauldian discursive approach (informed also by post-Foucauldian perspectives) can inform educational policy analysis.We conclude the paper with some reflections on what these examples might indicate about material-political and discursive conditions for turning regulation into resistance, highlighting not only wider discourses that challenge prevailing passive and determinist positions around poverty and educational policy, but also some specific discursive interventions made by those subject to those policies.

The 'Bedroom Tax' and/as Neoliberal Discourses of Work and Family
As already indicated, this paper draws on material generated from a project exploring the impacts on children and families of a current UK cut in housing -related welfare support, called the 'Removal of the Spare Room Subsidy' policy.This policy was introduced in April 2013 as part of a wider programme of welfare reforms, and it remains in force despite much controversy and agitation. 3Our project, the first study specifically addressing connections between welfare, wellbeing and education in relation to this new 'tax' (or welfare cut), was conducted in our locality (Manchester, UK).4It documented both the material and emotional effects of these cuts in welfare support (which are financially quite significant)5 on parents and children, which were accompanied by other welfare cuts affecting the same groups (see Bragg et al., 2015;Greenstein et al., 2016), with a particular focus on exploring their impacts on children's educational engagement. 6s its more common or popular designation as the 'bedroom tax' suggests, instead of official terms such as 'removal of the spare-room subsidy', 'social sector size criteria', or 'underoccupation deduction', this policy directly intervenes in the structure and composition of the domestic space, in terms of specifying age, gender and generational relations governing allocation and entitlement to space within the household.The colloquial name change points to the general perception of the policy as a tax rather than a removal of subsidy, while the topicalisation of 'bedroom' highlights a focus on intimate relations and sleeping arrangements.Through defining these, the 'bedroom tax' works to prescribe and proscribe family forms and ties, and in this sense can be read as going beyond mere economic considerations (of cutting welfare costs) to intensify the psychological gaze upon families (see Greenstein et al., 2016, for further discussion).This point is important to our later analysis, as such presuppositions are at play in particular within our final example discussed below, while it may also account for why the 'bedroom tax' is generally understood as both indicative of, and perhaps the most despised example of, general welfare reforms.As the (then) Shadow Work and Pensions Secretary Owen Smith said, 'If one policy sums up the cruelty of this Tory government it's the Bedroom Tax '. 7  This policy participates in the neoconservative political narrative of the post 2008 economic crisis as arising from too much public spending on welfare, and the moral/characterological deficits of the poor, rather than the mismanagement of bankers and their backers.Thus this, as other, welfare 'reforms' (we scarequote the term 'reform' to topicalise how this is really a cut) mark the discursive creation of certain types of individuals and families as deserving and valued, against which '...classed Others are produced and symbolically shamed for not being austere enough' (Jensen, 2012, p. 15).
Clearly, the 'bedroom tax' is merely one of a raft of cuts in welfare reflecting the neo-liberal economic reforms that have restructured the labour market (worldwide and also in the UK) to make work low-paid, precarious, fragile and often short-term (Bailey, 2016).This accounts for why the examples discussed below concern impacts of reduced income on households, rather than only what is happening to children's sleeping and housing arrangements.It is also worth noting here that, contrary to the public discourse of poverty as a matter of unemployment, even chronic unemployment, the most common picture is of in-work poverty (Rosso et al., 2015;Shildrick et al., 2010); that is, of periods of temporary, low paid employment generating incomes below minimum thresholds, which have been shown to particularly affect single parents and households with children (Padley & Hirsh, 2016).
Alongside this, current policy constructs poverty as a question of individual responsibility (Grabham & Smith, 2010;Pantazis, 2016), such that responses focus on 'activation'.That is, the neoliberal state encourages its citizens to be entrepreneurial and self-sufficient (Bhattacharyya, 2015;Lister, 2006;2011).This discursive turn 'explains' poverty as the result of reckless behaviour, so that its alleviation becomes a matter of re-educating the poor out of state dependency.In this discourse, questions of unemployment are transformed into discussions about strategies to increase 'employability' or to counter 'worklessness', both portrayed as a feature of intrinsic traits of the individual rather than socio-political conditions.Indeed notwithstanding the widespread discourse around cultures of worklessness and state dependency mobilised through narratives of povertystricken households composed of three generations of unemployed people, researchers were scarcely able to find any such households (Shildrick et al., 2012).In relation to the 'bedroom tax', and its impacts on parents of school-age children, the focus is on economic mobility, rather than on inequality, which occludes the role of structural changes to global markets in the creation and maintenance of poverty, in and out of work (Jensen & Tyler, 2012).
In terms of the broader concern with educational discourse outlined above, of pedagogical and psychoeducational imperatives operating outside explicit schooling institutions, we should note that the discourse of responsibilisation configures poverty as an educational issue in three ways.First, it is about educating the poor to find work through creating policies that penalise poverty and are presumed to 'incentivise' people into finding (and keeping) more paid work (Jones et al., 2013).Second, through the (false) portrayal of intergenerational worklessness (see Shildrick et al., 2012), these 'incentivising' policies are seen as useful in eradicating the cultures of worklessness and thus as an educational measure to prevent the spread of this 'culture' to future generations.Third, and crucially, these discourses place an onus on mothers to nurture resilient children who can maximise their 'human capital' in face of the adversity of poverty (Henderson & Denny, 2015).It is worth noting the contradictory policy demands on mothers to be simultaneously engaged in full time paid work (see Cain, 2016) and be emotionally available and ready to invest in the right nurturing response that would lead to such resilience.Hence our analysis extends feminist critiques of the individualisation of poverty in highlighting its deeply gendered aspects (e.g.Morini, 2007;Lister 2006), whose covertly racialised aspects (Bhattacharyya, 2013) and unequal impacts on disabled people (e.g.Duffy, 2013;Power et al., 2014) should also be noted.While unwaged care and reproductive work is disproportionately carried out by women, not only is this made invisible through rendering entitlement to many benefits conditional on actively seeking and gaining waged work and prioritising it over other commitments (Grabham & Smith, 2010;Pykett, 2012a), but cuts to social and educational services actually increase demands for this unpaid care (Abramovitz, 2012;Harrison, 2012;Roy, 2012;Roberts, 2014).
Notwithstanding its coexistence with other welfare cuts under neoliberal state policies, then, the 'bedroom tax', is of particular interest in relation to discourses of gender and class (including their intersections with racialisation and disability-status), which it both presumes and intensifies.As Murray (2014) noted, attention to this performativity suggests that the feminisation of poverty through welfare cutbacks is better framed as a feminisation through poverty.Hence not only does the 'bedroom tax' incite further state surveillance of the domestic sphere (in the name of neoliberal responsibilisation), it also performs particular -here particularly gendered and classed -acts of reification, or fixing of positions and identities.This conceptual-political point also has methodological implications, in terms of how narratives of its impacts can be analysed.It is the construction and negotiation of such positions by and through educational discourses and the accounts generated from and about educational practices that we attend to here.

A Foucauldian Discursive Approach
Our account so far has followed much of the critical social and educational policy debates that mobilise social constructionist or discursive frameworks.In particular, such discursive approaches have been helpful to explicate the repertoire of elements and positions brought to the fore by neoliberalism (Ball, 2015;Ball & Olmedo, 2013).These include not only active, and increasingly agile (Gillies, 2011) subjects, who are incited to maximise themselves to make good amid fluctuating markets (Masschelein & Simons, 2005) and a retrenched state, but also the responsibilisation (the making responsible) of citizens for functions and activities previously undertaken by and guaranteed by the state.The obverse of these can be seen in the rise of diagnoses of anxiety, vulnerability and other forms of distress associated with financial pressures, precarity and insecurity, that can lead even to suicide (Barr et al., 2015;Ellis et al., 2013;Isin, 2004 ).(Indeed it is worth noting that a key rationale for undertaking our study was because of the escalation in referrals to psychotherapy services in which some of the research team were involved, see also Winter et al., 2016).In particular, as with the discursive shift from 'unemployment' (a structural condition) to 'worklessness' (figured as an individual attribute or state), processes of psychologisation are at work (De Vos, 2012, 2014) that correspond both with the occlusion of global and local structural explanations and -as a correlate of responsibilisation -imply a retraction or minimisation of the social bond.Further, to promote this psychologisation there is a pedagogical state that extends schooling outside classrooms to guide and educate parents and families to make better choices, rather than fund resources (Jones et al., 2013;Pykett, 2012b).
Since several discursive approaches are employed by social and educational policy critics, we now explicate our specific approach.8Our analysis here engages Foucault's analyses of the ways modern states focus on disciplinary practices as elaborated by institutions governing the family and individual-state relations, and the corresponding subject positions they enable or ward off (Foucault, 1980(Foucault, , 1981(Foucault, , 1988a,b),b); that is, the way people are disciplined by the organisation of discourse.We draw on a Foucauldian model of subjectification or subject-formation -as elaborated both from within and in relation to the social (Foucault, 1970;1983a;Ball, 1990Ball, , 1994Ball, , 2005Ball, , 2015;;Bourke & Lidstone, 2015;Olssen, 2006;Popkewitz & Brennan, 1998).As Patton (1986) puts it: 'subjectification' refers to 'those forms of conscious and unconscious relation to the self which make us subjects of a certain kind' (p.24).Contrary to some criticisms of the ways Foucault's analytical frame has been applied in educational research (in particular as formulated by Scheurich & McKenzie, 2005), our focus is at least as much on the modalities of subjectivity promoted or proscribed by particular discourses as on the forms of regulation and control (Burman et al. 1996).This difference of emphasis perhaps arises from our own intellectual histories which, in addition to education, span critical psychology, community development, counselling psychology and disability studies, where the reception of Foucault has been oriented to challenging the 'psy complex' (Ingleby, 1985;Rose, 1985) with its production of forms of normalisation and pathologisation.Nevertheless, the project of elaborating socially-based models of the subject has remained a key socio-political project (see e.g.Henriques et al., 1984Henriques et al., /1998;;Adlam et al., 1978), including how (subjective and political) change is possible (Butler, 1997).While influenced by, we therefore also depart from, some governmentality approaches (e.g.Rose, 1990;Rose et al., 2006), which we read as offering too deterministic a reading of power relations.Instead we align with reformulations emerging within educational research of Foucault's earlier work that allow both for a greater political engagement and understanding of counterpractices of power (Allen, 2015, Ball, 2005;Pêcheux, 2014).As Foucault put it: …everything is dangerous, which is not exactly the same as bad.If everything is dangerous, then we always have something to do.So my position leads not to apathy but to a hyper-and pessimistic activism.I think that the ethico-political choice we have to make every day is to determine which is the main danger.(Foucault, 1983b p.356)For our purposes here, discourses are understood both as structurally elaborated and as situationally reiterated frameworks of meaning, with possibilities of shifts and renegotiations arising within specific local practices (Parker, 1992;Burman & Parker, 2016;Burman et al., 1996).Elaborating a model of power as relationship 'exercised from innumerable points' (Foucault, 1981: 94), Foucault highlighted complex reverberations and resonances that transfer across, between and within bodies and minds, and between the psychic and the political.So while discursive frameworks define and delimit what can be spoken, and in particular set out positions and relationships between elements and parties within each discourse, yet even if (or perhaps precisely because) there are predominating or hegemonic discourses, there are also counterdiscourses -suppressed or subordinated ways of speaking or perspectives.It is the job of the Foucauldian analyst to notice the gaps and shifts within and between discourses as routes into explicating those suppressed or hidden transcripts of power (Scott, 1990) or subaltern voices (Spivak, 1988).As Foucault's student and interpreter, philosopher Michel Pêcheux9 noted (in a recently re-translated paper): 'The object of discourse analysis, as it actually developed on the basis described, is precisely to explain and describe the construction and sociohistorical ordering of constellations of utterances' (Pêcheux, 2014, p. 95, fn6).We will return to Pêcheux's account later, as we have found this useful in helping to explicate and further evaluate our approach.
Foucault's famous claim, 'Where there is power, there is resistance ' (1981, p. 95), while amenable to multiple interpretations, nevertheless is fruitful in highlighting how taking seriously the relational character of power presupposes that resistance takes multiple forms: 'These points of resistance are present everywhere in the power network.Hence there is no single locus of great Refusal, no soul of revolt, source of all rebellions or pure law of the revolutionary.Instead there is a plurality of resistances.... the odd term in relations of power... distributed in irregular fashion... the swarm of points of resistance traverse social stratifications and individual unities.' (Foucault, 1981, p. 96).
As transindividual frameworks of meaning, therefore, discourses are both symbolic and material in effect (Parker, 2014;Burman, 1991;Burman et al., 1996).This approach helps to identify how participants are both subject to prevailing political and practice-oriented discourses (of educational "problems"), and also -at times -how they become the subject of -in the sense of reformulating -these in accounts of their everyday activities.
Having outlined our approach and the context in which the interviews that form the focus for our analysis were generated, we focus on four illustrative examples.These examples not only address the production and regulation of subjectivities via social and educational policy implementation, but also some glimpses of resistance to this as expressed by those who are its primary subjects.Or as Ball (2015: 310) puts it, as '…an agonism, a process of self-formation through engagement.'Our account moves between explication and analysis, in order to fulfil the two key aims motivating this analysis.That is, not only to, firstly, deepen understanding of the material presented in relation to its educational policy aspects, but also, secondly, to demonstrate and evaluate particular aspects of the discursive approach mobilised.Hence we also aim to illustrate how a Foucauldian-informed discursive approach engages with and offers educational policy-relevant readings of diverse kinds of material, including as framed by various local policy enactors (ranging from teaching and community practitioners to parents).As such, each example is situated in relation to particular debates and literature relevant to its topic.So we move from theory to empirical analysis, and back again.The examples we discuss below may be familiar, even routinely encountered, ways of speaking of students and their parents.To the extent that this is so, this works to support our wider aims of, first, showing how discursive analysis of everyday, assumed descriptions offers ways to explicate and interrogate the socio-political context they imply while, secondly, highlighting how such common resources combine with the specific and current case of welfare 'reform'/cuts to both inscribe and then intensify anew particular (usually vilified) subject positions for children and families from poor communities.
Four examples are interrogated, drawn from across the participant groups in our study.Firstly, from the school staff, how student 'bad behaviour' or 'going off the wall' becomes transformed into an understanding of the effects of hunger -with correspondingly different consequences for intervention.Secondly, the formulation used by school and community practitioners about services or amenities as 'gone' is analysed, as working to occlude and so naturalise cutbacks in state support and provision.While these practitioners struggle daily to alleviate poverty and deprivation, this indicates how they are nevertheless operating within a discursive field whose material environment and discursive resources work to reduce the social to the familial.Further, the de-legitimation of discourses around reduced service provision and material contexts gives rise to discourses that intensify the focus on parenting, as illustrated by our third example.In this, we move from the analysis of (the presence or absence of) specific words or phrases to broader discursive structuring of the accounts, focusing on how the topic of food (which has featured significantly in our first example) comes to be narrated by school and community practitioners within a particular, pedagogical, frame -with consequences for the moral-political positioning and evaluation of those (parents and families) who engage with such provision.These three indicative discursive tropes are discussed alongside a, fourth, example -this time arising from our corpus of parents'/carers' accounts of, and for, their circumstances.Here we, as researchers, were directly addressed by one particular parent interviewee, in terms of subject positions assumed, attributed and re-negotiated in their accounts to us.We interpret this specific example not merely as a request or even demand for 'validation' of her own perspectives (as some qualitative research assumes the function of interviewing to be), but also as setting in circulation an alternative discourse about herand, crucially, also other parents' and families' -competence and resourcefulness in dealing with adverse circumstances.
So we turn now to the examples, drawing from our corpus of interview material from structurally differently positioned participants interviewed for the 'bedroom tax' study, taken here as indicative of the impacts of welfare reform for children's schooling and their families.

'Off the Wall' or Hungry?
Interpretation-led descriptions have long been noted in educational research, giving rise to competing explanations for student (mis)behaviour (see Marks et al., 1995;Billington, 2006, Greenstein, 2015) which they both reflect and perform.The classroom, senior teachers and teaching support workers interviewed in our study offered as examples of the impacts of the 'bedroom tax' (alongside other welfare reforms) their difficulties in arriving at the 'correct' reason for 'bad behaviour' and disengagement at school.Their accounts overwhelmingly focused on children coming hungry and inadequately clothed to school, stressed, fatigued and therefore ill-prepared to engage in schoolwork. 10Examples included: We've had some young people that's come in and they haven't eaten or things like they're all off the wall they can't concentrate they can't settle down in their lessons and then that then you know ends up with the teacher ahm (.)I dunno the teacher having an issue with the young person the young person usually getting sent out and [this] usually has a trickle effect because if they get sent out of the class nine times out of ten they end up coming to me… (Teaching Support Worker) 'Off the wall' offers a non-technical description of apparently unacceptable behaviour that is itself ambiguous, conveying11 either eccentric, unconventional behaviour (which might even be positive in the sense of being 'creative') that extends to 'craziness'.Its informal character does some important work, as doubtless a more 'technical' description would sound not only more censorious but also imply some kind of diagnostic labelling -as in ADHD for the later reformulation 'hyper' (see below).Significantly, 'off the wall' also includes anger (see our discussion of 'defiant' later).
An interesting corollary of this ambiguity around hunger or being 'unsettled' (a further, somewhat anodyne, apparently value-free euphemism) is the way behaviour becomes framed as 'the teacher having an issue with the young person', which, significantly, positions the reaction of the teacher as the cause of the response.This interestingly sidesteps the question of the (mis)behaviour, which (through the 'teacher having an issue with') thereby implicitly threatens to position the teacher as responsible for eliciting as well as managing the 'young person''s behaviour.We might note the recurrence of (lack) of 'settling', a term apparently used to connote engagement in educational activity, but which of course invites other readings -in terms of (de)colonization that would render the role of schooling much more questionable (including whether education should be something one settles into or rather is unsettled by) (Tuck & Yang, 2012).The need for 'simmering down' (read here as becoming calmer, rather than the culinary sense of intensifying through gentle cooking) can be seen as a necessary supplement to the 'off the wall' narrative.The same support worker continued: 'usually after break when they had something to eat they kind of simmer down and they're not as I don't want to say off the wall but as you know hyper or they're a lot more they're able to concentrate a lot more now that they've simmered down'.
While its prevalence is perhaps overdetermined by the rise of behavioural therapeutic models as well as educational policy debates, the discourse of behaviour (including the desirability of 'settling') topicalises school management, rather than looking at wider socio-political context.It positions the child as a problem for the teacher and school, rather than as having problems or difficulties in relation to which the school could or should support them.It is part of a repertoire whose elements might include other subjects (teachers, other advisors, and perhaps other educational professionals including social workers or school psychologists) who have their own complex histories and power relationships with each other.Yet such discourses also lead to specific kinds of material actions (of censure, punishment, even exclusion).We might note that there may be covertly gendered and racialised features of this child-as-problem discourse, as well as classed (as particularly attending to boys and, in the UK context, to African-Caribbean boys particularly, see Coard, 1971; as also African-American boys in the US, Watts & Erevelles, 2004), to which even a child-centred discourse (of needs, developmental stages, or relational processes) might subscribe (Burman, 1996).
Such positionings are clearly not unique to neoliberal policies.Skrtic (1995) argues that the dominance of the functionalist worldview in the very early models of state schooling implemented the mutually reinforcing theories of organisational rationality and human pathology into practices, policies and theories of education.As a result, the problem of school failure was reframed as two interrelated problems -inefficient (non-rational) organisation and defective (pathological) students.This removed the problem of school failure from general educational discourse and compartmentalised it into two separate but mutually reinforcing discourses.The first discourse was in the developing field of educational administration, which, in the interest of maximising the efficacy of school organisations, was compelled to rationalise its practices and discourse according to the precepts of 'scientific management', an approach to administration designed to increase the efficacy of industrial mass production firms (Donaldson & Edelson, 2000).The second discourse on school failure was in the field of 'special education' that emerged as a means to remove and contain the most recalcitrant students in the interest of maintaining order in the rationalised school.
The move identified here from an explanation of 'bad behaviour' or disengagement with learning, each of which is too often framed under psycho-medical labels such as EBD (emotional behavioural difficulties) or ADHD (attention deficit hyperactivity disorder) to 'hunger' (via the trope of 'off the wall') therefore involves a major shift from a child or person-centred model concerned with intentions and actions in need of correction that are in danger of infecting others and disrupting the smooth functioning of school processes, to one of a subject whose actions arise from a condition (of hunger) or need over which they have no control, and which school staff feel an obligation to meet.
This shift opens up a space of resistance to the dominant discourse of neoliberal educational policy that seeks to promote what Masschelein & Simons (2005) call the 'entrepreneurial self' and which constructs success and failure as individual responsibility; with social participation figured as economic productivity.This has significant consequences.First, this changes the frame of the subject position of the (hungry rather than bad/disruptive/deviant) child, embedding children not only with a family or school context, but also in relation to wider social processes and policies.Second, it reframes the educational relationships, highlighting the role of care and concern rather than focusing merely on increasing measurable "achievements" (and it should be noted that this emotional labour extends the work of the teachers and teaching assistants very considerably).
What is noteworthy is how the two descriptions or explanations are acknowledged, in the teacher's account, as being coterminous or simultaneously available.Moreover, that her first interpretation was in terms of bad behaviour.Indeed this ambiguity was only resolved by the reported observation of the child's (re)action (of snatching food) or (in another example cited) of being uncharacteristically keen to go for an early lunch (prior to the time allotted to the child's year group).Discussing this specific case, the support worker reiterated: 'whether it's just him being defiant and just thinking I'm going to skip lesson and go or whether that is that he's genuinely hungry and going and you know I guess that needs we need to get to the bottom of that but I find that with him once he's had something to eat he simmers down.' In terms of the corresponding subject positions occupied by those reporting these incidents in our interviews, the school staff can be read as implicitly acknowledging the ease with which such misdemeanours could have been misclassified (in the above example, from 'defiant' to 'genuinely hungry') and so, presumably, giving rise to pathologising or punitive rather than philanthropic responses.Nevertheless a more covert performative effect of this narrative is the communication of a tacit understanding that many such other incidents could be misrecognised in these ways.

Interpreting Absences: 'They've All Gone'
A key (although not exclusive) feature of Foucauldian (and other, including psychoanalytic and deconstructionist) approaches to discourse analysis is identifying how what is said stands in relation to what is not said: that is, how every presence occludes another, or how what is spoken about necessarily silences other formulations (see also Billig, 1999;Burman, 2016a).While such absences cannot be known, since we cannot recover hidden meanings, and in any case meaning is a fraught -if not proscribed -feature of discourse analysis (Parker, 1992), it is possible to consider what positions and processes are generated from alternative readings, and their political consequences.As Foucault (1981) put it: There is no binary division to be made between what one says and what one does not say; we must try to determine the different ways of not saying such things, how those who can and those who cannot speak of them are distributed, which type of discourse is authorized, or which form of discretion is required in either case.There is not one but many silences, and they are an integral part of the strategies that underlie and permeate discourses.(p.27) In our treatment below we read for subjective orientation akin to Pêcheux's (2014) analysis of the cultural-politically sedimented understandings of the phrase on a gagnè![we've won!].One recurring trope or pattern we came to notice across the accounts of both the educational and community providers we interviewed was the way they referred to services as having 'gone'.This was their characterisation of resources or services that had been cut, or were no longer functioning.'Gone' is a passive description that does not name actors or causes.The focus is on effects, rather than processes or explanations, and -as a passive construction -it therefore does not implicate the speaker as having any particular role in having made it 'go' (or not).
This absence could be read relationally or interactionally, as arising from origins too 'obvious' (to the interlocutors) to mention; that is, as an appeal to mutually shared implicit knowledge.Nevertheless, this invites collusion in failing to name how the cutbacks have come about.Yet this reading implies a different affectivity.'Gone' is an evocative description, almost childlike in its simplicity and baldness-'all gone'.The childlike association suggests a sense of powerlessness or helplessness in relation to being subject to these measures, which may convey how people have experienced these cuts -as imposed irrespective of any involvement or consultation.The affect mobilised by 'all gone' invites a sense that no words can really capture the enormity of the effects and meanings of these deprivations.In this sense, the phrase 'all gone' reverberates like a lament, with grief so great that it cannot be put into words.
Nevertheless there are consequences for not putting these matters into discourse, including naming the absence of the presence of what it is that has gone, in this case key structures of neighbourhood belonging and support.Perhaps also the 'gone-ness', in its totalisation but also lack of linguistic/referential specificity (of what has gone, as well as how it came to go), could be understood as referring to how not only are some services totally cut but also still more are so functionally reduced as to exist practically in name only.That is, while some resources are still nominally available, their remit (whether eligibility of access, or opening hours etc.), access and provision have become much more heavily circumscribed (Fitzgerald et al., 2014).
The material significance of this framing was evident in the ways schools were documented as stepping up to fill in or compensate for missing services (ranging from direct resource provision, such as breakfast clubs and accounts with local shoe shops, to using Pupil Premium12 moneys to employ extra support and family liaison staff).Commendable (and even vital) as these measures may be, there is a danger that such practice on the part of schools succeeds in appearing to replace what has been missing, rather than providing a stopgap.For example, school staff reported referring to parents who could not buy food to foodbanks, yet these are crisis measures, rather than services, the demand for which and proliferation of in recent years is a noteworthy indicator of the chronic and escalating prevalence of poverty (Barnard, 2015).Moreover, as perhaps an even more striking indicator of this responsibilisation and activation imperative, staff reported that they were contributing resources -not only extra time (and stress in dealing with difficult problems caused by poverty) -but also buying materials (e.g.sanitary equipment for girls) out of their own pockets to give to pupils.
According to one reading, this implies a level of acceptance or lack of resistance.There appears to be no protest, nor any comment about this absence.Rather, the focus is on what can be done.While this, doubtless, also arises from the sense of urgency of responding to acute needs, nevertheless this absence -or overlooking of how and why these services and resources have disappeared -could be read as reflecting the 'success' of the activation discourse.As in Berlant's (2011) analysis of 'cruel optimism', the imperative to be positive and make things better generates responsibilised subject positions (Edwards et al., 2015) and colonises dissent (Henderson & Denny, 2015).This cruel optimism was also evident in the responses of community organisations, to which we will turn now as our third example.

Food as Pedagogy
Similar to the schools, community organisations have also responded to services having 'gone' by shifting their roles and activities within the community.Many organisations have made food distribution a priority by integrating it into other, often educational, activities such as cooking or gardening classes and groups.These activities were often framed not only as a response to the immediate material needs of the community, but also as educational opportunities: Doing the cooking: you're learning the cooking; you're learning hygiene; you're learning a lot of things.We're even getting young men setting the table properly and learning to eat with whatever and they're just learning certain social skills that you might not otherwise get.But now, it is actually a need now rather than something that we're doing for fun.(Youth worker, Voluntary sector organisation for young people, local resident) Note that 'need' trumps the more discretionary 'fun', yet as Walkerdine (1984Walkerdine ( /1998) ) highlighted, both partake of a (child-centred) pedagogy.The current version of this pedagogy transitions from positioning the children to the parents as 'learners': We tend to do meals that are cost effective for families, so we'll do something like rice with sweetcorn; tuna in it.So it's a very protein rich… We try showing that so they can go home and show the parents.But they say like they struggle 'cause the children don't want that.So we're doing a lot on trying to help the parents as well as the children.'cause obviously you can't just educate children, you've got to educate the parents.(Play worker, Voluntary sector organisation working with children, local resident since birth) Even social housing providers have come to position food support as a form of pedagogy for parents -with that pedagogy extending from practical (cooking, shopping) skills to social and community inclusion: We've got one programme of 'Healthy Eating / Cooking on a Budget' workshops that we're running … they are learning cookery skills; they're learning how to cook on a budget but also healthily; and they're learning lots of other tips as well about buying food, when to buy; where to buy -that kind of thing.They also have a Financial Inclusion Officer available.
[ …] Some of them are quite vulnerable.Some of them … feel socially excluded.They're coming together..., and also gaining confidence that is enabling them to go and help other people.… And then we're hoping that some community growing projects might emerge as well."(Director of Housing Services, Housing Association, born and raised in the area) In these examples, food provision activities by community organisations are framed as educational interventions rather than simply as charity projects that still leave the poor in the same state of dependence.Cooking on a budget and teaching children and parents to prefer high protein cheap food can be seen as part of the discursive promotion of the 'new thrift', the 'savvy' individual who can make more with less, as the ideal subject of austerity (Jensen, 2012).
Notwithstanding the value of such programmes to the actual survival of individuals and to their ability to draw on community support to mitigate to some extent the adversities of poverty and hunger, limitations of such approaches should be noted.As Henderson and Denny (2015) point out in relation to the discourse of resilience, the idea of positive interventions based on 'actionable knowledge' directed towards enhancing 'human capital' narrows the terrain of discourse and makes it '…more and more difficult to mount critiques of the realities of injustices which cannot be made visible via resilience's technologies defining what it is feasible to change: indicators and outcomes relating to the functioning of the person' (p.22).Clearly some youth and community projects do engage communities in political education and action that explore a wider social and structural context for poverty.For example, a youth worker interviewed quoted from a 14 year old young man's contribution to a project entitled Poverty Ends Now which provided opportunities for young people to develop their political understanding: [He said]'I think people are too quick to judge and stereotype, which will hold people back and cause less opportunities for people living in the area.'I believe some young people lose belief and give up, which may contribute to the area being financially deprived.It's all a cycle, which will not change unless people in power decide to do something about it.(Youth worker) However, those activities that focussed mainly on education relating to immediate needs may not have provided opportunities to explore wider or more structural change.Notions of resistance or challenge were present in some of the family members' discussions, which we will now explore as our final example.

'Tell Your Professor We are Good Mothers'
As various commentators on parenting and neoliberalism have pointed out, prevailing discourses of marketization have entered discourses of parenting, especially mothering.Current literature has highlighted parents' own subscription to entrepreneurialisation discourses (Edwards et al., 2015;Lupton, 2012;Thornton, 2011) as well as wider policy moves towards modes of governmentality through pedagogy (Pykett, 2012b).Recent UK Parliament proposals have focussed on educating (and regulating) parents rather than increasing investment in services (including education).While recent British research has attempted to explore and explain how and why members of poor communities appear to accept psychologised definitions of poverty/inequality (Shildrick & MacDonald, 2013), we end this article by discussing a discursive example we encountered indicating resistance.The indications of resistance included: (1) refusing an individualised model that positions particular individuals or families as subject to the welfare cuts, in favour of (2) claiming wider collective impacts.This last example also draws attention to the relational processes of this study, where we, as researchers, were directly addressed (and indeed challenged), in terms of subject positions assumed, attributed and re-negotiated in the interview.We interpret this example as not merely proving an audience by which this participant's perspective could be heard or warranted (as in the humanist model of qualitative research as 'giving voice'), but beyond this we interpret this performatively as working to set an alternative discourse in circulation about her, and others' like her, competence and resourcefulness in dealing with adverse circumstances.
The extract we analyse below was said by a woman who volunteered to be interviewed as a mother of school-aged children affected by the 'bedroom tax' towards the end of a (second) interview, conducted six months after the first and with the same interviewer. 13In considering how to read this comment, it is worth reiterating that our study was not explicitly about parenting, or mothering, but educational impacts of the 'bedroom tax' as a welfare cut to family income and wellbeing.Yet a subject position, and evaluation, of mother was inferred and interpellated (or hailed into being and identified with) through the research focus on children.Perhaps this is not surprising, given wider discourses around women and children that conflate their positions (Burman, 2008;in press), but also (as we have already indicated) in relation to the ways the 'bedroom tax' consolidates age, generation and gendered identities through its space allocations (Greenstein et al., 2016).
Several features are noteworthy about this participant's intervention.Firstly, it sets out key positions around knowledge and power; secondly, it addresses the prevailing models of poor (economically impoverished) mothers as bad mothers; and, thirdly, it comments on the process of knowledge production, as well as reception -including the institutional power relations within a research team.We will take each of these points in turn, exploring the forms of subjectification or subject positionings set in play, and their various relational and political effects.But first we offer some further material to indicate the immediate interpersonal context for this -in our view extraordinary14 -utterance.
Participant.We are good mothers (.) tell your professor we are good mothers (.) we are poor we live in council houses we have life skills (.) uff because we have suffered not because we are dummy …you're constantly battling they will reduce this they send this they will stop this eviction letters this and that bills coming left right centre.Just before the extract above, she said: '… these council houses women they are the most loving to their families it is the external environment which is causing because they want to change the family dynamic...' So in terms of subject positions, various are clearly identified, including professor and mothers, as well as the implied addressor and addressee ('you' and 'your'), and the less direct but present naming of the researcher/ed relation (via 'your professor').(A further positioning that remains ambiguous is whether the researcher's positioning is understood to include or exclude that of mother).Notably, the speaker names a collective subject ('we', not 'I') as being 'good mothers', that is sometimes generalised to include the inclusive 'you/your'.This is significant in particular as it, firstly, arises as a challenge to the usual, 'confessional' genre of one to one interviews (Freund, 2014) which thereby invites the presumption of a stable interiority to be excavated and scrutinised (Agnelli, 2015;Alldred & Gillies, 2002).An exclusive 'we' specifies mothers, or perhaps researched mothers, that does not include 'you' (the researchers).This implicitly topicalises the question of who comes under the policy/research gaze, highlighting how it is not all mothers, but rather poor, welfare-receiving mothers who are subject to scrutiny and evaluation.This also sets out the field of discourse as divided between 'good' mothers and the conversely implied figuration of bad mothers.
One key feature reflected here is the subjective investment for women as mothers in being seen as 'good'.This imports other -regulatory, disciplinary as well as self-disciplining -features ranging from the ways women's gendered identities are oriented around the perceived adequacy of their caring and relational capacities (Burman, 2012;Walkerdine & Lucey, 1989), to anxieties about children being taken away and put into institutional care if they are considered not to be good mothers.
Yet there are other considerations at play in the subject positions that are both present and absent.An absolute binary is implied, between (the topicalised) 'good' vs. (the implied) not good/bad.What is absent is the discourse of the 'good enough' mother, that was so prevalent in post-World War Two parenting advice discourse (initiated by the psychoanalyst Donald Winnicott, e.g. Winnicott, 1965), an absence that marks the high stakes currently surrounding parenting as well as school achievement (Blum & Fenton, 2016).In terms of the number and gendering of the subject positions topicalised, it is worth noting that there is an implied elision between parents and mothers in her account.The question of the role or responsibility of fathers does not arise.This appears to reflect how the dominant discourses still align responsibility for children with women as mothers, rather than men as fathers.Yet in fact fathers involved with shared custodial commitments emerged as a key party negatively impacted in our study, while foster parents and grandparents also emerged as negatively affected (see Bragg et al., 2015).
In terms of relational dynamics, 'tell your professor we are good mothers', leaves unspecified, and perhaps fruitfully open, the position of the interlocutor ('you'/'your'), other than someone relaying a message.This suggests some ambiguity of alignment, perhaps either to subscribe to conventions of politeness that discourage direct confrontation (was the interviewer like the unspecified others who 'want to change the family dynamic'?)or else a reference to academic hierarchies (from researcher to professor) that pass the power and agency of evaluation back to outside the interview.
Perhaps what is most noteworthy are the subject positions that are refused or warded off by this utterance.There are various refusals of identification or (after Althusser, 1971) interpellation.
Firstly, there is an explicit refusal to be positioned as not being not good mothers.But, second, there is also a refusal to subscribe to the dominant discourse that renders less available the position of identifying as 'poor' (Shildrick & MacDonald, 2013), in part because it has been invested (by prevailing political discourse) with a corresponding affective load of shame (Youngmie, 2012;Chase & Walker, 2012).Third, this speaker is also refusing to identify as the canonical uncertain, doubtful, confessional mother/parent (pace Geinger et al., 2014), perhaps since she assesses that in this context expressing anxieties would warrant further unwanted scrutiny/intervention and pathologisation of 'family dynamics' (in particular of losing custody of children, which is a major underlying anxiety for all poor families, see also Donzelot et al.,1979).Finally, and most importantly, there is a refusal of the singular voice 'I' and, instead, use of the collective first person voice, 'we'.This is very important in resisting individualisation as a dividing practice, as also resisting the calling into being of privatised interiority (as a key example of the incitement to psychologisation discussed above).
It is worth reflecting further on the relational refusals or resistant agonisms at play in this utterance.Far from displaying the mode of resistance of prevarication or obfuscation that has recently been heralded as almost the only available form in some circumscribed contexts (see Wagner's, 2012 overview;Hoy, 2005), by this statement this participant claims a voice that not only expresses an active, positive maternal subjectivity but also performs a demand.In fact there are three demands: firstly, to be heard; secondly, to make a collective/political claim on practices of representation, and beyond this, third, to make an ethical demand of the interviewer that she should use her authority, and the wider institutional relations of knowledge production that research participates in, to challenge the injustice of the ways the state stigmatises and blames poor people for their poverty (which in fact is precisely the ethical demand that prompted our study in the first place). 15e take this statement therefore as an example of the performativity of discourse that links power and knowledge.This analysis therefore not only addresses the current debates on the intensification of regulation of mothering under neoliberal, active citizenship regimes.It also highlights the importance of seeing speech as action: how the act of speaking can make a claim to harness authority in multiple senses: attending to the event of the interview as an act of speaking to a researcher that is connected to the act of reporting research, that in turn can be appealed to as an ethical, corrective intervention to counter an injustice.'Tell your professor we are good mothers' is a statement that makes a moral-political claim on what research is and does; and what we as researchers should do.It makes an explicit discursive transition from the micropolitics of the dyadic interview to institutional power relations outside, while claiming to speak in a collective voice, as one political constituency addressing another.

Discussion
We have presented these four examples from this recent educational research project on The Impacts of the Bedroom Tax on Children and Their Education to indicate the ways a Foucauldian-informed discursive approach can contribute and extend policy analysis, in particular to document and highlight the ways current neoliberal policies position poor children and families, including professionals and practitioners who work with them.We see this model of analysis as helping address the current urgent need to ward off deterministic and fatalistic models that threaten to overstate the reach or remit of neoliberal discourse into subjectivities and subjectifications.Instead, the examples we have discussed suggest how a discursive analysis can help identify the forms of subjectivities created by these social/educational policies (of welfare reform), their consequences, and the forging of some spaces for negotiation and resistance.
Like Ball (2005), our analysis has worked to 'chip away at bits of the social, always looking for joins and patterns but equally aware of fractures and discontinuities' (p.2).In the first place, our example of alternative framings of being 'off the wall' vs. 'hungry' indicates how professionals could navigate and negotiate spaces of resistance to a dominant discourse of individual pathology as an explanation for misbehaviour and instead embed child within a social-material context.This example not only illustrated how the interactional context of the interview could provide an arena for the formulation and perhaps reformulation of alternative discourses, but also how interpretations shape apparently equivalent constructions of reality to produce quite different material -and policyrelevant -responses.In relation to this, the second example addressed the question of theorising an absence -of services that have 'all gone', suggesting that such formulations risk naturalising a sociopolitical context of deprivation, with the domain of the social figured as out of individual or popular control.The emerging focus on food and feeding, discussed in our third example, can be related to cruel optimism, responsibilisation, and other responses that focus on changing the individual rather than social (fixing the small problem, not the larger ones) (Henderson & Denny, 2015).
The final example discussed is, we suggest, noteworthy by virtue of its refusal of prevailing configurations of both mothering and of the subject position of research participant.Beyond this, it offers a sorely needed reassertion or claim of collective subjectivity and solidarity.From this, following Pêcheux (2014), the interview can be seen as an event for the elaboration of discourse, rather than merely its rehearsal or repetition.Indeed Foucault (1981) himself pointed out that '[t]here is not, on the one side, a discourse of power, and opposite it, another discourse that runs counter to it.Discourses are tactical elements or blocks operating in the field of force relations' (pp.101-2).
Together, what these examples indicate is how policy discourse, in this case a housingrelated welfare reform that has widespread impacts on family and children's wellbeing and their education (Winter et al., 2016), is both reproduced and transformed by the various subjects who are both subjectified by, but also become subjects of, these discourses.We have claimed that such policies are educational in a double sense: highlighting the educational impacts of social and welfare changes, but also via the ways neoliberal state policies mobilise pedagogical strategies to incite and regulate subjectivity.Yet via this Foucauldian/ Pêcheuxian discourse analysis, the story here is not only one of co-option or ineluctable subscription to these dominant framings.Rather, our examples highlight considerable and considerably adept manoeuvring within, destabilisation and even contestation of, prevailing discursive constellations exemplifying Foucault's (1981) claim that: 'Discourse transmits and produces power; it reinforces it, but also undermines and exposes it, renders it fragile and makes it possible to thwart it' (p.100).
We suggest that such analysis not only illuminates better the current picture around the impacts of educational and social policies, but, as Pêcheux (2014) suggested, may also link to activism/intervention: …a discourse, by its very existence, marks the possibility of a de-structuring-re-structuring of these networks and trajectories.Any discourse is the potential index of a movement within the sociohistorical filiations of identification inasmuch as it constitutes, at the same time, an effect of those filiations and the work (more or less conscious, deliberate, and constructed or not, but all the same traversed by unconscious determinations) of displacement within their space (Pêcheux, 2014, p. 94).
A key conclusion Pêcheux draws from this is that by attending to (what would now be called) the performativity of discourses (Ball & Olmeda, 2013) not only enables attention to their movement rather than only structural determination, but also to how identification with societal prescriptions and images is never total or complete.Thus we see our final example as indicating the possibility of engagement and resistance in and via educational research; that is, what Pêcheux (2014) describes as 'moments of interpretation as acts that emerge in the form of positions taken and recognised as such; that is, as effects of identifications that are acknowledged and not denied' (p.64).Clearly our positions as not only researchers but also as moral-political (including aged, gendered, 'raced' and classed) subjects was not only being interrogated, but also interpellated, and called to action.As Pêcheux points out, this is also necessary: 'Before endless interpretations in which the interpreter acts as an absolute point without any other real, it is for me a matter of ethics and politics: a question of responsibility' (p.64).
Research not only documents discourse, it also produces it.Policy is formulated and also enacted and re-enacted through practices of subjectification.We have analysed examples of policy discourse as (re)formulated by differently positioned stakeholders (teachers, community social service providers, parents) within our educational policy-related research on a particular welfare 'reform'/cut (the 'bedroom tax'), offering a specific Foucauldian reading that also intervenes in prevailing Foucauldian approaches which emphasise governmentality, to attend instead to practices of resistance as well as regulation.The ethical-political project to document and enact such practice remains all too urgent.As Pêcheux's analysis highlights, what we do as generators and interpreters of this material puts into stark focus both our own understandings, as researchers, of the relations between policy and discourse, but also how we respond to the claims made of us as producers of such discourse to use this authority to change those policy discourses.