Disciplinary and pastoral power, food and poverty in late-modernity

Using a Foucauldian perspective, we explicate the systems of power which shape the lives of women in or at risk of ‘food poverty’. We develop a theoretical framework of power for analyses of contemporary food poverty, which we apply to data from focus groups with women on low incomes in two cities in the north of England. Our data underlines the repressive power of the state as well as the broader chronicity of state surveillance. We argue that, while disciplinary and pastoral power may characterise the majority of food banks, alternative logics of mutual aid are evident within some food aid providers. We underline the power of governmental discourse in constituting gendered subjectivities and find that the most potent form of coercion is derived from self-regulation. The article closes by exploring possibilities for praxis via discursive resistance.


Introduction
In 2010, poverty was endemic in the United Kingdom (UK) but 'food poverty', denoting a situation in which household income is too low to enable access to adequate food, did not receive much attention. Similarly, food banks were not considered a routine response to poverty. In 2020, writing during the Coronavirus pandemic, food banks have been incorporated as vital frontline responses to poverty; food bank staff have been re-categorised as 'key workers' and food banks have received unprecedented levels of financial donations.
This article explores the institutional, ideological, and discursive structures that allow food poverty and food banks to exist today (relatively) unchallenged. It draws upon post-structuralist theory, particularly Foucault, to consider the systems of power which shape the lives of those in or at risk of food poverty. It does this mindful of broader scrutiny of the nature of latemodernity in which decentred power is manifest amid heightened complexity and surveillance in a post-Fordist, globally-networked world. The paper looks beyond food banks to consider the mechanics of power throughout the whole social body, including individual biographical narratives which may be continuously revised in the context of multiple choices in late-modernity (Giddens, 1991: 4-5). Only a minority of those experiencing food poverty -hunger -access food banks and therefore this broader lens allows narratives of those who do not access this particular form of food charity to influence debates on poverty.
We begin by appraising literature on UK food poverty and food banks, examining the extent to which a critical perspective is adopted. We consider the relative absence of gender in discussions of contemporary food poverty, despite long-standing feminist social welfare literature, and note the importance of taking a critical approach to food, poverty and gender. We explain our theoretical framework, drawing upon Foucault to explicate contemporary manifestations of power, before describing our methodology. The latter half of the article considers our empirical data in light of our theoretical framework and explores implications for praxis.

A critical review of UK food poverty research
Since 2010, research on food banks and food poverty has grown into a rapidly expanding inter-disciplinary field. 'Food poverty' has been successfully constituted as an object of study across disciplines, predominantly shaped by an empiricist paradigm (e.g. Loopstra et al., 2015;Prayogo et al. 2018;Sosenko et al., 2019). Poverty is presented as an object of scientific study through a focus on -often behavioural -'risk' factors. The key concerns are precisely defining (O'Connor et al., 2016) and measuring (Smith et al., 2018) food poverty; critique is side-lined in favour of 'scientific' understanding. Policy impact is presented as a corollary of a more accurate characterisation of this new phenomenon.
Critical contributions to UK 'evidence' on food poverty, which challenge dominant assumptions or reflect upon the effects of a growing regime of experts, have been fewer in number. Important exceptions include Garthwaite's (2016a) ethnographic observations and interviews inside food banks. These document how shame among service users does not necessarily arise from their experience of the food bank, but from a wider stigmatising culture of 'Othering' (Lister, 2004;Jo, 2013) which constructs poverty as personal failure. Informed by ethnographic research with service users and providers, Cloke, Williams and May Williams et al., 2016;May et al., 2019;May et al., 2020) critically assess food banking in the UK. They highlight the moral ambiguity of food banking and the potential for politicisation of food bank volunteers; the growing convergence of bureaucratic practices of benefits officials and food bank organisations, and the production of moral distance that characterise both (May et al., 2019); and the notion of 'scarcity' that has become embedded at the level of common sense (May et al., 2020). Strong (2019) takes a similarly critical approach. Drawing upon ethnography inside a food bank in South Wales, and using Foucauldian theory, he argues that food banks exercise a power over life that seeks to transform hunger into a technical object through which it can be known and acted upon by food banks. In this analysis, Strong draws attention to the limits of this biopolitical and disciplinary system, emphasising the 'everyday power of living' that undermines techniques of power exercised over individuals.

Gender, food, poverty
Surprisingly, given the considerable literature on gendered aspects of the welfare state and the informal caring work performed by women (Lewis, 1983;Pascall, 1986;Glendinning and Miller 1987, among many), gender has been relatively absent from research on contemporary food poverty. Where it has been addressed, it has largely been from a positivist standpoint. Using quantitative data, researchers have shown that women are at higher 'risk' of food poverty than men (Loopstra et al., 2019) and that 'food poor' women have higher body weight than those who are not food poor (Yang et al., 2018). Echoing long-standing sociological literature on women's experiences of poverty, recent qualitative research (O'Connell et al., 2019) has illustrated that food is shared unequally within households where resources are constrained (Charles and Kerr, 1987).
In the UK, austerity-driven welfare reform (as set out in the Welfare Reform Act 2012) has had a gendered impact on familial relations, with financial, social and economic sacrifices disproportionately ascribed to women (Women's Budget Group, 2017). However, the gendered effects of this political economy have not been interrogated in relation to contemporary UK food poverty. Drawing upon qualitative research conducted in the 1990s in Canada, Power (2005) uses a governmentality framework to theorise the experiences of lone mothers living on social security, highlighting the salience of this approach in unpicking systems of power which shape lived experiences. She identifies two ways in which lone mothers may be constructed and disciplined as Other: as 'welfare bums' who are not in the labour market; and as 'flawed consumers' without the financial resources to participate in a consumer society. Informed by Power's study, we adopt a critical approach to elucidate the interplay of gender, poverty and food in contemporary society.

A theoretical framework of power for analyses of food poverty
We develop a Foucauldian reading of power -biopower in Foucault's lexicon -to explicate the micro-web of power which permeates the social body and may define lived experiences of poverty, including, but not restricted to, food poverty and food aid. Biopower is evident in the repressive power of the state, but it also persists in a subtle and continuous way, embedded within everyday practices. We show how techniques of power develop to make bodies docile so they can be, 'subjected, used, transformed, and improved' (Foucault 1977: 136) and illustrate how subjects of power are complicit in this process.

Developments in the exercise of power
Biopower, or power over life, is power employed or sponsored by the state to control individuals and populations. Techniques of power include repressive and constructive approaches. The former are exercised via disciplinary power, a form of power operating through processes of discipline and correction. This is exemplified by the punitive turn in social security policy, evident in many countries (including the UK), which sees poverty effectively criminalised by workfare approaches (Fletcher and Wright, 2017) and social security policy increasingly reliant upon deterrence and sanctions to modify behaviour (Blackmore, 2001;Wacquant, 2009). Reflecting on this punitive turn in social security, Wacquant (2009: 43) argues that a new type of neoliberal political regime has emerged, the 'centaur state', which is 'guided by a neoliberal head mounted on an authoritarian body'. The 'centaur state' retains strategies of consent towards corporations and upper classes but is authoritarian and coercive towards those experiencing poverty. Wacquant's portrayal of a centralised state devising and administering policy downstream, while reflective of the present moment, has been criticised for paying insufficient attention to the multi-layered web of power encompassing repressive and constructive approaches, effectuated via 'multiplex relations among government, business, nongovernmental organisations and hybrid organisationsal forms' (Marwell, 2016(Marwell, : 1095. Fundamental to Foucault's genealogy of power, tracing its evolution from the nineteenth to the twentieth century, is this shift from disciplinary power to a new architecture of power: governmentality. The latter is not only the domain of government but embedded throughout the social body (Foucault, 1990). 'Systems of subjection' (Foucault, 1977: 26) operate at the level of knowledge or discourse and/or through institutions, such as the welfare state, to coerce, correct and regulate bodies through the exercise of a seemingly limitless visual power, or surveillance: panopticism (Foucault, 1977: 201). An example of this form of power is evident in the British social security system. According to Fletcher and Wright (2017: 228), Universal Jobmatch (a Government online vacancy system which ran from 2012 to 2018 and was mandatory for most jobseekers) operated 'as a modern day panoptican with a disciplinary gaze that ensures self-administered surveillance and doubles as an online evidence-maker for sanctioning'. May et al. (2019) highlight the growing convergence of the bureaucratic practices of benefits officials and food bank organisations. However the extent to which food aid, like the social security system, coerces and regulates bodies through surveillance remains unclear.

Power and the self
The conduct of citizens is not only regulated by governmental activities of collecting and calculating data on population characteristics but also by individuals who engage in self-government: The subject constitutes himself [sic] in an active fashion, by the practices of the self, these practices are not something that the individual invents himself. They are patterns that he finds in his culture and which are proposed, suggested and imposed on him by his culture, his society and his social group. (Foucault, 1988: 11) The 'patterns' to be emulated within today's historical context are informed by the priorities of the economic market. Achievement of the ethical self -the autonomous, competitive, free individual, fulfilled by economic activity and engagement in civil society -'homo economicus' (Foucault, 2008) -is contingent upon application of certain technical practices of the self. Nevertheless, the government of populations does not rely upon certainties, there is a con-tinuous reciprocity between aggregate and individual actions and between the interactions of individuals. In asserting the ethical self via technical practices of the self -work, independence, activity -an individual demarcates themselves from 'Others' (Lister, 2004) who fail to regulate themselves according to the norms and requirements of the powerful. These Others may be stigmatised not only by those who successfully regulate themselves according to social norms but by citizens who themselves may be at risk of stigma because of their poverty (Lister, 2004). Garthwaite (2016b) found this practice of 'Othering' was evident across food banks in the North East and tended to be attached to those with substance misuse problems, 'immigrants and foreigners', and homeless people who used the food bank (Garthwaite, 2016b: 285).

Mediators: Pastoral power
Foucault has been criticised for his lack of attention to agency and neglect of 'how people respond to the external discourses and strategies that attempt to discipline them' (Lupton, 1997: 103). The posthumous re-publication of Foucault's lecture notes from the College de France go some way to addressing such criticism. The lectures detail the 'relational practices and technologies through which obedient and self-governing subjectivities are constituted' (Waring and Latif, 2018Latif, : 1070, outlining the role of 'pastoral power' or pastorship -an individualising power which requires detailed knowledge of the mental and physical attributes of its subjects -to governance of individuals. The concept of pastorship originates in early Christianity, where its characteristics included looking after every individual for the duration of their life; and exercising the need to know people's minds, souls and details of their actions (Waring and Latif, 2018). Although the importance of Christian pastorship diminished over time, the contemporary state may continue as a site of pastoral power. In this new configuration, the officials of pastoral power are disseminated throughout the 'whole social body', finding support in a 'multitude of institutions' including the family, social security, employers (Foucault, 1982) -and food banks.

Methodology
Data were collected in Bradford and York between 2016 and 2019. These north of England research sites differ in size and demography. Bradford, a metropolitan area with a population of over half a million (Office for National Statistics, 2019), has the largest proportion of people of Pakistani ethnic origin of any local authority in England (20.4%). Bradford is the 19th most deprived local authority in England, as measured by the Index of Multiple Deprivation (Gill, 2015). York, situated in the Yorkshire and Humber region, has a popu-lation of 210,000 (Office for National Statistics, 2019). It has a lower percentage of workless households and higher than average earnings compared with other parts of Yorkshire and Humber (Office for National Statistics, 2019). In the 2011 Census, 94% of York residents identified themselves as 'White' (Office for National Statistics, 2011). Nevertheless, persistent poverty and inequality also characterise York: in 2017-18, over 4,000 people in the city used a Trussell Trust food bank (Power, 2019).
Semi-structured focus groups were conducted with women in Bradford in 2016 (n=16) and York (n=19) in 2019. A women only sample was motivated by our intention to explore the potentially gendered interplay of food, poverty and power. In collaboration with community groups in both cities, women living in low income households were recruited to participate in focus groups lasting between one and two hours, held in a familiar location. The focus groups were moderated by the first author. The opening stage of the focus groups was conducted as an interview within a group. The moderator began by concentrating on a single participant, and subsequently requested group members to respond. This approach aimed to involve all participants in the group and encourage each participant to give a meaningful response (Morgan, 1997). As the focus group progressed, the researcher adopted a less direct role, maintaining the flow of the discussion and enabling group members to participate.
One focus group (Bradford) included participants with varying levels of English language ability. In this group, some participants (n=4) were bilingual (Urdu and English) and others spoke only Urdu (n=3). Because of this, the focus group was conducted as two smaller conversation groups within the larger group, with Urdu speakers spoken with separately via a translator.
The topic guide was produced collaboratively with community groups in Bradford and York. It was constructed to explore lived experiences of food in contexts of low income. Transcripts were coded and analysed thematically to elicit common themes related to the research question: 'how is power exercised in the lives of women in/at risk of food poverty?' The sample included 35 women on low incomes; all had children. There was some variation within the sample according to socioeconomic status, ranging from severe destitution to perception of financial security, despite a low income. There was no significant difference across the York and Bradford samples in socioeconomic status. In the York focus groups, 11 participants were in receipt of social security as their main source of income; the remaining eight described themselves as earning a low income but financially secure. Nine of the 19 women were lone parents. In the Bradford focus groups, a minority of participants lived in households where the only source of income was social security (n=3); others lived in households in which all adults were in junior managerial, administrative or professional employment (n=3). The majority of participants (n=13) lived in a household where at least one adult was in paid employment. In light of the focus group methodology and the stigma associated with food banks, participants were not asked individually or systemically whether they had used a food bank. However, in conversations surrounding food poverty, three participants in the Bradford focus groups described accessing some form of food aid, two of whom had used a food bank. Thirteen participants in the York focus groups had used food aid -possibly reflecting the time difference between the fieldwork in the sites and the development of informal food aid in these intervening years -seven of these participants had accessed a food bank.
Ethical consent was obtained from a university ethics committee. Given the vulnerability of some participants and the sensitive nature of the topic, ethical considerations were prominent in the design and conduct of the focus groups. The moderator asked participants about their personal experiences, however the line of questioning was discontinued if the participant appeared distressed. The moderator was also conscious of her position of power in her relationship with participants, in terms of academic knowledge and her role in setting the agenda of the group. The moderator attempted to address this imbalance by foregrounding the right of participants to withdraw at any time and providing participants with scope to determine the direction of the conversation. Participants were provided with full information about the study before agreeing to take part and informed consent was obtained before each focus group.

Results
We apply the theoretical framework outlined above to our empirical data to explicate the multi-layered manifestations of power in the lives of women in and at risk of food poverty. We start by considering the disciplinary power of the state, particularly in the administration of social security, and the chronicity of state surveillance. We assess the exercise of disciplinary and pastoral power within food banks, while noting possibilities for mutual aid in other forms of food aid. Finally, we address the technical practices of the self performed by participants to achieve the ethical self and distinguish themselves from the food poor Other.

Disciplinary state
The most explicit and comprehensive form of power exercised upon participants was that of the state. Lived experience was modified and behaviour manipulated by state policies and institutions. This coercive strand in state policy was directed with greatest fervour towards those in the most severe poverty, configuring in the place of a welfare state a disciplinary state (Jones and Novak, 2001). Conditionality, inherent to social security, intimately shaped the lives of women on low/no income. Daily activities oriented around the obligation to apply for a sufficient number of jobs per week while simultaneously managing a very small household budget and caring for children: Danielle: Well, if you're signing on, if you forget your book, or if you forget to put down your work. . . Jade: You're meant to search. Danielle: Searching for a job, three, four a week at a time. Jade: And you'll have no money, it will stop and you're just stuck with it. Bradford The bureaucratic and apparently inflexible system was also, at the same time, unpredictable and unreliable. 'Benefit sanctions' were portrayed as unwarranted, arbitrary and punitive, forcing claimants into destitution and eroding choice: Gail: If they sanction you, they take away your money. Danielle: Yeah, they do stop it. Moderator: So is that why people are going to food banks? Danielle: It is getting a lot worse from what it were, it is a lot worse. Bradford Universal Credit was described (by York participants) as 'not working': the digital claims process was complicated, impersonal, demeaning and untransparent. In line with the work of Cheetham et al. (2019), participants described how the Universal Credit administration process was poor at responding to queries in a timely or personal way and how system errors were exacerbated by 'moral distancing' (May et al. 2019) inherent to the system -for instance, online journals that claimants were required to complete appeared to be responded to inconsistently by different members of staff, despite the pretence of a personal service. Sophie, a lone parent of two children, described the long-term financial impact of the apparently inflexible system: When my partner moved out they expected me to live on about £400 a month taking into account that his wage was supposed to be in that month. But it didn't work out like that and they basically turned round and said 'tough that's how it works', that's pretty much their words. I was in minus, still in minus now, trying to play catch up with the companies, and paying off things, paying off my rent -I think I've got until March to pay it, two months' worth of rent, as well as bills that build up.

Sophie, York
The ability of individuals to manage the -punitively low -income received was monitored through a system of surveillance, which stretched beyond government to the economy, linking Jobcentre Plus to utility companies. For instance, failure to pay utility bills could result in a deduction of income from social security: 'I didn't pay mine [water bills] for a year but now it is coming out of my benefits. You get reductions off your benefits, which means you are lower.' Jade, Bradford

Chronicity of state surveillance
The absence of a direct financial relationship between other state institutions, such as schools and hospitals, and participants mitigated the immediate disciplinary power of the state. However, the power of the state continued in a subtle manner, embedded within everyday practices. Schools and hospitals were important sites of surveillance and potential channels through which the state could monitor and shape the private sphere. State institutions worked in partnership to intervene in cases of child food poverty, as described by Fiona who was both a provider of services, as a part-time nurse, and living on a low income: You do tend to have social services and stuff involved. A lot of the families where we think there are issues, social services are there. A lot of the time, it is better if there is a younger child who is under a health visitor because then we can get help through to them. Because sometimes the school pick it up as well, the children are not getting the meals that they need.

Fiona, Bradford
Financial relationships could exist with schools through obligated payments for school trips, meals, school uniforms, and classroom supplies. Participants were keenly aware of what they saw as their failure as parents -specifically as mothers -in reneging on these financial obligations and potential condemnation from the wider community of parents: The problem is that all these contributions get labelled as 'voluntary contributions', so you go on to the payment panel to deal with it and then you find out that the voluntary contribution is like seven to eight quid, when you get messages like this sent out saying stuff like, 'If we don't get enough contributions then the trip will be cancelled.' It just puts so much pressure on.

Food aid: Pastoral power/mutual aid
Power was exercised within and in relation to the food bank in direct and via insidious, indirect, ways. Most explicitly, receipt of a food parcel and associated support from the food bank -care, advice and sign-posting -was, in the majority of cases, dependent upon referral from a third-party agent. The criteria according to which an individual was deserving or undeserving of referral to a food bank was ambiguous and appeared to be tied to ideological constructs of the deserving -entrepreneurial, active, hard-working poor, and the undeserving poor (constructs discussed in respect of food aid elsewhere - Garthwaite, 2016a;May et al., 2019b;Power et al., 2020).
A series of bureaucratic, supposedly inflexible rules further governed access to the food bank once a voucher had been secured. Service users were limited to the receipt of three rounds of food parcels in six months. On presentation of the food bank voucher (itself disclosing extensive personal information, including the name of the service user, household composition, gender, ethnic group, age and reason for food bank use), the service user was required to explain to a member of staff why the food parcel was required before receiving food. In this way, the food bank emulated a key characteristic of both state social security bureaucracy and nineteenth century pastorship: 'it exercised the need to know people's minds, souls, and details of their actions' (Nettleton, 1997: 211).
Within the food bank, the service user was processed through a system which disassociated acts of care and welcome from receipt of the food parcel, allowing those distributing food to remain emotionally disconnected from service users: Now you go upstairs and talk to them so they get all your details, then you've got to take your voucher, go outside, around the back . . . So you go upstairs and they stamp your voucher and then you take it outside and around the back. This is the plan, you've got the thing there, you get your food and that's it, you're left on your own.

Jade, Bradford
Nevertheless, the 'more than food' identity of food banks, involving not only care but also sign-posting, budgeting courses and holiday hunger programmes alongside food parcels, endorsed a more holistic conception of the person. Food banks intended to support not only the physical self, through the provision of food, but also the intellectual or spiritual self. Self-improvement was an integral part of receiving a food parcel. The system of sign-posting and onward referral, accompanied by increasingly robust relationships between third party referral agents, including Job Centre Plus, schools and GPs, and local food banks, fostered the chronicity of the supervision made mandatory for service users.
Conversations with women in York about experiences of food charity, contrasted with those among women in Bradford. This may have been partly attributable to the time difference (three years) between the fieldwork in Bradford and York, a period which witnessed not only the steep growth of Trussell Trust and independent food banks, but also the development of other food aid providers, especially community food hubs and informal food banks. As a consequence, the food support network drawn upon by women in York was broader and more diverse. While many of the York participants used or had used formal food banks -albeit often only as a last resort -it was more common to visit community food hubs, involving communal meals, and community cafes, many of which operated informal food banks (free food was available within the building for anyone to take). The food and community/sociability available at some of these food aid providers was highly valued and considered essential to food security: Without them we'd seriously struggle. Like for lots of other people, they've been a huge rescue to us. Especially with that month where I couldn't have fed my children, if it wasn't for these food places. I always put money in, even if it's just a tiny amount, just to give that back. And even now I still need to use it.

Annie, York
Unlike in Bradford, where the idea of sacrifice was a notion only voiced by food aid providers describing their sacrifices in volunteering at food banks (see Power et al., 2017), in York, sacrifice was applicable to service providers and service users. Service users contributed in food or cash to support food aid (although not food banks) where possible. In stark contrast to 'morally distanced' and hierarchical interactions inside food banks there was, thus, evidence of mutual aid within some independent food aid providers and community food hubs. Food was shared by those providing and using the services; indeed, there was often little distinction between the two: the community serving and seeking food was interdependent.

Self-regulation and the Other
Notwithstanding the powerful systems and processes of coercion -the repressive and constructive power of the state, and the interplay of disciplinary and pastoral power in food banks -applied to participants, the arenas in which power was most effectively instrumentalised were those of the community and the self. The community was the site of a gendered form of surveillance, concerning the ability of mothers to care for their children according to certain (public health) standards. Participants described adopting surveillance roles surrounding potential child food poverty: I knew that she were embarrassed to say that she were a bit skint [no money]. Whatever we made, sometimes I would dish some up for her little boy and take it down so. 'Cos sometimes I used to wonder if he was getting enough.

Emily, Bradford
In this way, some participants entered into an inverted form of 'sousveillance' (Mann et al., 2003), surveying not state institutions but each other, and intervening to avert or correct deviant behaviour through social sanctions. Of particular interest to those conducting such surveillance was the financial affairs of neighbours and friends. While this could precipitate benign -albeit unsolicited -interventions to supposedly protect the welfare of children, it could also take a more malign form. Resentment against members of the community who were perceived to be receiving more favourable treatment could spill over into vindictiveness when mistrust was directed towards individuals (Manji, 2016). Social vindictiveness towards identifiable members of the community could be highly racialised -aimed at migrant neighbours and 'Asians': It is like my next-door neighbour, they're from -I can't remember where it is. They both work but they both claim [social security] as well which is quite annoying when there is me and my husband and he works every hour God sends to get money to bring home just to live off and next door they have got all this money, they have just got a brand new 60-inch telly, a U-shaped sofa.

Gemma, Bradford
Jade: Yeah, you can't go down to Asda and buy fresh burgers. They've now got up to three aisles, halal food. Don't get fresh burgers, you can get mince but they only put 16 packets of mince out in the morning and then 16 in the afternoon, if they have all gone there is no more mince.
Moderator: So why are they doing it? Jade: Because the Asians are complaining that our meat is next to their halal meat.

Bradford
The most potent channel through which control of the individual operated was the beliefs, behaviour and discourses of participants themselves. Feelings of shame in respect of poverty were common themes. Shame was most explicit in discussions around accessing food banks; in this context, shame was co-constructed through the convergence of an individual's internal sense of inadequacy and externally imposed disapproval for failing to satisfy societal expectations of economic self-reliance: I have degrees and I have qualifications; I have all this stuff and I can't get into jobs that I need to get into. I feel like I'm letting my husband down because I'm not really earning anything at work.

Dawn, York
The most widely adopted method to avert shame involved attempts to align themselves with dominant discourses, in particular the 'culture of poverty' (Lister, 2004), and define themselves in opposition to the demonised Other. Individuals who were apparently unashamed of accessing food aid were subject to disparagement. Openness about food poverty was itself assumed to reflect an absence of need and these women were considered to be 'playing the system': There is one kid that we know at school and the mum brags that she goes there to get things, "I don't go food shopping because I just go to food bank".

Emily, Bradford
This discourse was common to those who were not currently living through food poverty as well as to those who were -Sabira described episodes of severe food poverty, however when discussing food banks (which she avoided) she aligned herself with societal discourses individualising poverty:

Moderator:
Do you think it is a universal thing, people feeling shame and pride and not wanting to go to a food bank? Sabira: No, some people go because they want free stuff, people always want free stuff. Bradford Participants who expressed anxieties around food insufficiency but had not personally experienced hunger constructed a food poor female Other in opposition to themselves. The food poor Other was culpable for her food poverty, which was itself attributable to personal failings, notably her incompetent or selfish use of household income and poor cooking skills: I don't know them know them but I know of them, she's got six kids, she is on Income Support and the baby's dad don't help her at all but she goes out every weekend and she uses food bank because she ends up spending the money on clothes and beer.

Gemma, Bradford
The image presented of the food poor individual was reminiscent of Power's (2005) 'welfare bums' and 'flawed consumers'. Examples provided were of women, whose dependence on the state and selfishness were seen to deprive their children of care and food. This apparently gendered stigma surrounding food poverty, bound up with ideals of motherhood, induced guilt and shame in some mothers, unable to meet the standards of parenting in a consumer society: We've just got through Christmas and obviously had to use credit cards cos' we couldn't afford Christmas, which is not what I wanted to do but when you've got a little boy, you want to get him something and if you can't get him something then it's just awful.

Dawn, York
Participants contrasted their own work ethic and resource management with those of the constructed food poor woman. Central to this was the view that poor women respond to personal and societal changes in ways that other women do not. Thus, with the ascendency of individualistically determinist explanations, compassion and solidarity disappeared. The plight of the food poor woman was reconstituted as due to wilful attitudes or personal incompetence.

Mechanisms and knowledge strucures of power
Our data presented a complex and dynamic picture of the contemporary assemblage of power in late-modernity. In so doing, it highlighted the value of a Foucauldian approach to elucidate the variable processes which shape lived experiences of food poverty while also illuminating 'the empirical reality through which political and policy rationales actually play out' (Mckee, 2009: 478). We summarise explicit and subtle forms of power which define lived experiences of food poverty today: disciplinary power of the state and biopower, pastoral power, and the infiltration of people's -gendered, and sometimes racialised -subjectivities by governing discourses. However, we also argue that, consistent with Foucault's approach to power, there is scope for opposition.
State surveillance operated within and between distinct institutions -Jobcentre Plus, schools, hospitals. It comprised a 'whole set of instruments, procedures and levels of application' (Foucault, 1977: 215) operating across state institutions to monitor and coerce poor citizens. Disciplinary power exercised in relation to the receipt of social security defined the parameters of a claimant's life and constantly threatened destitution and food poverty. State surveillance via social security has a long history in the UK (Dean and Taylor-Gooby, 1992), however procedural surveillance via social security appeared to be increasing in line with government welfare reforms, as set out in the 2012 Welfare Reform Act (Edmiston, 2016). The threat of sanctions obliged benefit recipients to emulate 'homo economicus', engaging in a form of 'active unemployment' (Garthwaite, 2016b), submitting job applications and existing within the monthly structure defined by the Department for Work and Pensions. Failure to satisfy such expectations precipitated benefit sanctions, leaving the claimant with no money for food. Within (and between) hospitals and schools, surveillance was continuous and embedded within everyday practices; staff surveyed the welfare of children, contacting social services if a child was considered at risk of food poverty.
Notwithstanding Strong's (2019) Foucauldian reading of the technologies of bio-power in food banks, there has been limited consideration of governmentality in respect of food aid and apparently no literature addressing power in relation to the broader domain of food poverty. Our data identified forms of discipline and coercion inherent to systems of food distribution within some food banks. Technologies such as the referral system and moral distancing between staff and clients underpinned a highly moralised form of service delivery, inculcating a moral code -one closely aligned with a Foucauldian conception of the ethical self -in those assisted. And yet the provision of charity within food banks also represented a form of pastorship or pastoral power. Food banks, particularly Christian food banks, exhibited key characteristics of pastorship as it materialised in the sixteenth and seventeenth century Christian Church: it assured -or attempted to assure -individual salvation; it did not just command sacrifice but was prepared to make sacrifices for its subjects; and it exercised the need to know people's minds, souls and details of their actions (Nettleton, 1997: 211).
Nevertheless, there was a duality to food aid provision which has been relatively neglected. While formal (Trussell Trust) food banks aligned with a punitive state social security system, other forms of food aid, such as community food hubs, represented a different model of provision with a logic routed in mutual aid. These forms of provision, largely open-access, often including communal meals, and sometimes provided by people with experience of poverty, appeared to be motivated by ideas of human nature as social, co-operative and moral, and of society as interdependent. These forms of provision exhibited few of the hierarchical and coercive approaches of the state, rather they were sites of community and potential places of politicisation. Peterson (1997) examines the 'new public health' where there is a privatisation of risk management such that 'being a 'healthy', 'responsible' citizen entails new kinds of detailed work on the self' (Peterson, 1997: 204). Those individuals who do not conform to expected conduct are likely to be seen, and to see themselves, as 'lacking self-control, and as, therefore, not fulfilling their duties as fully autonomous, responsible citizens' (Peterson, 1997: 198). In our study, self-surveillance -modifying behaviour according to the expectations of society or the state (Foucault, 1977) -was highly prevalent among women on a low income. Self-regulation involved ascription to dominant narratives of the 'culture of poverty', opposing ones own attitudes and behaviour to that of the food poor Other (Lister, 2004), who was profoundly stigmatised and whose poverty was attributed to personal failings. This stigmatised Other was often presented by participants as female: her construction was reminiscent of Power's (2005) 'welfare bums', economically inactive and reliant upon the state, and 'flawed consumers', without the financial resources to participate on behalf of themselves and their children in a consumer society. There was a further way in which low-income women were constructed and disciplined as Other: as the negligent mother, failing to adequately feed her children because of her own incompetence or selfishness. This narrative conflicted with the structural obstacles to purchasing an adequate diet on a low income experienced by participants and it undermined commonality and trust within disadvantaged -and arguably exploited -communities. Societal and state power was thus a reciprocal process, mediated by external expectations and internally reinforced via a participant's comparison between themselves and the expected norm.
The stigmatised Other was, on occasion, not only female but also racialised. Young (2003) highlights the rise in widespread resentment and tension resulting from economic and cultural globalisation. He argues that globalisation exacerbates both relative deprivation and crises of identity: such a combination is experienced as unfair, humiliating and threatening and results in behaviour which is vindictive rather than instrumental. Such social vindictiveness was evident in Bradford, as an increasingly degraded working class surveilled and scapegoated migrants for declining conditions and their own precarity (Standing, 2011). Resentment within communities was thereby situated within a structural context, in which spatial and social developments contributed to divisiveness and anomie (Young, 2003).

Implications for praxis and resistance
Inherent in assemblages of power are limits, spaces for resistance. Strong (2019), for example, draws attention to the limits of disciplinary systems, emphasising the 'everyday power of living' that undermines techniques of power exercised over individuals. For Foucault, disciplinary power does not eliminate agency. Resistance is always an aspect of power (Foucault 1980: 142): dominant narratives can be opposed, solidarities can emerge, and alternatives can be offered (Foucault 1977). As Foucault moved from his early carceral focus to a greater concern with the reflexive self, the sphere for resistance increasingly became the discursive body -indeed, for Foucault, individuals are not docile bodies but reflexive, living, speaking individuals.
The scope for agency and opposition identified in our focus groups is evident in the emergence of different models of food aid and, critically, in possibilities for praxis via discursive resistance. It is through the exercise of agency via resistance to accepted norms and to the conduct of conduct that disciplinary power can be challenged. Such resistance may involve problematising dominant discourse and associated classifications -for instance, challenging behavioural explanations of (food) poverty, highlighting the inadequacy of social security income, and questioning the redefinition of poverty as food poverty. Disciplinary power utilises a medicalisation of (food) poverty via application of an epidemiological analytical framework of risk. This can be countered by arguing for critical examination of food practices among all income groups.
Such dominant discourses and accepted norms can be opposed by fostering coalitions of the marginalised to counter disciplinary power techniques of isolating and individualising people amid a wider socioeconomic system of wage inequality and insecure employment. In respect of food banks, this resistance may question their motivations and operations to reveal the exercise of control inside and surrounding the food bank. Crucially, resistance to the discursive and institutional disciplinary systems which coerce people in (food) poverty will require broader coalitions, beyond food, to oppose the use of stigma as a policy to facilitate compliance and promote economic activity.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by funding from the Economic and Social Research Council (grant number: ES/T006897/1). The funders had no influence over data collection, analysis, interpretation of results or the writing of this article. All views represented here are those of the authors.