Encounters with the centaur state: Advanced urban marginality and the practices and ethics of welfare sanctions regimes

This article examines the relationships between advanced urban marginality and new forms of state craft to regulate marginalised populations, specifically Wacquant’s concept of the centaur state and the use of conditionality mechanisms in the British welfare state. The article empirically explores the experiences and perspectives of welfare practitioners and subjects. It finds some evidence of an inculcation of elite narratives and understandings of urban marginality and incidences of antagonism. However, the orientations and ethical frameworks of those deploying or subject to processes of sanctioning within reconfigured welfare regimes are more differentiated and ambiguous than both governmental discourse and critical urban studies often suggest.


Introduction
Commentators have claimed that there is a new urban precariat (Standing, 2011;Wacquant, 2008Wacquant, , 2016 subject to forms of advanced urban marginality; driven by reduced stable and reasonably remunerated working class employment, the displacement of poorer populations from reconfigured urban neighbourhoods (Allen, 2008;Goetz, 2013) and a retraction of entitlements and wealth transfers within welfare states. The new regulation of urban marginality in nations such as the United Kingdom (UK) is framed within particular elite understandings of the causes of poverty. These emphasise a moral behaviouralism grounded upon the perceived irresponsibility and problematic conduct of the poor, to be countered by tutelary and increasingly punitive forms of workfare and penal policy (Wacquant, 2013), including the enhanced use of conditionality within welfare regimes. Wacquant (2016) requires us to establish the linkages between advanced marginality and new forms of state craft through new institutional mechanisms to regulate marginalised populations. In particular, he suggests (2009,2010) the emergence of a 'centaur state' in which deregulation for social and economic elites is contrasted with an expansive and disciplining 'mesh' thrown over marginalised groups to 'correct' their conduct and to inculcate a habituation to precarious low-wage labour, founded on self-blame and passivity (Wacquant et al., 2014).
Contemporary elite policy narratives of poverty and 'anti-social' conduct in the UK and new policy frameworks for disciplining the anti-social and unemployed through enhanced forms of conditionality appear to epitomise this centaur state (Fletcher et al., 2016). However, there is limited specific empirical understanding of how new institutional conditionality mechanisms to regulate advanced urban marginality are framed and experienced by welfare subjects or practitioners (Crane, 2016;Measor, 2013).
This article discusses the conceptualisation of advanced urban marginality and the emergence of the centaur state. Drawing on a study of welfare conditionality in England and Scotland, the article empirically explores the experiences and perspectives of welfare practitioners and subjects in the domains of anti-social behaviourbased interventions and employmentrelated benefit sanctions. The article finds some evidence of an inculcation of elite narratives and understandings of marginality and incidences of antagonism. However, the orientations and ethical frameworks of those deploying, or subject to, processes of sanctioning within reconfigured welfare regimes are more differentiated, nuanced and ambiguous than both governmental discourse and critical urban studies often suggest. This complexity offers one explanation for why growing urban marginality and apparently more punitive projects of governance appear to retain the consent and acquiescence of significant proportions of the actors affected by them.
Urban marginality and state craft Wacquant (2016) identifies relegation as a key process in contemporary advanced urban marginality involving the banishment of marginalised urban populations to particular socio-spatial formations. For Wacquant, such relegation is a collective activity and a form of relation, driven by class position. Wacquant (2008Wacquant ( , 2016) describes a postindustrial precariat (see also Standing, 2011) experiencing the spread and normalisation of social insecurity. Wacquant calls for linkages to be made between these changing forms of urban marginality and emerging modalities of state crafting (Wacquant, 2009(Wacquant, , 2016 and an examination of how institutional mechanisms produce, reproduce and transform the network of positions within which the relations driving urban marginality and relegation occur. While urban policy has contributed to advanced urban marginality (Goetz, 2013;Wacquant, 2016), there are also reconfigurations of state discourses, policies and practices in response to such marginality. It is argued, for example, that the contemporary British state has sought to redefine Beveridge's concept of shared social risks, enacted through the welfare system, as matters of individual personal responsibility and insurance (Lea and Hallsworth, 2013). Discourses and technologies of governance seek to inculcate an acceptance of, and habituation towards, precarious low-wage labour, premised on a 'philosophy of moral behaviourism [that] employs techniques of control including stigma, surveillance, punitive restrictions and graduated sanctions to ''correct'' the conduct of . clients [through a form of] authoritarian therapeutism' (Wacquant, 2013: 249).
This, it is claimed, generates a 'spectre of uselessness' or 'permutations of dishonour' (Sennett, 2006: 83;Wacquant, 2013: 244) in the characterisation and identity of marginalised urban populations and a 'noxious identity' that 'warps the perception and behaviour of operators within welfare regime systems (Wacquant, 2016(Wacquant, : 1083. The institutional mechanisms through which such processes are manifested are characterised by the concept of the 'centaur state' presenting 'a fearsome and frowning mug towards the lower class' (Lea and Hallsworth, 2013;Wacquant, 2010: 217). This is enacted through an alleged 'diligent and belligerent bureaucracy' (Wacquant, 2013: 248) imposing new forms of violence in which 'ferocity comes to be succeeded by other forms of violence, new forms of bureaucratic domination and asceticism' characterised by 'passionless, impersonal callousness' (Gouldner, 1981: 418, quoted in Rodger, 2013. The longstanding history of the functions of the welfare and therapeutic state being imposed on working class 'clients' by middle-class welfare workers (Polsky, 1989) is, it is argued, realigned through fracturing class lines and new configurations of relations, for example the increasing staffing of agencies of social control in the United States by middle-class blacks 'overseeing their unruly lower-class brethren' (Wacquant 2016(Wacquant : 1079see also McCarthy, 2011).
It is suggested that new forms of discourse generate 'a heavy imprint' of discipline on welfare recipients (Marwell, 2016(Marwell, : 1097 and that the mentalities of individuals subject to welfare interventions are 'transformed by their routine engagement with hostility, interpersonal threat and, too often, only the coercive arm of the state' (Rodger, 2013: 97). It is claimed that these interactions within the arenas of the welfare state apparatus are situated within a wider strengthening of class antagonisms that become heightened in periods of economic crisis (Hancock and Mooney, 2013).
However, Measor (2013) indicates that we do not have enough knowledge about the daily lives of the precariat, 'of those caught in the cracks and ditches of the new economic landscape' (Wacquant, 2009: xiv). This limited knowledge has resulted in the precariat often being presented 'as one undifferentiated mass, their individual characteristics and differences ironed flat . a colourless uniform group [of] Lowry-like stick figures' (Measor, 2013: 133-135). Although there is a growing literature on contemporary marginal working class lives (Bhattacharyya, 2015;Shildrick et al., 2012), their lived experiences of specific mechanisms of statecraft such as welfare conditionality are less well documented. Equally, accounts of the orientations and conduct of those employed within social control and social welfare agencies often neglect the ambiguities of exclusionary and inclusionary impulses, the diversity of class backgrounds and the importance of individualisation, personality and emotions framing encounters (Crawford and Flint, 2015;Flint, 2012;Lawson and Elwood, 2014;Stenson, 2013). Traditions of discretion, resistance and subversion are prominent in accounts of how governmental policy is actually translated and applied in localised daily practice (Barnes and Prior, 2012;McKee, 2015). There is, therefore, an urgent need for a further analysis of the everyday operation of state programmes and the 'unexpected encounters' within them (Crane, 2016(Crane, : 1111, and of how welfare practitioners and subjects frame their understandings of their encounters within the contemporary centaur state (Boltanski and Thevenot, 1991;Crawford and Flint, 2015).

Welfare conditionality in the United Kingdom
The rise of welfare conditionality in the UK is situated within a political project of redefining poverty and urban marginality as primarily stemming from individuals' inability or reluctance to take advantage of opportunity (Mead, 1991). Conditionality emphasises the behavioural elements of citizenship or non-citizenship: 'Conditionality embodies the principle that aspects of state support, usually financial or practical, are dependent on citizens meeting certain conditions which are invariably behavioural' (Department for Work and Pensions, 2008: 1). These rationalities have been applied across social policy fields, including employment search and benefits payments, where it is suggested that a policy apparatus requires 'new forms of deterrence for some people who are either not trying or who are gaming the system' (Department for Work and Pensions, 2012: 2).
Since the mid-1980s, culminating in the Claimant Commitment introduced in 2013, there has been a tightening of eligibility criteria (including medical assessments of claimants) and a requirement that claimants evidence their efforts to seek employment (applying for jobs and attending interviews) and their compliance with support (attending appointments with work advisors). Individuals failing to adhere to these conditions can face sanctions of a loss of benefits payments, now with a maximum duration of three years, representing the most punitive sanctions in the history of the British welfare state (Slater, 2014).
The use of sanctions and the simultaneous provision of support and disciplinary regulation has also been a feature of governing anti-social behaviour by successive UK governments since 1997 (Flint, 2006;Home Office, 2012;Millie, 2009aMillie, , 2009bRespect Task Force, 2006;Squires, 2008). Anti-social behaviour describes a spectrum of conduct from neighbourhood incivilities (excessive noise, graffiti, poor upkeep of properties) to conflict with neighbours and more serious harassment or criminal damage. Conditionality has been enacted through individuals being prohibited, or required to desist, from antisocial conduct, enacted through a range of new legal mechanisms. These included Anti-social Behaviour Orders (ASBOs) which stated a list of prohibitions on individuals subject to them. ASBOs were civil orders but a breach was a criminal offence, liable for fines or, ultimately, imprisonment (see Squires, 2008). Acceptable Behaviour Contracts (ABCs) are voluntary agreements between individuals and agencies that also listed prohibited conduct. Social housing management became a primary arena of anti-social behaviour regulation, with tenancies conditional on acceptable conduct and reduced security of tenure for anti-social individuals, in addition to injunctions and, ultimately, eviction (see Flint, 2006).
New mechanisms also required anti-social individuals to proactively engage with support services as a condition of accessing welfare state provision including housing and education, with technologies such as Parenting Orders mandating individuals to undertake counselling or parenting courses, with penalties for failure to do so.
Each of these legal and punitive mechanisms was envisaged as being deployed simultaneously with support packages that would enable individuals to adhere to the behavioural conditions imposed on them. This new configuration of support and sanction was epitomised by the growth of Family Intervention Projects (FIPs) that deploy key workers to holistically address the needs of individuals and households, linked to the continuing threat of sanctions if individuals did not engage (see Batty and Flint, 2012). The Conservative-Liberal Democrat Coalition Government's (2010Government's ( -2015 Troubled Families Programme in England deployed these approaches and aimed to 'turn around' the lives of 120,000 'troubled families' (Department for Communities and Local Government, 2017). The Anti-social Behaviour, Crime and Policing Act (2014) refined the plethora of existing legislative measures into six new powers, but did not represent a shift in existing governmental rationalities about regulating anti-social behaviour. The Scottish Government, which has devolved responsibility for tackling anti-social behaviour, established its own national framework, 'Promoting Positive Outcomes', in 2009, which placed a greater emphasis on early intervention and support.

The research
The findings presented in this article are based on an ESRC-funded study of the ethicality and efficacy of welfare conditionality (see: http://www.welfareconditionality.ac.uk/). The study comprised interviews with 44 national agency stakeholders, 24 focus groups with welfare practitioners and qualitative longitudinal research with 480 welfare recipients subject to welfare conditionality in 10 case study cities in England and Scotland who were interviewed on three separate occasions over a two-year period.
The findings presented here are drawn from a sample of 40 study participants subject to anti-social behaviour-related measures and uses data from the first wave of interviews, conducted between August 2014 and July 2015. Participants were recruited through local agencies and support organisations. Interviews took place in individuals' homes or the premises of local organisations, lasted between 20 and 90 minutes and were recorded and transcribed. The interviews focused on individuals' life histories and their experiences of, and views on the efficacy and ethicality of, forms of support and sanctions. All data in this article has been anonymised and pseudonyms are used throughout.
The sample is not statistically representative of the national population subject to anti-social behaviour interventions. Twentyone participants were male and 19 were female. The majority (28) were aged 25-49, although five were aged 18-24 and seven were between 50 and 64 years old. All but two participants stated their ethnicity to be White British. Half were in single person households, 11 were lone parents, eight were living with a partner and children and one had a partner but no children. Almost half the sample (19) reported a mental health issue. Thirty-eight percent had been in prison and the same proportion were homeless at the time of the interview (other individuals had also experienced homelessness in the past). This marginality was compounded by 36 percent of the participants having been subject to an unemployment or disability-related benefit sanction in addition to anti-social behaviour-related measures. Just under half (18) of the sample had been subject to ASBOs, almost a third (12) had been referred to FIPs, nine were subject to ABCs and five had received eviction notices or had been evicted (see Flint et al., 2016 for further details).
The article also uses data from two focus groups with practitioners, one comprising local authority solicitors from across Scotland expert in anti-social behaviour legislation, and a second focus group conducted in Bristol, England with practitioners from a range of agencies involved in addressing anti-social behaviour. It also presents data from interviews conducted with policy makers with responsibility for social housing and anti-social behaviour policy in the Scottish Government, a senior officer of a national organisation representing anti-social behaviour practitioners and a representative of a charity working with individuals subject to welfare conditionality mechanisms. This practitioner sample is not intended to be entirely representative of all forms of practice, or all orientations to practice (for example it was not possible to interview front-line Jobcentre Plus staff).

Findings
The urban precariat and the politics of the centaur state The welfare subjects interviewed would almost all be categorised as members of the new post-industrial urban precariat (Standing, 2011). They were engaged in lowpaid and often short-term forms of employment, including hotel work, labouring and car maintenance, and some individuals undertook voluntary work. However, their framing of their employment and housing circumstances, and the wider political contexts of these, challenged elite and policy discourses of a welfare dependency culture. Most participants articulated a strong orientation to work (see Shildrick et al., 2012) and an acceptance that this would involve low paid and often mundane jobs: I used to work in hotels doing waiting on silver service. I've done all kinds of work; do you know what I mean? All kinds. Whatever job come up I'd take really. Mostly factory work. Just boring work really. No skills in it. (Clare, Manchester, subject to eviction warning) Contrary to conceptualisations of the 'antisocial' as having no regard for their neighbours or communities, the research participants recognised that the districts of marginalisation (Wacquant, 2016) in which they often resided were subject to particular socio-spatial reconfigurations, comprising both societal stigmatisation and urban policies enacting forms of gentrification that categorised them as 'out of place': The estate that I live on is a very, very, very small little estate at the side of the motorway, and it's got a reputation for trouble, and I personally think that the housing [officers] don't want to really be going in there. (Harris, Manchester, subject to ASBO and injunction) They're trying to get the down-and-outs, and everyone else like for-you know like people that haven't got much or something like that. It looks to me like they're trying to get all them out for better people to come in . Like posh people. Like people that are not like on benefits . (Gillian, Manchester, evicted for ASB) That relegation is a relational process which occurs in socio-spatial arenas in which individuals are socialised and attuned to forms of shame was evident. For example, one individual spoke of the trauma of being required to undertake regulated access to methadone in a neighbourhood pharmacy: '. I'm having to go back to fucking supervised consumption in a chemist and that pisses me off that because it's embarrassing' (Clare, Manchester, subject to eviction warning). Reconfigured social relations, and their own declining status within these, which individuals recognised at the neighbourhood level, were mapped on to wider narratives about a broken social contract between working class communities and government (see Flint, 2015). Far from being 'asocial', the participants conceived themselves as being embedded in a set of social relations linked to a reciprocal social contract and national identity, enacted primarily through the political economy provisions of the welfare state. Within these narratives, the governmental discourse that the antisocial individual or the 'illegitimate' benefit claimant are a manifestation of 'Broken Britain' (Cameron, 2012) was refuted and reframed as a failure of government to deliver on a longstanding post-war social contract. Immigration and a lack of housing and employment were signifiers of this failure (see also Beider, 2015): 'Now don't get me wrong, I'm an equal opportunity person, help everyone. But at the end of the day the Government should have helped us English people first, who were here, then started letting people in' (Harris, Manchester, subject to ASBO and injunction); 'They're treating people like-wrong. They're trying to tell you there's jobs about there for everybody. If there were why are they at the Jobcentres? Why are people signing on? There's not jobs for everybody' (David, Sheffield, subject to Acceptable Behaviour Contract): My grandad fought in the war, do you know what I'm saying, and there's people here like in front of me and her who are in a hostel? I'm not saying, just come over here, come over here, been here four days and got a brand new house and we got thrown back on the streets, fit for the streets, not from this country, in front of my face. How disgusting is that? My grandad fought for this country are you with me? It's a joke. (Paul, Manchester, subject to an ASBO) This identification of wider macro-economic factors challenged the primacy of individuals' own 'employability' as the driving logic of conditionality and job-search mechanisms (see Crisp and Powell, 2017). The research participants' perceptions were situated within a broader belief that political and policy elites were increasingly divorced from the reality of working class experience and relied on stigmatising myths, such as young mothers dependent on welfare, to mask the failure of government: 'The government who run this country, they sit there drinking wine every night with steak and that' (Harris, Manchester, subject to ASBO and injunction); 'I wish they'd get their finger out and realise that there are people that want to get on with their lives, not people that just want to sit there and get their new pram or their buggy or whatever' (Joe, Edinburgh, subject to an ASBO).
This analysis also sometimes extended to critiquing the private profits being made in the administration of new welfare entitlement tests and benefit sanctioning: 'Sorry, to me they are paying out all this money for this company to come in. Oh, I don't like to get into politics. I really don't because it makes me mad' (Moira, Bristol, subject to FIP).

Encounters with the centaur state
The articulation of wider changing societal and governmental relations framed individuals' encounters within the welfare regime, especially interactions with JobCentre Plus staff. Several participants reported experiences aligned with the portrayal of a belligerent bureaucracy characterised by impersonal callousness (Gouldner, 1981): Well I think when you suffer from a mental illness you become more vulnerable really. There is a sense of vulnerability about it all and they [Jobcentre Plus staff] just seem so cold and horrible people. I don't particularly like them to be honest with you. (Michelle, Glasgow, subject to ASBO) Interactions could also evidence perceived class-based tensions (Hancock and Mooney, 2013): A lot of them sanction you and on the other end are stuck up gits, yes, who've got more than enough money, yes, and proper posh and they're just out to think 'why should they [benefit claimants] get free money off the government?'. (Chris, Sheffield, on probation and subject to benefit sanction) Many participants had experienced violence (as victims and perpetrators) in their own domestic arenas and in disputes with neighbours, with such violence often being a central focus of the anti-social behaviour-based interventions applied to them. However, other forms of violence pervaded their lives, including the symbolic violence inherent in stigmatising discourses and the indirect violence of housing displacement or dispossession. The threat of violence between welfare users and employment benefits practitioners was embedded in the new architecture of clinical sites of interaction, such as screens and alarms at Jobcentre Plus counters and the growing prevalence of security personnel: 'No wonder they have about 30 security guards in every Jobcentre . I'm surprised they don't get stabbed or something' (William, London, subject to Acceptable Behaviour Contract): I said 'Look man you're just taking the piss man. You're just blatantly taking the piss right in front of my face man and I'll tell you if you hadn't got that glass screen there, man, you wouldn't be talking to me like than man, because you think you've got some authority or some protection from me that you can treat me like a dog and get away with it yeah?'. (Nick, Bristol, subject to eviction) This reconfigured a remembered more benign, individualised and supportive framing of encounters between welfare recipients and practitioners to a relationship increasingly defined by perceived antagonism and distrust in which the risk of potential violence became more prevalent.
These findings support characterisations of a hostile and punitive state, with a consequential alienation of those subject to its interventions. However, Foucault's (1977) more nuanced understanding of an individualised and ambiguous duel between state actors and subjects more accurately captures the complex dimensions of contract and consent. Many individuals articulated recognition of government rationales: 'I can understand government, why they do it because there's a lot of lazy people around. If someone's just chucking money at you for just signing your name, then people are going to do that all the time' (Daniel, Bristol, subject  There was also an understanding of the necessity of interventions, including sanctions, to address their own periods or incidences of anti-social behaviour: It's more for people's safety and peace of mind and whatever. So, while many individuals remained opposed to the employment benefit sanctions regime they were more supportive of anti-social behaviour-related sanctions, especially where this could trigger individuals' engagement with support services. But there remained nuanced caveats about the limitations of sanctions and the need for these to be cognisant of individual circumstances: You're learning something aren't you? So I'm sure you are benefitting from it [having to attend classes to get a meal voucher] . I don't think they should sanction people because people are getting sanctioned for months and months and months aren't they? I don't think that's fair. (Liam, Sheffield, subject to ASBO) I think it varies from person to person really doesn't it? Some people work well when they're pushed, some people are forced to do it when they're pushed to do it. So I understand that, where others might sort of close up a bit when they feel that much pressure. (Cath, Manchester, subject to FIP) Where this personalisation of anti-social behaviour-related support, characterised by strong relationships between 'clients' and family intervention workers, was enacted, many individuals were positive about its transformative potential and an avoidance of the antagonism pervading benefit sanctions encounters: 'Oh I could sing their [family project workers] praises all day. I really can sing their praises all day . she wasn't judgmental at all. Not like some people are' (Neil, Sheffield, subject to FIP). In summary, these findings indicate that individuals subject to welfare system interventions have an economic status aligned with a new postindustrial precariat and many perceived their stigmatisation and increasingly precarious access to key pillars of the post-war social settlement including housing and employment-related benefits. But, contrary to elite political discourses, they are not asocial or apolitical. Rather, they are heavily socialised within a framework of social relations that they perceive, at individual, neighbourhood and national levels, to involve a breaking of a social contract arising from a governmental failure to deliver key elements of social security and protection.
Some individuals' experiences characterised the centaur state's 'frowning mug' (Wacquant, 2009(Wacquant, , 2010. Simultaneously, however, they articulated an understanding of government rationales and the need for a contractual reciprocity requiring their own activism and responsibility. In understanding how statecraft is related to new and distinctive socio-spatial formations of problematised territories and populations (Wacquant, 2016), an important distinction emerged in the accounts of the research participants about the techniques of governance. The enacting of unemployment benefit sanctions was inherently impersonal and aspatial (in which the interaction involved letters, or the virtual territories of the online form or email) or increasingly fleeting interactions in 'clinical' (Foucault, 1977) and securitised sites such as Jobcentre Plus. This was contrasted with the more personalised and sustained encounters with anti-social behaviour practitioners, especially FIP workers, that occurred in the socio-spatial arena of the domestic home and neighbourhood (Flint, 2012;Polsky, 1989). However, despite the importance of urban socio-spatial contexts in configuring individuals' perceptions of sanctions regimes and the broader socio-political contractual relations underpinning them, geographical locality at national (England and Scotland) or city scales did not appear to be a driver of differentiation in participants' experiences and orientations.
It should also be noted that, while Wacquant's concept of the centaur state was largely formulated in the racialised context of the United States, the participants in this study were overwhelmingly from White British working class backgrounds, although the ethnicity of the practitioners who they interacted with could not be established. It is to the experiences of these practitioners that the article now turns.

Challenging the principles of the centaur state
To understand the extent to which elite and policy discourses become embodied in practice and the everyday institutional operation of the welfare state apparatus, there is a need to first examine how practitioners framed the context and causes of the circumstances of their clients (Crawford and Flint, 2015;Polsky, 1989). Several practitioners articulated a focus on the inappropriate financial planning and management of the poor, inadequate parenting, intergenerational transmission of social problems and a poverty of aspiration, all of which have been key tropes of contemporary and historical social policy discourses (Welshman, 2012): 'The phrase used to be poverty of aspiration. We have that I'm afraid in many of our communities and that is a big, big, feeder for anti-social behaviour' (Local Authority Anti-social Behaviour Officer and Focus Group Participant, Bristol): There is something there that around about if you can't afford your rent and things like that, you might have to look at the things that you're spending money on. Can you afford this top of the range Sky package or whatever? (Senior Housing Policy Officer, Scottish Government) Part of the problem is that we're now on to a third generation of kids who met in care, so where are the parenting skills? . They don't know how to parent children effectively because they don't have those skills demonstrated to them when they were growing up. So, I think there is a much greater social aspect to it. (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant, Scotland) However, this apparent adherence to elite discourses was countered by an explicit critique of the failures of government, at both macro-structural and policy-operational levels, and several practitioners argued that forms of shame and disgrace associated with this should actually apply to government rather than welfare recipients: I don't like phrases about conditionality and welfare and sanctioning the poorest in society because of the actions of their parents or the economic sector in which they were brought up . we can remove your house. We do that anyway, but what other welfare? I mean welfare is under attack at this moment in time, our child poverty figures are a national disgrace. (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant, Scotland) At the end of the day the sanctions don't work because you're dealing with chaotic people who have chaotic lifestyles who don't know New York from New Year, you know. That's just saying to people 'Well, you've missed three appointments'. Some of them-look at that poor guy that was found dead down south in his house that was a severely autistic man who was sanctioned beyond belief and death! That's how effective sanctions are for me, as a civilisation we should be ashamed that that ever happened. (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant, Scotland) There was a particularly strong critique of contemporary social housing policy in the UK and of related housing benefits reform in which the social rights to affordable housing were perceived to be problematised and undermined by government: There's something about a tenancy being a right in property, and so forth, that doesn't sit comfortably with being regarded or described as a benefit for which you should be beholding. You should behave responsibly but there's a tenancy agreement. (Director, National Anti-social Behaviour Practitioner Professional Organisation) This is part of the issue of the private rented sector is you get a lot of vulnerable people there, they're people who fall through the cracks . the concern about linking conditionality with welfare when you've got the most vulnerable in society. (Local Authority Antisocial Behaviour Solicitor and Focus Group Participant, Scotland) Several respondents argued that the conceptual foundations upon which welfare conditionality and sanctions regimes, as an enactment of a social contract, were premised were fundamentally flawed: I think the government would say that they're trying to create a kind of contract in which people have rights but responsibilities . I think there's an atmosphere where there's no longer a belief in institutionalised poverty. It's the sense that people are poor because they have addictive or chaotic lifestyles and people are therefore responsible for their own situation and conditionality is a way of regulating or compelling them into a more orderly lifestyle which will be good for them. (Policy Officer, Rehabilitation and Support Charity, London) The irony, according to this respondent, of this encompassing project of ensuring that all populations, including the most marginalised, are subject to technologies that ensure their discipline is that such technologies may actually displace individuals beyond the reach of state intervention: 'I think unintended consequences are that people engage even less. That people disappear from the welfare system'. The importance, for practitioners, of ensuring that the most marginalised urban populations remained within the inclusionary reach of state intervention was cited as significant in the ethical justifications for enacting sanctions, to which the article now turns.

Dilemmas of practice and ethics in the centaur state
As noted above, several practitioners articulated drivers of urban marginality that adhered to dominant narratives of personalised inadequacy and intergenerational transmission of domestic and economic dysfunctionality. But this, in turn, generated ambiguity in the normative construction of the ethics of punitive technologies including sanctioning. A characterisation of the marginalised as not being autonomous rational actors imbued with necessary contemporary life skills served to undermine, rather than support, the ethical basis for a sanctioning regime based on individual responsibility and an alleged subversion of societal obligations: 'I think people sometimes just genuinely can't see any other way of being. It's just that's that bubble they are in .' (Housing Tenancy Manager and Focus Group Participant, Bristol); 'If they've not been parented, they've not been told how to run a household, then how are they expected to change their behaviour to match that?' (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant Scotland): A lot of them have such chaotic lives. I think a lot of them, it's not intentional, they didn't intend to end up in that way but they've reached a stage where maybe it's mental health or drug use, alcohol use, everything has escalated to the point where their life is so chaotic, they can't keep a handle in it without some support. (Local Authority Anti-social Behaviour Officer and Focus Group Participant, Bristol) Some respondents argued that a focus on the inadequacies of marginalised individuals, and the punitive interventions arising from this, could equally apply to the failings of state technologies, for example in delivering parenting duties for looked after children: 'We all make really crap corporate parents, can I just tell you that? We should-every local authority in this land should-be sued for its lack of parenting capacity ability' (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant, Scotland). These findings challenge the alleged passionless and impersonal callousness of operatives within the bureaucratic regimes of punitive interventions (Gouldner, 1981) or the 'frowning mug' that the centaur state presents to the working class (Wacquant, 2009(Wacquant, , 2010. Rather, for these particular groups of welfare practitioners the ambiguous and nuanced enactment of a philosophy of moral behaviourism (Wacquant, 2013) demonstrated above was accompanied by fundamental dilemmas of practice in deploying institutional mechanisms: 'No, well, no one likes making children homeless. No one likes making anyone homeless' (Local Authority Anti-social Behaviour Officer and Focus Group Participant, Bristol). Again, a contrast was made between the due process and almost agonised decision-making involved in evicting a household on the basis of anti-social behaviour and the new forms of sanctioning within unemployment and disability benefits regimes. In the quote below, a local authority solicitor specialising in anti-social behaviour legislation argues that the ethics of his professionalism would not enable him to operate the new sanctions regime in other social policy fields: God, are we really going to run this proof? Are we going to evict this family? Are we going to do this? Are we going to do that? That's taxing enough and it's troubling enough if you've got any sort of degree of social conscience . as a local authority lawyer, were I charged with implementing some of the proposals that could emerge from something like this, I'd have to say no. I would seek alternative employment, quite frankly, than implement that sort of stuff. (Local Authority Anti-social Behaviour Solicitor and Focus Group Participant, Scotland) It has been argued that a new socio-spatial stigmatisation of marginalised urban populations has resulted in a 'noxious identity' for welfare recipients that warps the behaviour of those operating the mechanisms of the contemporary welfare state (Wacquant, 2016: 1083; see also Rodger, 2013). This research suggests a more complicated process, whereby the professional practices associated with addressing anti-social behaviour (in which interventions have always deployed mechanisms of sanction) are contrasted with new forms of conditionality and punitive sanctions relating to employmentrelated benefits. So, for these practitioners, their orientations towards their 'clients' remain unchanged by new governmental regimes, but they recognise the risk of a 'warping' of relationships for other practitioners who may be applying sanctions in new fields of social policy and who were not included in this research.
There is also an important distinction in both the policy discourse and legislative framework between England and Scotland. For example, the Scottish Government and many Scottish local authorities have proactively sought to oppose, and financially ameliorate, the impacts of UK Governmentinstigated welfare reforms. One participant argued that the entire Scottish policy apparatus was based on the welfare of children and, therefore, that a punitive sanctions regime applied to marginalised households was fundamentally contrary to this: The whole approach is based on the welfare of the child. It is not based on sanction or anything else . The whole system deals with much broader problems than just criminal disorder and criminal behaviour, but it's all based on welfare of the child and support. (Senior Policy Officer, Scottish Government) Practitioners constructed three grounds on which technologies to address anti-social behaviour, including the use of sanctions and penal mechanisms, could be ethically justified. The first of these was a communitarian emphasis on the balance between individual rights and communal responsibilities. This was framed as being proportionate with an explicit rejection of personalised antagonism, whether class-based or otherwise, towards the subjects of intervention: The ethical thing for me is the balance of doing all this stuff, all this enforcement stuff to an individual balanced against the needs of the community to go about their lives, have their quiet enjoyment of their homes, be able to go down the shops without being hassled for money all the time. (Local Authority Antisocial Behaviour Officer and Focus Group Participant, Bristol) It's that qualified right thing, so that's the kind of consideration is are we, yes, we're going to be stopping people exercising these rights but are we doing it, so is it a proportionate means of achieving a legitimate aim and we're doing it to prevent crime and disorder and not just because we don't like them. (Housing Tenancy Manager and Focus Group Participant, Bristol) The second normative framing of practice was a belief in the progressive and transformative potential of interventions in which paternalistic and tutelary mechanisms were justified as being in the interests of the subjects of governmental mechanisms: 'It's quite a dysfunctional family at the moment and if you can try to help them become more functional, I don't see how anyone could have any ethical problems against that really' (Local Authority Anti-social Behaviour Officer and Focus Group Participant, Bristol); 'Quite a lot of them turn their lives around, don't they?' (Housing Tenancy Manager and Focus Group Participant, Bristol). Finally, interventions were framed within an inclusionary project of governmental practice that rejected the banishment of marginalised individuals beyond the reach of the state or their permanent relegation from wider society, explicitly using the language of social justice to support this: I think it's [conditionality] been tied very closely to the social justice agenda to behaviour change for people and it's seen as very noble, not giving up on anybody . A key plank of social justice is life change through work and the idea is that actually it's not socially just to leave people stuck on benefits and there is as way off of it. (Policy Officer, Rehabilitation and Support Charity, London) These findings suggest that, while practitioners articulate some discourses that are aligned with elite policy narratives in their problematisation of the orientations and conduct of the marginalised, they frame their practice in more nuanced and sophisticated ways than critiques of contemporary neoliberal statecraft usually give credence to. Rather than being ciphers of a punitive project further marginalising or relegating the working class, practitioners recognised that their interactions with the subjects of welfare regimes were embedded in wider socioeconomic processes, including the failures of government itself. They critiqued the conceptual underpinnings of contemporary policy and challenged its ethical basis, often repudiating a normative justification for sanctions mechanisms being applied against new populations including the unemployed, the disabled and social housing tenants. Their perspectives also challenge a characterisation of practitioners as emotionless operatives of an impersonal bureaucracy. Rather, they reported the considerable dilemmas of an authoritarian therapeutism (Wacquant, 2013), though they justified their practice on grounds of communitarian obligations, the transformative potential of paternalistic interventions and the social justice-influenced normative basis for inclusionary, rather than exclusionary, ambitions for government.

Conclusions
New forms of urban marginality, and their relationship with forms of state craft, vary between nations and across social policy fields. But the growing use of welfare conditionality in the UK appears to epitomise elements of what Wacquant has termed the centaur state, despite the very different racial context from the United States. If this centaur state is a 'splintered space of struggles over the selection, definition and treatment of social problems and . the result of the gradual and partial convergence of battles' (Wacquant, 2013: 253), then we need to understand these battles further, as well as how, across social and penal policy fields, the duel (Foucault, 1977) between welfare practice and its subjects is constructed and enacted. There is a further need to examine how advanced marginality, and state responses to this, results in institutional mechanisms that appear, to some extent, capable of manufacturing the continuing consent of actors (Marwell, 2016) and the quiescence (Pahl et al., 2007) of both the practitioners and subjects of new technologies of urban governance.
Elite media and political discourses, but also some academic commentary, has, to date, failed to empirically establish the lived experiences of those subject to a new form of statecraft embodied in welfare conditionality (Measor, 2013;Wacquant, 2009). The findings presented here, while rejecting the idea of the precariat as an undifferentiated uniform group, do challenge the governmental characterisation of the unemployed and 'anti-social' as essentially asocial, amoral and inactive individuals living in a social vacuum. Rather, these individuals, who, through being labelled anti-social, may be categorised as the antithesis of the normalised dutiful and self-reliant urban citizen, continue to demonstrate an orientation to (often low-paid and precarious) employment and situate themselves within a web of social relations and imagined social contracts at individual, neighbourhood, agency and national scales. While they often adhere to the principles of a sanctions regime, based on the same normative premises as policy narratives, they are also critical of the operationalisation of welfare technologies and the wider failure of government to address the causes of urban marginalisation. Their experiences provide some examples of antagonism and alienation within welfare regimes. But, crucially, they identify a distinction between the more fleeting, automated and impersonal encounters of employment-related support and sanctions and the more personalised, intensive and sustained interventions to address antisocial behaviour.
Further research is required into forms of territorialisation and new spatial configurations within welfare regimes, in which the virtual and impersonal mechanisms of the automated email and online form coincide with the new securitisation of the architecture of sites of encounter such as Jobcentre Plus. These may be contrasted with the centrality of domestic space as the continuing arena for many anti-social behaviour-and housing management-related interventions which increasingly necessitate practitioners in these fields also negotiating their clients' engagement with the new apparatus of benefit sanctions.
This research also challenges depictions of a new belligerent welfare bureaucracy characterised by a passionless and impersonal callousness (Gouldner, 1981;Wacquant, 2013). Rather, the 'frowning mug' that Wacquant (2009) suggests the contemporary centaur state presents to marginalised populations may for some forms of welfare practice be more accurately termed 'a worried frown'. Many practitioners recognise the limitations of the social contract and governmental achievements and wrestle with the daily emotional and ethical dilemmas of practice. Certainly, their orientations towards practice appear to support Wacquant's (2013: 253) statement that a contemporary centaur state remains constituted by continual and fragmented spaces of struggles and is certainly not 'the spawn of a malevolent design'. Practitioners' framing, explicitly in social justice terms, of these practices within an inclusionary and ambitious project of paternalistic and communitarian engagement with the marginalised may not fully capture the regressive elements of contemporary welfare and penal policy and antagonisms within them. But it does require urban scholarship to recognise, along with Foucault (1977), how individualisation, complexity, inclusionary impulses and ethicality continue to imbue technologies of urban governance and to frame the orientations and conduct of the actors within these regimes.