Social work in confinement: the spatiality of social work in carceral settings

Social work in prisons not only works with and for people who are confined; it also constitutes a form of social work that is carried out under conditions of confinement. This article draws on carceral geography to understand the corporeal and spatial aspects of social work in prison settings. Based on insights from two prisons in Switzerland, we argue that understanding carceral social work as a spatial and materially situated practice helps to gain deep insight into the intricate layers of meaning and powerful modes of functioning of prisons and of the people involved. In particular, it shows how the way social work is carried out in prison is supported and strongly structured by the spatiality of the prison itself, allowing for counselling, desk-type social work, rather than for social work that actively initiates and creates spaces for encounters or activities.


Introduction
Social work in prisons not only works with and for people who are confined; it also constitutes a form of social work that is carried out under conditions of confinement. We literally stumbled across this notion, when doing research, as we listened to the complaints of a social worker about her working environment. She criticised the building she was working in, stating that this novel concrete architecture was a response to the often-heard critique that prisons present a much too soft regime towards inmates. In her words, architects forget that there are people working inside these grey walls, and they (also) have to stay inside the whole day. At the same time, she later compared herself to the prisoners and stressed that she had better conditions than them while at the same time putting herself on a similar level, as they all spend the day within the same walls: "they have daily schedules, now it's time for showering, now free time in the courtyard, they cannot organize their day by themselves".
While there are many aspects of how confinement is enacted and experienced in prisons, the account of the social worker above stresses the aspect of time when talking about the rules while not mentioning that the time schedule is also strongly linked to a spatial setting of scheduled activities. The striking function spatiality plays for confinement, as also outlined in the example, has been famously detailed by Fou-cault (1995). Referring to the panopticon of Bentham he explains how the architecture of the prison is used as a dispositive of control and surveillance that transfers the aims and logics of the institution on and into the people living inside its walls.
Much literature, and in particular in geography (Mitchelson and Martin, 2009;Moran, 2015;Sibley and van Hoven, 2009), has been written about the spatialities of confinement in prisons, about how prisons immobilise people (Mincke and Lemonne, 2014) or how controlled mobility is used, to subject the inmates to the prison authority (Moran et al., 2012). What is only scarcely researched, though, is the effect that this confinement architecture has on the staff: how the spatiality of the prison structures the working conditions inside.
Our argument develops around the notion that carceral social work is not only social work that is conducted with people who are subjected to carceral regimes but one that inevitably also happens in carceral contexts and is therefore shaped by this specific spatiality that is linked to and was formed according to the specific logic of the carceral setting. We thereby focus on the spatiality of this setting and only where necessary make reference to other important current debates about how carceral social work is confronted with developments concerning increasing securitisation and con-trol (Garland, 2001), risk orientedness of the criminal justice system (Cummins, 2017), or the prevailing logic of administration and management (Bauwens and Roose, 2017). At the same time, these developments are also inscribed in the spatiality of the prison, as we will see in the course of the article.
The next section will give the necessary background to understand the position social work has in the criminal justice system. In particular, we will explore the literature on spatialities in social work settings and from this develop how spatiality is a useful analytical approach to explore conditions of confinement. We will explain the general setting of the project and give some background on our empirical approach and the data that constitute the basis for our analysis. We then use small scalar steps to develop an understanding of these spatialities of confinement: starting from spaces, and going on to bodies, and last to emotions and feelings, we analyse how carceral social work is shaped by these spatialities and at various levels.

Carceral social work and its spatiality
Social work has been linked to the field of criminal justice and the spaces of corrections since its early development. In the United States, the Hull House settlement, which was founded in 1889 in Chicago by Jane Addams, one of the early inspiring forerunners of social work, encouraged professional diagnostic work in juvenile justice at the end of the 19th century. The advocacy of Julia Lathrop, a Hull House member, led to the founding of the Juvenile Psychopathic Institute to diagnose offenders brought before the juvenile court. Therefore, the initiative of a Hull House member led to the introduction of social work and psychiatric knowledge in the criminal justice field (Brownell and Roberts, 2002). Since then, correctional or forensic social work has been an important element of the criminal justice system and its different spaces. Whether in courts, civil authorities, parole services or the correctional system itself, social workers have long played an important although shifting role. It remains a constant throughout this development that forensic social workers perform their professional duties in an ambiguous field (Maschi and Killian, 2011).
A development similar to that in the United States also occurred in Europe. For instance, in Germany in the early 20th century, the first institutions for children and young adults were founded, and Berlin opened its first juvenile prison (Borchert, 2016:33). The criminal justice system was criticised for providing poor conditions such as bad food and inadequate personnel, who often came from the military and received little education. Accordingly, changes were demanded for the system as a whole and for the youth sector in particular. The director of the juvenile prison that opened in 1912 in Wittlich (Rhineland Palatinate, Germany) based the prison's operation on pedagogic principles, which con-stituted a novelty at that time (Dörner, 1991:55). The personnel were trained in pedagogical work, and a new function was included amongst prison staff: the welfare provider (Fürsorger), who was responsible for preparing inmates to transition back into society. In addition, the new prison concept included elements of schooling and vocational training (Dörner, 1991:56).
Currently, carceral social work is accepted as a specialised field of social work and is defined, for instance by the US National Organization of Forensic Social Work (NOFSW), as "the application of social work to questions and issues relating to law and legal systems" (NOFSW, 2020). Therefore, just like the main profession of social work, carceral social work follows principles such as ethical codes, professional codes of practice or the principle of reflexivity (Healy, 2008;Becker-Lenz et al., 2013). In addition to the ethical debates in the social work profession in general, there are also specific ethical and professional discussions about the challenges carceral social work faces in particular and what its aims and principles are (Butters and Vaughan-Eden, 2011;Munson, 2011;Young, 2015). Classical social work divergences such as conflicting interests between clients and social workers or conflicting responsibilities of social workers are often accentuated in the carceral setting (Young, 2015).
Recent work about institutional settings of social work shows how the settings shape social work practice but also insists that social work is productive and produces space (Diebäcker and Reutlinger, 2018:5 ff.). The interlinkage between space and practice is for instance also debated with reference to the importance of corporeal co-presence in social work in contrast to electronically distant communication through current technology (Broadhurst and Mason, 2014). This echoes research in geography conducted about social work practice (Bondi and Fewell, 2003) and the practice of other professions that are situated in specific institutional settings such as nursing (Andrews and Shaw, 2008). In particular, when we look at street-work social work, it becomes apparent that social work also has a productive role in creating space. A recent study demonstrates critically how social work can become part of political initiatives to convert neighbourhoods into "clean-secure-social" (Dirks et al., 2016:123) areas. In establishing spaces for people who drink publicly, who take drugs or who are for some other reason not desired in the "new" public spaces, social work takes part in building a new spatial and social order (Dirks et al., 2016).
Analysing the spatiality of carceral social work therefore links the scarce literature on the spatiality of social work practice (for an example in the field of carceral social work see Maróthy, 2012) to the much more extended literature from carceral geography that points to a variety of dimensions of spatiality in the carceral setting. For instance, there are notions of the carceral as an immobilising power (Mincke and Lemonne, 2014;Philo, 2001) that controls bodies by fixing them in spaces, by counting them and confining them in certain spaces at certain times of the day, such as the cell dur-ing cell hours and the courtyard during walking time. Others point to the power of mobility within the carceral system, such as in prisoner transports (Gill et al., 2018;Moran et al., 2012), where the transports are used as a way to demonstrate power over the prisoners and their bodies by keeping them in suspense during the time between leaving one prison and arriving in the next one, without knowing where they would be brought. Spatiality is thereby understood in multiple scales, from the prison with its walls, buildings and yards to the cell or the very body. In such spatialities, there are various ways of exerting power or challenging it by creating spaces of resistance (Dirsuweit, 1999).
The analysis of carceral social work practice draws from such studies on carceral spaces and understands space, on the one hand, as the material context -a materiality that cannot be overlooked in prison, one that is not limited to walls and fences but also affects the layout of rooms, floors and the possibilities of encounters therein -exactly those spaces where carceral social work practice takes place. At the same time, these spaces are also produced through particular practices, in the sense that social practices always create, alter or reinforce spaces and spatial order (Lefebvre, 1974;Massey, 2005). As spaces are products of social processes, they materialise and reproduce social structure and norms (Cresswell, 1996). The prison building is a materialised expression of social norms about who should be punished and confined and in what ways. Such materialised expressions are experienced, felt and resisted with and in the body (Simonsen, 2007).

Empirical approach and project
The analysis is based on a short exploratory project conducted in two Swiss prisons 1 . The project aims at understanding social work practice in the carceral setting and focuses particularly on how the professional ethics and principles of social work can be maintained in a prison space. The close-up on carceral social work practice and its professionality showed multi-faceted practices of social work that depend on factors such as the history of the prison or the broader political tendencies and programmes on the criminal justice system (Emprechtinger and Richter, 2021). Increasingly, we became interested by the spatiality of this social work practice in these specific settings.
The initial example in the introduction points towards the immanent spatiality that transcended most of our research encounters. Entering a prison, even as a researcher, is a very physical and material act: in both prisons we had to leave most of our belongings outside in a locker and pass a body scanner; the massive doors, walls and fences mark the division between the outside world and the inside of the prison. At the same time, we rapidly observed that activities were not only scheduled in time but also took place in designated spaces.
This interest in the spatiality of carceral social work was certainly also fuelled by the methods employed: we started with shadowing social workers in the two prisons (four social workers in each prison, eight in total). The technique of shadowing is a type of observation that consists of following a person as a shadow. While the observer remains in principle a shadow, there are gaps, walks, pauses etc. where shorter or longer exchanges are possible between the social worker and those who followed (Czarniawska, 2012(Czarniawska, , 2014McDonald and Simpson, 2014). The method is often used in organisational studies or institutional ethnographies (Gilliat-Ray, 2011), as it provides insights into everyday actions and interactions during the working hours of a person. These observations were complemented by interviews with all the shadowed social workers to deepen the insights from the "shadowings". In particular, the interviews also provided the opportunity to reflect together with the social workers on the interactions and activities we had observed during the "shadowings".
Part of what we saw during these "shadowings" was planned or prepared in the sense that we depended on the head of the social service or the organising social worker for organising and scheduling our stay. Also the social workers knew in advance that we were coming and in part prepared some explanations or had documentation on hand to introduce us to their work. At the same time, not everything is plannable, and the meetings with the prisoners were far from predictable. While we had informed the social workers via their supervisors beforehand and signed a non-disclosure agreement, the consent of the prisoners was gained in verbal form. When they entered the social worker's office, the social worker presented the researcher, explained our presence as observing the social worker, not the prisoner, and asked whether they agreed to our presence. In both prisons, prisoners have the right to visit the social service and can ask for a meeting. The choice of not selecting prisoners by ourselves but leaving the selection to the daily work schedules of the social workers available allowed for a certain range of prisoners and meeting types. That we did no "shadowing" during therapeutic settings was due to the shortness of the project and is planned for a later in-depth study.
We were able to conduct our research in two Swiss prisons that represent, for Switzerland, rather large spaces with 200 and more prisoners. Prisons in Switzerland range from 5 to almost 400 places and the average amounts 73 places per prison in 2019 (BfS, 2019a). Compared to countries like the UK or the USA, Switzerland has a rather low incarceration rate, 50 per 100 000 inhabitants in 2019 (BfS, 2019b). The institutions for incarceration are organised according to the principle of progression. This means that prisoners, in particular those with longer sentences, enter closed settings and progress step by step to more open settings to prepare them for release. Shorter sentences are in general spent in open settings, where prisoners work for instance on the open fields. In the last years, risk orientedness has become a major element in the logic of the Swiss criminal justice system. While it is implemented differently in the various cantons, nevertheless the focus lies on reducing risk of recidivism and implies a special focus on prisoners who have been convicted because of severe crimes and who are supposed to have a high rate of recidivism. While forensic psychiatric is the driving profession in this development, social work has been assigned central tasks in this changed logic. The position and role of social work in the prisons varies. In German-speaking cantons the head of the social service is often also the vicedirector. The role also depends on the type of prison: while in pretrial units social work is reduced to a minimum, it constitutes one of the primary professions in institutions for young people.
Both studied prisons offer places for general sanctions, high-security sections, pre-trial areas as well as space for people with therapeutic measures. They are situated in different cantons in the German-speaking part of Switzerland, which allows for a certain variety regarding the legal and historical context, as the execution of sanctions depends on the cantonal legislation and politics.

Confinement 1: space and professional practice
Institutional spaces are often designated for certain functions they are meant to support. This is true for institutions in general (Philo, 2001), and prisons therefore are no exception. The panopticon by Bentham (Foucault, 1995) is a famous example how the institutional space of the prison is designed to support and even serve the function and aim of the institution to control. In more general terms, spaces, people and institutions are coupled in material and spatial terms (Diebäcker and Reutlinger, 2018:11). When social work takes place in an institution, this means that it takes place "somewhere", within a specific materiality and spatiality such as buildings, paths, walls, material artefacts. The spatiality of the prison is therefore designed to support and materialise the logic inscribed into the criminal justice system, a logic that ambivalently combines notions of custody and care: custody in the sense that it is meant to punish, control and provide security for society as well as the people within and care in the sense that it needs to take care of the people in custody. By restricting people's freedom, the institution has the obligation to care about their wellbeing (Marti et al., 2017). This first section focuses on the spatial context of the prison as the place where carceral social work practice is conducted or where this practice is confined to in order to become carceral social work.
During our observation, social workers we shadowed first took us to what we call their individual home base -their office space. It is a home base in several senses: there is all the material and instruments for working such as computer, telephone, files, books and other documentation; there are personal objects like cigarettes, family pictures, water bottle etc. It is space that the social workers have appropriated as their own within the prison. At the same time, it constitutes also a home base in the sense that social workers come back to it between meetings, visits and errands.
In most cases this office has a double function. Apart from representing a home base and a safe space, it is also used for meetings with prisoners. In general, prisoners are either called by social workers for meetings in the context of a rehabilitative training or as part of a programme or because the prisoner asked for a meeting via the prison communication system (usually small pieces of papers that are filled out by the prisoner and transmitted to staff). Apart from prison sections that are located too far away from the social workers' offices or where prisoners are not allowed to walk around by themselves, prisoners were called to come and meet the social workers in their office. The spatial organisation implies various notions. First, the prisoners leave other institutional settings of the prison to enter the institutional setting of the social service. Entering the social worker's office also points towards the power imbalance between prisoner and social worker, as the prisoner is called and dismissed by the social worker from this space. While this might not be directly intended, and put in place for practical reasons (as documents, computer, telephone are on hand), it implicitly mirrors the inherent and often discussed tensions within social work and its power imbalances, articulated in the dyad of controlling and helping (Bommes and Scherr, 2012;Emprechtinger and Richter, 2021).
Meetings with other professionals, prison administration or executive authorities and other authorities from outside the prison always take place in other spaces. These meeting spaces are located in neutral areas within the prison that are not attached to specific professions such as the office area of social work or the workshops where professionals supervise the prisoners' work. They are set up as rooms for meetings with large tables, perhaps a big screen for projections, and might have a piece of art for decoration but no personal objects. On the practical side, these rooms represent the ideal meeting space with respect to size and functionality. At the same time, they represent more than a practical institutional choice. They represent the institution and its professionality and therefore need an impersonal setting that underlines that decisions taken in these rooms about furlough, early release or other changes in the execution of the sanctions are meant to be objective and unbiased. Other meetings of the social work team take place in the areas attached to "their" part of the prison.
These different spaces define and structure how social work operates in prisons, and they are clearly designed for a certain type of social work, while other types of social work practice are inhibited by the spatial outline of offices and meeting places and by the ways in which they are used. For instance, one could imagine an approach where social workers initiate and organise activities and participate in everyday life such as in settings where groups of residents live together in a section as part of a more social pedagogical perspective. In the studied prisons, however, social workers take on the role of counsellors who sit at the desk and inform, help, evaluate and cooperate with other professionals.
Nevertheless, we also encountered examples of social workers who not only followed or used the spatial structure, but who also created spaces for new and different types of encounters. In both cases social workers initiated special projects to do with their practical work experience as part of their professional education. One consisted of a communal lunch, where prisoners ate lunch together at a table instead of sitting alone in their cells; another consisted of organising a Christmas celebration with prisoners: baking cookies, learning songs, reading A Christmas Carol and also having a dinner together, also including high-security prisoners without the usual handcuffs and shackles. Both initiatives were not institutionalised. Nevertheless they represent examplesalbeit rare -of how social work is not entirely structured by spatiality but also has the power and the potential to change this spatiality and create new spaces for different kinds of encounters that challenge confinement not only of prisoners but also of social workers.

Confinement 2: spaces and the availability of bodies
The spatiality of the prison also affects the intimate scale of the body. The body is a very contested space in prison: it is searched for drugs and other illicit objects when returning from visits or furlough, it is controlled and counted (Kantrowitz, 1996) when moving through the prison, and also its health is managed by a medical service that often is attached to the prison authority (Richter et al., 2019). Often the body also bears signs of the prison itself. It keeps the traces or the marks of being or having been a prisoner and signals the stigma by exposing missing teeth (e.g. Williams, 2007), tattoos or other signs of prison life including bad health conditions. At the same time, the body also constitutes a space of resistance. Early accounts such as the work by Dirsuweit (1999) show how in a South African women's prison, corporeal practices such as hanging blankets are used to create personal spaces of privacy within the spaces of control.
In such "total" institutions (Goffman, 1961), in prisons just as in nursing homes or psychiatric wards, the bodies of the detainees have a characteristic that distinguishes them from the bodies of other clients in different social work contexts: their availability. As described above, the prisoners are called by the social workers, when the social workers have time and are ready to see a prisoner and discuss their questions or concerns. Regular meetings are only scheduled for cases that are categorised as in need of a more intensive treatment because of the severity of their offence and the risk of recidivism (for a more detailed discussion of social work and risk assessment see Emprechtinger and Richter, 2021;Bauwens and Roose, 2017). Prisoners can ask for an appointment with their social worker when they need support. However, it is the social worker who calls the prisoner when it is convenient for the social worker, and in all our shadowings we did not experience once that a prisoner did not show up. Nevertheless, the availability of bodies, as they are fixed in the space of the prison, is, at least in principle, not equaled with a general availability of the person, as a social worker pointed out: The client does not have to come, he can refuse to come, when he is called. This does not have any disciplinary consequences. Our service is open to him, in contrast to other rules in the prison (social worker, interview, 12 December 2019).
The principle that social work constitutes an offer and not an obligation, however, remains more a principle than an actual option for prisoners. The availability of "bodies" was only discussed with other sections, such as the kitchen where the prisoner is working. Although the prisoners might not be obliged to follow the call of the social workers, they still have an interest in cooperating. Often the prisoners themselves have an inherent interest in a meeting and ask to meet the social workers because they have an issue they want to discuss or need support with. Even if this is not the case, though, they know that if they do not show up to a rehabilitative programme, this will generate a picture of someone who does not behave properly, even if it might not lead to disciplinary consequences directly. And when social workers give their judgement of the prisoners' behaviour for a report, this might affect them. The principle of social work as a free offer combined with the availability of bodies and the surveillance and judgement mode of the prison make carceral social work a "voluntary obligation". Social workers, therefore, exert soft power on the prisoners: on the one hand pointing to the voluntariness of their service while at the same time also counting on the institutional repercussions for not collaborating in programmes and rehabilitative work.
While we did not experience a prisoner not showing up when called, with no social worker mentioning such an example in the interviews either, the data nevertheless point towards another form of refusal: while bodies remain available in the prison space, this remains a mere material availability.
On the one hand, he comes to me to the meetings, on the other hand, I had to write a negative recommendation, as we could not work on the planned topics [relevant for offence treatment] (social worker, interview, 16 December 2019).
In particular, when prisoners are called in for regular meetings and are invited to "work" with the social worker and engage in a rather clinical setting of biographical work and other risk and offence treatment programmes, prisoners sometimes choose not to collaborate and refuse to continue. Although they are physically present, they counter the aim of the meeting by not "working" or engaging in the sense that the social worker had planned. While their bodies are fixed in their materiality and follow the prison rules, their minds are not as easily controllable. But as the above quote shows, there are indeed consequences for not collaborating while being physically present. There may not be direct sanctions, but the behaviour is recorded and will appear in future reports and serve as a negative recommendation.

Confinement 3: the spatiality of emotions
The geographies of prison life are not only material and corporeal but at its most intimate level also inherently emotional (Crewe et al., 2014) as prisoners, for instance, feel the loss of a loved one (Aday and Krabill, 2016) or other emotional distress. And there is a need for prisoners to find spaces where they can express their emotions, as showing emotions often is read as a sign of weakness, a form of opening one's intimate feelings to others, and giving an entry point for aggression and violence (Crewe, 2014). As even cells are often shared and can be inspected by prison staff at any time without the consent of the prisoners, there is little room for privacy. There is therefore a need for "emotion zones" (Crewe et al., 2014:64) where prisoners feel safe enough to show what really moves them emotionally. As these emotion zones are rare, managing emotions resembles walking a tightrope (Greer, 2002).
The fact that prisoners are in need, emotionally, was a topic that was raised in the interviews we conducted. Social workers explained that some prisoners needed someone to share their emotions with but also someone to listen to them, support them with their mere presence: Other [prisoners] are very much in need, emotionally. With one prisoner, for instance, if I just pull the paper and pen towards me to get up, then I see how he thinks of a next question, to keep me longer with him. This person is very much in need and this is also part of our work (social worker, interview, 14 January 2020).
Prisoners articulate their emotional needs differently. While in the previous example the prisoner rather asks questions to keep the social worker with him, the prisoner in the following example uses the strategy of talking to "discharge" all his accumulated emotions as in the following shadowing sequence: She tries to understand what he means by propaganda. The prisoner does not speak in an aggressive tone, but with time he gets louder and louder. . . . In total, I guess, he stayed there for 45 min. When he has left, P. [the social worker] tells me that this was not a typical social work conversation.
But she lets people speak, it is important for them. In the landing they are always under pressure to prove themselves, here [at the social worker's office] they can also discharge and talk freely (shadowing, 15 November 2019).
It therefore seems that carceral social work also explicitly provides space for emotions. The question is how this space is created in a setting, such as the prison, where showing emotions is usually equaled with showing weakness. The office of the social worker and the co-presence of both the social worker and the prisoner help to create a very specific space: a space that is marked by a prolonged bodily proximity of two people who usually are on different sides in the system while also effectively excluding other people such as prisoners or staff. In one prison social workers closed the door to their office and in the other they left the door ajar for protection because the concrete walls completely fenced off sound. The combination of these elements during a certain time creates a special space or an "intermediate zone" (Crewe et al., 2014:64) where emotions can have a place and where they are not judged but taken as personal expressions and where their articulation corresponds to a basic human need, which is recognised by professional social work.
The creation of these intermediate zones and temporary spaces of closeness is a material and bodily act. The proximity of the bodies creates a form of intimacy that is, as in the following example from a shadowing, often supported by other bodily expressions, such as smiling or leaning forward: There is a moment, when the prisoner and the social worker do not talk anymore about problems and possible solutions. The prisoner talks about his work at the prison and how he likes it. He talks about having to pay for his debt and no one else being able to do this for him. The social worker leans forward and starts smiling at the prisoner. I am not quite sure to understand what the prisoner is talking about exactly and what it has to do with the problems discussed before and I am neither sure, the social worker understands the content. There seems to be a sudden proximity between them. Then the social worker leans back, breaks the moment and they discuss again the prisoners' problems (shadowing, 14 November 2019).
In a certain sense, we glimpse a moment of humanity or maybe just a moment of good and well-trained professional understanding while observing social work in prison. What the moment exactly meant for everyone present is difficult to know, but it had clear markers of being a special moment different to the rest of the meeting. When the prisoner and the social worker stop talking about problems and concerns at stake and have an exchange that had nothing to do with the official reasons for this meeting, it is not "content" anymore that passes between them but emotions. The prisoner expressing his feelings about his duty, his sense of delight regarding the work he was doing and the social worker responding to this feeling with a smile. Whether the smile is a reaction of the social worker to cover up that he cannot understand what the prisoner is saying or whether it is a sign of interest in the person remains unclear and stays in the moment. Again, this emotional moment is broken corporeally by the social worker leaning back and orienting the conversation back to the topics at stake.

Conclusion
What are the material conditions of carceral social work if we conceive it not only as social work for incarcerated people but rather as a social work practice carried out in conditions of confinement? With this initial question we changed the focus from analysing social work as a practice with clients to analysing its conditions depending on the institutional and material setting. Analysing the conditions of work in the carceral setting follows a tradition of carceral research that focuses on prison staff, their role, working practices and conditions (Crawley, 2004;Liebling et al., 2011). It aims at understanding what it means to work in a setting of coercion, restricted options, strictly defined procedures and with people who are subjected to the total institution of the prison.
Our analysis outlined the materiality of social work in confinement at three scalar levels: spaces, bodies and emotions. This provides insights, on the one hand, for social work practice and for its professional stance towards reflexivity. On the other hand, this also shows, on a more general level, how a focus on materiality serves as a way to understand the (social work) practice in subtle and nuanced ways.
First, on the level of carceral social work and its professional claims, we showed how the material condition of the prison provides the setting for social work that is aimed at counselling instead of initiating activities, encounters etc. The materiality and the spatial outline of the prison are thereby an expression of contextual factors such as the resources employed for social work and the tasks assigned to social workers. Nevertheless, the spatial setting not only serves these institutional intentions, but it also enforces them. An active engagement and much energy is needed to initiate alternative usages of carceral spaces. Social work as a profession that understands reflexivity as a pillar of its professional claim remains thereby hardly attentive to the spatial and material setting and to the conditions this setting imposes on carceral social work. While both social workers who told us about the initiative of creating a space of encounter inside a prison as part of their education as social workers spoke of it as an important experience, they did not use this experience to question the low level of reflexivity of their profession towards its usage of space in the prison setting.
Second, on the level of materiality of institutions, it became apparent that the materiality is used and inscribed with meanings in subtle, uneven and changing ways. The space of the social worker's office provided a space secluded from the prison's eyes and ears, where more private conversations became possible while at the same time representing the institution of the social service with its powerful role of writing reports. At the same time, it represents the realm and power space of the social workers. While it defines carceral social work as counselling work, it also allows for distance with the prison as an institution. The importance that such safe spaces or "intermediate zones" (Crewe et al., 2014:64) acquire in a total institution resembles discussions on spaces of education (Richter et al., 2011) that provide a space for humane relationships between the prisoners as students as well as between the prisoners and the teacher.
Third, the analysis of materiality needs to focus on different scales within the institution. We outlined the level of the spaces, such as the social workers' offices or the meeting rooms with other professionals, the level of the bodies of the prisoners and how they are made available by the total institution of the prison, and finally the level of the emotions that are materialised in the body and need safe spaces to be pronounced. These scales are linked in complex ways. While bodies might be available, prisoners can refuse to cooperate and counter bodily presence with non-cooperation and emotional distance. At the same time the proximity of the bodies in safe spaces such as the social worker's office allows for emotional work in "intermediate zones", where emotions can safely be shown.
Finally, to focus on materiality and the social practices within and with it means to focus on complexity, subtlety and shades instead of drawing a clear-cut black-and-white picture. This also means to focus on possibilities that open up, once one has been able to dig deeper past the apparent institutional focus on rules, security and sanction. Data availability. Data are currently not publicly available as the project was of exploratory nature. Information about the data can be obtained through the corresponding author.
Author contributions. Both authors have equally contributed to data collection and analysis. MR holds, as PI, the responsibility for the project and is the lead author of the paper.
Competing interests. The authors declare that they have no conflict of interest.
Financial support. This research has been supported by the Haute école Spéialisée de Suisse Occidentale Valais/Wallis (Projet exploratoire).
Review statement. This paper was edited by Benedikt Korf and reviewed by two anonymous referees.