On 25 August 2017, the Swedish border police raided a weekend leisure camp organised by the Swedish church for families who lived under threat of deportation. The incident attracted significant media attention, since the police had breached the informal principle of church asylum; a principle that, with few exceptions, has historically been respected by the Swedish police.¹ Several among the apprehended families had open deportation cases, and one of the families was brought to the deportation prison where I was conducting fieldwork. One of the children, a young boy, wrote the letter to the Swedish king and the prime minister quoted above. He pleaded for the family’s release and for their deportations to be suspended, since his father feared for the family’s safety in the country they had fled from. The family had also appealed the deportation order, claiming there were new circumstances in their asylum case that had not been taken into consideration (in accordance with the Aliens Act, chapter 12, § 18–19). When I arrived at the deportation prison one morning, detention staff had just learnt that the family’s appeal had been dismissed. The police were already on their way to pick up the family for deportation. They would be there within ten minutes.

In the common room, the atmosphere is tense. I stay there with two detention officials while two others go to wake up the family, staying in the section for women and families. The family appears a few minutes later. The father is on the phone, unsuccessfully trying to reach their lawyer. The young boy starts making breakfast sandwiches for himself and his younger sister, carefully buttering the pieces of tin loaf bread, to save for their journey. The mother hugs some of the other incarcerated women who have stood up to say goodbye. Staff are quietly standing by. They are notified that the police have arrived, and the family is escorted to pick up their luggage and retrieve their personal belongings, which had been confiscated when they were incarcerated. When they leave, they will exit through the back door and enter directly into the police car, which is parked in an enclosed hangar attached to the building. From there, they will be escorted to the airport. Back in the common room, the staff slowly resume their routine tasks of filling up the coffee thermoses for breakfast and inspecting the detention premises. I stay with two senior detention officials, Greta and Livia, who are discussing what just happened.

I follow Zita, another detention official, as she goes to clean out the family’s room. Sheets, towels, and cups with the family members’ names on stickers are collected to be washed up. The name tags are removed, and plastic dinosaurs are put back into the box of toys that was brought in for the children. Zita tells me that this was a particularly difficult case. Not only did it involve children, which for her is always upsetting, but it was also mediatised because of the police raid in the church and the young boy’s letter in the newspaper. Detention staff who spoke to the family had also got the impression that the father’s fear of returning was genuine. Zita explains that she often finds deportation emotionally strenuous, ‘It’s hard to know how to react when they leave. Shall we wave goodbye and smile? Not really. Or look sad? No, you know, we must keep our act together.’ This time, it was hard. But, Zita concedes, as we carry the box of toys back to the storage room, her opinion doesn’t really matter. The outcome of the asylum case is not in her power to decide.

The incident, which took place at the beginning of my fieldwork, illustrates the contestations over deportation that take place within and beyond the walls of the deportation prison. The vignette exemplifies some of the strategies that people threatened by deportation may use in their struggle to remain, the suspicion that resistance evokes among state officials (who called the uses of legal and public appeals ‘dirty tricks’), and how deportation is always a matter of (highly asymmetrical) contestations until the very last minute of enforcement. It also shows the ambivalence with which detention staff move between serving coffee in personalised coffee mugs, ‘waving goodbye’ to people bound for deportation, and partaking in the enforcement of this turbulent process. This ambivalence encapsulates the oxymoron of a ‘humane and dignified’ detention regime (DeBono et al., 2015: 19), where welfarist care and compassion are mobilised to justify and smoothen incarceration and deportation enforcement.

If Ellebæk deportation prison was a spectacularly punitive institution, ridden by orchestrated neglect and violent anticipation, the Swedish detention regime exemplifies a rather different governing logic. In Swedish deportation prisons, care and compassion are mobilised for the purpose of facilitating ‘smooth’ and ‘humane’ deportations, and significant efforts are made to downplay the inherent violence of the deportation process. As critical researchers on detention and deportation enforcement have shown, sensitivity and care are ‘perfectly compatible with brutal systems of control’ (Gill, 2016: 17). The chapter describes a detention regime where state violence is meticulously calculated, codified, and performed under the guise of humanitarianism, for the supposed good of the people who are about to be deported. In the final section, I reflect on the continuities and the differences between the Danish and Swedish deportation prisons, and what they have to say about ‘the state’ and its performances of punishment and control.

A ‘humane and dignified’ detention regime?

Shahram Khosravi (2009: 41) has argued that the Swedish term förvar, which literally translates as ‘warehouse’, well captures the impersonal infrastructure set up not ‘to tend or to treat’ but to keep incarcerated people available for investigation or deportation. Yasmin, the official quoted above, agreed that the term had dehumanising connotations. Yasmin, who was in her mid-twenties and was working part-time in detention while finishing her master’s degree in human rights, was uncomfortable with the official name of her workplace. When friends asked her about her job, she told me, she would usually just tell them that she worked for the Swedish Migration Agency, or simply, ‘for the state’. Yet with all its uneasy connotations, the terminology is illustrative of the rationalising and depoliticising approach to deportations that prevails in Sweden. We can understand förvar as an ‘alibi term’, a political artefact (Lecadet, 2018), which encloses the paradoxes of a ‘humane’ and ‘dignified’ deportation regime.

The notion of a ‘humane’ deportation apparatus dates back to 1997, when the Swedish Migration Agency took over responsibility for running deportation prisons from the police. Prior to this, people detained on administrative grounds in accordance with the Aliens Act were held in police custody, supervised by private security firms. Following reports of violent incidents and degrading treatment by staff, responsibility for the operation of pre-deportation detention was subsequently handed over to the civil migration authorities (see Khosravi, 2010). In the preparatory works of the new law on detention (Prop. 1996/97: 147), it was underlined that the deportation prisons should henceforth retain a civil character and should ‘remain as close as possible to reception centres’ (Prop. 1996/97: 2), while any resemblance to prisons should be minimised. The vision of a civil detention apparatus was later codified in the 2006 Aliens Act, which stipulates that the prisons should be run in a ‘humane and dignified’ manner. This principle has been codified and inscribed into institutional practice. It is manifested in the setup of deportation prisons, which are low security facilities with relatively high material standards, and in the strong emphasis and preference that authorities place on voluntary over forced deportations (DeBono et al., 2015). In contrast to Ellebæk prison, Swedish deportation prisons are designed to decriminalise detained people, and ascribe staff a considerably more proactive role in their detention and deportation cases.

The use of deportation prisons for the purpose of controlling ‘illegal immigration’ has continuously expanded since the mid-2000s (Jansson-Keshavarz, 2016),² and the number of people incarcerated annually has been on a steady rise (from 1,645 in 2008 to 3,200 in 2014 and to 4,295 in 2019).³ The maximum length of stay is twelve months, while the average time of detention was fifty-five days in 2020 (Migrationsverket, 2020). In spring 2021, there were six operating deportation prisons with a holding capacity of 519 people, and there are plans to build a seventh prison in Northern Sweden. The deportation prison where I conducted fieldwork was inaugurated in 2011. At the time of my research, it had the capacity to incarcerate eighty people in two separate departments, with a separate section for women and families. A renovated warehouse building, half of which still served as storage for an electronic utility service provider, the prison was located next to the highway outside a de-industrialised small town. Its architecture reflected the political ambition to retain a civil-oriented detention regime: for unknowing passers-by, there was little revealing that it was a site of incarceration, unless they happened to notice the inward-tilting metal fence crowning the tall walls that flanked the main entrance.

Staff members, lawyers, and visitors would enter the prison through the front doors, which required an electronic key, and which led to a small reception, offering tea, coffee, and some sweets. On the reception desk there was a small sign saying, ‘You only have three choices in life: Give up, give in, or give it all you got.’ To reach the spaces where those incarcerated were held, one would have to pass through four additional locked doors. Incarcerated people would enter the building from an altogether different place; they would arrive and leave in police cars or with the prison service’s escorts and be shuffled through the electronically monitored gates at the back of the building. Inside the prison, secure doors locked with numeric codes and swipe cards and thick, plexiglass windows kept the people detained confined from the outside world. However – and again, in line with the decriminalising ambition of the detention regime – the prison’s interior design was supposed to induce a sense of relative freedom, permitting incarcerated people to move around between their dormitories, where up to four people would sleep in the same room, and the common room, where the meals were served. The common room, which was the room where staff also spent most of their working hours, was decorated with pots with plants, some framed IKEA posters, and pieces of handicraft and artwork made by incarcerated people. A TV was constantly on, alternating between Swedish news, cooking shows – and reality TV shows (see also Canning, 2019a). Incarcerated people also had unrestricted access to a computer room, a gym, and a library room containing a few books. Smokers could go outside into a confined space called the ‘smoking cage’.⁴

In her writing on the architecture of migration-related detention, Chak (2016) argues that the very idea that we can mediate and manage the endemic violence of these institutions serves, in fact, to reify it. The aesthetically appealing and comfortable architectures of confinement become ‘a tool of moderating violence […] so that we don’t question the logic behind their very existence’ (Chak, 2016: 17; see also Keshavarz, 2018). Marianne, the director of the deportation prison and herself a former prison officer, wished that the logic would go unnoticed, even to those incarcerated. Marianne explained that they therefore made efforts to create a ‘homely’ atmosphere that would ‘limit the number of occasions when detainees are reminded that they are confined [frihetsberövade]’. For detention officials, too, it was important that any association to prisons was eliminated. On my first visit to the prison, I met Joseph, who had studied public administration but spent the past decade working in psychiatric care. He initially worked ‘on the floor’ in the prison but had now assumed a position as decision maker and was monitoring detention decisions and undertaking quality evaluations. He took these legal responsibilities sincerely and liked to express this by making sure to dress smartly. He was the only detention official I met who wore a shirt and a suit jacket for work. Joseph sat me down in his office, located on the second floor of the building, took the Swedish Aliens Act from the bookshelf, and read in its commentary, ‘It is evident that detainees should not be equated with prisoners […] when it comes to our possibilities to restrict their everyday life.’ He continued,

Marianne and Joseph’s emphasis on the non-punitive, and even ‘homely’, character of the deportation prison, and the service-oriented role of staff, contrasts with the criminalising approach to incarcerated migrants characterising Ellebæk. With its prioritisation of caretaking and maintenance functions, the Swedish deportation prison represented what Khosravi (2009: 44) has termed a ‘hostile hospitality’, where care and compassion function as disciplining and responsibilising mechanisms. Hence, productive tensions were formed between care and coercion, hostility, and hospitality. These seeming tensions – which as Chak notes, ultimately work in a mutually reinforcing manner – were reflected in the design of the deportation prison, and in the role of the staff.

Deporting with care

The composition and role of the staff in the Swedish deportation prison differed significantly from Ellebæk. The Swedish deportation prisons employed people from a broad variety of occupational, social, and national backgrounds. Some were students of law, criminology, or human rights; they would usually work as caseworkers (handläggare) and were in charge of monitoring detention orders and maintaining communication with lawyers and other agencies involved in the deportation process. Three of the caseworkers were actively working on deportation cases, which entailed holding so-called return dialogues and pressuring detained people to cooperate with authorities. The deportation prison also employed ‘supervisors’ (handledare) who worked on the floor in the detention wings, and who oversaw the day-to-day logistics, such as serving meals, making the daily inspection rounds, and monitoring the people detained and answering their questions. Staff holding these roles had diverse occupational backgrounds. Some had previously worked within the Prison and Probation Service or the military; others had worked within the healthcare sector or the National Board of Institutional Care, and yet others were truck drivers or former warehouse workers.

Marianne, the director, took pride in the deportation prison being an ‘ethnically diverse’ workplace (even though all managerial and decision-maker positions were occupied by white staff members). Marianne particularly valued how staff members with ‘non-Swedish background’ (icke-svensk bakgrund, a term encompassing first- and second-generation migrants, in this context in particular denoting racialised people with what is commonly referred to as ‘non-European background’) held language skills and ‘cultural competences’, which they could use to smoothen interactions with detained people. Several detention officials also had personal experiences of arriving in Sweden to seek protection, of the asylum system, and of encampment. Jaromir, who had arrived in Sweden after he fled war in former Yugoslavia in the 1990s, was one of them. He said, ‘I’m empathetic to their situation as I was also a refugee once, and I know what it’s like to be forced to wait.’ Jaromir saw his own experience of forced displacement as something that facilitated a common understanding between himself and the people detained. In his research on detention and deportation in Sweden, Khosravi (2017a: 175) discusses how the notion of cultural proximity of racialised staff members with detained individuals is instrumentalised for deportation enforcement. He concedes, ‘the ‘potentiality of migrants’ cultural competences is recognised and valued only in the service of expulsion of other migrants’. The celebration of ‘diversity’ among staff is thus not antithetical to racism, but an affirmation of the importance of race as an ordering mechanism in the deportation prison (see also Bosworth, 2018).

In contrast to Ellebæk, detention officials in the Swedish deportation prison had a proactive role in deportation enforcement. For this purpose, staff were expected to mobilise all their legal, interpersonal, and psychological skills, language, and cultural competences – and, as a last resort, physical force. Their involvement in the deportation process could indeed entail anything from dusting off a bag of plastic dinosaurs and baking cakes with detained children (see Canning, 2019a), to engaging in actual casework, to assisting in deportations. Regardless of their formal mandates, all staff members were expected to partake in facilitating deportation enforcement. Joseph told me, ‘The prospects of return should be present throughout our conversations with them. If we come to discuss for instance passports, we can tell them that well, if you get your passport, you will be out of here sooner.’⁵ According to Joseph, the principal role of staff was to ‘help detainees to figure out their attitudes towards leaving Sweden and make them understand their situation’. One way of doing so was by manipulating their hopes, dreams, and future aspirations. Joseph put it bluntly, ‘detention is a graveyard for dreams. Here, we help them bury their dreams of a possible future in Sweden and plant a seed of a dream of another future elsewhere.’ Killing incarcerated people’s dreams was an act of care, he insisted, since the vast majority of those detained had no prospects of obtaining legal residency in Sweden. It was therefore better that they aligned their dreams with the state’s deportation fantasies as soon as possible, rather than maintaining ‘false hopes’ of getting a chance to remain (see Lindberg and Edward, 2021). Staff therefore considered detained people’s attempts at resisting deportation by filing appeals, refusing to cooperate, or physically resisting deportations a ‘waste of time’ (Eule et al., 2019), whereas swift and effective deportations were a more ‘humane’ way to an inevitable end. Charles, a rowdy caseworker who had worked in the prison since its inauguration, explained that his strategy was to make the inevitability of deportation clear to the incarcerated person from the very beginning, to discourage any attempt at resistance.

Charles not only holds high trust in the enforcement capacity of the Swedish police; he also makes it clear that incarcerated individuals carry the responsibility for their own confinement, and for the violence that might eventually be used on them, should they refuse to cooperate.

As Walters (2016: 438) has argued, the ‘political dream’ of deportation is ‘that the migrant places themselves on the plane, without the need for guards, restraints or any spectacle of enforcement’. However, officials’ own deportation fantasies of swift, rational, and humane deportations regularly failed to materialise, not only because of detained people’s resistance, but also due to bureaucratic or political hurdles, and to the uneven application of laws such as the Dublin Regulation. Like their Danish colleagues, detention staff encountered people who got stuck in repeated Dublin procedures, or who were oscillating back and forth across the Danish-Swedish border. Marianne commented on these cases with a mix of fascination and frustration, ‘It’s magical – magical! – You drop them off at the airport one day and two days later the police have apprehended them, and they are back here again’ (Eule et al., 2019: 158). Such circular mobilities were far from the smooth, linear deportation process that detention staff envisioned. The staff did not only feel unease with failed deportation attempts; Greta, a senior caseworker with several decades’ experience working in deportation prisons, also found ‘successful’ deportations taxing, especially when they involved coercive force.

For Greta, deportation was not necessarily a successful outcome of incarceration; unless the person in question agreed to travel, she said she preferred that they were released. When she sat down with the people awaiting forced deportation to Afghanistan (deportations that are prioritised politically but contested, and which have been met with fierce resistance from non-deported people and their support groups), she witnessed their fear and anxiety, and felt that fear sticking to her, too. Such confrontations were rare, and something her colleagues would selectively avoid. I was also told that authorities sought to alter the deportation prison from which chartered flights departed, as a way of diffusing resistance attempts but also of easing the emotional burden on staff. Indeed, the deportation prison was structured in such a way that it was easy for staff to neutralise the violent realities of detention and deportation, casting themselves as caregivers, and displacing responsibility for the harms of detention from the system onto the detained people themselves.

Seeing like a welfare state

Detention officials’ morning shift commenced at 07:00. After a brief meeting, where incidents from the previous shifts were reported and new arrivals and departures announced, the officials would usually sit down with a cup of coffee on the couches in the common rooms. The couches became the starting point of my fieldwork. Since staff were not allowed to be alone with detained people, they would always move around in pairs in the common room and in the dormitory corridors or sit down two and two (or as was often the case when the morning shifts began, five or seven) on the couches, placed in front of the TV. From their position, they had a good overview of the room.

Figure 3.1 Evacuation plan of a Swedish detention centre

Sitting on the couch did in no way equal passivity, they explained. From the couches, officials would ‘feel’ the atmosphere, identify changes in the behaviours and attitudes of the people detained, and, they explained, make them ‘feel seen’. Greta told me, ‘You have to be able to read the atmosphere. You’ve got to have your tentacles out, get to know them a bit, so you can get a sense of what’s going on … and maintain eye contact. I think that’s important, making them feel seen.’ Greta would greet every person who passed by the couches with a jolly ‘good morning’, and always address them by their names. This, she explained, made those detained feel recognised and enabled staff to establish ‘humane’ relationships with them. In addition, ‘seeing’ was also a means to exercise control. Apart from making observations from the couches, staff undertook inspection rounds in the prison five times a day, where they entered detained people’s rooms, conducted headcounts, and checked the facilities for any damage. Marianne explained, ‘some think it’s a lot of control, but it’s because we are responsible for these people, their health and well-being – what if someone lies dead in their room and we don’t realise until the day after?’ If a staff member observed anything they deemed suspicious, it was noted down and shared at the next staff meeting. These reports could contain information such as ‘Michel and Amadou seem suspiciously interested in the kitchen and the windows’ (interpreted as an indication that they were planning an escape attempt), ‘Hannah isolates herself and does not speak to anyone’ (leading to her being ‘flagged’ for mental ill-health and risk of self-harm), and ‘David is anxious about his deportation’ (warranting a meeting with a caseworker).

In her description of a Swedish deportation prison, Canning (2019a) elaborates on how the multifaceted practice of seeing detained people enabled staff to combine care with pacification and control. The welfarist gaze serves to make detained people intimately known to authorities, registering not only their attitudes and behaviours, but also their dreams and aspirations. While officials’ compassion and concern for the well-being of detained people were surely genuine, their recording gaze was also perceived as controlling and intrusive by those incarcerated. Reza, one of the detained men who had spent several months in the prison and who would occasionally sit down with us on the couches, noted, ‘They are always sitting around here, doing nothing but stare at us.’ David, the man who was anxiously awaiting deportation, shared that the routine inspection rounds stressed him out, because each time, he feared they were coming to pick him up for an unannounced deportation. Unannounced deportation procedures were activated if the person in question had previously resisted or declared that they would not cooperate in the deportation process (see Borrelli, 2021); the case of the family described in the beginning of this chapter is one example.

In addition to their caring gaze, staff sought to mediate the harmful impact that the deportation prison had on incarcerated people by encouraging them to partake in daily activities. Angelica, a physiotherapist who was responsible for the scheduled activities, explained that the idea of the activities was to make them ‘wind down and relax, get a sense of routine in their everyday life, and help them maintain the diurnal rhythm’. The activities on offer included circuit training, Saturday bingo, basic pottery activities (such as making pearl pegboards and bracelets), ‘krim’ yoga – which Angelica explained was a form of ‘calming’ yoga developed for locked environments – and baking cakes (which, I was told, was particularly popular when there were children detained; see also Canning (2019a) on officials’ use of sticky chocolate cake to cope with sticky time in confinement). At the time of my visits, the weekly schedule with activities pinned on a whiteboard in the common rooms contained the following:

Angelica admitted that the activities might seem banal and infantilising, and that they were usually poorly attended. Still, she insisted, activities could ‘help detainees take their minds off bad things’, such as the pains of imprisonment, or their anxiety over deportation. She continued,

Angelica rightly identifies the harmful nature of the deprivation of autonomy and control over one’s own life and daily routines that is part of confinement (Turnbull, 2018). Allowing detained individuals to refuse participation in infantilising activities might seem like a feeble compensation for this deprivation of ‘free choice’, but the activities also served to channel negative emotions such as anxiety, fear, and anger into manageable forms, such as a pearled bracelet stating ‘fuck the police’. This way, the activities served to displace incarcerated people’s attention from the structures and conditions causing their anger, anxiety, and deprivation of liberties in the first place towards ‘manageable’ feelings expressed in pacified forms.

The management of harms caused or amplified by imprisonment also took place through extensive documentation practices. Through standardised routines, forms, and protocols, staff were to ensure that risk behaviours (such as escape attempts, resistance to deportation, or acts of self-harm) were recorded and, ideally, prevented. The forms also legitimated expanded coercive controls: they authorised staff to confiscate the personal items of incarcerated individuals, which they deemed could be dangerous, and to enter their dormitories several times a day and wake them up for ‘headcounts’. They also authorised comprehensive visitations, where personalised spaces and items were inspected, ostensibly in the name of ensuring the ‘safety and security’ of the people detained, while those detained were held confined in a separate room. I witnessed a couple of those visitations during my fieldwork, which were conducted since staff were suspecting that some of those detained were plotting an escape attempt. All detained men (the section where women and families resided was not searched) were then ordered into the gym where they were offered lemonade while the staff, equipped with plastic gloves and bags, searched their rooms and personal belongings. Before being let back into their rooms, all detained people were asked to sign the visitation order.

The welfarist gaze, which was supposed to prevent disruptions, resistance, and harm, did not only scrutinise detained people’s belongings for potential causes of such disruptions. A contested issue at the time of my fieldwork was the recently introduced suicide screening forms, which were mandatory for detention officials to fill in for each detained person upon arrival. The protocol, I was told, had been developed after a detained man died from hanging, and was supposed to prevent similar incidents by making staff screen detained people for suicidal intent at the point of their arrival in the prison. The protocol included questions such as, ‘Have you ever attempted suicide? Have you ever been treated for depression?’ and, ‘Have you ever felt like life is not worth living?’ A suspicious answer to any of the questions could lead to solitary confinement or referral to a psychiatric clinic. Staff found the questions to be blunt, but straightforward. While understanding of their intent, they were ambivalent about having to use them in practice.

I sit down with Livia who is about to begin the registration screening of an incarcerated man. He is reserved, and she is tired, as she hasn’t had time to have her morning coffee yet. He speaks some Swedish, so Livia first tries to hold the conversation without an interpreter to get it over with. We have been in the room together for about two minutes, going through the formalities of the protocol, when she double-clicks on the suicide screening form on her computer. It instructs her to ask if the man suffers from any illnesses: ‘Physical? Psychological?’ – to which the man answers ‘yes’, ‘yes’. ‘Did you ever try to commit suicide?’ is the next question. The man doesn’t understand and asks her to repeat the question. Livia gets nervous, starts gesticulating, and tries to signal what she means: cutting wrists, swallowing pills? The man still looks puzzled, but answers, ‘yes’. Livia turns to me, ‘no, this doesn’t work. We need an interpreter.’ An interpreter joins us via telephone, and Livia tries to continue with the questionnaire. The man looks down at the table. He says he is stressed out and needs to have a smoke. He hasn’t smoked for twenty-seven hours, he says, so he needs it badly – and he needs his medicines. Livia persists and wants to finish with the protocol, as the man is not formally allowed to enter the prison facilities before the suicide screening is completed. She continues, ‘have you been treated for depression? … have you ever felt like life is not worth living?’ ‘I’m mentally ill’, the man insists, ‘I need my medicines. Why do you say you will help me but instead you put me in prison and want to deport me?’ He is now visibly upset and makes clear that he will not answer any more questions. Livia skips most of the remaining questions and chooses to focus on confiscating his phone, which also needs to be done before they finish. Eventually, the man is allowed to enter the prison and goes out for a smoke. Livia turns to me, ‘it’s totally worthless, this suicide screening. So intrusive. I think you just have to trust your own judgement.’

The vignette not only demonstrates the awkwardness with which the suicide screening was used in practice. As the detained man points out and Livia admits, it is also a shallow performance of care, unlikely to capture the state of mind or mental health of the detained person, and likely to create confusion, if not animosity. While ostensibly put in place for the safety of those incarcerated, such documentation practices also obscure the original cause of the violence. Indeed, the protocol can be understood as a paperwork performance (Borrelli and Lindberg, 2020), which serves to write off institutional responsibility for harms either directly caused, or likely to be aggravated, by incarceration and the threat of deportation. In her research on migration-related detention in the United Kingdom, Mary Bosworth (2016) comes to a similar conclusion regarding the role of bureaucratic forms, which primarily serve to ensure that staff have their ‘arses covered’ in case a detained person suffers any mental or physical harm in confinement. She concludes that bureaucratic forms magically transform detention from a violent realm into a humanitarian one (see also Fischer, 2015), effectively averting the question of what the original cause of harm is.

White lies, ruptures, and resistance

Through the mechanisms described in the previous section, detention officials carved out a ‘humane’ role for themselves in the Swedish deportation prison. This role enabled them to calculate, codify, and depoliticise the violence endemic to confinement. Uses of coercive force were principally outsourced to other state actors – the police and the prison service, who were authorised and trained to use such force – while the systemic violence of confinement was mediated through the forms and bureaucratic routines that served to mediate the harms it inflicted on incarcerated people. There were also bureaucratic scripts for how detention officials should manage their emotional reactions to their work. These scripts, I argue, can be understood as white lies. They were ‘white’ lies in that their purpose was to avoid hurting the feelings of staff, but they were also white in the sense that they were premised upon the assumption that the deportation prison was a fair, just, and necessary element of the border regime and the global mobility apartheid it serves to sustain. This assumption, in other words, was premised upon whiteness as the norm and affirmed and normalised the racial order (Bonilla-Silva, 2019) that the border regime is premised upon. To sustain the lie that detention was useful and legitimate required that staff developed stories about their work that foregrounded some of its elements while downplaying others.

Sara and Peter were two white junior detention officials who usually worked night shifts, since it was compatible with their academic studies, and since the nights offered more time to socialise with detained people. One evening, as we sat down on the couches in the common room, Sara, who was training to become a lawyer, told me that detention was the only part of the Swedish Migration Agency she would feel comfortable working for. Making life-altering decisions in the asylum office would have been way too emotionally challenging and morally problematic for her, she explained, but in detention, she was merely enforcing others’ decisions. Peter, who had studied philosophy at university, added,

Peter refers to the post-Second World War Nuremberg Trials, where high-ranking Nazi officials were tried for war crimes, including the Holocaust. Peter invokes Hannah Arendt’s (1963) famous coverage of the trial of Adolf Eichmann, which informed her writing on the banality of evil, where she develops a critique of the bureaucratised system that permits officials to thoughtlessly follow murderous orders. Yet Peter claims that, in contrast to Eichmann, the system he and Sara are working for is fair, democratic, and just. While not prepared to take decisions, they are confident that others are doing their job, permitting them to deflect responsibility for the ‘orders’ they were enforcing among the ‘many hands’ of bureaucracy (Thompson, 1980). Just like Livia and Greta discussed in the introductory vignette, Sara and Peter also felt misrepresented when the media and activist groups criticised them for controversial detention or deportation cases. Sara said,

Sara and Peter’s belief in the Swedish deportation regime is based on their trust in the infallibility of Swedish bureaucracies, on the one hand, and on the idea that their ‘humane’ work stands apart from the deportation system, on the other. In their research on staff perspectives on detention, Puthoopparambil and colleagues (2015) similarly find that staff variably positioned themselves as proud enforcers of a fair and just immigration control system, which they trusted, yet were not personally responsible for, or, as detached from the system, insisting that their role was only to care for people in a vulnerable situation, not enforcing deportations (see also Wettergren, 2010). This was something I discussed with Lars, the director of another deportation prison, where I spent a day conducting fieldwork. Lars had a background in the military and a degree in psychology. He had elaborated his own sociological theory of staff emotions, which he outlined for me over a coffee.

The walls of Lars’ office are covered with books. On the shelves, I spot titles about Swedish migration law, legal philosophy, political theory, prison research, and psychology. Lars says that many detention officials are struggling to make sense of their role in detention and the larger political and social structures they operate within. For him, it is important that his junior colleagues grasp these layers of complexity. Otherwise, they might end up in a situation where staff blindly ‘follow orders’ without further reflection – the ultimate implications of which, Lars explains, are captured in Zygmunt Bauman’s Modernity and the Holocaust, or Hannah Arendt’s Eichmann in Jerusalem, two books that Lars picks down from the shelf. He begins to draw an image on a whiteboard to illustrate his point. I copied the image into my notebook (see Figure 3.2).

Figure 3.2 Notes based on drawing by Lars

Lars contends that whether we are in Hitler’s genocidal dictatorship or in the Swedish democratic welfare state, officials should not follow orders blindly, or ‘thoughtlessly’. Therefore, Lars wished for detention officials to perform their tasks fully aware of their implications; they may – and should – be critical of the system, and it is through this critique that they will sustain its legitimacy. Indeed, the critique offered by Lars, Peter, and Sara remains within the self-referential logic of the benevolent and fair state system, and therefore, also builds on a reductionist and individualist understanding of this system. When comparing – and contrasting – their work to the horrendous acts of extermination perpetrated by state officials in Nazi Germany, they trivialise and render impossible a critique that highlights how contemporary formations of encampment and expulsion, and the bureaucratic management of life therein, take inspiration from and build on previous ones. They also denounce the possibility that such violence is inherent to and has indeed underpinned the historical development of ‘liberal democracies’, including the Swedish welfare state. This is not to say that Swedish deportation prisons are like Nazi extermination camps. Yet, when the officials evoke the comparison with the Holocaust, and the reflections of Hannah Arendt, only to mark their own difference and resistance to this analogy, they refute not only the moral implications of such analogy (and the critique of their position it would inevitably provoke) but also the existence of a relationality between ‘the modes of racial governance and orders of racist exclusion, humiliation and death’ (Goldberg, 2009: 1280) upon which border regimes ultimately are built.

The enclosed system of emotional and moral comfort, which staff created for themselves, enabled them to maintain the white lie that their work is good, humane, and possible to set apart from other historical and contemporary manifestations of state violence. Yet, the humanitarian inclinations of the staff were founded upon and necessitated a dehumanisation of those incarcerated, whether as villains warranting monitoring and control, or victimised figures ‘needing’ their care. Hence, their justification mechanisms must not only be considered as ‘coping mechanisms’ but as mechanisms of upholding racial hierarchies in detention. As such, this system of legitimation was fragile, regularly disrupted, and vulnerable to criticism. Staff members’ belief in a humane deportation prison was disrupted during forced deportations, which several staff members found difficult to handle and preferred not to partake in. It was also challenged by the processes of gradual securitisation of the deportation prison, which was underway during the time of my research, and which challenged officials’ insistence on working for an essentially non-violent detention regime. Marianne explained to me that while detention used to be considered ‘dirty work’ within the Swedish Migration Agency, the increased political focus on deportations had made it more publicly acceptable to talk about the prison ‘for what it was: a place where people are locked up’. In contrast to their colleagues in Ellebæk, the Swedish detention officials had limited training in using physical violence; instead, they built on what they called a ‘dynamic security’ approach, which entailed ‘building trustful relationships’ with detained individuals and using dialogue techniques to ensure their ‘calm’ and cooperation. As Livia once noted, ‘You are less inclined to beat up somebody you know and have talked to!’ However, this approach to security was changing with the ongoing restructuring and ‘professionalisation’ of the operation of deportation prisons, which essentially consisted in reinforcing control and security measures in the prisons. According to the new rules, which were implemented all over the country, staff were supposed to carry more sophisticated alarm systems, and be trained in self-defence and so-called pacification techniques, which staff were supposed to use if, for instance, a detained person tried to escape. Some staff members, notably those who had a background in the prison service, the police, or the military, welcomed this trend, while others remained sceptical. Greta told me, ‘I don’t like this new security logic and the harsher attitude that comes with it. I prefer talking to them, calming them down when they are upset, but these aspects are forgotten with all this focus on security.’

Officials’ discomfort with becoming directly involved in coercion can be illustrated by an event that took place on a night shift during my fieldwork. That week, a group of young men had staged a series of escape attempts, and one of them eventually succeeded in escaping through the courtyard. To prevent further escapes, staff stepped up control efforts, constrained the freedom of those incarcerated, and conducted several of the above-mentioned comprehensive visitations, where they searched through their belongings for sharp items or other prohibited objects that they suspected could have facilitated the escape. During one of these searches, staff found tools that they believed could be linked to the escape. Two young men were singled out to be punished for their alleged involvement in the incident. These are my notes from the evening after the comprehensive visitation.

I arrive too early for the night shift. Some detention officials are discussing the two men who have been placed in solitary confinement.⁶ One of them is suspected of involvement in the escape attempt; the other man has had a verbal conflict with staff over the opening hours of the smoking cage. The exchange ended with the man throwing his coffee cup on the ground, so that an official was speckled with coffee stains. Ronny, one of the supervisors, explains that this man also ‘has a history of being obstinate’. The officials are now discussing an incident that just occurred in the solitary confinement cell, when the man who threw the coffee cup called upon staff because he needed to go to the bathroom. The two officials who attended to his call had then ordered him to keep the door open so that they could keep an eye on him because, as they said, they wanted to make sure he did not commit any act of self-harm (even though he had not demonstrated any intention of doing so). The man had refused, since he found the situation degrading, so staff had let him back into his cell where he was compelled to urinate on the floor. Ronny sighs and tells me that they are uncertain of how they should deal with this. Should they wipe the floor for him or let him do it himself, with the risk that he uses the broom to threaten staff? Zita deplores that the man now complains about feeling dirty, and snorts, ‘He has caused this situation himself – and now he calls us racists! Ronny offers to go and clean the room if someone keeps him company and watches over the man. Suddenly reminded of my presence, he turns to me, boxes me on the arm and says, half-jokingly, ‘wait, perhaps we shouldn’t talk about this when you’re around’! I don’t say anything, but in an attempt to smooth over the conversation and change topics, Ronny starts to explain that he has nothing against those who are detained, and although they are often accused of racism, he really isn’t – he is the opposite. He does not define what this opposite means.

The incident drew my attention to the discomfort with which staff handled a highly intrusive, coercive task, which lay outside their formal training and experience. The humiliation of the detained man is the perverse result of their attempts to demonstrate care and caution, and to manage a perceived security threat simultaneously. An hour later, when the night shift staff arrived, I spoke to Karl, a more experienced detention official who had talked to the man in solitary confinement. Karl did not hide his irritation with his colleagues. ‘Why the hell would they ask him to keep the door open? He is not at risk of committing suicide! Of course, he gets pissed off.’ The incident illustrates how caretaking intentions and the securitisation of the detention environment clashed with detention officials’ own understanding of their roles, generated novel insecurities, and triggered new animosities between themselves and those incarcerated. Additional insecurities were, in this case, also triggered by my presence, and in the utterance of the word ‘racist’, which to staff was as contagious as to their Danish colleagues. Indeed, staff strongly resented when detained people questioned their caretaking intentions or called them racists – an allegation that they all firmly denied and denounced. In their understanding, racism equalled prejudiced attitudes and racist language, which they disassociated themselves from – even though both racialised staff members and detained people testified that such forms of racism prevailed in the deportation prison. Meanwhile, what was obscured or overlooked was how racism functioned as a structuring logic of the deportation prison, and how it was expressed also through staff members’ acts of care and compassion.

Detained people also regularly resisted and challenged the legitimacy of the detention and deportation system. An illustrative example is a conversation between several detention officials and Samir, a man in his early twenties, originating from a North African country that was unknown to the migration authorities. He had spent an unknown period of time in Sweden prior to his incarceration and had been confined in the deportation prison for some weeks at the time of my fieldwork. Prior to this, he had been shuffled around between different deportation prisons for several months. The police had attempted to deport him to different North African countries, but since he refused to cooperate with authorities and had no identity documents, he had been denied entry and subsequently returned to Sweden. One afternoon, I sat down on the couch in the common room with Samir and a couple of staff members. Samir turned to the officials and asked them, jokingly, and in fluent Swedish,

The detention officials looked at Samir, then at each other. Yasmin, one of the supervisors, rolled her eyes and leaned over to me, whispering, ‘it’s ok. He’s been talking about this all day.’ Samir stood up to go out for a cigarette, and she continued, ‘But we get that he is stressed. He was supposed to have travelled today but now he’s going to court instead. It’s not ok, we had another one like him who was here for nine months, was deported but refused entry at the border and came back here again … nobody wants them.’ Yasmin referred to other cases of people whose deportation order was complicated to enforce, because they had refused to disclose their identity or to cooperate with authorities in the deportation process, or because of practical obstacles to enforcement. Some of them ended up oscillating between living as destitutes, taking up jobs in the informal labour market, getting caught by the police, and being sent to regular prison or to deportation prisons. When Samir returned after his cigarette break, he started singing ‘Clandestino’⁷ by Manu Chao, first in Spanish, then attempting to translate it into Swedish.

Samir was caught in deportation limbo. His way of mocking authorities’ attempts at getting rid of him and while claiming to care for him was ingenious: by likening migration authorities to disputing parents, and answering the police’s statement of how much the circus of his repeated deportations cost them by promising to refuse their food, he challenged the authorities’ claim to benevolence, highlighting instead their limited ability – and willingness – to control and care for those who, like him, were ‘Clandestino’; illegalised, and unclassifiable. His case history also demonstrated the chaotic functioning of deportations (Hiemstra, 2014), which stood in sharp contrast to the imaginary of a rational, fair, and orderly deportation process that detention officials relied on to legitimise their work. Samir also knew that if the repeated efforts to deport him kept failing, he would eventually be thrown out on the street. He knew, in other words, the limitations of the welfarist caretaking gaze, and was familiar with the fact that incarceration was only one of several tools that would be mobilised to expel him.

Concluding remarks

In the Swedish deportation prisons, care, compassion, and the seeing gaze of the welfare state are utilised as techniques of expulsion. During my fieldwork, the often well-meaning officials were seeing, attending to, and trying to alleviate detained people’s immediate suffering, while ‘unseeing’ the violence endemic to confinement and deportation enforcement, and its role in causing their suffering. The staff registered attitudes and behaviours and identified ‘risk profiles’ among detained people but left unanswered the question of the risks they were exposed to as a result of their incarceration. Staff reluctantly partook in the gradual securitisation of the detention environment, accepting it on the premise that comprehensive visitations and solitary confinement were used for the sake of suicide prevention, and that twisting somebody’s arm was a de-escalating pacification technique, rather than an act of violence. Such codification of violence ‘institutes the oxymoron of an exercise of force stripped of any notion of violence’ (Makaremi, 2018). It does not ensure the safety of detained people against the violence of the state, but it facilitates its normalisation. Next to codification, the normalisation of violence was enabled through racialisation, which in the Swedish deportation prison took place through inferiorising tropes of detained people as objects of care, and detention officials as benevolent. This way, the Swedish deportation prison, if less overtly harmful than its Danish counterpart, exemplifies how violent state action is rendered more digestible and normalised through ‘humanising’ reforms. In his book on Manus prison, Behrouz Boochani (2018; see also Bhatia and Bruce-Jones, 2021: 84) identifies oppression to have been perfected where ‘everything is logical’: where an imprisoned man is banned from playing the guitar, because its strings might be used for suicide; and where smoking, as one of the few remnants of ‘freedom’, is prohibited for the safety of those imprisoned. Boochani explains how these bureaucratic rules are exercises in power that gradually but systematically deprive people of their autonomy and personhood. These forms of border violence, which operate through performances of care, humanitarianism, and rationalisation, do not fundamentally change the conditions of incarcerated people as much as they serve to ‘humanise’ the detention officials (see also Bosworth, 2016).

The incarceration and deportation of foreign nationals are carried out at the ‘threshold of public, political and ethical acceptability’ (Walters, 2019: 176). This threshold of acceptability is located on a continuum where violence can take multiple forms. However, whether it is manifested through overt, punitive violence like in Denmark, or in the form of codified ‘force’ redressed as care and rationalisation, like in Sweden, the violence in deportation prisons has similar effects: it reduces those incarcerated to threatening or threatened things, and breaks them down physically, mentally, and socially. These effects, and the critique and resistance against this violence articulated by detained individuals, were known to officials in Ellebæk and in the Swedish deportation prison. It manifested in the Danish physiotherapist’s calculation of the per cent of humanity left in a person after some time in confinement, in the suicide-preventive schemes put in place in both prisons, and in the Swedish detention official’s anxiety about what subversive words would be articulated on a pearl bracelet. For detention officials, it was not a matter of being ignorant of the violence of confinement. But their reactions to it – or lack thereof – were marked by a belief that these harms were inevitable, that detention was a necessary part of the border system, and that this system was the only one possible.

I sent two fieldnotes to Julia Suárez-Krabbe, including the preface to this book, which takes place in a Danish deportation camp – a context which she, as a scholar and activist in Denmark has also worked extensively on – and the above notes from my encounter with Lars. I asked her to help me think through how far denial is really a precondition or a necessity for sustaining state violence, as has been argued in the literature on bureaucratic violence (Arendt, 1963; Bauman, 2013), and in research on migration-related detention, respectively (Bosworth, 2014; Mountz, 2020). In our conversation that followed, Julia reflected,

Julia and I continued talking about the self-enclosed nature of this logic, which, as Julia added, relied on a rationality ‘that needs to be enclosed within itself, for it cannot bear reality’. On the one hand, this rationality precludes any critique which does not speak within that same framework, and which is therefore conceived as ‘irrational’. This includes perspectives that question the perceptions that non-deported people are imbued with difference and danger, and that what is at stake is the order built upon the violent policing of these differences. On the other, there are worlds beyond this enclosed logic. Julia concluded,