Shelters and clinics: sites where care and violence are mutually constitutive for migrant workers in Singapore

ABSTRACT This paper critically interrogates care and violence, demonstrating how they are both spatialised and mutually constitutive within the shelters and clinics of Singapore for migrant workers. While there are seemingly disconnected sites, both clinics and shelters are utilized to provide different forms of care, discursively represented as spaces of protection and healing. Drawing on ethnographic research with migrant domestic workers at these sites, this paper will, however, argue that shelters and clinics are also spaces where violence is enacted and experienced. Indeed, despite having lived through different forms of violence in their employers’ homes, this paper will reveal how domestic workers were subjected to further suffering and bodily harm; all while, paradoxically, receiving different forms of care. Building on social and cultural geographic debates, and particularly on feminist scholarship that foregrounds care, violence, and the body, this paper will argue that care and violence are mutually constitutive in these sites. Moreover, it will show that the migratory regime in Singapore creates spaces where the care that is practiced cannot be abstracted from violence. Beyond coexisting, in these geographies care and violence are shown to be inextricably connected.


Introduction
Rosamie, a former domestic worker (hereafter, DW) from the Philippines, sat on the floor of a shelter telling me about her experiences during the employment she had fled from only a week prior.Rosamie's employers had required her to stay up into the early hours of the morning, caring for a baby and awaiting her female employer, who would regularly come in loud and intoxicated, in need of oversight and care.During these nights, Rosamie's employers would often be physically violent towards one another.Following a more heated conversation with her female employer one day, Rosamie was physically assaulted.Feeling that this household was no longer safe for her, she fled and ultimately sought refuge with HOME (the Humanitarian Organization of Migration Economics), the organisation that managed the shelter where we were sitting and talking.
Rosamie reported this crime to the police, initially determined to get justice for the violence she experienced.After waiting unsalaried for several months, however, Rosamie felt both frustrated and angered by the way in which her 'case' was being managed.When I asked Rosamie about whether she felt she could get justice for the violence she experienced one afternoon, she responded: 'No, there's no justice here.There's no justice in Singapore for the helper.Why?Because I think the only decision that the MOM [the Ministry of Manpower] knows is send back.Send back.Whatever case, send back.' Rosamie was just one of the DWs who I met that felt that the care they were given after leaving their employers' homes was marred with further injustices and violence.Indeed, while the judicial systems, police, and Ministry of Manpower (the Singaporean governmental division that is responsible for all labour, hereafter MOM) are supposed to provide them with different forms of support, the processes by which DWs' 'cases' were managed demonstrated how violence was also thoroughly enmeshed within these systems and institutions.Drawing on the experiences of DWs like Rosamie, this paper will demonstrate how certain sites within Singapore -namely, shelters and clinics (meaning different settings where health-'care' is administered, sites such as hospitals, doctor's offices/ surgeries, etc.) -became geographies of both care and violence for DWs.Moreover, rather than being distinct or even coexisting practices/experiences, this paper will demonstrate how care and violence are mutually constitutive -that is, how they are inextricably connected and established in relation to one another -and how this manifested physically in certain sites.Moreover, it will demonstrate how the state enacts structurally violent practices of societal control and management under the guise of care.
Initially, this paper will provide an explanation of both how DWs are positioned within Singaporean society, and specifically the reasons why they might come to interact with clinics and shelters.It will also detail the methodological strategy that informed its arguments and provide insight into my positionality while undertaking this research.Following this, the paper will engage with existing literature on the geographies of care and violence, demonstrating the ways in which these concepts and practices have been understood as relational, embodied, and intersecting.It will also show where there are existing gaps in the literature, as violence and care are discussed as either discrete, overlapping or coexistent phenomena, but not as mutually constitutive.The two sections that follow utilise ethnographic data to unpack the ways in which clinics and shelters became geographies where violence and care were mutually constitutive for DWs.Finally, this paper will conclude by recentring Rosamie's assertion that there can be no justice for DWs in Singapore, as, even when purportedly enacting care, the migratory regime is shown to be one which subjugates migrant workers and disposes of them when they are no longer deemed of use.

Context and methods
This paper focuses on DWs' experiences in clinics and shelters, spaces that I gained access to, while I was primarily conducting ethnographic research in HOME's shelter: a refuge for migrant DWs who were no longer willing/able to work for their employers in Singapore.As a space with a high level of resident turnover, I became familiar with a variety of issues that different DWs faced during their employment while also getting to know the residents on a more personal and intimate level.While in a very different positioning to the residents -my nationality, white skin, and status as a visitor affording me huge privileges -being younger than most of the other volunteers and staff, led several residents to feel more comfortable speaking with me.Younger residents would discuss their lives out of the shelter with me, while several of the older residents would find ways to 'mother' and care for me, making me drink while I ran activities (even when I insisted that I was not thirsty) and checking I felt well if I ever looked tired.Significantly, I did not appear (according to several residents) as I could be an employer, an attribute that made me more approachable than others who entered the shelter.After an extended period in the field, the relationships that I formed with the residents, and with HOME, allowed me to extend my research parameters; enabling me to enter hospitals, doctors' offices/surgeries, and women's clinics.I also learnt about many other shelters in Singapore, by speaking with residents who had moved between them, and when dropping women off to Embassy shelters on a few occasions.This paper is, therefore, based on both observations and informal conversation on these sites, in addition to data from more formal interviews.In total, I interviewed 45 different DWs, of whom 29 were shelter residents (some of whom I interviewed on numerous occasions), along with eight of the shelter's staff/volunteers, and several other employers, activists, and employment agents.
Migration has always been central to the Singapore's economy, with its economy and development strategies heavily reliant on migrant labour (Koh, 2003).The state has produced what Yeoh (2006) has described as a 'bifurcated' system; actively recruiting people referred to as either foreign 'talent' or foreign 'workers' but affording them very different privileges.Significantly, the 'guestworker' scheme in place utilises a use-anddiscard approach, as migrant 'workers' are only able to reside in Singapore temporarily, rendering them a disposable population (Bal, 2015;Yeoh, 2004;Yeoh et al., 2000) with no routes to permanent residency.To fulfil reproductive labour needs and Singapore's care deficit, roughly 250,000 of these 'workers' are employed as domestic labourers, who today largely move from Indonesia, the Philippines and Myanmar.To manage this population, the state has created an employer-sponsored migratory system, despite it having been widely argued that these schemes create power inequalities that increase the likelihood of abuse (Parreñas et al., 2021).As such, and despite these concerns, DWs' work permits are tied to their employers, requiring them to live in their homes full time.
Within this regime, there are, then, a number of reasons why DWs interact with clinics and shelters.Having been positioned as DWs' guardians, and responsible for their bodily management and maintenance, employers are required to provide medical insurance for their employees in case of injury or sickness.DWs are also required to undergo testing for pregnancy and syphilis every six months (in addition to TB and HIV every 2 years), meaning that they regularly interact with doctors in healthcare settings.If DWs are found to be pregnant during these check-ups, they are required to leave Singapore.Significantly, the Singaporean state has also created specific policy surrounding safe working practices for DWs, as the number of fatalities following fall while cleaning external windows on tall buildings grew to concerning levels in 2012 (MOM, 2012).Despite these rules being in place, I still hear of DWs being required to clean windows while standing precariously on external ledges.Not all DWs who fall from buildings do this while cleaning, however, as there are also those trying to escape their employers' homes, those attempting suicide, and those who claim (or for whom it is claimed) they were pushed.Across the period of time that I spent in Singapore, I met three women who had fallen from buildings and been seriously injured as a result.As such, while all DWs interact with Singapore's healthcare systems regularly, some DWs have far more intensive interactions with hospitals.
While interactions with hospitals and other healthcare providers are fairly routine, DWs more rarely engage with shelters, as they are required to live in the home of their employer (or with their agency when they first enter the country and are completing their medical check-up and mandatory 'settling-in' procedures).Only when a DW is unable to reside with, or work for, their employer -for reasons such as physical and/or sexual assault, 'illegal deployment', over-work, and theft, and whether these actions are perpetuated by the employer, DW, or both -would they need to (or be allowed to) reside in a shelter.In these circumstances, it is required that a 'case' is filed with either the police or MOM, and shelters are obligated to give the state notice of their new residents' presence.As stated by the MOM (2021, online, emphasis added), 'the Singapore Government takes all cases of abuse seriously and thoroughly investigates all complaints.During the investigation period, FDWs are provided with care and support including shelter, food and medical treatment'.
There are many different shelters in Singapore: those managed by NGOs or religious intuitions; those attached to embassies; in addition to one that is run by the National Trades Union Congress (which is intimately tied to the state, if not directly under their control).Dependent upon the organisation, they offer different forms of support.While some of these shelters provide a very basic space for sleep (and healthcare if necessary), while I was conducting fieldwork, the HOME shelter provided legal aid, mental-health support, and daily activities such as cooking lessons (many of these activities had to be suspended through the COVID-19 pandemic).As HOME (2021, online) states: 'the shelter residents may stay for periods of a few days to over a year, depending on the speed in which their cases are processed.In many instances, HOME's shelter is their [the residents] only place of sanctuary at a time when many have no way to support themselves, nor anywhere to stay in Singapore . . .We seek to restore their dignity through holistic programmes and provide a variety of activities and services, from English classes to counselling sessions.'When moving into a shelter, DWs do not know how long they will be expected to reside there, or how quickly their cases will be resolved.They also do not get to choose whether their case will be taken to court, even if they are alleging serious criminal abuse.Instead, the MOM or the police, along with the Attorney-General's Chambers in some circumstances, hold these decision-making powers.
In this context, then, I encountered many DWs who had experienced a great deal of interpersonal violence, with their suffering often enduring.Unable to visit the sites in which the shelter residents had been living and working, I also conducted individual and group drawing activities, utilising arts-based methods to bring visibility to oft invisibilised spaces (Antona, 2019).While there are many approaches to writing about these forms and experiences of bodily brutality, with some feminist writers (understandably) advocating for exposing the truth of these realities (Ristock, 2002;Skinner et al., 2005;Westmarland & Bows, 2019), I remain cautious of providing unnecessary details of suffering, as this can also be consumptive and voyeuristic (Kasturi, 2007;Nnaemeka, 2005;Robbins, 2013;Sontag, 2004).Despite these concerns, several of the shelter residents told me that they wanted their 'stories' shared, often with the hope that they could create positive change for others.Following these desires, but without including details of bodily suffering that do not serve the arguments of this paper, I will first provide an overview of the academic literature that engages directly with geographies and spatialities of violence and care, in order to develop a theoretical framework to ground the remainder of this paper.To note, all of the names used in this paper are pseudonyms and ethical approval was granted by the London School of Economics and Political Science prior to conducting the ethnographic research on which this paper was based.

Geographies of violence and care
While seemingly oppositional, care and violence are not mutually exclusive.Rather, they are both experiences and practices that can be simultaneously (re)produced within space, and which also produce space.Indeed, both care and violence are said to have geographies (Conradson, 2003a;Middleton & Samanani, 2021;Tyner, 2012;Tyner & Inwood, 2014), and centrally relate to bodies; whether it be bodies that are enacting or receiving violence and/or care.While subject to wide-ranging academic investigation, care, and violence -as concepts, practices, and experiences -are most often discussed as either distinct or intersecting phenomena.This paper will, however, demonstrate that care and violence not only overlap and co-exist as complex experiences but are mutually constitutive; that is, they are established in relation to one another and are inextricable, developing and acting together.
Involving empathy, concern and responsibility (Green & Lawson, 2011;Tronto, 1993), and as 'embodied, emotional and relational' (Power & Hall, 2018, p. 304), Held (1995) argues that care needs to be understood as a moral practice.Indeed, whether it be practical or emotional, care is commonly conceived of as 'the proactive interest of one person in the well-being of another' (Conradson, 2003b, p. 508).In practice, however, these acts take place at every societal scale as a form of labour and involve a myriad of actors.The complex and multiple ways in which caring labour is governed, organised, and practiced across space have been framed as landscapes of care, or caring-/care-scapes (Bowlby, 2011;McKie et al., 2002;Milligan & Wiles, 2010).Today, caring labour is organised across international borders, both transnationalised and commodified, and with global care chains highly prevalent (Hochschild, 2000;Yeates, 2012).As Dyer et al. (2008) write, much care work is undertaken by female migrant workers and is both low paid and undervalued.In addition, mothering, community care-giving, and formalised healthcare systems have all been subject to geographic inquiry (Holloway, 1998;Milligan, 2003;Parr, 2003;Twigg, 2000); with private homes (Bowlby, 2012;Dyck et al., 2005), care homes (Milligan, 2005), or spaces in which more formal medical care is administered, such as hospices (Brown, 2003), or hospitals (Brown, 2012) being centred in this writing.
Care can also be conceptualised much more broadly.Practices of selfcare, friendship, international 'development', advocacy, activism, and welfare provision can all, for example, be considered forms of either individual or societal care (Atkinson, 2011;Bowlby, 2011;Green & Lawson, 2011;McKay, 2020;Power & Hall, 2018;Silk, 2004).As such, it can be argued that care is a broad set of practices, both multiscalar and multifaceted (Atkinson et al., 2011;Hall, 2011).Significantly, care/caring -whether it be informal, formal, and involving monetary exchange or not -involves bodies, and particularly 'a relationship between the giver and the recipient' (Bowlby, 2011, p. 607).As a form of labour, care has long been feminised and racialised across different national contexts, and is often considered to be menial (Roberts, 1997).Under colonial administrations, during periods of enslavement, and contemporarily, women of colour and migrant women are often required to fulfil caring responsibilities (whether remunerated or not).
Seemingly oppositional, violence can be conceived of as 'the intentional use of physical force to cause harm or physical injury, either actual or threatened' (Castree et al., 2013, p. 551) and has long been a subject of academic research.It can be 'directed towards oneself (e.g., suicide), towards others (inter-personal violence), or by one collective against another, including in the form of political violence' (ibid., p. 551).In their definition of violence, the World Health Organisation (Krug et al., 2002) also includes the word 'power' (in addition to the idea of violence involving the use of 'physical force') to broaden the scope to include coercion, intimidation, threat, and neglect, as well as other forms of psychological abuse.Tyner (2016, p. 9) has written extensively about violence, arguing that rather than being something that exists independently, or as an ontological category, violence needs to be understood as being 'historically and geographically contingent'.Like care, violence centrally involves bodies and the ways in which they relate to one another (Tyner, 2012).
Beyond interpersonal or directed forms of harm, violence occurs at different scales, frequencies, and timescales, ranging from acts that are tied to the individual, to actions which are enacted by/to a collective.Indeed, geographers have paid wide-ranging attention to different forms of violence: institutional; political; state; symbolic; cultural; silent; and slow, to name but a few (Galtung, 1990;Gregory, 2004;Holmes, 2013;Lee & Pratt, 2012;Nixon, 2011;Parreñas et al., 2021;Watts, 1983).Some of these forms are considered more 'extraordinary' -such as the violence of war, terrorism, and genocide -while others take place in more mundane and everyday formations.While not entirely isolated in its form, there has been significant scholarship that focuses on structural violence.Despite being harder to define and conceptualise, forms of structural violence are often deeply embedded in society, evasive in nature and without a distinct perpetrator (Galtung, 1969).It has been argued, particularly by medical anthropologists, that structural violence can still result in bodily harm, however, with certain bodies more likely to experience suffering than others (Farmer, 2001(Farmer, , 2004;;Holmes, 2013).Significantly, Gardner (2010) argues that structural violence can produce/exacerbate events of interpersonal violence.Indeed, he demonstrates that the migratory system in Bahrain produces and enables employers' direct violence against migrant construction labourers.Similarly, in Singapore, the lack of social protections afforded to migrant workers, and racist perceptions that they are disposable and of lower worth (Parreñas et al., 2021;Tadiar, 2004), results in unjust working conditions and abuse (Huang & Yeoh, 2007).While both internationally acclaimed and self-professed as one of the safest urban environments globally (MHA, 2021), there is, then, both a geography and a corporeality to violence in Singapore.
Despite being their own fields of academic investigation, then, theorisations and conceptualisations of care and violence do share similarities and continuities.They are both multiscalar and multifaceted practices that are organised spatially, involving relationships between bodies.Beyond these overlaps, however, there is literature that draws together the confluences between these two concepts/practices more concretely.Tyner (2016), for instance, discusses the provision of healthcare in the USA and how its delivery/distribution is assessed by cost-benefit analysis systems that rate patients and their potential 'quality' of life thereafter.He argues that the process by which healthcare is rationed is violent, as it leads to personal harm and potentially death.Similarly, Mulla (2014) argues that the medical and judicial institutions that respond to sexual assault survivors have systems that mean the 'care' they provide also enacts violence.She explains that the practice of collecting DNA as evidence, for instance, requires survivors of assault to comply and be passive, obliging to forensic examinations and investigation in the immediate aftermath of a traumatic event.Where Mulla's conclusions are that caring can be violent itself (and while I agree with her conclusions), in this article I build on these arguments to show that care and violence can also be mutually constitutive.
Beyond the hospital, homes have also been written about as sites where care and violence are both, sometimes simultaneously, experienced/enacted.Feminist scholars have, for instance, long argued that women regularly experience the home as a site of violence and confinement (Blunt & Dowling, 2006;Massey, 2004), in addition to being a space where they may both receive and provide care.Migrant DWs' experiences during employment are particularly interesting in this regard, as they are paid to perform caring labour while working within systems and sites where their experiences of violence are highly prevalent (Constable, N, 2007;Huang & Yeoh, 2007;Parreñas, 2001).
While not directly engaging with the concept of violence, many scholars have also written about care and its relationship to discipline and/or control.For Foucault (trans. Foucault, 1977), clinics -like schools, factories, and prisons -are an important site through which societies are disciplined, as social norms are taught and both docility and normativity engrained.Other scholars have also explored the connections between care and control, with Johnson and Lindquist (2020, p. 195) arguing, for instance, that they are demonstrated to have 'entangled relationships . . . in Asian migrations.'With contested meanings, practices, and relations, and drawing on Constable's (2020) analysis, Johnson and Lindquist (2020, p. 203) suggest that peoples' perceptions and differentiated experiences mean that 'what one person might perceive as controlling, another might regard as evidence of care'.Indeed, systems of control -structures of healthcare, legislation, and education, for instance -can (and do) also provide different forms of care.While not always straightforwardly, the systems which privilege some and disadvantage others map on to care and control.As Johnson and Lindquist (2020) contend, it must be acknowledged that 'the way care is organised, practiced, understood and imagined shapes and is shaped by conditions of im/mobilities, processes of stratified citizenship and forms of differential inclusion in complex, and often contradictory, ways.'While control can be a form of violence, it does not encapsulate the physical harm, emotional turmoil, and bodily suffering that some DWs in Singapore endure.Nor does it directly encapsulate the kind of structural and racialised violence that they are subjected to.For these reasons, violence has been used as the lens of analysis in this paper.While social, cultural, and feminist geographic scholarship engages with the confluences between these care and violence then, the ways in which it is also mutually constitutive have not been addressed.This paper will now engage with ethnographic research to demonstrate how clinics and shelters were sites in which this materialized for migrant DWs in Singapore.

The clinic
Canessa told me of her experiences with doctors in Singapore while sitting in a small room at HOME's shelter.We had started the conversation because I had asked if her hands were less sore after applying a cream that she was prescribed following her arrival at the shelter.Opening her hands to show me, she said that her palms felt much less sore, but that this had also been the case before.Canessa explained that this was the fourth prescription she had received, as her employer had taken her to a local doctors' surgery every time that her skin had reacted this way.She described how her hands started to get blistered and swollen for the first time after working at her employer's home for a few months, with irritated patches appearing over her palms, knuckles, and wrists.Her employer took her to the doctor who prescribed a cream and told her to wear gloves while working.Following some improvement, only a few weeks after the tube of cream had depleted, the issue started again.Following another visit to the doctor's surgery, and it having been established that allergies to the cleaning products she was using were the cause of her skin irritation, Canessa was instructed to wear gloves while cleaning.In an interview, Canessa mimicked the doctor saying to her employer, 'if the gloves aren't working, I am not sure if she is going to be much good with this condition'.
Following this visit, Canessa's skin did not heal as quickly, and she returned to the surgery to be prescribed some more cream.The doctor issued it but warned that each time her skin reacted, it may take longer to heal.This time, Canessa's employer asked if there was nothing else that could be done, to which the doctor joked, 'unless she can work with none of these products, not really'.While wearing gloves helped initially, after several more months, the skin on Canessa's hands and forearms continued to cause her pain.Having waited sometime, the sores became too great and once more, Canessa showed her employer the condition of her skin.Rather than immediately book a doctor's appointment, however, on this occasion, Canessa's employer asked her if she felt she could continue working in Singapore, reasserting that the irritation would likely continue to return.Like the doctor, her employer was concerned that a job that required cleaning products was not suitable.While Canessa's employer did not explicitly say that she was going to terminate her contract, Canessa felt that this was inevitable.Already aware of the HOME shelter, Canessa fled her employer's home soon after.
When speaking with Canessa about this decision, she explained that she had no opportunities to work in the Philippines so felt that it was better to try to stay in the country even with this skin irritation.She said that her employer was 'good and fair', but she knew that the multiple trips to the clinic had made her feel that Canessa was not a suitable employee.Canessa also expressed regret at asking for further treatment for her hands, as she had felt that the doctor's comments and advice had impacted her employer's decision.To Canessa, it seemed as though the doctor and her employer placed more value on ensuring that her labour was conducted in a particular way, and with products that were causing her damage.Rather than asking Canessa about the importance of this work to her or considering if alternative natural products could be used to clean, she was treated as exchangeable.
While Canessa had initially been grateful for the medical support she received, the clinic she attended became a place of anxiety for her.It was a site where the care she received came together with advice that she felt would lead to her return migration.When we spoke about the situation, Canessa was aware that the doctor and her employer were looking out for her best interest too, but she felt that they were prioritising the use of cleaning products over her.She said to me, 'the doctor say no cleaning products as a joke, but why not?It's still possible to clean'.Despite her condition improving, Canessa's employer cancelled her work permit while she was in the shelter, and she was eventually required to leave Singapore.While Canessa's skin condition had not initially prevented her from continuing her job, developing slowly and not arising in an instantly traumatic way, there are many DWs whose experiences in hospitals and clinics arise from life-changing accidents and events.Indeed, across the period I was in Singapore, I met three women whose lives had been forever changed after falling from buildings.Along with two women I met at the shelter, I met Mitali at a hospital with another volunteer, Kamala, having received a series of concerned calls from different parties asking for HOME's support.Upon arrival at the hospital, and after conversing with some of Mitali's friends who were visiting, it became clear that she did not have full cognitive function.Her injuries were severe and neither Kamala nor I had prepared ourselves adequately for this encounter.We left Mitali with her friends and went to ask a nurse where she would stay when she was well enough to leave the hospital.Angered by our question, the nurse stated that Mitali's employers were her legal guardians and that they were the only ones eligible for this information.While letting slip that it was likely Mitali's cognitive state may never get better, the nurse said that we need not be concerned as Mitali's care was being handled.The nurse added: 'she's leaving Singapore soon anyway because she's now fit to travel'.
In the days that followed this visit to the hospital, I was forwarded an email that Mitali's husband had written, asking that Mitali remains in Singapore so that he could come to represent her and pursue a legal case against her employers.He also shared voice recordings of Mitali, which, when translated, stated that she had experienced sexual assault and was treated badly by her employer.Despite her husband's desires, however, Mitali was ultimately returned to a hospital in India only a few days later, where her employers, and their medical insurance providers, would no longer be responsible for her care.
Despite being a site in which Mitali received life-saving surgery and different forms of health and bodily care, through these experiences, the hospital also emerged as a site in which violence was enacted.Rather than being able to seek justice or compensation, Mitali, her husband, and her friends' concerns were not met or (seemingly) even considered.As a migrant worker with no legal right to remain in Singapore after her work permit had been revoked, not only was justice not sought/given, but Mitali was ultimately removed from the country, unlikely to ever be able to work again and with any ongoing medical costs falling onto her family.Despite being a recipient of formalised care, Mitali's societal positioning, and the ways in which her body related to those who were involved in its provision (her employers, insurance providers, medical staff, and the state), left her 'uncared' for in other respects.While the informal care of her friends, family and husband endured, when the mandated responsibility of the state, her employers, and medical providers ended, so did their material/physical care.
As a form of structural violence (Galtung, 1969), there was no clear individual perpetrator of Mitali's (and her husband's) ongoing suffering.Rather, Singapore's migratory regime, and the systems of care that intersect with this, led to Mitali's enforced removal from the country.As such, the hospital not only existed as a site of care -where her friends visited regularly to provide emotional support, where NGO volunteers/staff came to offer assistance and advocacy, and where nurses, doctors, and others enacted medical interventions -but it also became a site from which Mitali was removed from Singapore.Indeed, once it was determined that she would no longer be able to work, and once her case was 'resolved', there was no longer a place for Mitali in the country.Rather, the state reaffirmed its use-and-discard approach (Yeoh, 2004).Now that she was unable to provide care for Singaporeans, her ongoing care ceased.
Both Mitali and Canessa's experiences in clinics demonstrate the complexity of practices of care and violence.While they both received forms of medical and interpersonal care at these sites, their personhood was also diminished as both were ultimately deemed unfit labouring bodies.Indeed, the migratory regime in place in Singapore ensures that all migrant 'workers' are temporary and disposable (Bal, 2015;Tadiar, 2004), a societal positioning that was reasserted and prioritized, while healthcare was being provided.While Canessa maintained that her employer was always good to her, her allergies became too great a burden in the household.Rather than the labour she performed being adapted (or there even being a suggestion of that), Canessa's suitability for the role of DW was called into question in the space she visited for healing: the clinic.
While healthcare is a necessary facet of capitalism, as workers reproduction is necessary for production, this care was only extended so far for Mitali and Canessa.When it became clear that these DWs were no longer able to adequately perform the labour they were employed for -for Mitali that being upon her fall, and Canessa after months of skin treatment and it being established that the use of cleaning products would always be damaging -they were removed from the country.While necessary, then, the care that Mitali and Canessa received was inseparable from the structural and racialised violence of the migratory regime and the ways in which the Singaporean state controls its population.Violence and care were, for Canessa and Mitali, mutually constitutive within these clinics; that is, they were inextricable and enacted together.

The shelter
The home shelter was a dynamic space with a huge amount of flux.While some residents only stayed in the shelter for a few days or weeks, others lived there for months or even years.Khin Aye was one of the HOME shelter's longest-standing residents, having moved there following a short period at another refuge, and after reporting the severe physical and emotional abuses that she suffered from her employers.When I asked Khin Aye about her initial impression of the HOME shelter, and how it compared to the previous one she resided in (which was run by a religious organisation), she explained: 'First time . . .I saw so many plugs . . .You know it means, free to charge the phone . . .It means they don't control you . . .And then also the staff . . .they say, "don't call me ma'am, call me sister".That thing also made me really like, wow . . .I feel like, really, it's family . . .Even in meeting, if we all sitting on the floor, even volunteer, staff also sit on the floor.So, it's like, that's how they show us, I am same as you' (Interview with Khin Aye, 2017).Khin Aye's relationship with the shelter was complicated.While she was grateful for HOME's support and care, she also wanted to leave and to return to Myanmar.Once she had reported the abuse that she suffered, however, and with the state having then decided to pursue a legal case, she was required to stay at the shelter while awaiting trial.During this time, she was unable to work, and had no income.She was rarely updated on the progress of her case, with the uncertainty of her situation leading to many periods of suffering and depression.
Understandably, Khin Aye was not the only resident whose mental health suffered because of the ambiguity of their situation.Merilyn, a resident who lived in the shelter for five months, while her sexual assault case was investigated by the police, also professed to feeling both safe and thankful for the care she received in the shelter, while also trapped and desperate to leave the country.While in the shelter, Merilyn also took on the responsibility of caring for another resident, Aye Aye Swe, who was unable to walk for a prolonged period.Sleeping in the bunk above her, Merilyn would help her to the toilet, get her food, and help her change clothes.Beyond this caring responsibility that she very willingly took on, she also enjoyed participating in different activities that volunteers ran and loved teaching others to dance.
Like Khin Aye (and many other residents), however, Merilyn grew weary of shelter life after a few months.In conversation with me one day, she explained that while she never wanted her male employer to be able to be in a household with a DW again (and that she was glad that she had reported the assault), she was desperate for the police investigation to finish, as she wanted to return to the Philippines.Grateful to the HOME staff for their support with her case, Merilyn would regularly ask them for updates from the police, only to return to her bunk disappointed as there was rare news.While Merilyn was aware that moving on from the shelter would not erase the memories of the abuse she suffered, she wanted to move on with her life, and knew that she would not feel safe working as a domestic worker again.For Merilyn, the wait became exhausting and left her feeling hopeless.
Indeed, the enforcement of waiting and the control of people's lives through occupying their time has been discussed by different scholars.Anderson (2004), for instance, writes of the temporality of detainment and the ways in which states can control time.He suggests, with reference to 'irregular' migrants attempting to enter Europe, that people are subjected to 'extended periods of waiting' and 'an active usurpation of time by state authorities through serial expulsions and retentions ' (ibid., p. 795).He argues that migrants' time, their present and future, becomes colonised.As Schwartz (1974, p. 856) writes, '[t]o be kept waiting -especially to be kept waiting an unusually long while -is to be the subject of an assertion that one's own time (and, therefore, one's social worth) is less valuable than the time and worth of the one who imposes the wait'.While justice was supposedly being enacted in both Khin Aye and Merilyn's names -to protect other DWs and to ensure the rule of law is maintained, acts/intentions which can be considered forms of care (Held, 1995;Tronto, 1993) -the extended periods that they were required to live in the shelter, to wait and to endure, resulted in their suffering.Though it can be argued that Khin Aye and Merilyn were recipients of care from the state, this was all performed within a paradigm in which they were always, and necessarily, experiencing violence.Their care was never separate from the interpersonal violence they experienced, nor from the structural and institutional violence of the legal process that they were required to endure.Indeed, once they had reported these incidents, abuse that was in many ways created and/or enabled by the form of migratory system in place in Singapore (Parreñas et al., 2021), neither Khin Aye or Merilyn were able to influence the process or outcome, nor control their own lives.Rather, the system in place required them to live unsalaried in a shelter, with great uncertainty.Despite all the care that HOME's staff, volunteers, and residents provided -be it formal legal support, the provision of free food and refuge, or simple acts that demonstrated friendship and solidarity -this too was embedded within structures of violence.Despite my personal care for Khin Aye and Merilyn, my role, and actions -as a researcher and friend -were also inextricable from this violence (and perhaps continue to be so).
The shelter was, then, a physical space where it became evident that both systems and individual acts of care and violence were mutually constitutive.Managed by an NGO that advocates on behalf of migrant 'workers' and has core aims of supporting/providing welfare, empowerment and advocacy for the population, HOME was founded with caring motives at its core.Nonetheless, the shelter exists within the paradigm of a stratified labour market and economy that only functions by subjugating 'foreign' workers, necessarily operating in tandem with a racialised migratory regime and imbued with structural violence.By working within this system, even if making efforts to alter it for the benefit the residents, the HOME shelter was a site where (now former) DWs experienced both care and violence.Khin Aye and Merilyn's momentary enjoyment of daily activities, the legal support, the listening ears, and the care they both provided others and were also afforded was inescapably entwined with the structural violence of their societal positioning, and the interpersonal violence they had experienced.Like the hospital for Mitali, the shelter was also a site from which many residents were removed/deported -where their flight tickets were issued, and where they left to go to the airport -whether or not they wanted to leave Singapore, and whether or not they felt that they had received justice.
As Constable (2020) and Johnson and Lindquist (2020) contend in relation to care and control, violence and care were not straightforwardly two sides of the same coin here.Nor was the care that was being practiced at the shelter felt to be violent in and of itself (Mulla, 2014).Rather, within the Singaporean migratory regime, caring institutions and practices are also necessary to ensure the functioning of the economy and labour market.Bringing together Foucauldian assertions that hospitals, prisons, and schools are sites which function to discipline society with Marxist arguments that the healthcare and education afforded to factory workers and their families was necessary for maintaining production, in Singapore, it can be argued that shelters are now required for the migratory regime to function.Indeed, they are sites in which refuge and care are afforded because the potential for interpersonal violence has been created; fundamental in this context, with the care that is practiced, inextricable from the violence enacted.Significantly, embassies also have shelters as sending states want to be seen to be providing care for their nationals who experience abuse and violence, all while encouraging their citizens to migrate (and branding them heroes) as their economies rely upon remittances.As such, any care that is provided in shelters is inextricable from the structural violence of the migratory regime and the racialised violence of a transnational caring labour market that subjugates DWs and situates them as disposable (Parreñas et al., 2021;Tadiar, 2004).Violence and care were, here, mutually constitutive, practices and experiences that are enmeshed in systems that are both developed and enacted together.

Conclusion
While it has long been established that care and violence (and control/discipline) can be coexistent (Foucault, trans. 1977;Johnson & Lindquist, 2020;Tyner & Inwood, 2014), and that caring practices can be violent (Mulla, 2014), this paper has demonstrated how they can also be understood as mutually constitutive in the clinics and shelters of Singapore for migrant domestic workers.Within an economy that functions by subjugating migrant 'workers' -tying their visas to their employers and situating them unequally in both the rights they are afforded and their societal positioninglandscapes of care still exist (Milligan & Wiles, 2010).Moreover, it is not a coincidence that care is practiced and experienced within this migratory regime.Not only do workers require bodily maintenance and support, with medical insurance being something that employers are mandated to provide, but they also have human relationships, and both give and receive emotional care as a result.Within migratory regimes that create/enable abuse and interpersonal violence (Gardner, 2010;Parreñas et al., 2021), as with that in Singapore, judicial procedure, mechanisms for seeking justice, and refuge (given that workers are required to live with their employers) must also exist; not just as an act of kindness/care to migrants, but to ensure that state's maintain control of their society and to maintain a supply of workers.Indeed, sending nations have previously banned their nationals from migrating to certain countries (including Singapore) because of the treatment they were experiencing.
In this context, systems of care are, then, utilised by the state to enact practices of societal and population management, maintaining a racial order, and hierarchies.Systems that could appear caring, such as providing regular medical check-ups and refuge when required, are ultimately a tool for the state to monitor bodies; to ensure that DWs do not give birth in Singapore, that their whereabouts are known, and that they remain temporary.As Rosamie stated (in a comment at the opening of this paper), the end result will always be that migrant workers are sent back.The care that DWs receive is always undergirded by structural, institutional, and racialised violence, and the violence (of all forms) that they were subjected to was braced for varying degrees of care.While it can be argued that some migrant 'workers' receive no form of care at all -individuals who are removed from the nation before earning a wage, or who obtain injuries and are not treated, for instance -clinics and shelters were geographies in which care and violence were mutually constitutive for some.
While demonstrated through specific empirical cases here, the arguments developed in this paper have important wider implications.Acknowledging that care and violence can be inextricably connected -and uncovering where, and for who, this is so -is the first step that is needed to dismantle systems, structures, and institutions that are violent.This is not to say that because violence is prevalent within certain sites and institutions that efforts to provide care are meaningless or futile, however, as (in this case) HOME's work and care have profound and lasting impacts.Rather, it is to acknowledge that spaces such as shelters and clinics operate within paradigms that are imbued with racialised and structural violence, and that working to dismantle these, and to reduce harm further, might require additional and/or novel processes and procedures.Indeed, the ultimate goal of organisations such as HOME, may well be to stop the need for its own existence, and for care to be distinct from the varying forms of violence, which it is currently inextricable from.