State violence against migrant women: Ontological security, threat, and legitimacy

Abstract

This research examines the ways the hostile environment in the UK utilises domestic violence as a deterrent measure, weaponising this endemic form of interpersonal violence against migrant women. I argue that the state’s own processes of accountability and responsibility for domestic violence fatalities, and the active exclusion of migrant women from state-provided services that are key in intervening in cases of domestic violence, are sufficient for domestic violence against migrant women to be constituted as state violence. I frame this in the context of what an ontological security approach can offer to our understanding of the multiplicity of encounters and experiences that migrant women have with a state apparatus that is designed to offer both security and accountability to address the particularly gendered insecurity of domestic violence. The active exclusion of migrant women from these monitoring mechanisms embeds both an affective and a very real empirical insecurity in the lives of migrant women. This ontological insecurity is both inside and outside of state, making ontological security for some while unmaking it for others.

Video Abstract

Introduction

Migrants in insecure status are more likely to experience violence than the general population.¹ In part this is due to constructed vulnerabilities along migration routes, and in part it is due to state enforcement mechanisms both at the border and internal to the state. The security–immigration nexus has been well established in academic literature,² but attention specifically to violence tends to be absent. This is true of International Relations (IR) literature generally,³ but also of ontological security studies specifically. In this research, I consider what an ontological security approach can offer to our understanding of the security–immigration nexus, with particular attention to state violence against migrants and the intersectional inequalities that are reproduced by the operation of state violence. In particular, I attend to the way in which domestic violence has been adopted as an immigration deterrent mechanism, a phenomenon that conventionally has not been attended to in migration studies. Work has focused on where migrants are more vulnerable to escalating and prolonged situations of domestic violence than the citizen population.⁴ I develop this empirical work to critically assess how this violence forms part of a continuum of state violence against migrants. Ultimately, I suggest that the state weaponises domestic violence as part of the arsenal of immigration deterrence, which directly targets migrant women.

To build this argument, I adopt the state’s own framework for assessing state responsibility for domestic violence due to missed opportunities for intervention in fatal domestic violence, in the form of Domestic Homicide Reviews. I consider two tandem processes of ontological security that together use violence to underwrite state sovereignty and identity: that which is endogenous to the state, making identity and offering insight into sometimes counterintuitive security acts;⁵ and that which is exogenous to the state, explaining state action through the position of the state relative to its self-understanding in the international community.⁶ My approach holds that limiting discussion to either the state or to understanding the state–society nexus as the driver of state action has the effect of reproducing extant inequality and of making insecurity within society in order to present a cohesion outside of society. Drawing critical attention to the relationship between these processes I argue that ontological (in)security is violent in its expression, and that this violence calcifies existing inequality and division. This phenomenon can be most plainly seen in violence against migrants.

We are familiar with the shocking violence against migrants happening at the border, and there have been many reports through the years of horrific deaths in the Mediterranean, on migrant routes through the Sinai desert, and through the Northern Triangle of Central America. There is evidence that migrants in insecure status experience violence more frequently than the citizen population, and when they do experience violence, they have little recourse to the law, or to the necessary support services.⁷ States are responsible for or complicit in this violence: either as the perpetrator in the form of policy and practice, or as the responsible agent through withdrawal of normal violence prevention measures that are afforded to citizens (such as policing and social services).

State violence against migrants is a symbol of the state’s legitimacy: an expression of sovereignty and therefore of the right to use fatal violence in the name of security. This is an exogenously reproduced marker of the state’s ontological security as a conceptual unit around which global politics are organised and legitimised. These performative acts of violence solidify the state in a world of human movement that challenges the state. The authorisation for these acts of violence comes through appeal to endogenous insecurity: ontologically insecure populations accept violence as a necessary means to alleviating threat, and this violence reinforces the illusion of a cohesive society connected to the territory and identity of the state to continuously make an endogenously and exogenously recognised ‘self’ that is nevertheless incomplete. Internal (in)security is remade and maintained though ritualistic acts of identity-making, and through micropolitical bureaucratic acts that serve as daily reminders of the integrity of the state and the need for constant vigilance. The case in point for this study is the UK’s hostile environment, referring to the collection of immigration policies first introduced in 2012 by then UK Home Secretary Theresa May, who also coined the referential term. These policies grew in scope over the subsequent decade, most recently in the Illegal Migration Act 2023. The hostile environment, and similar deterrent mechanisms adopted in other states, play on an endogenous ontological insecurity and are used to justify performative violence at the borders that exogenously declares ontological security not for an individual state, but for the state as the legitimate authority. I argue that forms of violence against migrants that are sanctioned – either actively or tacitly – by the state are weaponised as part of the arsenal of deterrence. By knowingly and intentionally withholding protection from violence from migrants with the objective of immigration deterrence, the state is not just complicit in that violence but is weaponising it. When the state withholds protection, there are two consequent assumptions to consider: the first is that the state is willing to sacrifice migrant victims of violence in order to deter future migrants; the second is that the state intends to compel existing migrants to regularise and secure their status (which is often not possible) or to leave. Both of these assumptions use violence to compel a preferred outcome. I argue that the weaponisation of domestic violence against migrant women can be explained in the context of an understanding of performative ontological security.

In what follows I examine the scope for turning attention to (state) violence against migrants in the context of ontological security. I look at two dimensions of ontological security: (1) performative acts of violence as a mechanism of exogenous ontological security at the conceptual level of the state; and (2) the Möbius ribbon (an analogy borrowed from Didier Bigo)⁸ of ontological insecurity that remakes security through ritualistic practices and ‘little security nothings’⁹ that rely on exclusion and simultaneously justify acts of violence to maintain that exclusion. I then consider the intersection of these two dimensions, where ‘little security nothings’ constitute violence and appropriate existing violence. I argue that internal, societal security-making practices endogenous to the state that rely on maintaining a discursive threat and an anxiety-laden insecurity are a continuation of the external security practices that make the state by performing the legitimate use of violence. As will be shown below, both of these routes to ontological security are violent in their expression: performative violence exogenously constitutes the unit of the state as holding the authority on the use of force, while these units host societies that pursue an always-ontologically-insecure selfhood, using violence to stymie forces that threaten or transgress such a sense of self.

Ontological security and violence against migrants

Without rehearsing the ontological security sub-paradigm, and the plethora of interventions it has harvested, my approach here focuses on narrative and performative ontological security and looks at two dimensions that together elucidate the continuous and evident violence against migrants both inside and outside of the state. The first considers ontological security exogenously derived, in which state identity and actions can be explained by exogenous forces that act upon state identity and shape the state’s way of being in the world. However, as will be clarified in the next section, my approach departs from the assumption that states can be agents with a sense of selfhood¹⁰ and instead considers this observable behaviour as a way of continuously remaking legitimacy for the system of power we refer to as the state. Rather than looking at the ontological security of one state that is produced or protected through state actions that respond to its relational position in the world, I instead consider how these actions serve to maintain and reinforce the notion that violence is legitimate when authorised by the state; that is, the legitimate use of violence to reinforce the modern Western ideology of sovereignty, citizenship, and legal belonging. Investment in this ideology forms the basis of state-based political identities, and therefore maintenance of the ideology is crucial for ontological security for the state, practised in the state–society nexus, and reliant on the legitimacy of the state generically. It is particularly relevant in the context of migration because migrants, most acutely when travelling without state authorisation, are actively excluded from state-based political identities. By claiming unauthorised international mobility, migrants transgress the norms of the global system,¹¹ and their presence calls to question the unit of the state. When violence against migrants is uncontested or legitimised, it remakes the ontological security of the sovereign state unit.

Legitimacy locates power and permits some actors and ideologies to retain power over others. As Zarakol argues, the state both provides and seeks ontological security.¹² The modern state is the primary ontological security provider (although this hasn’t always been the case), and sovereignty derives from this role. If challenge to that role arises in the form of institutionalised authority, then the state must seek ontological security to maintain its political authority over the monopoly on the use of violence. The state as a unit is steeped in violence: states hold the right to kill individuals to protect their existence and have normalised sacrificial death in the form of standing armies. Performative violence against migrants is one mechanism of state violence to maintain the state as the primary ontological security provider. Yet such a state-centric approach homogenises the population at multiple levels of abstraction. It locates ontological security provision by states for people, and therefore populations within states have shared defining characteristics connecting them to the national territorial unit through which ontological security is sought and reproduced on a societal level.

This leads to the second dimension: ontological security produced endogenously to the state and attached to identity-making within the state. Here, I turn to the insights of the Lacanian approach to ontological insecurity as a form of existential anxiety that is made by the constant lack of the solidified self,¹³ requiring the constant remaking of it; for example, through acts of identity-making, memorialisation, rituals, and icons.¹⁴ That connection to the national territorial unit is always incomplete and therefore societies are always ontologically insecure, in ways that might be normalised as existential anxiety or pathologised as ontological insecurity.¹⁵ Kinnvall and Mitzen suggest that anxiety precedes fear and leaves space for political possibilities that are not only characterised by fight-or-flight security responses driven by fear.¹⁶ While fear produces an emotional reflex, anxiety asks how can I stop feeling fearful, a more reflective response. Yet an anxious society is responsive to a politics of fear. Bolton draws on Durkheim and the nation as a quasi-religious affective environment to suggest that the ontological aspect of security is produced through a sacred moral order that belongs to the state.¹⁷ The discursive reproduction of threat to this affective and sacred moral order produces fear responses. While work in ontological security attends to this is the context of trauma and crisis,¹⁸ this is in parallel with a shift to consider the political potential of ontological insecurity in the context of a normalised existential anxiety.¹⁹ Nevertheless, I problematise that political potential by returning to the problem of the homogenisation of the population, which reveals that normalised existential anxiety and the underpinning ontological insecurity relies on violence against minoritised populations to provide the illusion of stability and cohesive selfhood. This violence is a continuation of the explicit violence against migrants at the borders that remakes ontological security for the unit of the state itself, that is, not any state in particular, but the state as the ordering unit of international and domestic politics. These forms of violence are bound together to protect the sanctity of the state and to justify that protection in the making of a homogeneous identity. The constant practice of ontological security seeking for society maintains an anxiety that justifies violence against groups that are discursively constituted as a threat to that affective security.

I juxtapose these two dimensions of ontological security seeking in the context of attending to the violence they produce in the continuous maintenance of ontological security for the sovereign state in the first instance, and the continuous seeking of it within state-based societies in the second. Here, acts and narratives of ontological security are not a response to violence, or an explanation for violence, but are violent. The exogenous unit of the ontologically secure state can be identified through the uncontested acts of violence that the state is sanctioned to perform, that the state relies on to reproduce its role as ontological security provider, and which intensify if that ontological security is under threat for a single state actor, not because of the particular actor but because the threat is against the unit of the state and its sovereign authority to protect its borders applicable to all states. The endogenous processes of ontological security map more closely to the Lacanian conceptualisation of a constant existential anxiety.²⁰ The violence in the latter is one of exclusion, including exclusion from the normal forms of protection from violence provided to society by the state via its various agencies, institutions, bureaucracies, and services. Violence against and exclusion of outsiders is sanctioned as an extension of state sovereignty. Violence is perpetrated against challenges posed to the state, both challenges constituted exogenously and endogenously and exemplified in forms of violence against migrants and, specifically in the context of this study, migrant women. In the next section, I develop the concept of state violence and articulate the normalised state violence against migrant women to show that violence and inequality are necessarily the substance of ontological security.

State violence

It is worth offering a brief conceptualisation of the state in order to specify the term in the context of agency and accountability for violence. The state is of course central to IR, often as the unit of analysis.²¹ For Lake, simplifying the state into a unitary actor can render complex analysis readily explicable.²² Lerner adopts a similar logic in his argument for a pragmatic notion of state consciousness, because the entity of the state is ‘irreducible to individuals’.²³ In contrast, I suggest that simplifying the state obscures, or even uses, the violence that the state relies on to secure its continued existence. States do violence against people, yet if those people are not housed under the umbrella of a state, the violence is hidden, or visible but legitimised beneath the veil of state sovereignty and the legitimate use of force. Here, I use the shorthand term ‘the state’ throughout this paper but refer to a complex system of power that comprises an international identity perceived as a cohesive unit and unitary actor, with a fragmented internal institutional structure including agencies where state power is devolved into particular services (for example, these might include social services, policing, and education) or geographic areas, and key decision-making individuals or collectives of individuals. The state cannot bear agency as a unitary actor, but the system of power that comprises the state can be held responsible and accountable for actions that can be traced to institutionalised systemic processes rather than unauthorised individual choices. For example, this might involve recognising that police violence, deaths in immigration detention, or health inequities that lead to early and unexpected death are systemic rather that the result of isolated deviant cases.

Modern sovereignty constructs the system of power that is manifest in the unit of the state as holding a monopoly on the use of force. Force – and, if resisted, violence – is generally accepted as part of the role of the state. The state is dually constructed as maintaining public order and providing security. Those who threaten and transgress state-constructed order are met with a display of force, that is, violence. Torres characterises state violence in three areas: political violence and repression, genocide, and legal violence.²⁴ Political violence is usually against specific political or religious groups that the state perceives as a threat, or against minoritised groups (and, simultaneously, minoritising processes might be understood as a form of state violence). Genocide is totalitarian violence against an entire group or population, ‘most simply defined as the denial of the right of existence to an entire human group’.²⁵ Genocidal violence is targeted, systematised, and intended to be fatal both at the individual and collective level. Legal violence for Torres is based on the exclusion it engenders. Drawing from Agamben,²⁶ she pinpoints where the rule of law constitutes sovereignty and in doing so enacts sovereign violence. Yet if we conceptualise the state as a system of power, the idea of state violence becomes necessarily more complex. For example, one could argue that violence against migrants could be simultaneously characterised as political violence (reinforcing the political unit of the state); genocidal in the characteristics of being targeted, systematised, and often fatal²⁷ against a group the perpetrator characterises as a collective, and targets because of their membership in that collective; and legal violence as it is endorsed by the rule of sovereign immigration law.

Violence itself might be understood as an essentially contested concept.²⁸ Violence can be constituted as physical, structural, and systemic. A theorisation of everyday violence suggests placing an individual within the context of their circumstances and experiences. Violence includes both structures and agents: structural violence was conceived in peace studies and theorised by Galtung, arguing that structural violence is present when social injustice prevents peace.²⁹ In other words, violence is not just the result of an agent enacting violence against an individual, but it happens as a result of injustice and inequality in the world.³⁰ While this violence is certainly important and relevant, it can also be useful to delimit the definition of violence for the purposes of a particular study: not to undermine the relevancy of the myriad forms of structural violence but instead to add clarity to a specified form of violence as an outcome, even while acknowledging that such an outcome is the result of a complex system of circumstances, forces, institutions, and individuals.

The conceptualisation of violence I adopt in this argument is informed by the World Health Organization’s (WHO) Violence Prevention Alliance.³¹ The WHO conceptualises interpersonal violence as enacted by one individual against another. This violence can then be broken down into categories of physical, sexual, psychological, or deprivation/neglect. This is further divided into domestic space where the violence is carried out by a family or partner against a family (child or elder) or partner; and community space where the violence might be carried out by either a stranger or an acquaintance.³² The WHO also defines collective violence that is differentiated from interpersonal violence that happens in the community. Collective violence is identified as being enacted by a group against a group. State violence would fall into community violence: international or internal conflict, persecution or genocide, or targeted police violence all offer examples.

Here, I use these definitions as a starting point but consider where interpersonal violence can be attributed to the state, that is, a collective actor. Drawing on Walby, who locates physical interpersonal violence within a regime of violence that both deploys and regulates the use of violence,³³ I consider state violence as part of the same regime as interpersonal violence. This regime relies on legitimate state perpetration of violence; here, I pose that the deployment of state violence is a habitual practice of ontological security. In this way, I speak to IR’s tendency to personify the state as a means of elucidating how the state is a violent actor in ways that blur the categories of interpersonal and collective violence. Yet I maintain that the state is a system of power itself that must reproduce its legitimacy to use fatal violence through the use of fatal violence. While there are means in international law of identifying state violence against their own citizens, such as the Responsibility to Protect (R2P) doctrine, it is more complex to identify the violence that states in general exercise against people without citizenship or status in the context of international migration. This falls plainly at the intersection of legal violence (constituting a group outside of the protection of the law) and political violence (targeting a particular identity). This happens at the systemic level, with the system of states holding the monopoly on legalised violence and perpetrating violence against people who transgress these geographic containers. This systemic level violence falls within the regime of violence, deploying the violence of the state (ostensibly) to regulate violence against the state. Violence against migrants is necessary for the ontological security of the system: migrants transgress by failing to comply with the sovereign bordering processes of the state. People who migrate without state authorisation cannot be contained by the state and therefore are constituted in this regime of violence as a threat to the state.

State violence against migrants

Borders are sites of violence, and that violence is inextricably tied to state-making. This encompasses ‘the modern geopolitical founding violence’,³⁴ whereby historical border-making established and consolidated power over territory that ultimately manifests in the contemporary world of sovereign states. It also includes current conflict in defence of borders, where threat to a territorial state legitimises the use of force, even when land appropriation is historically contested.³⁵ In the context of migration, state violence is legitimised in the form of protecting borders from unauthorised entries and from transnational crimes such as substance and people smuggling and trafficking. These things are framed in security objectives, whereby the discourse of security gives licence for the use of violence.³⁶ Where the authority of the state to adopt violence is contested, the ontological security of the state-as-being, as the unquestionable authoritative actor in international politics, is under attack. This can be conceptualised as an exogenous force that provokes the state to reinforce its authority in the world. Thus, acts of violence against migrants can be understood as performative acts of ontological security. These performative acts may involve active violence such as pushbacks,³⁷ physical assaults on migrants by border enforcement including via weapons or beating,³⁸ attacks by dogs,³⁹ detention, imprisonment, and removal from state territory⁴⁰ that often involves physical and/or sexual violence⁴¹ or torture.⁴² Even explicit and coordinated bombing attacks on migrant boats were sanctioned by the UN Security Council in 2015.⁴³ Active violence against migrants by the state has therefore been authorised by the highest international authority.

In addition to active violence, the performative violence of the state also includes conspicuous passive violence, such as allowing boats to flounder and sink, resulting in mass drownings, preventing rescue efforts at sea⁴⁴ or on land,⁴⁵ and maintaining conditions of destitution, ill health, and often violence against women and girls in informal migrant camp settings.⁴⁶ That the state has licence to carry out these acts of violence against migrants, and that this violence is endorsed and accepted by other states, demonstrates ontological security maintenance at the level of the state in international relations. This violence must remain uncontested and incontestable for the state to retain its authority on the use of (fatal) violence. Hence, violence is a legitimate means of preventing the transgression of sovereignty inherent in unauthorised border crossing While the endogenous identity–security nexus may be subject to anxiety, constant ontological insecurity, and identity-making practices, this exogenous process of state legitimacy constructs and maintains the state as ontologically secure in the world, yet it is saturated with violence that is not declared as violence but instead as security and propagates violence at the (global) systemic level.

Turning to the endogenous identity–security nexus in the context of violence against migrants elucidates a constant reproduction of a cohesive internal identity narrative. While this internal identity narrative cannot account for inequality, it also works to suppress non-normativity.⁴⁷ This is most obvious in policies that seek to deter migrants. These policies appropriate violence in society as a means of preventing immigration, while simultaneously justifying these policies as a necessary means of constituting security in multiple forms: physical, biological, economic, and cultural. Security-making practices include rituals such as processions, national anthems, and memorial ceremonies.⁴⁸ They also include everyday reiteration of the normative values of the state, such as teaching ‘British values’ in civics classes, wearing poppies (and castigating those who do not) during October and November in the UK, repeating the Pledge of Allegiance in US classrooms, or more insidious practices such as rewarding normative behavioural ‘achievements’ (for example, rewarding marriage with tax breaks).

In the context of migration, the UK’s hostile environment is an exemplary case in point for the appropriation of violence by the state. This violence is enacted unequally across society, rewarding ‘good’ normative behaviour and punishing transgressions. It triangulates identity, citizenship, and economic stability by preventing meaningful participation in society for anybody who cannot successfully perform all of these aspects of belonging. The ontological security is remade through identity, and that identity is securitised because only those who perform successfully in these other aspects form part of the cohesive state identity. This includes being able to demonstrate ‘legal’ belonging to be able to access basic services such as bank accounts, providing evidence of pre-existing economic stability (such as a fixed abode) to be able to access employment, and being able to demonstrate your historical eligibility for secondary and tertiary healthcare.⁴⁹ In this way, the punitive measures of the state in the context of the hostile environment can be understood as a manifestation of an endogenous ontological security seeking. This is ontological (in)security in nature, because it responds to the constant lack of a cohesive self. That this violence is uncontested and framed as incontestable is what provides the crucial link between these violent bordering practices and the ontological (in)security of the state, both internal and external.

The case study included here explores the state’s appropriation of violence by actively withdrawing from migrant women standardised forms of violence prevention that are acknowledged as crucial by the state. These are evidenced through a short case study of domestic homicide reviews (DHRs).

Violence experienced by migrant women

It is worth acknowledging that there is an implicit understanding that the system of power that is manifest in the state plays a role in protecting people at risk, and particularly vulnerable members of society. Gender-based violence (including intimate partner and domestic violence) and sexual violence is pervasive among people in insecure migration status.⁵⁰ That it not to say that every woman in insecure migration status is a victim of violence, but it is to say that probabilistically the risk of experiencing violence is higher if you are in insecure migration status,⁵¹ and that it is common for women who experience violence while in insecure migration status to associate that violence with their insecure status, that is, they consider violence to be a direct result of their migration status.⁵²

The state is rarely considered to be the perpetrator of this type of violence, in part because the state is a system and violence against women, particularly intimate partner violence, is theorised as interpersonal rather than systemic violence. Violence might be attributed to agents of the state, but this is often produced as deviance.⁵³ The state might be considered a direct perpetrator in cases of the most direct forms of violence such as we see at the border or in detention settings where agents of the state use violence against migrants in their care, or in the context of structural and systemic violence.

Here, I focus on interpersonal violence and tracing the complex intersections to build an argument that interpersonal violence, specifically intimate partner and domestic violence against migrant women, can and should be in part attributed to the state. While I do not intend to personify the state, I do attribute state responsibility for permitting this form of interpersonal violence. I base this on the exemplary case study of the UK, to suggest that the state, embodied in Home Office decisions and policies over time, (knowingly and deliberately) entraps migrant women in violent circumstances as part of the series of immigration deterrence policies known as the hostile environment.

There is very little in the scope of international law and organisation to protect women against gender-based violence, with the exception of UNSC Resolution 1325, which offers women protection from conflict-related sexual violence. The Council of Europe Convention on preventing and combating violence against women and domestic violence, known as the Istanbul Convention, recognises the need to articulate protections against domestic violence, although several signatories, including the UK, have signed with a specified reservation against extending protections to migrants. According to available data, a staggering 36 per cent of murders of women worldwide are committed by an intimate partner.⁵⁴ Domestic violence (DV) and intimate partner violence (IPV) of course can affect any woman. Studies of risk factors for DV propose several different risk factors such as youth, lower level of education, alcohol and drug abuse, and historic victimhood, yet there is no accepted risk profile.⁵⁵ An estimated 35 per cent of women worldwide have experienced either physical or sexual intimate partner violence (IPV) or sexual violence by a non-sexual partner.⁵⁶ This is of course a huge proportion of the global population, yet there is no immigration law to protect a person fleeing gender-based violence. For example, gender-based violence in asylum and refugee claims must be constituted and delimited in the ‘social group’ category. Gender is not one of the five protected nexus reasons for persecution, which are race, religion, nationality, membership of a particular social group, and political opinion.⁵⁷

While the state provides security from violence, this tends to be in the context of violence in public spaces and conflict-related violence. Limiting security from violence to violence that is typically biased towards men and/or constructed as masculine obscures the effects of intimate partner violence and violence against women and girls. In this way, security studies reproduce gender inequality.

Domestic Homicide Reviews

In several high-income countries, including the United States, Canada, Australia, New Zealand, Portugal, and England and Wales, the state adopts specific responsibilities in the context of protecting women from domestic violence. These responsibilities are expressed in Domestic Violence Fatality Reviews (DVFR) or, in England and Wales, Domestic Homicide Reviews (DHR). The practice emerged in the United States in the 1990s, then expanded internationally to a handful of countries.⁵⁸ The aim of a DHR is to ascertain what interventions could have been taken place to prevent domestic homicides, ultimately to draw attention to domestic violence and prevent future deaths.⁵⁹ This involves scrutinising state-provided services such as police and healthcare, social services, community organisations that had contact with the victim and/or perpetrator such as religious organisations, or community services such as children’s centres. In this context, the definition of ‘the state’ is dispersed and fragmented; indeed, as Rowlands demonstrates in an analysis of policy documents, DHRs ‘were not represented as the act of a singular sovereign state’.⁶⁰

In the statutory guidance for England and Wales, state actors include local constabularies, GPs and urgent care or accident and emergency settings, and agencies that are governed by the relevant local authority.⁶¹ The local authority is the level of government that is responsible for carrying out and publishing the DHR. Nevertheless, the Home Office establishes the statutory guidance, reviews the finalised documents for compatibility with the objectives, and is currently in the process of standardising the publication of DHRs to better achieve comparability across cases (as of June 2023).

A DHR is typically initiated by the police force in the area in which the homicide happened, who must action the review in communication with the Community Safety Partnership for the local authority in which the victim was resident or habitually residing, at the time of the homicide. The statutory guidance names agencies ‘with a key role to play in the review process’ including ‘housing associations and social landlords, HM Prison Service, HM Courts and Tribunals Service, GPs, dentists, and teachers.’⁶²

The DHR represents the point at which the state adopts and reflects on its responsibility to protect people who are victims of DV and IPV. Since 2011, there have been four clear objectives applied to DHRs: these are (1) establish what lessons can be learnt from each homicide; (2) identify lessons for collaboration between agencies; (3) apply these lessons to service providers; and (4) prevent future domestic homicides.⁶³ It should be noted that while this is applied in all cases of domestic violence, the victims of domestic homicide are disproportionately female: for example, in a recent mixed methods analysis of DHRs in England and Wales, 81 per cent of victims were female overall,⁶⁴ and 89 per cent of victims were female in cases where either victim or perpetrator was of a Black or minoritised identity group.⁶⁵

The relationship between migration – or insecure migration status – and risk of domestic violence is not well established with statistical evidence. The reason behind this is that reporting of domestic violence when in insecure migration status is disincentivised by the associated or perceived risk of loss of status or removal.⁶⁶ This is also enhanced by intersectional race and ethnicity characteristics that are linked to a distrust of authorities, and therefore fear and mistrust of support services,⁶⁷ and is compounded in the UK by policies associated with the hostile environment that pertain specifically to named agencies in the DHR statutory guidance. For example, across DHRs it is common to see health records from GP visits, interactions with police, interactions with social services such as adult social care, and interactions with mental health services. A study of 141 DHRs in England and Wales identified the following agencies as informing reviews: Domestic violence and abuse (DVA)-specific services, police, Refuge (the largest UK domestic violence support service, comprising a network of specialist services including refuge accommodation), housing support services, legal support and advice services, physical health care services, education, women’s centres, substance misuse services, children’s services, adult social care services, and Multi-Agency Risk-Assessment Conference (MARAC). Two of the main points of intervention against domestic violence on the part of the state are through policing and healthcare, yet the state actively restricts access to these things for migrant women. The next subsections details what these restrictions look like in the UK specifically.

Restrictions on policing

Between May and August 2020, there were 357 police telephone referrals to the Home Office on the basis of immigration infraction. All of these referrals were victims of domestic abuse who reported that abuse to the police.⁶⁸ While there is little verifiable data available, a recent report in The Detail suggests that the Police Service of Northern Ireland alone reported around 40 people per day to the Home Office on the basis of suspected immigration infraction, including victims and witnesses.⁶⁹ This suggests the scale of reporting across the UK could number in the thousands per day.

The police guidance document on sharing data with the Home Office tells us: ‘Sharing information on an individual may assist the police and Home Office in the discharge of their respective functions. There is a clear public interest in a joined-up approach to upholding the law.’⁷⁰ In this context, the public interest is assessed at the national level. The national level does not take into account the needs of people in insecure immigration status who are vulnerable to abuse. The ‘joined-up’ approach is in the sense of upholding the law, rather that protecting people experiencing violence from explicit harm. People in insecure migration status are positioned as outside of, rather than part of, ‘the public’, and therefore their interests are not considered as relevant to the public interest. Indeed, as detailed above, migrants are often constituted as a threat to the public interest.⁷¹ The police guidance is clear that their first purpose in responding to the report of a crime is to investigate that crime – not to investigate the reporter’s, witness’s or victim’s immigration status. Nevertheless, if they come to suspect that a person is an immigration offender, the operational guidance directs them to share that information with the Home Office. This does not mean that the Home Office will act on the information. Nevertheless, even if the Home Office does nothing, the person is left less secure in their status than they were before and less likely to take the same action – contacting the police – a second time. Policing is a key service that women can access to find protection against domestic violence and is a key place where the state recognises its own role in protecting women from domestic violence and IPV. The availability of policing is restricted for women in insecure migration status.

Restrictions on healthcare

The Home Office has implemented various reaches into the NHS to access health data for people in insecure migration status. The NHS Information System started providing identity data to the Home Office in 2005. This data sharing was suspended in 2018 due to the obvious health risks of people failing to seek needed healthcare due to a fear of removal. Nonetheless, a fence around the NHS has been constructed via a paywall for people in temporary or insecure immigration status. The 2014 Immigration Act introduced an NHS visa surcharge of £200 per year. This was subsequently raised to £624, and then in February 2024 to £1,035 per year of the visa. People with no recourse to public funds must pay the surcharge, and people in certain statuses or without status can also be charged for secondary or tertiary care. Anyone who is considered ‘not ordinarily resident in the UK’, which would include people with undocumented statuses, must pay 150% of the cost of care for NHS hospital treatment. The NHS can report any unpaid charges to the Home Office, creating a link between immigration policing and access to healthcare.⁷²

Between 2005 and 2018, NHS Digital and its precursors (the NHS Information Centre and the Health and Social Care Information Centre) passed patient personal data to the Home Office, including present and past addresses and information on GP registration, for the express purpose of immigration enforcement. The Home Office insisted that tracing ‘illegal immigrants’ was crucial to the public interest.⁷³ However, public health experts and health practitioners argues that this is in fact detrimental to public health: data sharing creates a disincentive for already highly vulnerable people to access healthcare. This puts their own health at risk but also affects the treatment of infectious diseases, hence producing a greater public risk. Ring-fencing healthcare fosters inequality, which is also well established as detrimental to health.⁷⁴ Avoiding care for progressive conditions also increases the need for emergency medical interventions later. Pregnant women in particular are left particularly vulnerable if they feel unable to access healthcare and support.

While data sharing was halted in 2018, pending review, charges for secondary and tertiary care for immigrants remain in place. Charging immigrants for ongoing and necessary healthcare produces a similar dynamic to data sharing in the sense that it deters people from seeking healthcare or causes people to postpone seeking healthcare until the situation is an emergency. Furthermore, the role of the NHS – providing necessary healthcare – is explicitly weaponised against migrants. Using the NHS without authorisation or having an unpaid charge outstanding (even if an invoice was never issued or an immigrant was not told they were eligible to pay) are grounds for immediate refusal of an application for leave to remain in the UK and have also been used as a grounds for refusal of leave to enter.⁷⁵ Hence, while formal data sharing between the NHS and the Home Office has been curtailed, the Home Office can still compel access to NHS charges and payment and uses this as a means of excluding immigrants.

GPs are a named entity in DHRs and are often a key point of contact for identifying domestic abuse. While people in insecure status can safely access GP services, they may be deterred from doing so because of the risk of being charged in the case of referrals for secondary care, or the risk of facing immigration enforcement as a result of unpaid charges, even when these charges were unexpected.

Restrictions on specialist services

Specialist services are the leading form of protection for women experiencing domestic abuse, and while they are not provided by the state, they are delivered through and underpinned by social welfare services. Women in insecure migration status with no recourse to public funds (NRPF) have severely limited access to these services, meaning that they are often unable to leave an abusive relationship without risking destitution and homelessness. NRPF is a mandatory visa stipulation for most types of visa, including spousal visas. The UK actively blocks support for migrant women victims of domestic violence; for example, as a signatory of the Istanbul Convention, also known as the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence, the UK maintains a reservation on Article 59. This article specifies that states must take the necessary legislative measures to protect victims whose residence in the country depends on that of their (abusive) spouse or partner.⁷⁶ Currently, the UK has no commitment to protect women in insecure migration status, including women on spousal visas or dependent visas that rely on the status of an abusive spouse. While there are potential routes to protection available to some women with no recourse, these are complex and rely on legal advice, and legal aid is limited and difficult to access. Hence, migrant women face a significant barrier in accessing protection even at this crucial point of seeking it.

Specialist services do advise on the pathways to support for migrant women that are available, often referring women to specialist by-and-for services. Yet these pathways are often complex, baroque, and opaque: they both depend on knowledge of current status, which victims of abuse may not have access to, and permit a concession route to lift NRPF for victims of abuse. However, the concession limits access to public funds to just three months, during which time an application for leave to remain must be launched if the individual wants to remain in the country. The application must also include substantial evidence of abuse, and if made after a status has expired can be subject to refusal with no right of appeal. These limitations on the concession involve a degree of risk that must be assumed by the migrant victim of abuse and therefore have limited use for women who have been subject to immigration abuse that has restricted their access to immigration documents and passport or that has resulted in lapsed status.⁷⁷ In summary, the laws preventing access to public funds for migrant women in insecure status further expose them to prolonged and escalating abuse in cases of domestic violence.⁷⁸

Weaponising domestic violence?

Public Health Wales recognises the fact that there is no Home Office policy to recognise and support women seeking asylum or in other insecure statuses who are experiencing domestic violence as a key public health concern.⁷⁹ Women with NRPF can often not seek support from specialised services unless those services are designed specifically for women with no recourse. In these cases, services are often under-funded and lack institutional economic security.⁸⁰

Facing the threat of charges for NHS secondary and tertiary care often propels women away from healthcare, which means there are fewer opportunities for detection of violence. Police report people suspected to be without secure migration cases to the Home Office (at officer/constabulary discretion) even when they are the victims of crimes. This means women are unlikely to report domestic violence if in insecure status or if their status relies on the relationship with the abuser. Police records of domestic violence are considered a key point of potential state intervention to prevent domestic violence and consequent fatalities, yet these are knowingly restricted for migrant women. The key informants that identify, recognise, support, and act to help women who experience domestic and intimate partner violence are not available to migrant women. The state, using law, policy, visa stipulations, standardised practices, and data-sharing agreements between agencies actively prevents migrant women from accessing these services.

Violence against migrant women sits within the intersection of (a) domestic and intimate partner violence, which is well recognised as a problem at the state level and has been identified within policing and other systems and institutions, and (b) border violence – the fact that migrant women are not offered the same (even if meagre) protection that is offered to citizen women is another means of devaluing the lives of migrants. Most of the policies that lead to increased risk of violence are part of the hostile environment in the UK. They are actively designed to produce a hostile space. The DHR process recognises particular interactions with agencies, institutions of the state, and services subject to funding from the state as mechanisms of intervention in cases of domestic violence that are points at which a domestic homicide might be prevented. The state’s own review mechanism underscores the importance of access to these agencies and services. Yet policies, laws, and standardised practices within stage agencies actively deny these key interactions to migrant women. Acknowledgement of these crucial interventions and their lack of extension to migrant women was presented and debated by the Joint Committee on the Draft Domestic Abuse Bill 2019, but the Act that was passed in 2021 failed to incorporate the suggested amendment to protect migrant women.⁸¹ The state prevents migrant women from access to these possible points of intervention that are determined as crucial by an arm of the state itself (in local authority Domestic Homicide Reviews), and acknowledged in parliamentary discourse preceding the most recent Domestic Abuse Act. Migrant women are vulnerable to ongoing domestic abuse and face prolonged exposure to violence and escalating risk. In this way, the state weaponises domestic violence as an immigration deterrent, adopting violence against migrant women as a tool of the state.

State violence and ontological (in)security

This case study of state violence against migrant women is a means of speaking across the exogenous and endogenous forces reproducing the state, via the ontological (in)security that (1) undergirds the unit of the state as the sovereign with the authority on the use of force, and (2) is constantly sought by the various agencies and bureaucracies of the fragmented state, in pursuit of a coherent national self that is justified in its use of force.

The exogenous process of ontological security is expressed through wanton violence against migrants, and the use of this violence reinforces the state as an authoritative international actor, providing the illusion of a self with agency to act against threats to its existence. The endogenous insecurity that is constantly manifest in the lack of a self able to reflect the cohesive identity unit that is performed internationally makes violence an acceptable security practice internal to the state in addition to its use in border, transnational, and international zones. These two forces work together: state authority to use violence is performed in explicit acts of violence against migrants, and in conspicuous passive acts of refusing to prevent violence and death. Migrants are constructed as a threat to state security to explain the use of violence against migrants, yet the actual threat is always potential rather than physical, evoking the anxiety narratives of ontological security.⁸² The use of violence against migrants within the state is a sanctioned response to the potential threat, undergirded by the performative violence of the state against migrants at and beyond the borders that, along with other manifestations of violence, maintains the position of the sovereign state as the primary provider of ontological security.

The endogenous security narrative within the state is a product of anxiety, constituting threat and ontological insecurity because state identity at the level of society must suppress non-normative identities to maintain the ever-incomplete illusion of the cohesive unitary self. The remaking of national identity is always ontologically insecure because the self of the state is a process rather than a static entity. Acts of violence both reproduce and respond to the constitution of ontological security and insecurity simultaneously. This replicates Bigo’s Möbius ribbon analogy of internal and external security that together produces migrants as a multidimensional threat to the state that exists continuously at and beyond the borders as an external Other, and inside the state as a scapegoated internal Other.⁸³

Ultimately, the ontological security approach in this context offers an understanding as to why this obvious and endemic violence against migrants and violence against migrant women in particular is (mostly) uncontested and unproblematised. The example of Domestic Homicide Reviews makes this clear: these are the state’s own assessment criteria for missed interventions in violence. Yet the key possibilities for intervention – policing and healthcare specifically – are actively and consciously denied to migrant women as part of the deterrent policies of the state. While there are of course various manifestations of state power at play here (immigration policy, the Home Office, the local authorities, police constabularies, the National Health Service, and the various healthcare trusts), they each form part of the fragmented system of power that leaves migrant women more vulnerable to domestic violence and domestic homicide than non-migrant women. As stated above, the vulnerability of migrant women was raised unequivocally to policymakers by the Parliamentary Joint Committee on the Draft Domestic Abuse Bill in 2019 and 2020 in the UK.⁸⁴ Yet the amendment to the bill that would extend the same protections to migrant women as were afforded citizen women was not included in the final Domestic Abuse Act of 2021. This cannot be explained by oversight: it was policy by design, with the objective of immigration deterrence in the context of the hostile environment. This is an example of the state as a system of power weaponising domestic violence as mode of immigration deterrence.

The state ‘self’ relies simultaneously on the ontological security of the state as the sovereign unit in the world, combined with the ontological insecurity of an internally remade and remaking state identity, which in this instance creates a recipe for continuous and incontestable violence against migrant women. Migrant women collectively become a symbolic non-normative identity that is produced in opposition to the normative state identity, necessary for the process and performance of ontological security.