Conceptualizing displacement: the importance of coercion

ABSTRACT The paper conceptualises displacement as a process of coercive disruption to valued ways of living and functioning. It does so by engaging with coercion scholarship from moral philosophy, supported by empirical research about displacement in occupied Iraq. Coercion and force are largely taken for granted in refugee and forced migration studies, and in migration studies. The paper also considers recent scholarship on voluntariness in migration and argues that approaches to coercion are conceptually more satisfying. To build the conceptualisation of displacement, a systematic approach is used which involves identifying a phenomenon’s constitutive components, and their relationship to each other, before integrating them into an explanatory framework. It argues that displacement is a process that begins in place, during which time valued ways of living, conceptualised as important baselines, are disrupted in coercive ways. These coercive disruptions relate to power imbalances created, or leveraged, by coercers and coercive structures. Displacement’s processual nature, consisting of phases and counter phases, including evasion and resistance, is incorporated in the framework. The different forms of coercion, as well as the phases and counter-phases, that constitute displacement, are explained and illustrated from the perspectives of Iraqis who lived through the occupation.


Introduction
Refugee Studies scholars agree that conceptualisations of displacement have been surprisingly elusive in the past three decades since the emergence of the field (Chatty and Marfleet 2013). In a recent and wide-ranging review of such attempts, Zetter (2018) concluded that conceptual engagement has been limited, lamenting the over-emphasis on asserting terminology and the search of more embracing labels. There remains an absence of a conceptualisation of displacement which simultaneously involves compulsion and agency (Zetter 2018). An event-centred understanding of displacement prevails. However, displacement (a process) and forced migration (an event) are often conflated, concealing the processes occurring before movement begins. Displacement is a process that begins before people are forced to leave their places of residence and it can consist of an array of different pressures and constraints, as well as strategies of evasion and resistance, which are obscured by conflating displacement with the event of forced migration (Lubkemann 2008;Marfleet 2011;Celestina 2015;Vaz-Jones 2018). This paper uses Jabareen's (2009Jabareen's ( , 2014 guide to building a conceptual framework, identifying interlinked and constitutive concepts of a phenomenon, before integrating them into an explanatory conceptual framework. I hope to build an understanding of what displacement is through the integration of approaches to coercion (Wertheimer 1987;Anderson 2010), process (Stiegler in Lemmens 2011), and baselines understood potentially as rights and/or instrumental freedoms (Sen 1999). I offer a conceptualisation of displacement as a process of coercive disruption. Displacement is processual, coercive, and disruptive to valued ways of living and functioning. Importantly, it precedes the potential outcome of the processforced migrationwith which it is misleadingly conflated.
The article provides material for conceptual thinking on displacement across different contexts, but is underpinned by 29 extended narrative interviews (over 300,000 words of translated and transcribed data) about displacement in occupied Iraq, some of which I quote from directly, as well as observations from visits to Iraq in 2001Iraq in , 2016Iraq in , 2018Iraq in , and 2019. The interviews were conducted in Syria during 2010 and 2011 over 13 months. I interviewed Iraqis who had sought refuge there, through personal relationships formed during fieldwork, and through volunteering as an interpreter for the Iraqi Refugee Assistance Project, conducting research for UNHCR in Syria and interviewing willing participants outside of the UNHCR. I also interviewed students, and the parents of students, enrolled in an educational programme run by volunteers from the US in Syria, the Iraqi Student Project, where I volunteered as a teacher. Their testimonies related to transformations in daily life resulting from the US occupation, and earlier years of sanctions, and how they addressed consequent pressures and constraints. This included exploring decision making around whether to leave or remain in their homes, neighbourhoods, or their city or country of residence. I asked what factors shaped those decisions, who was consulted and involved, and whether others' decisions to leave or remain shaped their own choices. 1

Displacement: a process that begins in place, and differs from forced migration
In refugee studies, forced migration studies, and migration studies, most scholars refer to displacement and forced migration interchangeably. Bakewell (2011), seeking greater semantic clarity, wisely advises us to distinguish processes from conditions and categories, although he also conflates displacement and forced migration. Displacement is a semantically misleading term because it suggests being physically dislodged from a geographical location and is conflated with the event of being forced to leave a place of residence. Displacement happens first in place. This sounds contradictory, but many people experience the pressures and disruptions of displacement but are unable, or unwilling, to relocate. This semantic confusion relates to an event-centred perspective, the idea that displacement starts with the migratory movements from one point (home) to another (destination of refuge), an understanding which can obscure antecedent processual complexities (Marfleet 2011, 281). Displacement begins in place, and it can force a person to remain in place. Not all who wish to move can do so and may be involuntarily immobilised (Carling 2002). They may lack the resources to move, or they may be prevented by more powerful and better resourced agents. In effect, they are forcibly immobilised: rather than forced to migrate, they are forced to remain in place, potentially in a situation of diminished capacity for action, a problem for social groups for whom mobility is essential to income generation (Lubkemann 2008). Similarly, Penz (2006) refers to 'locational captivity', where people are compelled to remain where they are by threats, or by having their means of mobility taken away from them. These claims are supported by empirical research, including my own.
Conflating displacement with movement not only obscures the processes that lead to people leaving their homes, it also renders invisible the stratified, gendered, and generational ways in which transformative processes, such as armed conflict, impact on those who remain (Lubkemann 2008). This is a conclusion Lubkemann (2008) reached after a decade of researching conflict's effects on mobility in Mozambique. In occupied Iraq, displacement was a series of complex episodes: pressures and threats accumulated, compelling individuals, families, and even communities to take evasive measures, such as changing routes to and from places of work and education, to evade threats, before changing residence (Marfleet 2011). Other scholars acknowledge the in-situ processual nature of displacement. Reflecting on sustained ethnographic fieldwork about conflict and displacement in rural Columbia, Celestina (2015, 368) asks when the 'clock' of displacement starts, noting how the official focus on the moment of departure as the starting point of displacement 'hides the complex political, social and economic contexts that have generated displacement processes in which the state is complicit'. Vaz-Jones (2018, 714), considering land struggles, recognises displacement as 'ongoing and processual rather than a single moment', involving myriad ways in which people's livelihoods are destabilised, and their access to land and resources lost. The scholarship on displacement reframed as a process happening in place is limited but growing, says Vaz-Jones (2018), alerting us to pertinent literature: Mollett's (2014, 30) displacement-in-place does not involve physical movement but is constituted of displaced futures resulting from constraints on livelihoods and cultural practices. Nixon's (2011, 19) displacement without moving, is people losing the land and resources beneath them. Furthermore, Safransky (2016Safransky ( , 1089 understands displacement as a process entailing loss of access to public services and resources within people's communities, of 'being left behind rather than forcibly moved'. For Feldman, Geisler, and Silberling (2003, 9), people experience displacement in situ as 'relations of exclusion that set new boundaries for people's physical and social movement'. In situ displacement can affect specific social groups in disproportionate ways, such as when socially constructed norms and boundaries dislodge women from the means to attain economic stability (Wessels 2003). Focusing on the event of physically exiting a place can render such processes invisible.

Forced relocations
Furthermore, where it offers greater precision of an aspect of the process, I refer at times to forced relocation rather than forced migration because it identifies in the displacement process the short-distance physical relocations that migration misses. Forced relocation can be a potential outcome, and component, of the displacement processin that people may be forced to relocate multiple times before 'settling' so to speak at a 'final' destination. Migration is too closely associated with movement across long distances and borders to capture shorter journeys common in the displacement process. Forced migration is the most visible type of forced relocation because it involves people crossing international borders. But before this event, a person or household, may have already moved from one city to another, or to a provincial area. And before that, they may have moved to neighbourhoods within their city of residence. Movements within cities to evade threat are rarely documented compared to those across borders (some exceptions include Lindley [2010], Harling [2011]). As social processes, forced intra-city relocations have much in common with forced relocations to other cities and countries. Methodological nationalism and sedentary bias still dominate perceptions of human mobility, creating the distorted impressions that migration is a 'problem' (Wimmer and Schiller 2002), and consequently, that movements across borders are distinct social processes from movements within them (Marfleet 2011).

Building a conceptual framework of displacement around coercion
A conceptualisation of displacement that considers both agency and compulsion, and the related issue of choice under extreme constraint, remains absent (Zetter 2018;Chatty and Marfleet 2013). I offer one here, based on Jabareen's (2009Jabareen's ( , 2014 approach to building conceptual frameworks by identifying the constitutive concepts of a phenomenon, with their own attributes and functions, to understand and explain it. A conceptual framework is an interpretive approach to social reality, identifying a network of interlinked concepts to provide understanding rather than causative analysis (Jabareen 2009). I identify the concepts that constitute displacement from my research about Iraq, developing existing scholarship, to conceptualise it as a process of coercive disruption. Its constitutive concepts are coercion (Wertheimer 1987;Anderson 2010), process (Stiegler in Lemmens 2011), and the idea of baselines (Wertheimer 1987) which can be conceptualised differently, possibly as instrumental freedoms (Sen 1999), but also as human rights, acknowledging contextual variation. The aforementioned examples of in situ displacement represent coercive processes through which people are threatened to be left worse off in relation to baselines including personal safety, land rights, livelihoods, cultural rights, and their rights to secure residence (Carling 2002, Feldman, Geisler, andSilberling 2003;Lubkemann 2008;Penz 2006;Marfleet 2011;Nixon 2011;Mollett 2014;Celestina 2015;Safransky 2016;Vaz-Jones 2018). In the following pages, I set out the utility of different approaches to coercion, and how they link to process and to instrumental freedoms. Rigidity is not prescribed: researchers may identify more suitable notions of coercion and baselines. But the framework is intended to address a conceptual deficit in the study of displacement.

Side-stepping coercion and its moral aspects
Whether talking of displacement, migration, or forced migration, what constitutes force is typically taken for granted, with a reluctance to engage with the meaning of coercion. Considering that displacement is usually conflated with forced migration, this is puzzling as it relates to what distinguishes forced from voluntary migration, on a scale or between binaries. Symptomatic of this reluctance to address coercion is an assumption, underpinning some existing conceptualisations of voluntariness in migration, that because all migrations involve varying degrees of volition and constraint, investigating what constitutes force or coercion is unnecessary. Richmond (1994), whose work influenced similar positions, contends that all actions are constrained to some extent, and thus focuses on autonomy levels. However, his typology of 'reactive' migrations (with low autonomy), includes persons 'forced' to migrate because of economic collapse and 'threats to lifestyle' (Richmond 1994, 67-70), despite his disinterest in making a distinction between forced and free, or voluntary and involuntary movement (Richmond 1994, 48). Van Hear (1998, 42) supports Richmond's view that almost all migration involves different levels of compulsion and choice, arguing to focus on the extent of choice availablemore/less/little choiceand on directional aspectsoutward/onward/return/inward/ staying put (Van Hear 1998, 43-47), expanding upon Richmond's reactive/proactive matrix. Carling's (2014, 7) position is that all migration involves both choices and constraints, thus there is 'no categorical analytical distinction between "forced" and "voluntary" migration'. Erdal and Oeppen (2018, 982) also note that the complexities of migration observed in fieldwork convince them of 'the impossibility of describing someone's migration decision to migrate as entirely voluntary or entirely forced'. Yet such conclusions are drawn without engaging with what constitutes force or coercion.
There is a surprising dearth of systematic engagement with coercion in refugee and forced migration studies, and in migration studies; although engagement with coercion's conceptual counterpart, voluntarism, has recently emerged in the latter. Ottonelli and Torresi (2013) ask when migration is voluntary while engaging only tersely with coercion. For migration to be voluntary, one of the conditions is 'non-coercion', but it is defined with a conceptually limited policy definitionthe UN Protocol to prevent traffickingdespite recognising the need for scholarly interrogation of voluntarism. They and other scholars consider issues of choice and the availability of 'acceptable alternatives', influenced by Olsaretti's (1998) work on voluntarism, in assessing whether migration is in/voluntary (Erdal and Oeppen 2018). Erdal and Oeppen (2018) consider the distinctions between forced and voluntary migration, as well as the significance of labelling, but the discussion begins at the point of leaving, only briefly considering processes that precede migration decisions. Olsaretti (1998) considers the importance of acceptable alternatives for choices to be voluntary, but she is reluctant to engage with the 'thorny' issues around coercion which coercion theorists (Wertheimer 1987; Anderson 2010) address head-on. These include the potentially coercive nature of proposals, and how and by whom such situations are brought about (Olsaretti 1998, 75-76). Her reflections on voluntarism are valuable, but she acknowledges omitting detailed explanation of what is required for an alternative to be 'acceptable' (Olsaretti 1998, 76). This is something that Wertheimer (1987) argues is a moral issue: the moral aspect of coercion relates to the coercee being limited to a set of immoral proposals, resonating powerfully with my empirical research with displaced populations who were confronted with choices such as abandoning the family home to a militia or risking the safety of loved ones. Importantly, despite Olsaretti (1998:74) arguing against a moralised approach to voluntarism, she still refers to 'morally worrying limited choice situations' that are 'dire', suggesting tensions in her reluctance to consider the moral aspects of constrained choice situations.
Furthermore, both Olsaretti and Wertheimer focus on the mechanics of coerced or voluntary choices. These are important but epiphenomenal aspects of coercion (Anderson 2010). Focusing exclusively on these means overlooking the wider issues of the power imbalances that allow choices to be constrained by an agent, institution, or consequent structural transformation (Anderson 2010). Understanding why a person is limited to unacceptable alternatives, requires 'zooming out' so to speak, to analyse the wider power dynamics upon which coercion, in its different forms, rests (Anderson 2010). The power behind coercion, as I hope to show with examples from research, is as pertinent to understanding displacement as are the mechanics of choice.
Coercion: a constitutive component of displacement Building a conceptual framework requires identifying interlinked constitutive components of a phenomenon to provide an interpretive understanding of it (Jabareen 2009(Jabareen , 2014. A key constitutive component of displacement is coercion. My purpose here is not to present an extensive discussion of the debates about what coercion isalready done elsewhere (Anderson 2021). Rather, it is to explain the relevance to displacement of two approaches to coercion using examples from fieldwork. I will also show why it is conceptually more suitable than approaches centred on voluntarism.
Threats, constrained choices, and immoral proposals in displacement As long ago as 350BCE, Aristotle wrestled with the issue of what constitutes voluntary and involuntary action, and the deliberations which precede decisions under threat, considering them alongside acts taken under compulsion in Nichomachean Ethics. His musings seem particularly relevant here. Aristotle defined clear-cut involuntary actions as those where the origination is external to the agent. The agent contributes nothing, as if carried away by a wind or as though men had power over his person. But Aristotle also said that actions can be 'mixed' in nature (Aristotle, NE, iii, 53, 54). Even in severely constrained circumstances certain actions taken under compulsion are more like voluntary acts for Aristotle because the agent physically undertakes them. Forced relocations relating to displacement, if seen through this lens, are actions of a 'mixed kind'. They are done: … from fear of greater evils, or from some honourable motive, as, for instance, if you were ordered to commit some base act by a despot who had your parents or children in his power, and they were to be saved upon your compliance or die upon your refusal … (Aristotle, NE,iii,53,54) Another example is a man throwing possessions overboard in a storm to prevent a ship from sinking (Aristotle, NE, iii, 1:54): a 'mixed kind' of action and 'choiceworthy' at the time. He was the originator of the motion of his limbs, and with himself being the origin of the action, 'it rests with himself to do or not to do' (Aristotle, NE, iii, 1:54). In both cases, a choice is made, and the ends and objectives of these actions need to be understood with reference to the actual occasions. 'Such actions then are voluntary, though in the abstract perhaps involuntary because no one would choose any of such things in and by itself' (Aristotle, NE, iii, 1:54). People abandon their homes for fear of their own and their loved ones' safety. In Iraq, a common saying is that regime change removed Saddam Hussein and replaced him with multiple despots, such as corrupt ministers with personal militias. Some of my participants left their homes after receiving threats from militias. They were not literally forced into taking each step of the journey, such as packing up their belongings, in chains or at gunpoint. They would not have otherwise done such things without assessing the credibility of the threat, and enacting a threat evasion strategy. Considering their experiences, the Aristotelean perspectivethat abandoning homes in response to fatal threats is voluntary, and only involuntary in the abstractseems inadequate.
Wertheimer moves the discussion forward. While a person may take action, voluntarily, with their choices constrained, the person's confinement to constrained alternatives is not voluntary (Wertheimer 1987). Involuntary actions take place when a person must choose from options that are contrary to that person's moral will (Wertheimer 1987, 301-302). 2 It is not merely about disliking the available options, as reluctance and voluntariness can go side by side, rather: One acts involuntarily because one has a deep aversion to having to choose in response to immoral proposals. Coerced choices are not unwilled, but they are, it may be said, against one's will. (Wertheimer 1987, 302) Moralised approaches to coercion are contested by philosophers. A satisfactory discussion is beyond the scope of this paper, but they deserve some attention. Olsaretti (1998) argues against a moralised approach to voluntarism. But perhaps this relates to the purpose of her argument which is to critique Nozick's rights-based libertarian approach to voluntarism. Nozick says actions are not involuntary if the persons limiting the alternatives of others are acting within their rights, a problematic minimalist approach which Olsaretti critiques. The point here is not to contest this critique. Rather it is to say that: Wertheimer's moralised approach to coercion differs from Nozick's to voluntarism; the moral aspect of coercion should not be dismissed on the basis of issues with Nozick's approach; immoral coercive proposals are evident in the displacement process; Olsaretti's insistence on a non-moralised approach to voluntarism contains tensions. For Olsaretti, voluntarism rests on an agent having acceptable alternatives from which to choose, which are either 'things which one very much likes', or 'things which one does not particularly like but which are acceptable' (Olsaretti 1998, 72). Unacceptable options are 'thoroughly bad … which no rational agent may reasonably be expected to choose. An acceptable option, by contrast, is one which is not thoroughly bad and is, therefore, choiceworthy.' (Olsaretti 1998, 72). This is vague. There are conceptual issues with conflating non-voluntarism, or coercion, with things which one very much dislikes (see Wertheimer 1987, 192-201), implied in Olsaretti's account. And, while arguing against a moralised account, Olsaretti (1998, 74) refers to 'morally worrying limited choice situations', which are not unlike Wertheimer's moralised coercive proposals and hard choices, outlined in what follows.
Moralised baselines chimed with important elements of Iraqis' displacement narratives. Here are some examples. For long periods in Baghdad, leaving one's home meant potentially abandoning it to militias who were seizing properties of the people they threatened, with little prospect of regaining possession. The home is often a household's most significant material asset with huge emotional value. However, refusing to abandon a home after being threatened presented another immoral proposal: of risking loved ones being harmed. Other examples included the proposals that the security situation imposed on the pious among Iraq's Mandaeans, an ancient sect following the teachings of John the Baptist. Insecurity in Iraq meant that for long periods they were unable to practice baptisms in flowing rivers if they remained. Regular baptisms are compulsory and relate to spiritual regeneration and not only to births and conversions. However, leaving Iraq also risked international dispersal which could threaten the existence of the sect as Mandaeans are forbidden from marrying non-Mandaeans. When schools were caught in the crossfire of armed clashes, some parents transferred their children to inferior schools to avoid violence. Sending their children to inferior schools had moral implications, but not doing so could be more immoral with the risk of being shot in crossfire. These are not only options that people 'very much dislike' which are 'very bad', in line with Olsaretti's (1998, 74) unacceptable choices, they relate to immoral proposals that constitute threats (Wertheimer 1987).

Threats and coercive proposals in displacement
Threats can be constitutive of both coercion and displacement. But to avoid taking this concept for granted, I shall explain what a threat is. It is a proposal which one agentthe coercermakes to anotherthe coercee. The proposal limits the coercee to a set of choices all of which will leave the coercee worse off to different degrees, depending on whether the proposal is accepted or rejected. In this approach to defining coercion, being worse off is considered in relation to a standard or a baseline. An offer is a proposal from A that will leave B better off in relation to a baseline if accepted, and no worse off if the proposal is rejected. An example might be if B is securely and happily employed by one firm, and A is the director of a rival firm. A offers B a job in her firm, with improved pay, located closer to B's home than in his current post. There are no consequences for B if he rejects the proposal; he is assured of being better off in relation to an economic baseline, and a temporal baseline, with time gained from a shorter commute. That is an offer. But if A's proposal was that B should take a job in A's firm that involved similar working conditions for a decrease in pay, that means the offer makes B worse off in relation to the economic baseline. It is coercive if A also threatens to make B worse off should he reject A's proposal; such as if the director then says that she will spread allegations relating to historical misconduct on B's part if B refuses the offer, which could lead to B being dismissed from working altogether. B is damned if he accepts the proposal, and potentially more damned if he rejects it. The proposal is coercive when either way the agent is left worse off in relation to a moralised baseline and this is the essence of threats (Wertheimer 1987).
Residents of Baghdad were regularly confronted with coercive proposals in daily life after the occupation. On the city's roads, they varied in intensity depending on levels of violence which varied over time and across neighbourhoods as conflict dynamics shifted. To remain in occupied Baghdad, one had to tolerate longer, costlier, and more dangerous travel arrangements because firefights, explosions, and armed forces at checkpoints added threat to daily commutes. US soldiers often killed occupants of vehicles who misunderstood their English orders. Militias, motivated by ransom acquisition, sectarian agendas (or both), kidnapped or executed people at checkpoints. Evading these threats turned Raz' journey to school from a 30 to a 75-minute commute. The importance people attach to affordable and straightforward mobility certainly varies. But those manning the checkpoints created coercive proposals in the way they deformed Baghdad's roads. Pass through a checkpoint and risk violence, or take another, likely longer route? Both options leave people worse off in relation to the baselines of personal safety and increasing time and fuel costs. Raz, a teenaged boy growing up in Baghdad when the US invaded Iraq, also spoke of how US forces used his school's grounds to park armoured vehicles, creating the possibility for the school to become a site of conflict should those fighting the occupation choose to attack them there. Jasim, also a teenager when the invasion began, and whose school neighboured a police station where US forces parked their vehicles, was left in shock after an attack on a US military vehicle shattered the school's windows. His parents transferred him immediately to another school where another explosion occurred on the day he registered. His parents decided to leave Iraq after that incident as they believed that remaining there would jeopardise his safety. The threat was not issued directly to Jasim or his family, they were not the intended targets, but the violent circumstances presented a coercive proposal to them to which they responded by leaving Iraq.
It is worth acknowledging that context and subjective perceptions can shape coercion in displacement. In Baghdad, while some felt protected by the presence of certain armed factions, others felt coerced by them. In Farah's neighbourhood, a mixture of Christian and Shia households, the Mehdi Army militia became active. It was the loosely controlled armed wing of the Sadrist movement whose social base at the time was Iraq's marginalised Shia underclass. It was engaged in theft, sexual violence, and the violent eviction of Sunni and othermostly but not exclusivelynon-Shia households. The Mehdi Army used the houses for different purposes: often to rehouse Shia families who had themselves been evicted by Sunni militias, but also as offices and makeshift prisons. A dentist I interviewed recalled how, in the quiet of the night, he could hear the screams of men being tortured in a house seized by the Mehdi Army close to his home. Members of the Mehdi Army attempted to enter Farah's home on one occasion but were blocked by the quick thinking of Farah and her sister: they quickly locked all the doors and windows and refused to let the men in as her parents were not home. They were both in their early teens at the time. Men from the militia entered their home during a new year's celebration, claiming to be patrolling for illegal alcohol consumption. This was a common pretext the militia used to evict Christians from their homes. A neighbour of Farah's, from a working-class Shia household, told her that she should feel safe and protected by their presence. Unsurprisingly, Farah did not. Remaining where they were meant that further militia intrusions were a possibility.
Coercion in displacement does not always result from obviously identifiable individuals. Rather, it may be structural circumstances and transformations that present coercive proposals to a population or specific group within it. Galtung's (1969) work on indirect or structural violence comes to mind, where there is not an identifiable person who acts, but where harm that prevents people from reaching their somatic and mental potential is built into the structure, from unequal resource distribution, possibly in unintended ways. In the displacement process, coercive proposals can be evident as transformations or situations that leave people worse off in relation to baselines, but which do not necessarily compel people to leave their homes in and of themselves, but add to the pressures of daily life. Certain infrastructural examples from occupied Baghdad come to mind, such as the collapse of Iraq's national electricity grid after the occupation. A system that offered energy in alternating two-hour slots after the 1991 war, and 13 years of international embargo, barely survived the post-2003 era. The embargo was lifted and Iraqis could purchase consumer electrical goods again, particularly air conditioning units for increasingly unbearable summer daytime temperatures of 50c, and 30c evenings. But the grid was never updated. In the early years of the occupation, the national grid provided two hours of supply daily, compelling dependence on expensive and unreliable private neighbourhood generators. Again, this in and of itself may not be enough to compel a person to relocate permanently (although some Iraqis who could afford to do so began spending their summers in Syria and Turkey). But add to it the many infrastructural failures that Iraqis have had to endure over decades, and it can be seen as a displacement that entails a loss of public services, a loss of resources in a community, 'being left behind rather than forcibly moved' (Safransky 2016). Dewachi (2017) notes the degeneration of medical infrastructure, particularly under international sanctions in the 1990s, and of Iraqis compelled to travel to Beirut for treatments unavailable in Iraq. In Damascus, until the very early years of the Syrian uprising, it was common to see Iraqis there for the same reason, as care was more affordable than in Lebanon. It is tempting to see the above circumstances in a non-moralised way, as people acting nonvoluntarily in their confinement to unacceptable bad alternatives (Olsaretti 1998, 74), especially where alternatives are consequences of indirect/structural violence (Galtung 1969). However, structural violence can have identifiable perpetrators, often politicians who disallow life to others through budget decisions (Tyner 2016). Iraqis living through the sanctions of the 1990s were acutely aware of the US and UK governments' insistence on perpetuating, via the UN Security Council, international sanctions, punishing Iraq's population for the actions of its dictator Saddam Hussein long after Iraqi forces fled Kuwait in 1991. They are also acutely aware of their politicians stealing oil and tax revenues for personal gain, neglecting infrastructural regeneration. With this in mind, it is possible to see both the immoral proposals in having to seek expensive private medical care abroad, and paying for expensive and unreliable private generators as a result of Western-backed interventions and elite corruption.

Freedoms to function as disrupted baselines
A conceptual element of displacement is the baseline, which can be understood in different ways and can relate to infrastructure and public services. The idea of 'instrumental freedoms' from development and welfare economist Amartya Sen (1999) offers a framework to think about what these baselines might be. The point is not to produce a rigid typology of what is disrupted, but to encourage us to think of what mattered to displaced subjects when they were making decisions to address coercion through different means. Sen speaks of substantive human freedoms that development should be concerned with expanding. They include 'elementary capabilities' like: … being able to avoid such deprivations as starvation, under-nourishment, escapable morbidity and premature mortality, as well as the freedoms that are associated with being literate and numerate, enjoying political participation and uncensored speech and so on (Sen 1999, 36).
The expansion of these elementary capabilities, or freedoms, is intrinsically important as a primary end of human development. Sen argues that development itself should be a process that enlarges human freedom, or capabilities, in general. (Sen 1999, 37) Different kinds of rights, opportunities, and entitlements contribute to development in two principal ways: their expansion contributes to the expansion of human freedom more generally, and to the expansion of other interconnected instrumental freedoms. Sen identifies five distinctive, complementary, but not exhaustive groups of instrumental freedoms, impacting directly and indirectly upon the abilities that people have, to live in ways that they value (Sen 1999, 38-39). They are summarised as follows, but not in order of importance. First are instrumental political freedoms: to be able to choose governments from a selection of parties, and to be free to express political views, to dissent and to organise. Second are economic facilities, meaning the opportunities for individuals to make use of economic resources to generate income and wealth. Social opportunities refer to arrangements that society makes for the welfare of individuals to be able to participate economically and politically. Transparency guarantees give people the freedom to operate on some sort of presumption of what they are being offered, and can expect to get, in their social interactions, preventing corruption. And protective security refers to the presence of a social safety net that prevents the vulnerable from falling into abject destitution. Instrumental freedoms can interconnect. In the displacement process, we can think of these instrumental freedoms as baselines that can be disrupted in ways which constrain choices in morally unacceptable ways. This could be because of the risk of harm to loved ones, or to their own future prospects. Mohsin was still at high school when the occupation began and disrupted his education. His father, a former teacher in a military university, relied on pension payments which stopped during the first year of the occupation after the US dismantled Iraqi state institutions. Mohsin had to work to support his family. Between 2003 and 2005, long power cuts made it difficult to study, as did insecurity in Baghdad, which also led teachers in his school to miss classes, while others left altogether. Violence in his neighbourhood led his parents to move him several times to the homes of relatives elsewhere in Baghdad until Mohsin completed his studies in June 2005.
Instrumental freedoms are not suggested here as a rigid or comprehensive framework for understanding baselines. Human rights may in some instances better capture the baseline that is being coercively disrupted. For example, Sen's framework is development-focused, omitting religious freedomsprotected explicitly in Article 18 of the Universal Declaration of Human Rights (UDHR). For significant periods after 2003, religious groups in Iraqincluding the Mandaeans and Christianswere not able to practice their faith in safety; and it is now five years since the Genocide of the Yezidis at the hands of Da'esh. But thinking about instrumentality can sometimes assist our understanding of displacement better than rights, because it is not only the denial of rights that is important, but the implications of their denial for the ability to do things. This is clear in the case of property. Article 17.1 of the UDHR cements the right to own property, and 17.2 states that no one shall be arbitrarily deprived of their property. A militia depriving a household of their home is of course significant, but in a displacement process, we also need to think about the implications of this deprivation for how persons can function in a society after being dispossessed. This is particularly important in a city like Baghdad where rents and property prices spiralled after regime change. Additionally, a crucial baseline to consider is the integrity of social support networks which can be undermined by the departure of others. People rely on cultural and social ties for psychological, emotional, and material support. They are based on important familial ties, language, religious practices, and traditions. As more and more people leave their homes, these networks begin to break down and eventually the difficulties of remaining increase (Marfleet 2007, 407;Moore and Shellman 2004, 728). Their depletion at times of insecurity and crisis becomes particularly unsettling and is itself a form of coercive disruption. A final addition to the freedoms that I suggest is secure mobility. Iraqis recalled the consequences of losing the freedom to move around Baghdad securely and freely. Even today, many continue to use trusted private drivers instead of public transport or hailing random taxis on the street. Being able to generate income, access healthcare and continue in education, all require secure freedom of mobility.

Hard choices
'Hard choices' are another constitutive component of displacement, and evident in forced relocations. Alluded to in earlier empirical examples, they are an extreme form of coercive proposal. Some proposals are too unpalatable to accept, but if rejecting them means remaining in dire straits, then a person is faced with a hard choice situation (Wertheimer 1987, 233). The important difference between hard choices and other choices, including coercive proposals, is the 'particularly severe constraining effect' they have because each of the few options leaves the agent much worse off in relation to a specific baseline. But even in a hard choice situation, there can be autonomy and agency, however highly constrained. In a hard choice situation, the scope of choices is extremely limited, but: the choice among those options may be of especial importance. In the latter sense, making a hard choice may constitute an important and positive assertion of our autonomy. Nonetheless, just as the prospect of hanging is said to focus the mind, (very) hard choices produce too much focus and not enough scope. (Wertheimer 1987, 233, emphasis in original) The intensity of their focus and the limited nature of their scope qualify hard choice situations as forms of coercion, duress, force and compulsion (Wertheimer 1987, 5-6). Wertheimer uses these terms in a rough way without drawing important terminological distinctions between them, as I also do here.
The concept of hard choices captures many situations which confront individuals in the displacement process, particularly before forced relocations. Common in occupied Baghdad was the threat letter, purporting to be from an armed group, being delivered to a home (sometimes as wrapping for a bullet). Letters contained threats to harm the occupants unless they left their homes, forcing a hard choice upon them. The recipients had to decide whether to abandon their homes to preserve the lives of loved ones or remain and risk their own safety in order to protect their home. Each of those options is morally unpalatable and could leave participants in dire straits. Some risky options are available. Perhaps the threat was a bluff or a prank. Perhaps mistaken identity was at playcould the armed group be reasoned with, or could it anger them? Hard choices often relate to the most fundamental baselines: the safety of oneself and loved ones. Returning to Mohsin's experiences, after completing his studies in 2005, he returned to live with his parents after living with relatives elsewhere in Baghdad. But they were confronted with a hard choice situation in October that year. US forces raided their neighbour's home, arresting several young men from the family, and killing two others. One was Mohsin's dear friend. US soldiers had taken photographs of Mohsin with his friend in the raid. His parents feared they would return for Mohsin. Awareness of torture in US detention, like in Abu Ghraib prison, was widespread by then. Confronted with a 'hard choice' between leaving him at risk of violent harm if he remained in Iraq, and moving him elsewhere, they called upon an uncle in the United Arab Emirates (UAE) who sponsored and hosted him until Mohsin moved to Syria in 2006, to where his parents had fled sectarian violence in Baghdad, and where living expenses were lower than in the UAE. Neither choice was morally palatable. Remaining in Iraq entailed the risk of death, and his economic circumstances in the UAE and Syria precluded a university education there, but the latter was less unpalatable than the former.

The power behind threats: the enforcement approach to coercion
The forms of coercion in displacement thus far have centred on an approach focused on threats and pressures (Wertheimer 1987), which assumes that 'coercion always involves the use of threats' (Olsaretti 1998, 54). It takes for granted the reasons why an agent's choices can credibly be constrained by threats, and avoids the 'thorny' issue of how and by whom the situation was brought about (Olsaretti 1998, 75). This is inadequate because we must interrogate how and why threats and pressures are credible, and because the pressures approach misses certain forms of coercion (Anderson 2010(Anderson , 2016, which, I shall show, are also constitutive of displacement. A summary of the enforcement approach follows. Coercion centres on power differentials that are created or leveraged by the coercer to constrain the possibilities for action of the coercee. Importantly, the coercer may be an institution, like the police, or the state, enforcing its laws in different ways. The coercer may constrain the coercee, compelling him to do something, or may threaten to use powers against the coercee to necessitate action. The coercer may also use a power imbalance to disable or shackle the coercee. For coercion to occur, a coercer must be willing to use a power imbalance and not simply possess that power; and the coercee is unable or unwilling to counter the coercer with similar or greater powers. Coercion then impacts on the coercee's situation by foreclosing some possibilities for action immediately. The coercee then sees that it is necessary to do what the coercer demands, for practical reasons, because it is practically necessary to avoid the situation the coercer threatens to bring aboutor to exit the situation once it has been brought about. The enforcement approach to identifying coercion does not require the success of the attempt to constrain or change the actions of the coercee as a condition for coercion to have happened. The attempt to coerce may affect many, some, or none of the coercee's activities. A coercer's aims may be foiled by defensive measures, or by retaliation with another threat. Failure, on the part of the coercer, to achieve the intended outcome does not mean that coercion did not occur. Bluff threats can be coercive even if they fail (Anderson 2010). This is a point of difference between the enforcement approach and the threats and pressures approach, particularly for Olsaretti (1998, 54). Olsaretti argues that coercion always involves threats, but that threats are only coercive if they constitute 'forcings'consequent actions of an agent who does something because there is no acceptable alternative to doing it (Olsaretti 1998, 54). In addition to the vague explanation of what is un/acceptable, the notion that coercion requires a forcing to occur is symptomatic of the threats/pressures success condition of coercion. If the situation of constrained choice does not produce the intended actionthe 'forcing' in Olsaretti's termsthen coercion did not occur. But the experiences of people who receive threats suggest we should not dismiss the coercive effects of threats on the basis that they did not produce the intended actions. Receiving a threat letter can produce fear, distress, sleepless nights, and other harmful effects, whether or not the threat leads to the action intended by the coercer. The dentist I interviewed had many sleepless nights, with a gun under his pillow, after an altercation with militiamen who entered his garden in an intimidation attempt. Those who received threat letters might find it puzzling for their feelings of being threatened or coerced to be dismissed on the basis that they did not do what the coercer demanded. The enforcement approach can help us to identify coercion that the threats/pressures approach misses, and to understand some of the power imbalances underpinning displacement in Iraq.
Actions which are not overtly communicative of threat are also constitutive of displacement as a form of coercion. In occupied Baghdad, certain actions took on a coercive character that they would not have had if undertaken before the occupation. These resulted from the violent transformations of the occupation: the US military and various militias were empowered and competing against each other for power. In the context of Baghdad before the occupation, spraying the word matloob, meaning 'wanted', on the outside wall of a house was more likely to be read as an act of rude vandalism, prompting inquiries to neighbours for witnesses, possibly a call to the police if the hassle of removing the graffiti was less than the hassle of involving the police. After the 2003 invasion, the violent dismantling of the state, and the ensuing widespread insecurity, such acts of vandalism took on a menacing character because of the shift in power relations. Under US auspices, the new order in Iraq failed to provide security and public safety. US occupiers failed in their own obligations to do the same. The consequence was a security vacuum which led to militias infiltrating neighbourhoods and the Iraqi police. Militias were tied to political parties, and to criminal gangs, with lines between them blurred. Armed militias were able and willing to use their power to expel unarmed populations from specific neighbourhoods, often constrained only by other militias or ad-hoc neighbourhood defence forces. Militias used threatening graffiti or threat letters to expel households. These methods are more efficient than physically restraining or shackling members of a household, or frog-marching them off the premises at gunpoint. Amina's father was a retired engineer in the Iraqi Army and when an unknown group sprayed the word matloob in large letters on the outer wall of their home, he decided that they should leave. The graffiti appeared on their home at a time when former army officers were being murdered, and when others who had had such words sprayed onto their homes were confronted by militias if they did not leave their homes.
This type of coercion, designed to vacate a property, requires a history of violent coercion for it to succeed and is a common component of displacement. It is called the commons of enforcement powers (Anderson 2010, 3) because it relies on a history of past acts to be credible. Organisations like mafias and states rely on populations understanding that they have the means and the will to enforce their powers, especially when defied. This also makes it possible for agents who resemble those organisations, superficially at least, to enact bluff threats by free riding on the commons of enforcement powers (Anderson 2010, 3). In Baghdad at the height of the widespread violence, it was possible to free-ride on the commons, to send threat letters and spray menacing words on people's homes that, if resembling closely enough the words of the militias, could force a household to leave their home. Um Ahmed lived in Ghazaliyya, a suburb of west Baghdad, with her husband and teenaged children. She recalled how one of her neighbours, with strong sectarian prejudices, delivered a threat letter to another neighbour who promptly vacated his home. Her son witnessed the incident, and the malicious neighbour spoke openly of his dislike for the other sect. He did not have the means to physically expel his neighbour, nor was he a member of a militia. But he took advantage of the commons of enforcement powers during the height of sectarian violence in 2006, when such threats were frequently carried out.

Immobility as a component of displacement
Additionally, coercion can immobilise as well as forcibly relocate people. At the peak of violence in Baghdad, especially in 2006-2007, people left their homes only when absolutely necessary. The fear was worse for those who had received threats, and sometimes led people being as afraid to venture out of their homes as they were to remain in them. Hadi, a civic rights activist in Iraq, told me that 'sometimes I did not leave my home for days, or weeks' because of a barrage of intimidating phone calls. Raz' father narrowly escaped militiamen targeting him at work after receiving a warning but was confined to his neighbourhood which was off limits to that militia. Neither Hadi or Raz's father were able to generate income or engage in civic activism while immobilised. Raz explained that this confinement was the final coercive pressure that led to the family leaving Iraq, showing how disruptions to economic baselines negatively impact other instrumental freedoms.
Another component of the displacement process is coercion where there really is no choice. Paradigmatic of this is someone being shackled, physically restrained, drugged into incapacity, imprisoned, or a combination of those. This is something missed by understanding coercion as threats and pressures (Wertheimer 1987;Olsaretti 1998). The enforcement approach is interested in this choiceless coercion because of the contingent power imbalances. Such coercion can occur in the displacement process in influential ways. For example, Adnaan was arrested and detained in Abu Ghraib prison before he decided to leave Iraq. He was not deported, but the prospect of political imprisonment in Iraq shaped his decision to leave. 'Iraqi security forces did of course treat me with total respect' he said with a morbid sarcasm. In another setting, the type of forced migration where this type of coercion is starkly apparent is in deportation, with individuals being detained against their will until they can be forced onto flights, often handcuffed (Gibney 2013).
Coercion that is not overtly communicated by threats is another component of displacement. It is, accordingly, overlooked in the threats/pressures approach, but the enforcement approach sees it because it is contingent on recognisable power imbalances (Anderson 2010). The despot in Aristotle's example may not need to kidnap a man's relatives to coerce him into action. A despot could likely just 'ask' him to do something which then prompts action from anticipatory fear of the consequences of refusal. Populations can perceive threat without its overt communication. An example is when a militia takes over a neighbourhood, and signals its presence with banners, flags, sonic/acoustic intrusions/interventions, and other signifiers which some residents perceive as threatening. In Baghdad, militias sprayed slogans on walls to announce their presence and stake claims to territories. In some neighbourhoods, some mosques added loudspeakers to increase the volume of sermons, which themselves changed character in ways that signalled changing power relations, including a militia's presence. Residents may decide to move to avoid potentially threatening interactions and experiences with that militia, even if it has no history of harming their social group. The mere presence of a militia in a neighbourhood can be enough to produce displacement, creating a feeling of insecurity, and accelerating relocations. Residents may feel coerced even if militiamen do not directly or interpersonally communicate threat. A store owner who refuses to discount goods for a customer, may still refuse to discount the same customer even if she returns with a burly man who pointedly asks the store owner: 'Is there a problem?' The store owner may counter the intimidation attempt by saying no and threatening to call the police. But what if a militia has infiltrated the police? And what if the burly man carries a gun? And what if the man is in the attire of a militia that has signalled its presence in the area? When that man asks the storekeeper 'Is there a problem?', and the storekeeper is not able or willing to use a gun, and is not protected by another militia, we can understand if the storekeeper feels coerced. In Baghdad, where Farah's father ran a store, a woman who was refused a discount returned with a Mehdi Army militiaman who asked her father if there was a problem. He replied 'no' and sold the item at a discount.
Amplifier coercion I also suggest considering what I call amplifier coercion as a potential component of displacement. This is when coercion is targeted at a specific group with the intention of also coercing another, usually larger and interdependent, group. It differs from acts of exemplary coercion (often violent), applied to a small number of people from a specific group to coerce members of that group by making an example of them. The killing of a Christian priest sets an example to his congregation of what can happen to them if they remain, but is unlikely to affect other social groups with the same intensity. Amplifier coercion targets a group, like shopkeepers in a Baghdad neighbourhood, in order to coerce that group and another larger interdependent group, like the residents of that neighbourhood. Methodical attacks on shopkeepers made other shopkeepers flee and in turn disrupted the lives of residents in the area, forcing them to flee for lack of supplies (Harling 2011, 50). Anwar, in his twenties during the occupation, witnessed the assassination of a shopkeeper in a drive-by shooting close to where he had opened a store.
There was a time when they were killing the ice vendors. Then they started killing people who sold fruit and vegetables. That was an incident that really made me afraid to stay in my work. I was thinking 'are they going to kill shopkeepers now?' He closed his store, understanding the hard choice between losing his income or losing his life. The coercion was amplified because the store closure also disrupted residents' access to food and other supplies, prompting them to consider if they ought to remain in a neighbourhood with limited access to food at a time when moving around Baghdad involved crossing checkpoints manned by trigger-happy US soldiers and violent militias.

Process as phases and counter phases in displacement
The concept of process is constitutive of the displacement phenomenon. A process involves phases and counter-phases (Stiegler in Lemmens 2011) and in displacement it can mean evading or countering threats and pressures. Resistance to the pressures that seek to displace people, be it to constrain, exclude, expel, or immobilise them, is possible. It is, however, contingent upon resources and power relations which relate to social identities and class. In this way, displacement is a dynamic process which involves phases and counter phases (Stiegler in Lemmens 2011). These phases are not neatly sequenced nor evenly balanced; power imbalances can mean that people may not have the effective means to counter their displacement (Penz 2006).
Counter-phases were evident in Iraq as people sought ways to evade and resist attempts to expel or harm them and sometimes their neighbours. Raz, a teenager during the invasion, participated in self-organised nightly neighbourhood watch shifts, keeping watch from rooftops with guns to hand. The neighbourhood was populated by retired army officers' families (among them Raz's father), some with connections to resistance fighters active in the area, making it a hostile place for looters, militias, and US forces, a social capital providing valuable security. Hadi, a businessman and prominent civil rights advocate for the Mandaean sect in post-Ba'ath Iraq, was forced to take a number of threat evasion manoeuvres in order to remain in Iraq. In his apartment in Damascus in 2011, he explained how he kept his meetings secret, varied his routes, and sometimes made no outings at all, in the face of persistent threatening phone calls. He used the premises of one of his inactive businesses as a secret meeting place until coercers set off a car bomb outside. They targeted him in 2009 for criticising Iraqi authorities' exclusion of Mandaeans in political processes, and for security failures. After Hadi's son was kidnapped by the same people threatening Hadionly released upon payment of a $50,000 ransomthey fled to Arbil in Iraqi Kurdistan. From there he again criticised the government in Baghdad, but was again targeted by Kurdish Asaysh security forces, allies of the government at the time. They detained his eldest son while Hadi made an exploratory trip to Damascus to register with UNHCR, releasing him only in exchange for Hadi's imprisonment. He was detained for weeks, but an influential contact in Arbil secured his release. However, it was a threshold point; he no longer had the appetite to resist his displacement and he then moved to Syria with his family. Access to different endowments of capital influence trajectories of mobility (Vaz-Jones 2018). Hadi's social and material capital allowed him to evade threat and resist what Feldman, Geisler, and Silberling's (2003, 9) conception of displacement in situ as efforts to expel him from Iraq as 'relations of exclusion that set new boundaries for people's physical and social movement'. Eventually, more powerful coercers were able to expel him from Iraq. And while Hadi showed agency and resistance to displacement, his exit from Iraq was very much forced.

Conclusion
In building a conceptual framework of displacement based on the experiences of Iraqi subjects, I have shown that it consists of several conceptual components: different forms of coercion; contextually variegated baselines, such as instrumental freedoms and rights; and processes consisting of phases and counter phases. I have also shown the relevance of moralised baselines in displacement. The conceptualisation shows that many forms of coercion constitute displacement, and that power imbalances often underpin them. In addition to threats to personal safety, there are threats to valued ways of living and functioning, and other constraints on people's capacity for action, such as the effects of amplifier coercion. Concurring with recent scholarship on displacement, I have also shown that it occurs first in place, before people are forced to relocate. Focusing on coercion rather than on voluntarism helps tackle some of the thorny issues that approaches to voluntarism avoid, and, it offers precision in understanding what is forced about forced relocations and migrations. Coercion also offers something else. Understanding the coercer, or the coercive structures, behind displacement requires identifying the power underpinning them, at least when adopting the enforcement approach. It is possible to do so when looking at volition and voluntary action, as migration scholars often do; but it is not built-in to the analysis. Identifying the coercer has a moral imperative to it. Displacement, forced relocations, and forced migrations, are often linked to acts of injustice and oppression by the coercers. Too often, media and policy discussions and representations of displacement are made without reference to the power dynamics behind displacement, forced migration, and forced immobility. Focusing on coercion highlights the role of the coercers, such as states, warring parties, and unjust economic distributions, as well as the coerced, drawing attention not only to the coercive power imbalances that displace and expel populations; but also to restrictive border-regimes which prevent people evading the coercive disruptions of displacement.