The collective victimization of a civilian population
As for the suffering, any human being of any race, ethnicity, colour, or religious conviction suffers the same.1
This chapter examines how a collectivity of civilian persons is understood to comprise a civilian population. It builds on the analysis of the complexities of the adjudication of the identities of individual civilians set out in the previous chapter to examine how collectivities of civilians are ‘recognized’ as a cohesive population and have their legal protections enforced as such by the processes of international criminal justice. The chapter argues that acts of collective victimization are most often understood through the frame of ‘ethnic cleansing’, that is, the victimization of a group of persons because of their national, ethnic, religious or political identities or affiliations. This model of victimization tends to understand a civilian population as comprising a collectivity of persons with a shared social identity. It assumes that the individuals that comprise this population have common bonds or characteristics that shape their existence as a social group as well as a civilian group of persons. However, analysis of testimonies given by civilian victim-witnesses in the cases of D. Milošević and Galić call this form of ‘social’ groupings of civilians into question. Notions of ‘solidarity’ and ‘allegiance’ are shown to underpin civilian identities and the collective experience of conflict as a cohesive populace, without necessarily being fixed or shaped by ethnic categorizations. Taking account of this narrative of the form of a civilian population, the chapter considers the Trial Chamber’s judgement of the social and legal relations of the civilians that comprise the populace of the city of Sarajevo. It argues that the Trial Chamber not only uses ethnically based depictions of civilian groupings that reassert the ideologies of the warring parties, but fails to represent the ‘positive’ social relations of civilians beyond such groupings before, during and after the siege.
In its Opening Statement at the trial of The Prosecutor v. Stanislav Galić, the Prosecution began with a description of the atrocities committed against the civilian population of the city of Sarajevo during the Yugoslavian conflicts. Depicting the events under adjudication as causing such prolific and systematic harm to the civilian population that they were without similarity in contemporary armed conflicts, the Prosecutor described how:
[t]he siege of Sarajevo, as it came to be popularly known, was an episode of such notoriety in the conflict in the former Yugoslavia that one must go back to World War II to find a parallel in European history. Not since then had a professional army conducted a campaign of unrelenting violence against the inhabitants of a European city so as to reduce them to a state of medieval deprivation in which they were in constant fear of death … there was nowhere safe for a Sarajevan, not at home, at school, in a hospital, from deliberate attack.2
Indeed, read alongside the casual disregard for civilians in Antony Beevor’s ‘Berlin: The Downfall 1945’ (2002), the siege of Sarajevo strikingly parallels depiction of those persons experiencing the violence of conflict, that is, all civilians without distinction of age, gender or ethnicity. Contrary to the traditional image of ‘armies on the battlefield’, this state of violence does not figure combatants as the sole participants of conflict. Neither can it be understood in terms of harmful interactions between individual combatants and civilians, or as instances of collateral damage ‘producing’ civilian victims. Rather, the siege of Sarajevo requires recognition of the perpetration of widespread and systematic violence against a civilian population. It necessitates an understanding of the collective victimization of this civilian populace, that is, against all civilians of the city of Sarajevo.
Commentators point out that Sarajevo is home to persons of multiple ethnicities (Glenny, 1996; Donia, 2006). As the Prosecutor in D. Milošević emphasized, it was ‘a flourishing multi-ethnic community’ prior to 19923 and continues to retain this ‘mixity’ in its populace in the present day (Bose, 2002; Vulliamy, 2012).4 However, during the trial proceedings of Galić and D. Milošević it soon became apparent that the notion of the civilian population of Sarajevo comprising a cohesive collectivity of persons was to become a matter of dispute and contention. On the one hand, there was a strong narrative put forward by many of the victim-witnesses that the civilian population was a multi-ethnic populace that suffered the harms of the siege as a unified entity of persons. The ‘mixity’ of the population was often strongly emphasized by these witnesses and fiercely defended if called into question. On the other hand, there was also a counter-narrative put forward by a small number of the victim-witnesses that the civilian population was not a cohesive entity, but instead comprised multiple civilian collectivities that were differentiated by bonds of ethnic identity. During the trial proceedings these opposing views of the form and actuality of civilian–civilian relations, and civilian–military relations, provided very different accounts of the social fabric of the city during the siege. They called into question how we should understand civilians as a collectivity of persons harmed by the violence of armed conflict, and how that collectivity can figure as a victim of unlawful conduct.
In its contemporary focus on the protection of civilians in situations of armed conflict, the UN identifies the ‘presence’ of these persons in collective terms. It is a ‘civilian population’ or civilian ‘persons’ in the plural that figures as the entity in need of safety and security during the violence of hostilities and redress in its aftermath. As the UN describes, it is the ‘well-being of civilian populations’ rather than simply that of isolated individuals that suffers from the harms of war.5 While attention is often drawn to certain persons as ‘more vulnerable’ in this social context, most often women, children, the elderly and refugees, the overall category of a civilian population is understood to comprise civilians. As such, the ‘civilian’ identity of the individuals of this collectivity supersedes any other social characteristics that these persons may hold.
However, if we consider the terms through which civilian populations are understood to constitute the subjects of acts of collective victimization, it can be seen that other notions of identity become influential. Analyses of collective victimization tend to shift from a notion that it is a civilian population as a ‘neutral’ group of civilian persons that is the target of attack to an understanding that these populations hold a cohesive ‘social’ identity. For example, Kiza et al. describe that collective victimization ‘emerges when victimization is directed against a group or specific population’ (2006: 37). However, this conception of collective victimization does not consider that a civilian population merely defined as such constitutes a group. Rather, as they describe, this ‘specific group is often identified through religion, ethnicity, and other means of exclusive categories of “otherness”’ (Kiza et al., 2006: 37). Similarly, Jauković notes that social communities and groups are ‘targets of victimization in conditions of war conflicts … it is the family, but also the communities in which people live, such as the cities and villages. Ethnic groups and the state become objects to victimization’ (2002: 115).
These conceptions of collective victimization understand the ‘social’ identity of a group to be central to the motivations of the perpetrator in enacting violence against its members.6 As has been seen in many contemporary, and historical, conflicts, this form of collective civilian victimization does not figure as the perpetration of violence directed against a civilian population as a group of persons simply because they are civilians. Rather, it is directed at a group of civilians because of their social characteristics. This conception of a civilian population tends to assume that a civilian population holds a cohesive social identity. That shared social identity is ‘other’ to another group of persons. It forms the basis of the boundaries of the civilian population as a group of persons to another, and serves as the reason for their victimization as a collectivity. In this model of collective victimization, civilian populations are targeted because of their particular social characteristics and not just their civilian identity.
This form of collective civilian victimization is most commonly defined as ‘ethnic cleansing’. The Commission of Experts defines ethnic cleansing as a ‘purposeful policy designed by one ethnic or religious group to remove by terror-inspiring means the civilian population of another ethnic or religious group from certain geographic areas’.7 This type of warfare is predicated upon, or constructive of, ethnic divisions (Kaldor, 2001). The methods and means of the removal of a civilian population include acts such as the murder, torture and sexual violence of persons of a given ethnic or religious group as well as the destruction of property and physical infrastructure. While the term ‘ethnic cleansing’ was first employed to describe the forms of violence being perpetrated in Bosnia, such policies and practices have also been seen in contemporary conflicts such as Rwanda and Sierra Leone. Commentators, however, have also pointed out that techniques of ‘ethnic expulsion’ such as mass shootings, rape, pillage and plunder can be traced back at least as far as the fifteenth century (Wimmer and Schetter, 2003: 250). Rather than being a new form of violence then, the perpetration of actions to remove a civilian population of a particular ethnic or religious identity from a specific region has been a long-standing trait of large-scale acts of violence perpetration.
In the ‘ethnic cleansing’ model of collective civilian victimization, it is a social group that constitutes the ‘victim’ of acts of violence. While individuals are subjected to direct or indirect acts of violence, it is a collectivity of persons as an ethnic or religious group that is the overall target. As Aptel points out, when identity-based crimes are committed, ‘individuals are then merely defined as members of a group, reducing their multifaceted individuality to a monolithic, collective identity’ (2014: 154). Individuals come to ‘stand in’ for the group as a whole. Their status is reduced to their ethnic or religious identity, and that social status is considered to represent the very identity that requires destruction.
This reduction of individuals and groups to a particular social identity has another important effect. As it is the social group that constitutes the ‘enemy’ to be destroyed, the ‘enemy’ persons that constitute this collectivity may be civilians or military personnel (Shaw, 2007). This form of collective victimization therefore appears to obliterate the civilian–military distinction (Kaldor, 2001; Shaw, 2007). As Shaw sets out, ‘people are targeted for their particular identities rather than because they are “civilians”’ (2007: 113). In this conception of ‘social groups as enemies’, the status of its members as either civilian or combatant is deemed irrelevant by the perpetrator of violence (Shaw, 2007: 114). Accordingly, the protective rules of humanitarian law do not shape the forms of violence perpetration toward civilians and combatants. There is not a ‘sparing’ of civilians due to their status as such, since the enemy group is defined by ethnic or religious bonds and targeted as such. Perpetrators thus ‘deny the civilian idea by imposing on the target population a singular, absolute, death-justifying identity’ (Shaw, 2007: 114). They ‘regard unarmed civilians as part of an aggressive, combatant enemy’, and so ‘deny the difference between fighters and civilians’ (Shaw, 2007: 114).
For these reasons, this form of collective victimization can be understood to constitute a ‘relational harm’ to the social group subjected to violent acts by the perpetrators. It is a relational harm as all civilians (and military personnel) of the given collectivity, as defined by ethnic, political or religious bonds, are the potential targets of intentional or indiscriminate attacks because of their social identity. Perpetrators seek to shatter the economic, social, political and cultural power of the social group through targeted power and violence (Shaw, 2007). They aim to destroy the group in its existence as such. Downes points out that repeated attacks on civilians, as a matter of policy, are undertaken to terrorize the group as a whole through the creation of an environment of violence and fear (2006). That environment of fear is not simply intended to affect the civilians that constitute the populace at that time. Rather, as Slim points out, such attacks are ‘a deliberate intent to inflict widespread wounding so that the horror of the attack lives on and is embodied in the community long after the event’ (2007: 59). This futural intent of harm seeks to go beyond the current generation of individuals directly affected to those of successive generations.
This form of collective victimization may, in turn, shape the form of military–civilian relations in situations of conflict. Using the Bosnian conflict as an example, Marie-Joëlle Zahar develops an account of the form of identifications and interactions between militia groups and civilian populations (2001). Zahar argues that there is often an ‘identification’ made between militia groups and particular members of a civilian population (2001: 46). Such identification arises from ‘social constructs, such as ethnicity, religion, language’ (Zahar, 2001: 46). Members of the same group, both civilians and militias, understand themselves as an ‘in-group’ in accordance with these social constructs, in opposition to the ‘out-group’ of persons that do not hold these same characteristics (Zahar, 2001: 46). Following the terms of the ethnic-cleansing model of collective victimization, social categorizations and identifications are often influential of the perpetration of attacks by militias upon civilians understood to be ‘other’ to themselves. In contrast, ‘militias are often involved in the protection and promotion of the rights of their own civilian populations (in-group members)’ (Zahar, 2001: 47). Civilian–military interactions are thus shaped by social constructs and groupings. These social constructs determine the relations of either protection or victimization afforded by military personnel to civilian populations.
However, there are two key issues with this model of collective victimization and the form of social relations and allegiances between civilians and military personnel that it assumes. First, as Shaw points out, a significant number of the victims of acts of collective victimization are civilians (2007). Yet the ‘ethnic cleansing’ model of victimization appears to remove the ‘civilian’ identity of these victims from its conceptual framework. While it emphasizes that civilians come to be seen as the ‘enemy’ due to their particular social identities, it does not engage with this identity in any detail. Instead, the particular social identity of a group of civilians and military personnel is understood to dominate the views of the perpetrators and so serve as the foundational ‘reason’ for their violence. More broadly, analyses of this form of collective victimization have yet to adequately engage with how civilian populations themselves, rather than perpetrators or bystanders, understand their identities and relations to work to constitute a collectivity of persons subject to attack. They provide an account of civilian–military interactions, but not of civilian–civilian relations.
Second, this model of collective victimization does not provide an adequate framing for all situations of armed violence. In particular, it does not allow for an understanding of the collective victimization of a civilian population of heterogeneous composition. As the siege of Sarajevo exemplifies, there are instances of collective victimization that do not ‘fit’ the ethnic cleansing model of one social group seeking to expel or destroy another defined as ‘other’ to itself. As such, it is necessary to both recognize the existence of such populations and consider how the law addresses the forms of their victimization.
Cynthia Cockburn employs a conception of ‘mixity’ to describe the composition of social groups along national, ethnic, religious or gender lines that retain their distinctive character, but also interrelate and intermingle (1998: 6). As Cockburn suggests, the societal constitution and relations of the persons of Bosnia and Herzegovina often reflect the ‘extraordinary ethnic mixity of this population’ (1998: 29). The composition of Sarajevo can be understood through these terms of ‘mixity’. For these reasons, it is necessary to consider the form of civilian–civilian interactions, and civilian–military interactions, in broader terms than those put forward by the ‘ethnic cleansing’ model of collective victimization. It is important to consider how, or whether, the collective victimization of a civilian population can figure as both destructive, but also constructive, of communal relations and allegiances between civilians across ethnic divides.
To more adequately understand the nature of suffering in conflict, Ewald points out that it is necessary to take ‘into account the fact that victimization is shaped by the collective modes of experiences and individual interactions’ (2002: 95). The conduct of the siege of Sarajevo compels us to consider how collectivities of persons understood this experience of collective victimization, and how their individual interactions and relations served to constitute a populace that suffered from the harms of such violence. It requires both social and legal recognition of a state of victimization perpetrated against a civilian population of mixed composition, and of relational harms sustained by persons holding these different forms of social identity.
How, then, do the legal processes of adjudication and judgement represent collectivities of civilians and the terms of their victimization in situations of armed conflict? The cases of D. Milošević and Galić provide a useful case-study for analysis into how the law understands a collectivity of civilians to comprise a civilian population. As noted in the previous chapter, the accused were charged with, and found guilty of, the war crime of ‘terror against a civilian population’ for conducting a campaign of shelling and sniping of the civilian population of Sarajevo. In the indictments brought against these accused, the Prosecutor does not provide any ‘content’ as to the composition of this civilian population. There is no reference as to the ethnic identity of the persons that comprise this population, nor any determination that the victims of the crimes are, or should be, ethnically ‘other’ to the accused. Rather, the indictments broadly describe that the military campaigns were intended to ‘kill, maim, wound and terrorise the civilian inhabitants of Sarajevo’.8
However, analysis of the rules and practices of humanitarian law, and of the adjudication of the cases of D. Milošević and Galić, indicate that there are significant complexities in understanding the composition of a civilian population and the terms of its protection. This section of the chapter considers the legal definition of a civilian population, and then examines the testimonies of civilian victim–witnesses to consider the forms of relations, identifications and experiences that worked to comprise such a collectivity of persons in this context. It explores the terms of the adjudication and judgement of D. Milošević and Galić to consider the models of civilian collectivities the Trial Chambers draw upon to understand the victims and perpetrators of the siege of Sarajevo.
Article 50(2) of Additional Protocol I sets out the legal definition of a civilian population. It describes that a civilian population ‘comprises all persons who are civilians’. Mirroring the ‘neutral’ definition of a civilian, this definition of a civilian population does not attribute any further markers of identity to the individuals that constitute this collectivity of persons. This definition does not assume that a civilian population is made up of persons holding similar social characteristics or ties. Nor does it assume that this collectivity of persons holds similar allegiances to a particular armed force or has a unified stance on the policies or practices of a conflict. The principles of humanitarian law therefore determine that it is the civilian identity of a collectivity of persons, and no other aspect of their status, that constitutes such persons as a civilian population.
Additional Protocol I further determines that the ‘presence within the civilian population of individuals who do not come within the definition of civilians does not deprive the population of its civilian character’.9 As referred to by the Trial Chamber in D. Milošević,10 the Commentary of the International
Committee of the Red Cross on this point states that:
in wartime conditions it is inevitable that individuals belonging to the category of combatants become intermingled with the civilian population, for example, soldiers on leave visiting their families. However, provided that these are not regular units with fairly large numbers, this does not in any way change the civilian character of a population.
(Sandoz et al., 1987: 1922)
In the adjudication of crimes committed against a civilian population, however, the question of whether the presence of military personnel or targets within a particular geographic area challenges the civilian status of a population or area has been the subject of debate. In D. Milošević for example, the Defence argued that the entire area of Sarajevo was a military zone.11 It put forward an argument that the presence of military targets and military fighters meant that the area was not civilian in status. In its judgement, the Trial Chamber concurred that military targets existed inside the confrontation lines.12 However, it found that these military targets could not ‘be said to be of such great numbers that they would deprive the entire urban area within the confrontation of its civilian status’.13 The entire area within the confrontation lines of Sarajevo was therefore found to be civilian in status.14 Concurring with the terms of API and the ICRC Commentary, the Tribunal’s jurisprudence determines that ‘the presence within the civilian population of individuals who do not come within the definition of civilians does not deprive a population of its civilian character’.15 For the Tribunal, a civilian population ‘broadly interpreted, refers to a population that is predominantly civilian’.16
However, not all of the categories of crimes that comprise the framework of humanitarian law protect civilian populations in their status as such. It is not necessarily the civilian character of a group of persons that is the object of protection. Rather, certain categories of crimes seek to protect civilian populations, while others seek to protect ‘social’ or other groups of persons. Crimes against humanity represent the only category of crimes that protects civilian populations in their specificity as such from attacks directed against them. In D. Milošević, the Trial Chamber points out that in order to constitute a crime against humanity, the acts of an accused must be part of a widespread or systematic attack directed against a civilian population. It determines that this phrase encompasses five elements: (i) there must be an attack; (ii) the acts of the perpetrator must be part of the attack; (iii) the attack must be directed at any civilian population; (iv) the attack must be widespread or systematic; (v) the perpetrator knows that his acts constitute part of a pattern of widespread or systematic crimes directed against a civilian population and knows that his acts fit into such a pattern.17 This category of crime seeks to protect civilian populations from offenses including murder, extermination, enslavement, deportation, imprisonment, torture and rape (Zahar and Sluiter, 2008). Crimes against humanity appear in the Statutes of the ad hoc tribunals, as well as the Rome Statute of the ICC, and are now considered customary law.
Kittichaisaree points out that the rules of this offense understand that the ‘individual victim is victimized because of his membership of a civilian population targeted by the accused’ (2001: 95). This category of crimes figures a civilian population as being the object of attack, and so ‘emphasizes the collective nature of the crime, thus ruling out attacks against individuals and isolated acts of violence’ (Werle, 2005: 221). It is the membership of a civilian population, as a civilian, that figures as the aspect of the individual’s identity that affords them protection. For this reason, as Cryer et al. note, the ‘law of crimes against humanity not only protects enemy nationals, it also covers, for example, crimes by a State against its own subjects. The nationality or affiliation of the victim is irrelevant’ (2007: 192). While the discriminatory element specific to the crime of persecution (that the crime is undertaken on political, racial or religious grounds) is an exception (Boas et al., 2008; Cassese, 2005),18 there is no determination that the other acts that fall under crimes against humanity must be directed against a civilian population that also comprises a social group. Rather, this category of crimes protects any civilian population in its status as such. As Cryer et al. set out, the ‘international interest arises from the attack on a civilian population’, and it is this population that is subject to the protection of the rules of humanitarian law (2007: 190).
Unlike crimes against humanity, the rules that set out those offenses constituting war crimes protect a number of different categories of person. They do not protect civilian populations alone, but also establish prohibitions on the perpetration of particular forms of harm against combatants and those hors de combat. Zahar and Sluiter point out that international humanitarian law has often been divided into ‘Hague law’ that regulates specific means and methods of warfare, and ‘Geneva law’ that primarily focuses on the protection of civilians and other persons who are not active combatants (2008: 222). The Hague Regulations of 1907, for example, regulate specific means and methods of warfare to reduce the unnecessary suffering of combatants. As noted in Chapter 3, these regulations determine that the ‘right of belligerents to adopt means of injuring the enemy is not unlimited’.19 They prohibit actions including the use of poison or poisoned weapons,20 the ‘treacherous’ killing or wounding of individuals belonging to the hostile nation or army,21 and the use of arms, projectiles, or material calculated to cause unnecessary suffering.22
The Geneva Conventions of 1949 protect a number of ‘categories’ of persons present in situations of armed conflict. As noted in Chapter 3, Geneva Convention IV is the only convention that focuses specifically upon the protection of civilians. Geneva Convention I deals with sick and wounded in the fields, Geneva Convention II with the wounded, sick and shipwrecked at sea and Geneva Convention III with prisoners of war (Roberts and Guelff, 2005). The later Additional Protocols of 1977 combine elements of ‘Hague law’ and ‘Geneva law’ and so set out rules that protect civilians and other categories of persons (Zahar and Sluiter, 2008: 222). For example, the crime of ‘terror against a civilian population’ that is brought as a charge in both D. Milošević and Galić takes a civilian population as the group of persons it seeks to protect, while other rules prohibit the causing of superfluous injury or unnecessary suffering of combatants and prisoners of war.23
The crime of genocide does not protect a group in relation to its civilian (or military) character. This category of crime does not seek to ensure the safety and security of a collectivity of civilians, that is, a civilian population in its status as such from being the object of attack. Nor does it protect an individual on account of his or her individual qualities or characteristics. Instead, the crime of genocide relates to the targeting of persons ‘because he or she is a member of a group’ (Cassese, 2003: 103). It prohibits particular acts (such as killing or causing serious bodily or mental harm) committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group.24 As such, this category of crime seeks to protect four ‘groups’ that are understood as such because of the shared characteristics of their members (Werle, 2005: 193). While commentators point out that genocide was first envisaged as a sub-category of crimes against humanity (Cassese, 2003: 96), it is important to note the nuances of the different notions of victims inherent in these categories of crimes.25 The crime of genocide, unlike crimes against humanity, does not protect a civilian population purely in its status as such. Rather, it protects particular social groups from being the object of attack. That group can comprise both civilians and military personnel. It is the membership of a national, ethnical, racial or religious group that determines the parameters of the protection afforded by this category of crime, and not the civilian or military personnel status of the individual or group.