Exploring inclusive victimhood narratives: the case of Bosnia-Herzegovina

Abstract Narratives are essential tools for communicating thoughts about competitive and inclusive victimhood socially and politically. In reconciliation processes, promoting narratives of inclusive victimhood (an understanding that ‘we all suffered together’) has been suggested as one way to overcome competitive victimhood (the idea that one ethnoreligious group or nation is the sole or primary victim in a conflict or war). However, the notion of inclusive victimhood remains understudied in post-war contexts in which exposure to violence was relatively imbalanced between former adversaries. This article traces the potential narrative variation from competitive to inclusive victimhood in post-war Bosnia-Herzegovina. It draws on (1) the competitive victimhood typology as an analytical tool and (2) a mapping of narrative sites as a methodological tool for tracing collective victimhood. The article scrutinises less competitive and inclusive accounts of victimhood identities in Bosnia-Herzegovina by examining the narratives that recognise outgroup victimhood and acknowledge ingroup responsibility for harmdoing. It suggests that there is potential for peaceful coexistence realised through the narrative of shared suffering, especially in post-war contexts where the exposure to violence was not entirely unidirectional. However, shared responsibility is less likely to be observed when the exposure to violence was highly asymmetrical.


Introduction
Constructing a shared/single narrative across conflicting sides is considered crucial in addressing the past as a component of reconciliation processes in the aftermath of conflicts (Schaap 2005;Auerbach 2009;Strupinskiene 2012). Narratives concerning past suffering are utilised to strengthen collective identities and shape subjective perceptions about who is right or wrong, who is the victim (ingroup), and who is blameworthy (outgroup) in post-war contexts, potentially impeding reconciliation processes.
Agreeing on a single narrative is complicated when victimhood becomes a part of ethnonational identity on both sides of a conflict (Volkan 1997;Ignatieff 1998;Kaufman 2001;Kelman 2004). 1 When both parties consider themselves the sole or primary victim, they might undervalue outgroup victimhood and justify violence by the ingroup that causes harm to others (ie competitive victimhood [CV]; see Noor, Brown, and Prentice 2008), thus impeding reconciliation processes. Narratives of suffering are crucial tools in constructing ethnonational identities and forming thoughts and beliefs about ingroup victimhood and outgroup blameworthiness, thereby reinforcing CV (Shnabel, Halabi, and Noor 2013). Several studies have shown that recognising others' suffering and acknowledging ingroup responsibility counter competitiveness without imposing a single narrative or an identity change (Adelman et al. 2016;McNeil, Pehrson, and Stevenson 2017). Correspondingly, I previously developed a CV typology to observe narrative variation from highly competitive to less competitive and inclusive victimhood by identifying critical characteristics of competitiveness (Demirel 2023). earlier literature has indicated that inclusive victimhood narratives between ethnonational groups are more common in cases where experiences of violence are relatively symmetrical than in contexts where one side of the war is exposed to greater violence (Penic et al. 2018). In the former, there are often victims and perpetrators on both sides, enabling mutual recognition and apologies. 2 However, the notion of inclusive victimhood, which entails recognising others' victimhood and the ingroup's responsibility in harming them, is understudied in contexts of asymmetrical violence. exploring inclusive victimhood in such contexts seems complicated. Nonetheless, it is necessary to more deeply understand the nature of group relations when competition over collective victimhood is observed between the two sides of the conflict. Accordingly, the context of Bosnia-Herzegovina, where exposure to violence was imbalanced, has substantial implications for understanding the notions of competitive and inclusive victimhood in post-war inter-ethnic relations.
The International Criminal Tribunal for the Former yugoslavia (ICTy)'s estimates of casualty numbers based on ethnicity during the war in Bosnia-Herzegovina in the 1990s indicate the Bosniak side as the primary victim, which suffered greater losses and genocide, and Bosnian Serb forces (ie the army of Republika Srpska [VRS] with military support from Milošević's yugoslav army [JNA]) as the primary perpetrators. 3 However, Serbs in Bosnia-Herzegovina predominantly deny the crimes, and each party claims to be the main victim of the war (Božić 2019;Hajdarpašić 2010;Moll 2013;Sokol 2014). 4 Additionally, a growing body of literature problematises the construction of victim hierarchies in post-war Bosnia-Herzegovina through the imagination of an 'idealised/innocent' victim (Ticktin 2020;Rudling 2019). Defining the boundaries of the most victimised ethnic group as Bosniaks seems to result in disregard for Bosnian Serb victims and war-time harmdoing committed by the Bosniaks and hamper inter-ethnic collaboration (Korac 1998;Helms 2013). 5 Moreover, conflicting sides' contested narratives overshadow the inclusive expressions of victimhood. Therefore, exploring inclusive victimhood, a rare phenomenon in post-war Bosnia-Herzegovina, can be theoretically and empirically revelatory in understanding the dynamics of inter-ethnic recognition and fill a gap in the literature.
I argue that a potential space between conflict and a thick form of reconciliation (ie the acceptance of a single narrative of the past) can be observed in Bosnia-Herzegovina by tracing and mapping the narrative sites to show illustrations of a variation from competitive to inclusive victimhood. A shared intergroup victimhood stimulates peaceful coexistence when experiences of violence were not entirely unidirectional. yet, agreeing to shared responsibility is more challenging for Bosniaks due to the significant disparity in exposure to violence. I use two frameworks: (1) the CV typology as an analytical tool for exploring potential narrative variation from competitive to inclusive victimhood and (2) the mapping of narrative sites as a methodological tool for illustrating victimhood narratives. The analysis examines selective narrative examples, such as public apologies, political statements and memory places, alongside 21 semi-structured interviews with local actors (non-governmental organisation [NGo] workers). The interpretive analysis of the materials aims to bring nuance to the phenomena of CV and inclusive victimhood in a context of asymmetrical exposure to violence.
The findings contribute to the peacebuilding and reconciliation literature by showing that tentative narrative examples of less competitive and more inclusive victimhood do exist in post-war Bosnia-Herzegovina despite the imbalance in violence experiences between ethnonational groups. Further, the mapping provides a novel tool for analysing victimhood narratives by combining different narrative sites, diverging from earlier studies that emphasised one narrative form or site and enabling a richer analysis (Forde 2016). The article also confirms the usability of the CV typology in a new context, and the findings from Bosnia-Herzegovina feed new characteristics into its categories.
The article first presents the theoretical framework for the study of CV and inclusive victimhood by outlining the role of narratives in reconciliation processes and the CV typology. It then introduces the research context before describing the methods, materials and mapping of the narrative sites analysed. The subsequent sections showcase the mainstream CV narratives based on primary and secondary resources and discuss less competitive and more inclusive victimhood in various narrative sites in Bosnia-Herzegovina. Finally, the article examines the broader implications of the findings.

From competitive to inclusive victimhood
Collective victimhood is narratively constituted. Narratives are subjective meaning-making tools for shaping people's thoughts and beliefs, which provide a framework for explaining and justifying social and political events, thereby constructing the boundaries between 'we' and 'others' (Cobb 2013;Winslade and Monk 2008;Pilecki and Hammack 2014).
In identity conflicts, the notion of collective victimhood is often a core component of group identity, which is imagined by all members of a group even if they were not physically harmed. When each party suffers human losses, collective victimhood often results in intergroup CV. Changing conflicting narratives (Garagozov and Gadirova 2019;Auerbach 2009) and adapting a single narrative about the past by redefining national identities are considered reconciliatory in such contexts (Staub 2006;Schaap 2005;Kelman 2004). However, most post-war contexts prove that adapting a single narrative and a transcendent identity is not seen as desirable by all parties in the conflicts.
Arguably, I think, an all-encompassing narrative that could melt former adversaries' subjective understandings of the past can only succeed if it is oppressive (see eastmond and Selimovic 2012). Alternatively, former adversaries may communicate inter-ethnic inclusive victimhood by accentuating (shared) suffering and recognising their responsibility in harming each other as sub-categories of their ingroup identities. Inclusive victimhood suggests the development of a shared sub-notion of collective identity rather than a complete replacement of broader ethnonational identities. Inclusive victimhood provides a space for accepting the validity of some parts of the other's experiences (Auerbach 2009;Rosoux 2004;Kelman 2004) and offers a more realistic path for overcoming CV and achieving peaceful coexistence than the adoption of a single narrative (Shnabel, Halabi, and Noor 2013;Adelman et al. 2016). Although challenging, promoting ethnically inclusive victimhood narratives can facilitate peaceful attitudes (Adelman et al. 2016;Vollhardt and Bilali 2015).

Typology of CV as an analytical tool
In an earlier study, I developed a typology of CV that presents potential narrative variations in conflict-to-peace transitions. The categories (revengeful victimhood, strong CV, mid CV, weak CV, and inclusive victimhood; see Table 1) help us understand the intertwined components of CV: recognising/denying outgroup suffering and ingroup responsibility for the harms against the outgroup (Demirel 2023).
As Nguyen (2016) argues, individuals/groups/nations may acknowledge other individuals/ groups/nations as victims in post-war contexts to encourage reconciliation. This normative expectation of acknowledging others' suffering inevitably links to ingroup perpetratorhood if the ingroup is the source of others' pain. When victimhood is associated with pure innocence and non-harm against others, a victim who perpetrates an act of violence is often not considered a 'deserving' , 'innocent' victim. Thus, recognising ingroup responsibility for harmdoing can potentially undermine the image of ingroup suffering, thereby leading to the failure to acknowledge others' victimhood.
The typology presents a globally translatable spectrum of recognising outgroup victimhood and ingroup guilt/responsibility for harmdoing, thus enabling the exploration of context-dependent characteristics of collective victimhood narratives. This analytical tool can be used to assess inter-group recognition when suffering of human beings occurs on all sides of a conflict regardless of the nature, scale and causes of the conflict. Contexts in which CV typology might be implemented range from post-imperial Turkish-Armenian relations to post-colonial Rwanda or ideologically divided Korea, with changing levels/forms of external influence.
The typology was first instrumentalised in a context where power relations were relatively balanced, and each side of the conflict offered apologies in the peace process. In such cases, victims and perpetrators are often on both sides of the conflict, and the boundaries between victim and perpetrator identities on the group level become ambiguous. This article utilises the CV typology to understand the potential narrative variations in Bosnia-Herzegovina, where violence was relatively asymmetrical. For this context, I zoom in first on the mid CV category to briefly show the level of competitiveness in predominant victimhood narratives in Bosnia-Herzegovina, then on the weak CV and inclusive victimhood categories to show how a narrative variation is conceivable. The focus on these categories provides a perspective conducive to reconciliation, unlike the revengeful and strong CV victimhood, which are more commonly associated with armed and frozen conflicts. 6 Mid CV in the typology is the most common category in recent post-conflict contexts where former adversaries mainly emphasise their own victimhood and only partially recognise that of others. even though sides do not glorify or deny outgroup suffering as in the initial categories, others' suffering is belittled, and ingroup guilt/responsibility for harming outgroup is justified.
In weak CV, the victimhood of others is recognised, but not completely, because ingroup guilt/responsibility is silenced or excused. There is a difference between justifying and excusing violence. Justifying violence (mid CV) means that 'harming others was the right thing to do under the circumstances' , whereas excusing violence (weak CV) acknowledges that the harm was wrong but frames it as a consequence of others' actions. The perception of the wrongness of ingroup guilt/responsibility for harmdoing can also cause disassociation from the ingroup members who committed the crimes. Accordingly, even though others' suffering is recognised in weak CV, the ingroup's potential to harm and kill can be denied/omitted. Inclusive victimhood ideally refers to an inclusive understanding and recognition of others' victimisation, in contrast to CV (McNeill and Vollhardt 2020;Shnabel, Halabi, and Noor 2013). extending this definition, I include the full recognition of ingroup responsibility/guilt in harming others because in the contexts of mutual harmdoing, acknowledging outgroup victimhood is associated with understanding ingroup responsibility for it. Acknowledging ingroup potential to harm others refers to a complete recognition of outgroup suffering because the ingroup is not taken for granted as rightful. However, in most cases, recognising ingroup guilt/responsibility is challenging because it impairs the ingroup's image of innocence and blamelessness (Rudling 2019;Helms 2013). 7 Inclusive victimhood differs from selective inclusivity in solidarity with other victim groups worldwide (Vollhardt 2020) because it refers to relational situations. Furthermore, inter-group recognition of harmdoing does not necessarily claim that the scale of harm was the same on all sides of the conflict, but it would be susceptive of all victimisation and perpetration experiences regardless of ethnicity. 8 It should also not be confused with universal victimhood, which makes almost no distinction between victims and perpetrators (linklater 2009;levy and Sznaider 2005). universal victimhood can relativise all war involvements in one category of suffering irrespective of specific individual experiences. Finally, even though external actors' universalist approaches can potentially have a considerable effect on victimhood narratives, the inclusive victimhood category in the typology does not equate experiences of suffering or committing harm.

Background to the context of Bosnia-Herzegovina
The socioeconomic and political drivers for the Bosnian War in the 1990s were far too complex to be reduced to 'ancient hatreds' and ethnic conflict. Nevertheless, the construction of ethnic identities during the war did contribute to the ethnocentric interpretation of history since then (Campbell 1998). Indeed, the Dayton Peace Agreement, imposed by the international interveners, ended the war in 1995, and constructed an ethnocentric political structure that divided the newly formed Bosnian state into three political entities: Republika Srpska (RS [majority Bosnian Serbs]), the 10 cantons of the Federation of Bosnia-Herzegovina (FBiH [majority Bosniaks and Croats]), and Brčko District, an autonomous entity. RS attained all the regions where Bosnian Serbs constituted an ethnic majority, including those places where their dominance results from ethnic cleansing during the war.
In Bosnia-Herzegovina today, the political structure is deeply fragmented, historical questions are highly politicised, and the dominant narratives are still divided along ethnonational lines 28 years after the Dayton Peace Agreement. Accordingly, a 'memory war' among ethnonationalist groups has been ongoing both on a political level through official narratives (Subotic 2013) and on a social level, through commemorative discourses/practices, monuments, and ethnically biased curricula in history education in schools (Božić 2019;Hajdarpašić 2010;Sokol 2014).
Despite attempts to establish the truth about what happened during the war, why it happened and who was responsible for it, no single official truth regarding the past exists in Bosnia-Herzegovina today. The legacies of war, which position suffering and responsibility within ethnic boundaries, force one to think of inter-group relations and reconciliation in ethnic terms, just like differing conceptualisations of reconciliation determine what sides expect from it. Bosniaks' understanding of reconciliation, a moral remembrance model of addressing the past, generally overlaps with that of the international community. However, Bosnian Serbs predominantly anticipate reconciliation as abandoning the narrative that presents them as exclusive aggressors (Karić and Mihić 2020).
Moreover, the international community remains highly engaged in Bosnia-Herzegovina's politics and memory space, as evidenced by the latest decision of the former High Representative banning genocide denial in Bosnia-Herzegovina, which also triggered secessionist narratives by the RS (Muslimović 2021).

Mapping inclusive victimhood -material and method
Potential variation from competitive to inclusive victimhood is best observed where inclusivity is found. When a phenomenon is rare in a context, looking at different settings where it can emerge enriches our understanding of its characteristics. Therefore, mapping relatively inclusive victimhood narratives in settings where mainstream narratives are contested is challenging but revelatory. It requires analysis of extensive narrative units and multiple settings where inclusive narratives develop. The mapping of narrative examples from Bosnia-Herzegovina (Table 2) outlines the multiple sites analysed (ie official statements, civil society, and memory sites and practices).
Political speeches and statements have a narrative function due to their selective/performative references to past events (Schubert 2010). They materialise ethnonational victimhood, fear and hatred of others. At the same time, they can turn conflicts into peaceful relations by referencing apologetic and reconciliatory gestures. To trace political narratives by political leaders, I mainly used the Sarajevo Times, Srpska Republika News Agency (SRNA) and Balkan Insight. Sarajevo Times and Balkan Insight are the leading organisations for reports and articles about Bosnia-Herzegovina. Furthermore, SRNA is the leading media organ in RS and broadcasts entity-based news in english. Finally, for public apologies, I used the Political Apology Database, which maps political apologies by states or state representatives worldwide for past violence and human rights violations. The database provided two statements by the former leaders of the union of Montenegro and Serbia and the Republic of Serbia, and a declaration by the RS parliament directed at Bosnia-Herzegovina. 9 I also used media sources to identify two more public apologies by RS leaders.
Additionally, civil society is an aggregation of sites useful for scrutinising potential narrative variations, and peacebuilding NGos and research centres are the primary promoters of alternative narratives (Williams 2019). Before a field trip to Sarajevo, I identified critical NGos in the debate on dealing with the past and peacebuilding from media outlets and public debates and contacted them via organisational email addresses. I also used a list of NGos provided by Peace Direct. 10 I reached out to all the organisations working on inclusive cross-community projects and conducted 21 semi-structured qualitative interviews with their staff members. All interviews were confidential and took place at the time and place that the organisations suggested. I also drew on early literature and NGo reports to reflect on narrative examples from the inter-ethnic dialogue groups of women's and veterans' organisations.
The qualitative strategy of selecting critical informants to collect examples of uncommon narratives should not be misjudged as biased (Suri 2011). The staff of peacebuilding NGos are very knowledgeable about such narratives since they are promoters of inclusive ideas among the communities in Bosnia-Herzegovina. Further, they are local actors who work with local people to find new ways to facilitate communication and change the nature of relations across the conflict lines (Chigas 2007). 11 Thus, interviews with them are essential for discovering the existence and nature of inclusive victimhood narratives. 12 Finally, memory sites and commemorative practices are well-known empirical sites for observing references to the past and analysing collective victimhood narratives. Thus, I made in-person visits to several memory sites in Bosnia-Herzegovina. The article provides inclusive examples I have encountered, such as Slana Banja Memorial Centre and Sarajevo Roses. I also utilised the narratives about memory sites from the interviews, as in the instance of the Vrbanja Bridge and the memorial dedicated to child victims in Konjic. I used the media mentioned above for information about commemorative practices in Žepče and Bosanski Petrovac.
The interviewees were asked to discuss specific research topics, defining features of CV (recognising others' suffering and ingroup responsibility for harmdoing), as well as the broader research themes (dealing with the past and peace and reconciliation). 13 each interviewee was also asked follow-up questions depending on their expertise. I conducted and transcribed all interviews in english. Categories in the CV typology (mid CV, weak CV, and inclusive victimhood) provided a predefined framework for attaching interpretive meaning in the analysis of the empirical material (yanow and Schartz-Shea 2006; Kvale 2007). The typology is well suited to explore inclusive victimhood because the CV categories outline key characteristics of inter-ethnic construals of victimhood, providing features that help compare competitiveness and inclusiveness. 14

Analysing the ethnonational victimhood narratives
The findings are described in two sections. The first section addresses the dominant CV narratives that demonstrate the division in Bosnia-Herzegovina and analyses public apologies and statements that correspond to the mid CV category in the typology. The second section analyses narratives from civil society and memory sites to show variation towards weak CV and inclusive victimhood.

Interviews
Many interviewees mentioned comparisons between the numbers of victims with different ethnic backgrounds. An interviewee highlighted the disagreements over the numbers: each side claims that they are the main victims of the conflict. one of my colleagues from Belgrade openly called it 'victims olympics' . So, 'we are the main victims, and our population suffers the most, and we have the highest number of the victims' . So, we can discuss the motives and the reasons, but they claim that they are innocent victims, and someone else started the conflict. (Interviewee 15) The numbers are manipulated when facts are not established because greater losses are often associated with innocence and blamelessness. Most interviewees emphasised the ongoing 'numbers game' among politicians and among the broader ethnonationalist groups: There is always the manipulation of numbers. For example, someone will say, 'oh, in Srebrenica, there was not a thousand. There were like 300' . Then, a similar narrative would be for Jasenovac, 15 that four million people died in Jasenovac. No! There were 100,000 people. They think that the numbers should be something people should race like 'who has the most victims' . They keep forgetting and ignoring that the people lost their lives. one or 100, and still, every mother cries the same. (Interviewee 13) one of the respondents underlined that some Bosniak politicians do not differentiate between official victimhood narratives although they openly talk about Bosnian Serb and Bosnian Croat victims, leading the struggle for the victim position to silence others' suffering:  -16) one interviewee reflected on a narrative conditioning the Bosniaks' recognition of the Bosnian Serb victims on the Serbs' prior acknowledgement of their own harmdoing: People are so connected to it [victimhood] because of its trauma. Because they were in a war, they were fighting for it. They will not be critical of their people, but they will be very critical of others. If I ask, 'oK, why don't you support the memorial for Kazani and so on?' , they will say 'oh, we will do it when they do it' . The mentality is like, 'they would never do it, so we will not do it' . (Interviewee 5) The overall narratives about multiple truths and the 'numbers game' can be associated with mid CV. The Bosnian Serbs' attempts to introduce historical Serbian victimhood into the debate also resonate with the same category. only ingroup victimhood is emphasised, and ingroup crimes against others are justified via earlier ingroup victimisation.

Public apologies and political statements
Political leaders' public apologies and statements are crucial official steps towards recognising others' suffering and accepting ingroup responsibility for harmdoing. The official apologies issued by state leaders reveal that the politicians could not face the past without emphasising their ingroup victimisation. The political leaders who have issued apologies have either not labelled the crimes in Srebrenica as genocide or highlighted that they expect similar apologetic gestures from the political leaders of other ethnic groups. Many local Bosniak politicians, scholars and experts on public apologies found these apologies strategic, insincere and pragmatic, except for a few members of the international community and Bosniak politicians who welcomed apologies at the formal level and regarded them as contributing to the development of mutual understanding and neighbourly relations (Horelt 2016 Despite the RS's refusal to call the crimes committed in Srebrenica genocide at the official level, Bosnian Serb leaders do occasionally acknowledge that there was genocide. However, they undervalue Bosniak suffering by claiming that everybody, regardless of ethnic background, was a victim of genocide. The narrative that losses were experienced on all sides on an equal level reveals why the Bosniak side has not accepted these attempted apologies. Moreover, apologetic RS President Čavić subsequently lost political support and was defeated by Milorad Dodik in the 2006 elections. Dodik, formerly known as a moderate and cooperative politician for having recognised the Srebrenica genocide in private meetings with members of the international community before his election, adopted ethnonational populist rhetoric and began to express denialism (Barton-Hronešová 2021).
In another example, former Serbian member of Bosnia-Herzegovina President Council Mladen Ivanić referred to the Srebrenica genocide as a 'terrible crime' that should be defined as 'preventive genocide' . yet he argued that this terrible crime was necessary to prevent a new Jasenovac for the Serbs (Slobodna Bosna, 2017), justifying the genocide based on the historical suffering of the Serbs at the hands of the Croats during World War II. As a result, the rhetorical features of Serbian apologies to date fit in the mid CV category because they do not go beyond equating the losses and crimes of the different sides, thus belittling Bosniak suffering.

Political statements
Apart from apologetic political statements corresponding to mid CV, the political actors who promote a shared narrative of suffering are multi-ethnic political parties in Bosnia-Herzegovina like our Party (Naša Stranka) and the Social Democratic Party of Bosnia-Herzegovina (Socijaldemokratska partija Bosne i Hercegovine [SDP BiH]). Nevertheless, they are often perceived by the other groups as supporting the Bosniak war narrative (Interviewee 21). A reflection of the controversy is evident in the Kazani pit killings debate. The mayor of Sarajevo, the SDP's Benjamina Karić, visited the massacre site to honour predominantly Bosnian Serb victims of the Kazani killings by stating that all the victims of the siege of Sarajevo deserve respect and a dignified memorial (Balkan Insight, November 15, 2021). However, the City Council's refusal to inscribe the memorial with the names of convicted members of the Army of the Republic of Bosnia-Herzegovina (ARBiH) provokes criticism on the side of Bosnian Serbs, thereby manifesting weak CV.

Civil society
The Book of the Dead is a local attempt to include the names of all victims of the war. Names of civilians from all sides regardless of ethnonational background (Bosniak, Bosnian Serb, Bosnian Croat) and members of the three armies (ARBiH, JNA-VRS, Croatian Army-Croatian Council of Defence [HV-HVo]) paint a broader picture of the numbers. The book is described by its author, Mirsad Tokača, the head of the Research and Documentation Centre (RDC), as a 'unique monument to all human losses caused by the Bosnia War' and was criticised by some politicians because the numbers it presents did not accommodate their claims. 17 Although emphasising shared suffering, Tokača argued against balancing responsibility with a 'we are all guilty' narrative and negation of facts: At the most general level, it is not difficult to agree that war as a social phenomenon brings disaster and suffering to all people, but it is unacceptable to say that we are all identical victims, ie that causes and consequences of the suffering and pain in Bosnia are the same for all of us. (Tokača 2012, 25) The Book of the Dead was mentioned in almost every interview with the NGo representatives. An interviewee described his encounter with the book as mind-opening: From 1992 to 1995, you have their [people who died during the war] last names, names, date of birth, place, nationality, and status. you know, you have it all. It was documented systematically with verified resources. There were cases, you know, people coming up and saying, 'I am sure you do not have this person's name' . And you say, 'What is the name? It is here' . It brings the number of victims in it. It demystifies what has happened and who died, so you can see this is the available data … It is a way of acknowledging that they were victims of war. Were they the military or civilian victims? That is the question. And that is limiting the numbers game. (Interviewee number 6) As articulated, the Book of the Dead highlights joint wartime suffering beyond ethnic boundaries while pointing to the guilt and responsibility of the JNA-VRS forces as the offenders. This distinction between shared suffering and non-shared responsibility in harmdoing reveals the difference between weak CV and inclusive victimhood.
In addition to similar reports produced by research centres that emphasise the multi-ethnicity of the dead, multiple NGos in the Bosnian context have cooperated across ethnic lines and facilitated the emergence of a common suffering narrative (weak CV) without touching the question of responsibility. Helms's (2013, 150) extensive work on women's organisations from RS and FBiH reveals that common interests between female activists on 'women issues' has helped women create a general narrative that 'we all suffered during the war' due to their non-involvement in violence.
The Centre for Nonviolent Action (CNA) stands out as an NGo that brings war veterans of different groups together to engage in dialogue, deconstruct enemy images, and address ingroup responsibility, if only implicitly. The Bosniak, Bosnian Serb, and Bosnian Croat veterans' associations attend the other sides' commemorations to pay their respects to the communities and their victims and constitute a counter-narrative to mainstream competitive narratives. The commemorative action is loaded with shared victimhood and shared responsibility narratives due to the war veterans' realisation that they may simultaneously be heroes or veterans for their community and perpetrators in the eyes of other communities (Franović 2015). Being a war veteran is unlikely to have a harmonious meaning for the project participants. Nevertheless, veterans often share a common understanding of the general catastrophic consequences of the war (Ignatieff 1998), and the project's results confirm it. The common understanding is achieved not by internalising perpetratorhood on each side but rather by respecting the others' perspectives. War memories often constitute stories of veterans' heroism and sacrifice, promoting the idea of 'military victims' within Bosnia-Herzegovina (Barton-Hronešová 2020, 313). 18 Hence, their participation in reconciliation processes is significant.

Memory sites and practices
Many interviewees referred to a 'blame game' to describe dominant narratives and pointed out memory sites where it can be observed. one of the most discussed recent examples, the inscription of Kazani monument reads, 'We shall forever remember with sadness and respect our fellow citizens who were killed' (Balkan Insight, october 27, 2021). The acknowledgement of Bosnian Serb victims alongside two ukrainians, two Bosnian Croats, and one Bosniak and the inclusion of their names in the memorial marked a departure from the predominant ethnonational Bosniak narrative. 19 No reference is made to the names of the perpetrators or the responsibility of the Bosnian Army. Thus, the narrative embodied in the Kazani memorial suggests weak CV as although the other groups' victims are acknowledged, the responsibility for ingroup crimes is omitted.
Furthermore, the dominant Bosniak narrative regarding the Kazani pit killings excludes the members of ARBiH from the ingroup as 'a bunch of criminals' . This disassociation facilitates the recognition of others' victims while enabling ingroup 'purity' . For example, a police officer, Dragan Mioković, who witnessed the crime at Kazani, said, 'if we confirm the court verdict that a crime was committed in Kazani and carve it in stone, that it is something that will distance us even more from it' (N1, october 27, 2021). Therefore, not naming the perpetrator as ingroup also articulates weak CV. 20 other memory spaces promoting a shared victimhood also exist in Bosnia-Herzegovina, although they are few. These venues and monuments create a more unifying and inclusive historical narrative. For example, Sarajevo Roses (Sarajevske ruže) is a form of memorial that marks the flower-like shapes left in the city by the mortar shell explosions (Kappler 2017). These traces can be observed in various parts of the city, drawing attention to the devastating suffering caused by the war without referencing the identity of either the victims or the perpetrators (Sokol 2014).
one of the interviewees reflected on her disappointment with the representation of the ethnic groups in official memory policies and provided an example of shared suffering: one example would be Boško-Admira. Boško is a Bosnian Serb, Admira is a Muslim woman. They were trying to run away together, and they were shot dead on the Vrbanja Bridge. They are called the Romeo and Juliet of Sarajevo because when they were shot dead, they crawled to each other and hugged, that is how they passed away. However, there is not much recognition. That is a story everyone knows, but there is nothing like a memorial or picture or anything in the city. (Interviewee 9) The new name of the Vrbanja bridge, the Suada and olga Bridge, 21 represents a counternarrative to the dominant ethnonational politics in Bosnia-Herzegovina (Björkdahl and Selimovic 2016). However, the removal of the sign commemorating the well-known Romeo and Juliet story is controversial. The omission of a story depicting 'Serbian love for a Bosniak' likely restricts the victim image to Bosniaks and Croats. This official mnemonic selection of one narrative over another, intentional or not, indicates the potential erasure of specific memories from the public memory, if not now, then through new generations.
The Slana Banja memorial complex in Tuzla is also a significant example of manifesting shared suffering. The Kapijana massacre in Tuzla on 25 May 1995 left a profound mark on the city's memory with seventy-one civilian victims, primarily including children and adolescents. All adolescent and child victims -regardless of their ethnicity -were buried in this memorial complex, in contrast to divided cemeteries where ethnic groups buried their soldiers and civilians who died in the war. The families agreed to bury their lost children in the same cemetery as those of other religions. The absence of religious symbols on the headstones contrasts with prevailing memory policies. Although ultra-nationalist politicians and clergypersons strongly opposed it, Tuzla citizens commemorate their children together, preferring to avoid any symbolism (Armakolas 2015). A similar example is a monument dedicated to the memory of murdered children of war in Konjic. The memorial includes child victims' names from all sides, even though their perpetrators were from different sides of the conflict (ljubas 2018).
Finally, two contemporary commemorations were held on the municipality level to promote a narrative of shared suffering. Newly founded monuments (the Peace and Tolerance monument in Bosanski Petrovac and a memorial dedicated to all civilian victims of war in Žepče) are rare examples of municipality-level shared commemorations dedicated to all victims, irrespective of their ethnoreligious background. In these places, the scale of violence was not comparable to mass killings elsewhere in Bosnia-Herzegovina, and violence was symmetrical between the sides of the battle. The current mayor, Mato Zovko, who initiated the memorial for all civilian victims, belongs to the main Croatian party, Croatian Democratic union of Bosnia-Herzegovina. A similar initiative was carried out by the SDP in Bosanski Petrovac, where Bosniaks are the majority and Bosnian Serbs are the largest minority group.

Discussion
The empirical findings support the earlier literature's emphasis on the division of mainstream victimhood narratives that manifest 'a numbers game' between the sides. More importantly, the findings also show that examples of a narrative variation from competitive to inclusive victimhood is observable even in cases where exposure to violence was asymmetrical. The extent of the asymmetry of violence seems to matter in this observation. For example, one of the places where expressions of weak CV were found was Sarajevo, which experienced the highest number of human losses (military and civilian casualties from both sides) during the war. However, no traces of inclusive victimhood occurred in narratives regarding Srebrenica as there were only civilian victims from the Bosniak side.
A thick reconciliation approach prioritises the cross-cut victim-perpetrator roles by constructing a single narrative of the past (Strupinskiene 2012) and generates the expectation of an apology by the perpetrator side and forgiveness by the victimised side. However, the rare narratives presented in this article show that this is not the case for Bosnia-Herzegovina overall since there were Bosnian Serb and Bosnian Croat losses as well. Therefore, dealing with the past requires going beyond the thick reconciliation approach. Some scholars of reconciliation and agonistic memory have also pointed out the need to acknowledge other's emotions and political passions in remembering the past (Rosoux 2004;Rumelili and Strömbom 2022;Bull and Hansen 2016). Correspondingly, I propose that when experiences of violence are not entirely unidirectional, there can be space, beyond the formulations of thick reconciliation, for peaceful coexistence. Acknowledging outgroup suffering alone (weak CV) or recognising others' suffering and ingroup responsibility (inclusive victimhood) can enable this space and help overcome the construction of hierarchies of ethnonational victimhood.
The CV typology has proven helpful in understanding the varying levels of competitiveness in victimhood narratives in Bosnia-Herzegovina. The analysis of the Bosnian context has revealed a number of narrative examples of weak CV and inclusive victimhood, which goes beyond the emphasis on dominant CV narratives. The empirical findings suggest that mourning human losses provides a space for crossing ethnonational boundaries. Nonetheless, admitting ingroup responsibility/guilt is less likely for the time being. The only example that emerged in this study concerns a limited number of veteran groups who voluntarily joined reconciliation projects and visited other communities' commemorations of victims. Their experience indicates that veterans can understand shared responsibility by not internalising guilt but recognising others' perspectives and why they are perceived as perpetrators by others.
Preconditioning others' recognition/apology prevents acknowledgement of ingroup responsibility for harm inflicted upon others and resonates with the mid CV category in the typology. Some elements that promote variation from mid CV and promote inclusive victimhood are the locality of conflict, indicating power symmetry/asymmetry between groups, and proximity of violence. For example, in Bosanski Petrovac and Žepče, the legacy of low-scale symmetrical violence suggests a higher likelihood of promoting a shared suffering narrative based on the broader influences of war or a narrative more dedicated to all victims. Proximity is important because people cannot deny what happened easily if the violence was happening nearby (Interviewee 21). Furthermore, the nostalgic feelings for pre-war social relations, being the adult children of mixed marriages, non-involvement in war, and anti-nationalist ideological stance reveal the components of narratives that are more likely to create understanding for both sides' claims for victimhood/harmdoing on a social level.
Some respondents argued that inclusive victimhood is a Serbian-Croatian approach to levelling suffering and avoiding recognising responsibility for harmdoing during the war. The levelling of victimhood is linked to the notion of universal victimhood (levy and Sznaider 2005), which is criticised for failing to differentiate between victims and perpetrators, minimising responsibility for crimes, and belittling the victim side's suffering. In contrast, the concept of inclusive victimhood acknowledges the boundaries between outgroup suffering and ingroup responsibility for harms inflicted upon others in inter-ethnic relations rather than constructing hierarchies for individual victims and approaching victims and victimisers as equal categories.
Moreover, insisting on a primary victim status can be equally risky when victimhood is associated with 'pure innocence' , ie not being guilty/blameworthy for any crimes. Thus, recognising others' victimhood is often regarded as degrading ingroup victimhood or acknowledging ingroup blameworthiness. The previous literature established this well for the Serbian side (Bandura 1999;Ramet 2007). The analysis in this article shows that there has also been resistance to recognising others' victims from the nationalist Bosniak front, as in the Kazani example. Despite recent developments in recognising Kazani victims, the ingroup's responsibility is denied by claiming that a paramilitary group committed the crime. Not being critical of ingroup crimes is not unique to Bosniaks or Bosnian Serbs. It is even more understandable when catastrophic violence occurs on both sides (even in different places and time frames).
The empirical analysis uncovers the potential for recognising outgroup victims on a social and local political level, in contrast to the divided mainstream narratives on the national political level. Although Bosniak politicians do not openly accept Bosnian Serbs' victimhood during the war, their references to it in private conversations and the recent development with the Kazani memorial indicate that this is a political choice. The same is evident on the Bosnian Serbs' side, as illustrated by Dodik's references to the Srebrenica genocide. Nonetheless, commemorations dedicated to all victims of war (organised by the Žepče and Bosanski Petrovac municipalities) as well as the Sarajevo Roses (presently overseen by Sarajevo canton) and the youth Armor in Slana Banja memorial (endorsed by the Tuzla municipality) indicate that there is potential to promote a narrative of shared suffering in local politics.
Furthermore, the analysis demonstrates that various actors promote inclusive victimhood narratives in the Bosnian context, such as international and local NGos, politicians on a national and local level, victim and veteran groups, and everyday people. These actors often do not create narratives exclusively by themselves but rather in conversation with one another. For example, in Bosanski Petrovac and Žepče, when local politicians promoted these narratives, they received visible support from the international community. However, one should not ignore local endeavours, as can be seen in CNA's efforts to bring multi-ethnic war veterans together. Their dialogue work has developed independently and in conjunction with participants' expectations (Franović 2021;cf. David 2020).
Recognising mutual suffering may incur a risk when it is imposed by external intervention actors; however, this does not necessarily mean there are no locally produced shared victimhood narratives. Among the sites where recognition emerges, official-level apologies are more likely to occur when externally imposed, but one must also recognise the potential differentiation between pure enforcement and granting space to local motivations for local NGos such as the RDC and CNA.
Acknowledging others' victims involves either seeing them as victimised by the same perpetrator, as in the example of Slana Banja memorial complex, or distancing themselves from the ingroup members who harmed others, as in the Kazani example. Those who are remembered also have a potential place in promoting inclusiveness. In the Slana Banja memorial complex, adolescent and child victims of the war constitute an example of multi-ethnic remembrance of suffering. The memorial dedicated to child victims in Konjic presents the only example that includes child victims from all sides, that perpetrators from different ethnic backgrounds killed.
Finally, the current study is not free from limitations. For example, whether the ideas/ perceptions of contemporary power parity/disparity between Bosnian Serbs and Bosniaks affect the tendency to recognise ingroup crimes or accept others' apologetic gestures remains unaddressed. Moreover, the article has not been able to more deeply discuss different degrees of CV used by diverse victim organisations.

Conclusion
In this article, I have traced the potential narrative variation from CV to inclusive victimhood in post-war Bosnia-Herzegovina by mapping narrative sites and analysing the boundaries between recognising others' suffering and acknowledging ingroup responsibility. I contend that even though exposure to violence was asymmetrical between the Bosniak and Serbian sides, there are noteworthy narrative examples of weak CV and inclusive victimhood in post-war Bosnia-Herzegovina. However, weak CV (recognising outgroup victimhood without emphasising responsibility for ingroup crimes) offers a more likely space between conflict and reconciliation for promoting peaceful coexistence than inclusive victimhood (recognising outgroup suffering and ingroup responsibility) in this context.
The contribution of this article is twofold. First, the findings contribute to a theoretical understanding of the typology's weak CV and inclusive victimhood categories and their relation to reconciliation in a context of imbalanced exposure to violence. Furthermore, the empirical evidence from Bosnia-Herzegovina highlights the importance of further investigating the nuances in accepting others' suffering and ingroup responsibility for harmdoing in other contexts. The investigation of these notions elsewhere can deepen our understanding of conflict-to-peace transitions.
Second, the article presents an empirical attempt to go beyond the existing literature's focus on mainstream narratives of contested memories. It reveals potential for exploring narrative variation in Bosnia-Herzegovina. The rarity of evidence enables an innovative method combining multiple narrative sites to understand the features of an uncommon phenomenon -inclusive victimhood.
Finally, the identified new characteristics of competitive and inclusive victimhood can enable further research in Bosnia-Herzegovina and other contexts by considering the effects of different pillars of reconciliation, such as religion, education and dialogue, on variations from competitiveness to inclusiveness.

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributor
Cagla Demirel is a PhD candidate at the Department of Political Science at Södertörn university. Her research interests are peace, reconciliation, inter-ethnic relations, memory politics and collective victimhood.

Notes
1. The article clearly distinguishes between 'victimisation' and 'victimhood' . The former emphasises experiences of harm and suffering on a personal or group level. Conversely, victimhood refers to an identity that embraces the sense of victimisation (Jacoby 2015). Hence, the article does not focus on individual victims or specific victim groups, makes no normative claims about victims' understanding of their or others' victimisation experiences, and does not place any normative expectations on victims. It deals with ethnonational victimhood as a notion of collective identity. 2. For example, power relations were relatively balanced in Northern Ireland, and each side of the conflict (the state, and the Republican and loyalist paramilitaries) offered public apologies during the peace process. 3. The ICTy estimated the number of civilian casualties to be 25,609 on the Bosniak side and 7480 on the Serbian side. Military casualties amounted to 42,492 on the Bosniak side and 14,298 on the Serbian side (Zwierzchowski and Tabeau 2010). 4. The Serb-dominated Republika Srpska entity in Bosnia-Herzegovina and the Republic of Serbia have strong ties founded on common ethnicity. Given that this article's primary focus is Bosnia-Herzegovina, the term 'Bosnian Serbs' refers to the Serbs living in Bosnia-Herzegovina who comprise the majority in Republika Srpska and the minority in the cantons of the FBiH. 5. For example, this ethnonational victimhood hierarchy placed non-Serb rape victims and mothers of Srebrenica at the top of the victims' hierarchy by leaving Bosnian Serb rape victims aside (Berry 2017). 6. The narratives of revengeful victimhood and strong CV are evident in the Bosnian context and manifest in glorifying ingroup war criminals and denying outgroup suffering, respectively. However, these two types are beyond the scope of this article. 7. In this article, the inclusive victimhood category concerns only the inclusion of inter-ethnic groups. It does not define specific victim categories or construct hierarchies among individual victims of violence. 8. This approach is conducive to understanding complex victim identities (Bouris 2007) because it challenges the dichotomous conception of victim and perpetrator roles in post-conflict contexts by giving special attention to individual experiences.
9. The importance of the apologies by the former union of Montenegro and Serbia, and by Serbia, derive not only from the special ethnicity-related ties between Serbs in RS and Serbia but also from an apology-related dialogue triangle among these political entities in which Serbia is positioned as the offering side, Bosnia-Herzegovina as the receiving side, and RS as expecting an apology from Bosniaks and Croats in return. For example, then-president Dragan Čavić of the RS criticised the apology issued by Svetozar Marović (the president of the union) and underlined that the region needed three Willy Brandt(s) and three apologies (lind 2008).