• No results found

Variations of sexual violence in Syria and Sri Lanka : a comparative case analysis on the patterns of sexual violence perpetrated by government forces

N/A
N/A
Protected

Academic year: 2021

Share "Variations of sexual violence in Syria and Sri Lanka : a comparative case analysis on the patterns of sexual violence perpetrated by government forces"

Copied!
77
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Master Thesis Political Science

MSc International Relations

Variations of sexual violence in Syria and Sri Lanka: A comparative case

analysis on the patterns of sexual violence perpetrated by government forces

Supervisor: Dr. Jana Krause

Second reader: Dr. Vivienne Matthies-Boon Eshara Kohli - 11221917

June 22, 2018

(2)

2

Acknowledgments

First and foremost, I would like to thank the brave people who share their horrific experiences while this was incredibly difficult and at times placed themselves in jeopardy. These accounts are of an incredible value and without the testimonies it would not have been possible to conduct this research. Also, I am grateful for the researchers, activists and journalists who interviewed the victims.

Further, I thank my supervisor Dr. Jana Krause for introducing me with the topic of conflict and guiding me through the whole process of researching this difficult topic. Her mentorship throughout the this year has greatly enhanced my understanding of scholarly work and how to conduct a research. Also, I want to articulate my gratitude towards Dr. Vivienne Matthies-Boon who was willing to be the second reader.

Lastly, I would like to thank my fellow students; I learned a lot throughout the classes and they served as a true helpline while reassuring me on my own research process.

(3)

3

Abstract

Conflict-related sexual violence perpetrated by state forces is a complicated matter and the existing literature and theory fail to sufficiently explain its variations. Accordingly, this field of research needs more in-depth study to contribute to the existing theories and build new ones. This research provides a comparative case analysis on the differences and similarities between two cases - Syria and Sri Lanka - using secondary literature. The findings indicate a strategic use of sexual violence by both regimes, the nature of the violence differs, which can be explained by different group ideologies.

(4)

4

Table of contents

Chapter 1. Introduction ... 6

1.1 Research structure ... 9

Chapter 2. Analytical framework ... 9

2.1 Definitions of key concepts ... 10

2.1.1 Conflict-related sexual violence ... 10

2.1.2 Sexualised torture ... 11

2.1.3 Government forces ... 12

2.2 Literature review ... 12

2.3 Theoretical framework ... 18

2.3.1 Causes and motives ... 18

2.3.2 Group ideologies ... 19

2.3.3 Patterns of sexual violence ... 23

2.4 Research design ... 28 2.4.1 Methodology ... 28 2.4.2 Case selection ... 29 2.4.3 Analyses ... 30 2.4.4 Operationalisation ... 31 Chapter 3. Syria ... 32 3.1 Background ... 32 3.1.1 History ... 33 3.1.2 Revolution ... 34 3.1.3 Syrian War ... 34

3.2.2 Patterns of sexual violence ... 36

3.2.3 Conclusion ... 37

3.2 Analysis ... 37

3.2.1 Key numbers ... 38

3.2.2 Patterns of sexual violence ... 39

3.2.3 Group ideologies ... 42

3.2.4 Causes and motives ... 43

3.2.5 Remarks ... 45

Chapter 4. Sri Lanka ... 46

4.1 Background ... 46

4.1.2 History ... 46

4.1.3 Civil war ... 47

4.1.4 Patterns of sexual violence ... 48

(5)

5

4.2 Analysis ... 49

4.2.1 Key numbers ... 49

4.2.2 Patterns of sexual violence. ... 50

4.2.3 Group ideologies ... 53

4.2.4 Causes and motives ... 53

4.2.5 Remarks ... 55

Chapter 5. Discussion and conclusion ... 56

5.1 Summary of findings Syria ... 56

5.2 Summary of findings Sri Lanka ... 57

5.3 Research questions ... 57 5.4 Remarks... 62 5.5 Conclusion ... 63 5.5.1 Limitations... 64 5.5.2 Further research ... 65 Bibliography ... 67

(6)

6

Chapter 1. Introduction

Conflict-related sexual violence is likely as old as war itself. It used to be seen as a tragic but inevitable by-product of armed conflict (Brownmiller 1994). Conflicts without the occurrence of sexual violence are rare (Bensel & Sample 2014).

Conflict-related sexual violence only gained broad attention from social scientists belonging various schools following the Bosnian war of 1992. During this conflict, mass rapes were perpetrated by government soldiers. Soldiers impregnated thousands of Muslim women during house raids and in detention centres and they even set up ‘rape camps’ (Stiglmayer 1994). Shortly thereafter, the Rwandan conflict attracted attention not only because of the horrific genocide but also due to the mass rapes of Tutsi population by Hutu’s (Russel-Brown 2003). More than a decade after the Bosnian and Rwandan conflicts, Wood (2009) found that some conflicts display high levels of sexual violence whereas others do not. This insight encouraged scholars to study the phenomenon of wartime sexual violence, as the variation in conflict-related sexual violence indicated that some factors increased or decreased its likelihood. In other words, the occurrence of sexual violence during war and armed conflict was no longer seen as inevitable, but to some extent preventable. As Wood (2009) argues, “if some groups do

not engage in sexual violence, then rape is not inevitable in war as is sometimes claimed, and there are stronger grounds for holding responsible those groups that do engage in sexual violence” (Wood 2009: 136). This influential finding by Wood resulted in an awareness that

scholars have a lot of work to do to explain variations of sexual violence.

Another development resulting from the broad attention of wartime sexual violence was the recognition of it as a problem that concerns the international community as a whole. This led to the establishment of an international policy framework of protection and prevention and a legal framework to hold perpetrators accountable. Both the International Criminal Tribunal for the Former Yugoslavia (ICTY) and the International Criminal Tribunal for Rwanda (ICTR) classified conflict-related sexual violence as a crime against humanity (Chun and Skjelsbæk 2010) and in some instances as genocide (MacKinnon 1994).

Furthermore, six United Nations Security Council (UNSC) resolutions have contributed to the international framework of variations of sexual violence. Besides the six UNSC resolutions, other events similarly helped build the international framework. In 2013, the Declaration of Commitment to End Sexual Violence in Conflict was adopted and endorsed by 122 countries,

(7)

7 which indicates a willingness of these countries to act to halt conflict-related sexual violence. A year later, the first Global Summit to End Sexual Violence in Conflict took place, which was the largest event on this topic, with delegates from 123 countries. During the summit, member-states agreed on two main issues: to undertake more measures to punish those who used rape as a weapon of war and to start changing the global attitudes on crimes concerning conflict-related sexual violence.

The aforementioned developments have been an incredibly valuable contribution to the international policy framework of prevention and protection, but unfortunately they leave plenty of room for improvement. To enhance the international policy framework of protection and prevention, a better comprehension of variations of sexual violence is necessary to discover why sexual violence is widespread in some conflicts and nearly absent in others and under which circumstances and context the likelihood of sexual violence increases or decreases and to identify the prevailing patterns of violence. In other words; which factors influence the occurrence of sexualised violence and which factors lead to restraint.

Numerous factors lead to an increased probability of conflict-related sexual violence. The number of deaths (Wood 2006), ethnic conflicts, pre-war data on sexual violence (Tomkins 1994) and patriarchy (Brownmiller 1975) are some predictors of conflict related sexual violence. Although all these elements influence the probability of wartime sexual violence, a substantial part of the variation of sexual violence remains unexplained. More specifically, it is unclear why in some conflicts only one side of refrains from engaging in sexual violence.

For instance, the existing literature does not explain why some of the deadliest conflicts display such a low level of sexual violence; Sri Lanka and Israel/Palestine (Wood 2006). To comprehend the occurrence of sexual violence, a solid understanding of variation between and within conflicts is necessary. Part of the variations in conflict-related sexual violence can be explained by the perpetrating actors (Cohen and Nordås 2014).

The traditional understanding of perpetrator actors stated that rebels and militia members are most likely to engage in sexual violence in a conflict. However, Cohen and Nordås (2014: 425) showed that state actors commit roughly twice as much sexual violence as rebel groups and militias. A possible explanation could be that rebels and militias are more dependent on civilian support and therefore refrain from committing sexual violations (Cohen 2014). This relatively novel insight of state actors as the main perpetrators of conflict-related sexual violence requires systematic study. Outcomes of such researches can be utilised to shape and form policies of

(8)

8 prevention and protection. To “turn policy into action, a good understanding of the causes of

sexual violence in warzones should shape the implementation of resolutions and protection policies” (Krause 2015: 9). In addition, Baaz and Stern (2010:17) emphasise that the lack of a

solid comprehension of the causes and motives of sexual violence is one of the main reasons for the obstruction of effective measures. Accordingly, more research on the variation of wartime sexual violence is required.

Since a major part of sexual violations perpetrated by state forces occur inside detention centres, these atrocities could be identified as sexualised torture, as the types are often gruesome. To comprehend this specific type of violation, it is necessary to gain insight into this topic as well. According to the Istanbul Protocol, sexualised torture starts with forced nudity as “an individual is never as vulnerable as when naked and helpless. Nudity enhances the

psychological terror of every aspect of torture, as there is always the background of potential abuse, rape or sodomy” (Istanbul Protocol p. 41). Further, the illegality of torture and thus

sexual torture finds its foundation in the United Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment. This brings up the question of whether sexualised torture could be considered as a weapon of war.

To shed light on the variations of sexual violence, I focus on two cases displaying high levels of sexual violence that have similar and different characteristics; Syria and Sri Lanka. Although, both these cases display high levels of sexual violence the differences of the causes and motives, group ideologies and patterns of violence could lead nonetheless to new insights. The complex character of the Syrian civil war results in lots of warring parties. ISIS attracted most media publicity regarding sexual violence, whereas the sexualised atrocities committed by the regime remained undercover. One of the reasons may be that ISIS purposely broadcasts their atrocities as part of their media campaign, whereas the regime does not. However, the level of sexualised violence perpetrated by the regime is significant and unfortunately the international framework of policy and protection does not seem sufficient to protect civilians. The second case - Sri Lanka - was chosen as it is a suitable case for comparison with Syria on the basis of differing characteristics (Gerring 2017).

Since the majority of sexual abuses perpetrated by government forces occur inside of detention centres, this thesis focusses on the specific conditions in detention centres. Accordingly, the aim of this thesis is to describe the differences in patterns of sexual violence between the two cases, specifically that perpetrated by government and security forces, as both cases show high

(9)

9 levels of sexualised violence inside detention centres. The results could contribute to the broader framework of variations in sexual violence during armed conflict as new insights could help build new theories or develop existing ones to explain variations of sexual violence. This objective resulted in the following research question;

What are the similarities and differences of patterns of sexual violence perpetrated by the government forces of Syria and Sri Lanka?

To answer the research question, three sub-questions were formulated, which are presented in the method section. Ultimately, the data reveals that perpetrators of both cases are motivated by strategy, although the aim of the strategy and the regimes group ideologies result in varying patterns of violence.

1.1 Research structure

The remainder of this thesis is structured as follows. In chapter 2 I define the terms used in this thesis. Subsequently, a literature review is outlined containing leading literature regarding sexual violence. Then I elaborate on the theoretical framework that serves as a foundation for the analysis and is divided into three sections corresponding with the sub-questions: causes and motives, group ideologies and patterns of violence. Next, the design of the research is discussed including a justification of the cases and a method section. Chapter 3 on Syria provides first a background section of the history, the civil war and an overview of sexual violence according to human rights reports. Chapter 4 on Sri Lanka outlines the background as well starting with the history, the civil war and maps out the perpetuation of sexualised violence based on different reports. And lastly, the first section of chapter 5, provides a summary of each analysis, discusses the results by demonstrating how the literature explains the finding and elucidates alternative explanations. The second section of the last chapter draws conclusions of the study as a whole.

2. Analytical framework

This section maps out the analytical foundations of this research by first discussion the relevant concepts, by providing a literature review on the development of scholarly work, a theoretical framework and lastly, the design of the research.

(10)

10

2.1 Definitions of key concepts

To understand the phenomena in this thesis the relevant terms are discussed and defined. These involve consecutively conflict-related sexual violence, sexual torture and government forces.

2.1.1 Conflict-related sexual violence

Various scholars and institutions define sexual violence in several ways. Therefore, I examine some definitions before presenting the one I use. The World Health Organisation defines sexual violence as “any sexual act, attempt to obtain a sexual act, unwanted sexual comments or

advances, or acts to traffic, or otherwise directed against a person’s sexuality using coercion, by any person regardless of their relationship to the victim, in any setting” (WHO; 149). In

addition, the ICTR, for example, states that sexual violence is “any act of a sexual nature which

is committed on a person under circumstances which are coercive” and can be broadly

understood as “acts of a sexual nature imposed by force, threat of force or coercion, or by

taking advantage of a coercive environment or a person’s incapacity to give genuine consent”

(De Brouwer 2005) . According to the ICC, sexual violence includes rape, sexual slavery, forced prostitution, forced pregnancy, and forced sterilisation.

The obvious and distinguishing mark between sexual violence and conflict-related sexual violence is the context of armed conflict. Armed conflict is defined under International Humanitarian Law, which distinguishes between international armed conflict (IAC) and non- international armed conflict (NIAC) (ICRC 2008). The first type of armed conflict involves two or more opposing states, whereas a non-international armed conflict has two or more parties to the conflict in the same state, for example, government forces and non-governmental armed groups. Altogether, conflict-related sexual violence is defined in this thesis as a combination of the definitions mentioned above; acts of a sexual nature imposed by force, threat of force or coercion, or by taking advantage of a coercive environment or a person’s incapacity to give genuine consent, including rape, enforced rape, sexual slavery, forced prostitution, forced pregnancy, forced sterilisation, or passive and indirect sexual violence in the context of armed conflict, either NIAC or IAC. Further, I use the terms sexual violence, sexualised violence, conflict-related sexual violence, and wartime sexual violence interchangeably.

(11)

11

2.1.2 Sexualised torture

The difference between torture, sexual violence and sexual torture could lead to confusion. Some scholars agree that sexual violence is considered sexual torture (Oosterhoff et al. 2004; Mitchell 2004: 671). This seems straightforward as the nature or cruelty of the violence does not distinguish ordinary violence from torture but the aim and context of the violence. In other words, violence perpetrated by a guard in the prison is perceived as torture because of the perpetrator and the setting. Taking this into account, sexual violence, committed for the same goals and context as torture is regarded as sexual torture. Mitchell argues that sexual torture is mostly motivated by personal goals (ibid). Further, the ICTY recognised rape as a form of torture (ICTY 2009). For the sake of this study, this definition is not specific enough therefore, I elaborate on it by exploring other definitions. For instance; “The use of any form of sexual

activity with the purpose of manifestation aggression and of causing physical and phycological damage” (Lira & Weinstein 1986: 1). The latter combines the psychological and physical goals

of torture, resulting in sexualised torture.

These definitions do not indicate the context in which sexual torture occurs nor the nature of the act. The United Nations Convention Against Torture maintains the following definition which could provide helpful insights:

“Torture means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. It does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions.” (The United

Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment).

Thus, the emphasis lies on the fact that violence is systematic and is deliberately used to collect information, as punishment for alleged committed actions, as punishment on discriminative grounds, or as a deterrent tool. In this regard, sexualised torture concerns sexual violence with the objective of obtaining the aforementioned goals. This implies that sexual torture would have a strategic goal, although it is likely that this is not always the case.

(12)

12 After conducting preliminary analysis of the data I found the level of cruelty remarkable and hence worthwhile to research more in depth. However, the existing definition does not allow for distinguishing between sexual violence and sexual torture therefore I decided to use the following definition: sexual acts perpetrated with the purpose of manifesting aggression and causing both physical and psychological damage, regardless the motive. The level of cruelty demarcates the line between sexual violence and sexual torture. When the perpetrator goes beyond the ‘normal sexual acts’ by using objects, electrocution, and sexual mutilation I consider this sexual torture. By ‘normal sexual acts’ I mean oral sex, vaginal penetration and anal penetration. Maintaining this definition enables me to measure the concept of sexual torture while distinguishing between sexual violence.

2.1.3 Government forces

Although the meaning of government forces is straightforward, it does no harm to define it:

“all government actors with special status as representatives of the state” (Cohen & Nordås,

2013:6). I also consider plainclothes militias working for the government as government forces. Both Syria and Sri Lanka have active plainclothes militias working for the government which I refer to as plainclothes militias for the Sri Lankan case and shabiha for the Syrian case. Furthermore, the terms government forces, security forces and military are used interchangeably.

2.2 Literature review

The literature on conflict-related sexual violence found its beginning with the publication of

Against our Will: Men, Women and Rape (Brownmiller 1975), in which the author argues that

men engage in sexual violence during wartime due to the absence of social norms and judicial sanctions that exist in peacetime. According to Brownmiller, soldiers convince themselves and others of their militaristic identity-which is strongly characterised by masculinity-by demonstrating their power over the enemy and women in general. Based on this theory, conflict-related sexual violence is inevitable and inherent to the male biological nature and corresponding instincts.

Contrary to Brownmiller’s argument, psychoanalytical theory claims perpetrators engaging in sexual violence during armed conflict show “deviant and abnormal behaviour” which some scholars thought this theory was unconvincing as it would mean that a substantial number of

(13)

13 soldiers were suffering from psychiatric conditions (Skjelsbæk 2001: 212). Brownmiller’s argument was disputed from the cultural anthropologic angle, claiming that rape is a result from cultural and social norms instead of merely biological instincts (Sanday, 1981).

During these early years of academic research on sexual violence in armed conflict, the topic received little attention, however, this dramatically changed after the Bosnian and Rwandan conflicts (Skjelsbæk 2001). These horrible conflicts demanded broad attention initiated by Stiglmayer’s (1994) influential book Mass Rape: The War against Women in

Bosnia-Herzegovina containing contributions of other leading scholars, such as Seifert, MacKinnon

and Enloe. Stiglmayer interviewed victims of the Bosnian war, documenting the gruesome experiences women had to go through, notably, the existence of rape houses, which were used to impregnate women to ensure that the victims would give birth to Serb children (Stiglmayer 1994). Consequently, the topic of sexual violence in armed conflict received considerable attention from scholars, although the literature was characterised by confusion about what sexual violence entails and whether it can be classified as a weapon of war (Skjelsbæk 2001).

To date, the range of literature on wartime sexual violence concentrates primarily on either opportunistic or strategic accounts (Leiby, 2009: 447). This is probably a result of the Rwandan and Bosnian conflicts with their documentation of systematic rape. The violations during this conflict have resulted in theories of mono-causal explanation, which should be treated with caution, as concentration on one characteristic assumes that conflict-related sexual violence is used for a single objective (Leiby 2009).

Besides strategic and opportunistic sexual violence, a third category, sexual violence as a practice, was introduced by Wood (2009), as she discovered that sexual violence occurred because perpetrators were socialised within the armed group. Identifying whether sexual violence involves opportunistic or strategic motives can be extremely challenging, so few studies have been conducted on this topic (Leiby 2009).

Some feminist researchers argue that sexual violence in conflict can be explained by the deficiency of social, political and economic women’s participation (Brownmiller 1975; MacKinnon 1994; Seifert 1994). Skjelsbæk (2001) classified feminist literature into three epistemologies and possible corresponding accounts: the essentialist, the structuralist and the social constructivist perspective. From the essentialist perspective, every woman, without an exception, could be subjected to sexual violence in during conflict. The explanation for this does not deviate from explanations during peacetime and is caused by “inequality,

(14)

14 discrimination, male domination and aggression, misogyny and the entrenched socialisation of sexual myth” (Tompkins 1994: 851). One of the main shortcomings of this perspective involves the lack of explanatory potential. Moreover, this approach fails to explain variance between conflicts and between men and women (Krause 2015 10).

The structuralist perspective argues that women are potentially subjected to sexual violence as a consequence of patriarchal structure combined with group identities (Krause 2015: 10). Sexual abuse of women is seen as a disgrace for the whole social identity, because men failed protect the women (Bastick, Grimm & Kunz 2007). This perspective has two major shortcomings: identifying ethnic violence as the main cause of sexual violence neglects other explanatory factors and the perspective fails to explain sexual abuse against men.

The third perspective explains the establishment of gender relations composed by human reciprocal action. From this perspective, both men and women can be subjected to sexual violations. By sexually violating someone, the violator confirms their masculinity and feminised the victim (Skjelsbæk 2001: 215). According to Skjelsbæk, the social constructivist perspective is most convincing due to the shortcomings of the other two perspectives (Skjelsbæk 2001: 214).

Several other scholars claim that rape is considered as a weapon of war (Kirby 2012: 798; Leiby 2009). Although mass rape in war received little attention until the 90s, the occurrence of mass rape during war is certainly not new (Card 1996: 5); the first documentation of conflict-related sexual violence dates back from the Ancient Greeks (Fishman 1996). While there seems to be a broad consensus of the existence of strategic, systematic, and widespread sexual violence during wartime, scholars disagree on whether to define this practice as a weapon of war.

Krause warns researchers to remain cautious in labelling mass rape during conflict as a weapon of war because of a complexity of causes: “social dynamics within armed groups and groups’

institutions, ideologies, leadership and recruitment patterns are major explanatory factors”

(Krause 2015: 18) and doing so would neglect these explanatory, variables resulting in an oversimplification of the matter. Furthermore, there is a risk of focussing solely on rape and women, thus excluding other forms of sexual violence and men as victims (Krause 2015: 23). Krause demonstrates that wartime sexual violence occurs due to several factors both on the micro and meso level (Krause 2015: 23). She continues to argue that the Rwandan and Bosnian cases were exceptional although it was of major influence on the existing literature on wartime

(15)

15 sexual violence. Skjelsbæk argues “it is too widespread, too frequent and seemingly too

calculated and effective not to be part of a larger political scheme and hence a weapon of war”

(Skjelsbæk 2001: 115).

However, Wood has an alternative explanation and claims that rape is not necessarily a weapon of war even if it is frequent. Rather it could be explained as a social practice resulting from the horizontal and vertical social dynamics within a combat group (Wood 2018: 3). In support of Wood’s argument, research shows state military in the DRC have frequently engaged in sexualised violence against civilians. Wood’s Rape as a Practice Theory takes into account the warnings about the mono-causal explanation (Leiby 2009; Wood 2018, 2012; Krause 2015), so this theory has a lot of potential, especially for explaining sexual violence by government forces. Wood proceeds by stating that the weapon of war debate has contributed to the international protection and prevention framework, yet it is important to compose theories that can explain the excessive use of sexualised violence when it is not ordered or authorised, as was the case in the DRC.

The question remains when sexual violence can be perceived as a weapon of war. The distinctive characteristic to deem rape as a weapon of war is that it is “used as part of systematic

political campaign which has strategic military purposes” (Skjelsbæk 2001: 213). This was

the case in Bosnia where rape was systematically committed by Serb forces with the aim of impregnating Bosnian women to prevent them of having children within their own ethnicity. While there is a general agreement of this unique case as an example of the use of rape as a weapon of war some scholars go even further by claiming that forced impregnating can be classified as a genocide (Barstow 2000: 45).

An important relationship is that between principals and agents, as it involves two dilemmas that could influence the occurrence of sexual violence by armed groups: goal variance and information asymmetry (Mitchell 2004; Butler et al. 2007). The former involves deviating goals and motives between principals and agents. The latter indicates misinformation as a commander leans on his soldiers for information from the field. Simultaneously, commanders could take advantage of the sexual violation perpetrated by their soldiers to serve their own goals. Among other things, it weakens the opponents and instils fear without the direct involvement of combat. Allowing this form of abuse stimulates soldiers to perpetuate sexual violence and is considered a reward, also known as “The Count Tilly’s reward” (Leiby 2009).

(16)

16 By not officially ordering agents to engage in sexual violence, principals have the opportunity to deny knowledge of this abuse (Mitchell 2004). Additionally, Mitchell claims that sexual violence is generally motivated by personal and opportunistic goals, so the principal-agent logic plays a crucial role (Mitchell 2004 : 671). In support of the principal-agent theory, cross-national statistical research has produced findings consistent with this theory; agents who experience more control are less likely to engage in sexual violence (Butler et al. 2007). Despite these findings, the principal-agent logic is disputed by Cohen and Nordås, as they claim the theory is too limited and thus neglects other important explanatory factors that lead to sexual violence committed by militias (Cohen & Nordås 2015: 878). To adequately explain the variation in conflict-related sexual violence, it is imperative to take other variables into account.

According to Wood (2006), the degree of perpetuation of sexual violence differs between actors involved in a single conflict. In support of this claim, statistical research has shown that state forces are considerably more likely to commit sexual violence during conflict and war (Cohen 2013; Cohen & Nordås 2015; Leiby 2009). This can be explained as follows: state forces generally do not depend on civilian support, as they possess sufficient means – information, artillery, intelligence - to exercise control whereas rebel groups rely on civilians for information and support. Therefore, rebels are less likely to engage in sexual violence (Wood 2009).

Another statistical study finds that militias that received government training have a higher probability of engaging in sexual violence using the SVAC dataset. This was in contrast with the expectation of the researchers, as training is often associated with discipline (Cohen & Nordås, 2015: 887). This curious finding can be explained by Wood’s Rape as a Practice Theory (Wood 2012: 393, 2018). During training, soldiers are subjected to violent situations which involve the demoralisation of opponents and similar socialisation processes. Moreover, they are formed not only by the leadership but also by their fellow-combatants.

Sexualised state torture

The cultivation of torture stems from establishing a tool for state forces to determine the authenticity of a testimony (Evans & Morgan 1998). However, with the United Nations’ Convention Against Torture (CAT), member states endorsed the urgency of putting an end to torture and consequently torture became illegal (Ryals Conrad & Moore 2010: 459). Nonetheless, torture still occurs inside detention centres (ibid). It is plausible that a wide range of sexualised torture exists under the disguise of torture. Accordingly, this part of sexual

(17)

17 violence could be seen as neglected in the existent literature as scholars and researchers to date have not studied this in depth.

Generally, torture is divided in two distinct categories: physical and psychological (Agger 1989: 306). While the main objective of the former is to cause various degrees of pain, the latter aims to destroy the victim’s ‘defence mechanism’, which eventually leads to a break down. Psychological torture includes, among other, solitary confinement, sleep-deprivation, forced nakedness and sexual torture. While all torture is incredibly traumatic and comparison is impossible, sexualised torture leaves few or no evident traces (Agger 1989: 306). Although this claim may be true in many instances, the severity of sexualised torture determines whether this act leaves visible traces. Perhaps the changing character of sexualised torture comes to play with Aggers’ argument, as the nature of sexualised torture has become more brutal. Taking this into account, sexual torture can be seen as an intersection of psychological and physical torture. Moreover, this could be related to other factors, such as institution forms, activity of human rights organisations and civil society. Additionally, there is a general understanding of state torture is unusual, yet this is not the case (Ryals Conrad & Moore 2010).

Whereas the definition of sexual violence can be perceived as static due to the clear definition, this is not the case as the meaning changed over time. In colonial times, rape was defined as non-consensual relations with married women when the man was someone inferior to the husband of the victim (Skjeslbeak 2001: 212). This changing character of sexual violence is important to acknowledge, as it could have implications for researching this topic. For instance, the objectives or patterns of violence may change over time. The changing character might be specifically important for the nature of sexualised torture in this research, as Agger claims this leaves little traceable evidence, yet numerous human rights reports have published horrific accounts of sexual torture leaving the victim in desperate need of medical care. This emphasises the importance of systematic and continuing study coincided by temporal bounds.

2.1.1 Literature gap

Numerous factors have been identified that influence the occurrence of sexual violence and scholars have come to valuable insights, of which arguably the most important is that wartime sexual violence is to some extent preventable. However, to date, scholarly work failed to produce a comprehensive explanation for variation between wars and armed groups, particularly regarding sexual violence perpetrated by government and security forces. While most scholarly work focussed on sexual violence perpetrated by rebel groups and armed

(18)

18 militias a solid framework to explain conflict-related sexual violence perpetrated by security forces is missing, although some theories may be applicable to sexual violence committed by government forces. Likewise, the changing character of sexual violence and torture should be taken into account, requiring continuing systematic research on this topic. Possibly, the changing character could lead to varying patterns of violence.

Several scholars argued the existing theories have a substantial limitation in explaining variation, therefore, more theory building is necessary to provide new insights. Identifying differences and similarities in the patterns of violence through a case study which aims to provide an in-depth understanding of this variation could contribute to theory forming of variations of wartime sexual violence. While studying this topic, it is important to take the changing character of sexual violence into account. Obviously, the comparison of these specific cases alone would not be sufficient to generalise the results, however, it may serve as an impetus to study some specific causal mechanisms more in depth. Additionally, some findings may be interesting to investigate whether other cases have similar patterns which could lead to the explanation of variations of sexual violence.

2.3 Theoretical framework

This section discusses the relevant literature that serves as a baseline for the analysis. The section is divided into three subsections: causes and motives, patterns of violence and group ideologies.

2.3.1 Causes and motives

As stated in the literature review, determining whether sexualised violence is caused by opportunity or strategy or is a practice is extremely difficult. Nonetheless, examining the conditions under which it occurs might give insights. Opportunistic sexualised violence involves violence that state forces have not been ordered to enact and therefore commanders may not be aware of its occurrence. It is primarily practiced to satisfy personal needs and self-benefit. Whereas strategic use of sexual violence could be commissioned by the state or high-ranking commanders. In some cases, it is allowed by commanders as the occurrence of violence could benefit their goals and therefore they fail to intervene (Wood 2012: 393). To prove sexualised violence is a strategy, “its purposeful adoption for military objectives should be

shown, not presumed” (Wood 2018: 2). Meaning, convenient side effects resulting from

(19)

19 Furthermore, senior officials may implement sexual violence as a strategy by engaging in sexual torture of detained victims and public rape as a means to terrorize civilians (Wood 2014: 472). In other instances, sexual violence could serve as a reward; commanders may have calculated that it is beneficial although it has some side effects, for instance less cooperation’s and loyalty to civilians and decreased discipline (ibid). Additionally, Leiby claims that regime forces commit sexual violence in areas where rebels are active although do not exercise effectively authority. Under this circumstances, they make use of sexual torture and gang rape to terrorize the locals and gather information. Her research in Peru confirmed this theory (Leiby 2009). Although the weapon of war debates has had a valuable contribution to the understanding of rape being an integral part of warfare, and the Strategic Rape Theory can have a lot of explanatory potential, various scholars warn against the oversimplification of wartime sexualised violence leading to a neglect of other forms of sexualised violence (Krause 2015; Leiby 2009; Baaz & Stern 2013; Boesten 2010, 2014; Wood 2009).

The third category - sexual violence as a practice - is gradually gaining ground as the weapon of war debate continues, but also increasingly meets with critique. In this category, commanders do not order or encourage their agents to actively engage in sexual violence. This category involves opportunistic rape but also rape resulting from horizontal group dynamics (Wood 2018: 3). In support of the horizontal group dynamics claim, research shows that soldiers claimed that social pressure from their peers influenced them to take part in the violations. Another characteristic of this theory is that leadership fails to adequality punish their agents for taking part in sexualised crimes (ibid). Several scholars claimed that sexual violence is frequently used in order to bond relationships within the military (Krause 2015). The occurrence in order to create cohesion can be perceived as bonding technique (Wood 2014). Consequently, militias which rely on forced recruitment make use of this bonding technique which is effectively (Cohen 2013). In this sense, it is often used during gang rapes in front of peers. Cross-national research shows militias practicing forced recruitment have a higher probability of performing gang rape (Cohen 2013). Commanders take part in these gang and frequently encourage or even force their combatants to do the same (Cohen 2013).

2.3.2 Group ideologies and strategies

Leaders of armed forces are burdened with the task of forming a cohesive combat force. Armed forces with low levels of cohesion are more likely to commit more sexual violations, as engaging in sexual violence strengthens the group cohesion (Krause 2015). This leads to

(20)

20 another influential factor: socialisation. Combatants socialised to high levels of violence are more likely to use violence in future situations (Wood 2018). This mechanism could apply to sexual violence too.

A consensus exists that governments play a crucial role in the extent of violence during war (Chiozza & Goemans 2011; Horowitz et al. 2015; Midlarsky 2005: Valentino 2004). While most of the violence can be regarded as instrumental and leaders will use this when benefits outweigh the costs (Valentino, 2004), variation does exist. This can be explained based on ideology, which is defined as “a set of more or less systematic ideas that identify a constituency, the challenges the groups confronts, the objectives to pursue on behalf of that groups and a program of action” (Gutierrez Sanin & Wood 2014: 214). Sometimes these ideologies also involve strategies to achieve the perceived objective (ibid).

The role of ideology can be analysed through a ‘weak programme’ and a ‘strong programme’. The former follows an instrumentalist view and regards ideology as a means to achieve the group objective. The latter has a normative commitment to the ideology itself. Depending on the nature of the ideology, members of the group can refrain from engaging in sexual violence. Thus, analysing violence through the weak programme only would fail to explain variation (Gutierrez Sanin & Wood 2014: 223). The LTTE forces are a good example of the strong programme, as the group ideology required that members should at all times restrain from engaging in sexual violence (Gutierrez Sanin & Wood 2014). Analyses of this example through the weak programme only would fail to explain this variation as the LTTE seems to have had a normative commitment to restrain of engaging in sexualised violence. In this sense, normative commitment can be understood as the commitment of recruits to conform the group norms and thus ideology. However, the leadership could follow a certain ideology for instrumental reasons (Gutierrez Sanin & Wood 2014)

Additionally, variation within a conflict can be to some extent explained based on different ideological beliefs of state officials within a government (Scharpf 2018: 206). Therefore, both group and individual ideologies influence the decisions that determines the extent of violence used.

Ideology could play a crucial role in the beliefs of the perpetuating party and can be used to obtain certain goals or the perpetrators could have a normative commitment to the ideology itself. In the former, it has an instrumental value-the means justify the goals. Sexual violence

(21)

21 can be used to attain the goals. In the latter sense, the ideology itself is characterised by sexual violence or exercising restraint.

Shame and honour culture

Another aspect of group ideology involves the concept of honour, defined as “the right to be

treated as having a certain worth” and “a right to respect” (Stewert 1994: 21). Societies which

places high value on women’s honour and purity have an increased incentive for using sexual violence as a distressing means for dishonouring families (Leiby 2009: 449). While victims of sexual violence in all kinds of societies feel ashamed, perpetrators in more conservative societies have an extra stimulus to dishonour victims. Victims of sexual violence often feel ashamed of themselves, which results in their silence (Skjelsbæk 2001: 212), which in turn could result in the impunity of perpetrators. Additionally, it is probable that victims in societies where women’s sexual purity is paramount for the honour of families and the community will avoid speaking about their experiences. According to Bjarnegard et al. (2017: 751-752), honour is intertwined with patriarchal norms and masculine toughness. Patriarchal societies often hold the family/sect/community honour dear, meaning that a disgrace of an individual could harm the whole community. Accordingly, a women’s virginity determines her stature and position in the ‘marriage market’. Once her status is compromised, reversal is impossible, often resulting in the community perceiving the victim as a vagabond. Consequently, the women’s status is perceived through her relations (Weitsman 2008: 563). This fosters a culture marked by shame and stigma, which causes victims to fear to speak about certain topics. Besides the feelings of shame and guilt, victims subjected to sexual violence could face resentment from their family or community. This resentment, rejection, and in some instances violence could tear whole families and communities apart.

Remarkably, the act itself is far less devastating than the shaming by community. Several scholars agree that sexual violence stems mostly from patriarchal ideologies (Cohen, Hoover Green & Wood 2013; Shalhoub-Kevrokian 1999).

Additionally, ‘masculine toughness’ plays a central role in the notion of honour, as it requires men to engage in violence to protect their honour. In support of this theory, researchers found that in cultures that place high value on honour, men “are prepared to fight or even to kill to defend their reputations as honourable men” (Nissbet & Cohen 1996: 4). Consequently, failure to respond to insults results in the perception of a powerless man, unable to safeguard himself and his family (Nisbett & Cohen 1996).

(22)

22

Social identities

Human beings tend to organise themselves in distinct categories (i.e., social identities), which can cause conflict between different identities (Weitsman 2008: 562). Social identities could serve as an indicator for specific targeting of victims. Social identities can be understood as distinguishing characteristics, which include gender, political preference or other similar traits (Wood 2011). Additionally, selective targeting of specific individuals often serves as retribution or punishment for (alleged) committed acts. If violence is carried out based on selective or collective targeting, it can be seen as a strategy. The opposite of selective and collective targeting is indiscriminate violence, which occurs when military makes no distinction based on specific traits, characteristics, social identities, neighbourhoods or similar categories (Masullo 2017).

Militaristic masculinity

In order to behave as a ‘real man’ on the battlefield and similar environments requires to overcome uncertainty and other feelings generally regarded as feminine. This internal development might make ‘it easier’ to engage in sexual violence (Seifert, 1994: 59-62). Enloe explains that “in the war-zone men are not only set to be men, but to be militaristic men.

Understanding of militaristic culture is key in attempting to understand the gender dimension of the war-zone.” and “The glue [of militarism] is camaraderie, the base of that glue is masculinity” (Enloe 1993: 52). Conflict-related sexual violence can be perceived through

re-enhanced patriarchal relations between men and women (Skjelsbæk 2001: 217). While initially patriarchal relations would explain sexual violence against women, this theory is helpful for explaining sexual violence against men too. Direct sexual abuse of men feminises the male victim and fosters the perpetrator’s masculinity, whereas indirect sexual violence against men, for instance, sexual violence against female family members, feminises the victim through the inability to protect the direct victims (Leiby 2009: 450).

A good soldier needs the ability to engage in violence. This often requires soldiers to witness violence, change their moral beliefs and dehumanise their victims. Further, desires and practices can change dramatically due to continual witnessing of violence, feelings of insecurity for their own life, which in combination with the process of socialising may lead to increased levels of violence against civilians. This also includes sexualised violence (Wood 2018: 7). Research shows correspondingly that nearly half of the military men of DRC who participated in the research found it satisfying to abuse others (ibid). Evidence from perpetrator

(23)

23 testimonies reveals that violence was an integral part of training, as the new recruits were subjected to beatings and solitary confinement and forced to commit severe acts of violence against dogs. This last part was to transform the soldier into a perpetrator (Boesten 2014: 39).

Taking all these group ideologies into account, it becomes clear that ideology plays a decisive role for explaining variations of sexual violence. First, the dominant ideology of the specific group-in this case the military-results in certain behaviour. Second, a more general military ideology-militaristic masculinity-defines the military identity of soldiers, which may lead them more easily to engage in sexual violence. And third, the collective ideology of honour can be an explanatory variable.

2.3.3 Patterns of sexual violence

Sexual violence perpetrated by state forces is undoubtedly a form of violence against civilians. To explain variation in sexual violence, it is useful to understand variation of violence against civilians. Because the existent theory of variations of violence against civilians focussed on fatalities instead of violence, it failed to a large extent to adequately explain variations (Valentino 2014: 90). Balcells et. al. claim that political factors are essential in understanding why political violence occurs. When political mobilisation occurs before the war and a large crowd evolves into a politically active one, the likelihood of violence against civilians increases significantly, as they can be seen as a threat (Balcells et. al. 2015: 4). Thus, variation of violence against civilians can be explained by rational choice to suppress civilian antagonists. The more resistance civilians show, the more violent a government is thought to respond. Further, the degree of violence can also vary based on the amount of support in a particular area; when armed forces do not control the territory and fail to persuade locals to support them, they may resort to violence (R. Wood 2010).

While some armed groups refrain to a large extent from engaging in sexual violence, other groups use wide-spread and systematic sexual violations. Accordingly, the patterns of conflict-related sexual violence varies greatly within and between conflicts as well between armed forces (Wood 2009). How armed forces engage in sexual violations can be seen as a type of violence, meaning “a set of practices that a group routinely engages in as it makes claims on

other political or social actors” (Wood 2009: 133).

This set of practices – repertoires - take place within certain patterns of violence. Wood identifies “two patterns of state-sponsored sexual violence” (Wood 2009: 145). The first is

(24)

24 sexual violence of detainees, which is probably more rampant than the second pattern. These sexual violations can take many forms, such as rape, electric shocks or forced sexual actions. The second is sexual violence at checkpoints and similar spots, which is quite challenging to monitor. Besides the difficulty of measuring the prevalence of this form of sexual violence, it has another challenge; identifying which factors lead to this pattern (Wood 2009). Possibilities include a lack of control, allowance or perhaps strategy.

To determine whether recruits were ordered, motivated, or allowed to engage in sexual violence, the testimonies of perpetrators are necessary (Leiby 2009: 448). In addition, it is likely that soldiers held captive by the opposition would state that sexual violations were ordered, as it reduces their individual blame. These factors challenges make it challenging to determine whether the state was or should have been aware of the sexual violations.

If sexualised violence during armed conflict is a result of loss of control, it would be non-existent in front of the leadership, the greater part would happen in the field where supervision is difficult to establish, and targeting based on social identities would be rare (Leiby 2009: 448). Although the third element is debatable, as personal grievances could lead to targeting individuals with a specific social identity. This grievance could be based on one’s social economic status resulting from inequality, political identity or gender. This last social identity could be based on the punishment of women, or punishment of LGBT people.

As discussed in the literature review, the principal-agent logic may play a crucial role in explaining variations of sexual violence. This theory empathises the relationship between the commander and the soldier, which is characterised by two main problems; information asymmetry and deviating goals. As the agent possesses more information from the field, the commander relies on the agent for accurate information. Furthermore, the goals of the commander and agent may differ. According to this logic, Mitchell suggests that sexual violence is a result of loss of control (Mitchell 673).

Further, Hoover Greens’ theory (2016) of the Commander Dilemma may influence group ideology. This theory emphasises restraint rather than violence and claims that combatants who received political education practiced more restraint than those who did not. Restraint is achieved by matching soldiers’ preferences with those of their commanders. This can be accomplished through maintaining effective group institutions. Hoover Green argues that after completion of the training, soldiers’ preferences of violence are on an average far more than of their commanders. Commanders must persuade their soldiers to practice restraint while

(25)

25 avoiding decisions that may jeopardises the individual or group survival. Because commanders tend to differ in preference of levels of violence, it is likely that soldiers will often disagree with their leader (ibid). In other words, commanders need to align their soldiers’ preferences to their own. When this is for some reason not the case, commanders needs to make sure that soldiers practice restraint without causing outrage towards themselves or group members.

Additionally, commanders often provide their combatants with vague orders when it comes to non-lethal violence (Hoover Green 2016: 620). For the use of sexual violence, which is in many instances is seen as non-lethal, this means that combatants often do not receive explicit order to practice restraint. Therefore, variation within an armed group may depend on the extent of allowance from the commander and the preferences of combatants.

Rape regimes

Boesten suggest that ‘rape regimes’ are used strategically for multiple reasons. These include “public mass rape, sexual torture and war booty rape” (Boesten 2014: 26). Whereas public

mass rape is primarily used to instil fear and send a message to the community, sexual torture is often used to gain information (Leiby 2009: 450). War booty rape is seen as a reward for

“hardworking soldiers who need sex” (Boesten 2010: 119).

Also, soldiers who are new to the type of violence may need some guidance or forcing, as becomes clear by a Peruvian veteran perpetrator’s testimony that he did not need orders to

“detain, beat and kill people” “anymore” indicating that soldiers determined by themselves

whether to kill and rape (Boesten 2014: 26). Moreover, the findings suggest that recruits familiarise themselves with committing rape and other forms of brutality. Additionally, forcing soldiers to engage in rape cements their loyalty, obedience, and complicity (Boesten 2010)

State prison torture

One way of explaining the occurrence of torture involves the principal-agent theory, in which the prison guard or interrogator is the agent and the state the principal (Ryals Conrad & Moore 2010: 461). The principal delegates the interrogation to the agent, who receives orders to torture the detainees and generally obeys this order. The obedience to authority study supports this claim (Milgram 1974) in which individuals thought they gave someone a deadly dose of electric shocks because they were told to.

In addition, a hypothetical experiment was conducted in Hawaii, in which a professor suggested that a substantial portion of the population had to be murdered to safeguard the survival of

(26)

26 those ‘most fit for survival’. Interestingly, 91% percent agreed that “under extreme circumstances it is entirely just to eliminate those judged most dangerous to the general welfare” and a staggering 29% approved this ‘final solution’ if this involved their family members as well (Zimbardo 2007).

However, when the principal orders the agent not to torture, the agent might not obey (Ryals Conrad & Moore 2010: 461), as shown by the notorious Stanford prison experiment (Zimbardo 2007). This experiment showed how normal students can turn into violent ‘guards’ and abuse the ‘inmates’ without receiving specific orders to engage in violence or torture.

Although these studies demonstrate how effortlessly individuals engage in violence and thus torture, especially if it is ordered, Enloe claims that “Torturers are trained – in military

doctrine, chain of command, social psychology and anatomies” (Enloe 2000: 129). By making

this claim, she refers to a variety of factors leading a soldier from an ordinary citizen to a trained torturer. Combining these arguments of ordinary people easily engaging in violence and the social dynamics of a military, it seems soldiers inevitably become torturers. Fortunately, this is not always the case and restraint is exercised in prisons. Several factors linked to restraint are democratic institutions, regime type, and activeness of human right organisations.

Besides human nature and social dynamics other factors like regime influences the likelihood of torture inside prisons as well. For instance, autocratic states have a higher probability of engaging in state torture as a frequently used instrument (Rejali 1994). Other scholars show that the process of detaining, torturing and the required bribe which secures releasement serves as a state income (Rothman & Neier 1991).

While in some cases commanders deliberately embrace violent policies, in other instances the leadership implicitly authorises violence exercised by their soldiers with claims like “we are

engaged in total war” by commanders in Bosnia and Sudan and Bush’s “war on terror” (Wood

2018: 8). These expressions serve as licence for excessive violence, such as sexual violence and torture, while effective punishment leads to an environment of restraint (Wood 2018). This findings shows sheds light on the potential influence social dynamics inside prisons and remarks of the leadership. Further, the repertoire of sexualised torture, indicates systematic, prepared, and deliberate actions. Thus, this could be labelled as systematic and strategic use of sexual violence if sexual torture is regarded as sexual violence.

Although I have distinguished various types of sexual violence some of these types overlap or follow from each other. This is because the patterns of violence, group ideologies and

(27)

27 strategies explain what the causes and motives of sexual violence are. Conversely, it is likely that the initial causes and motives of the regime determine which patterns of violence they utilise. The graph below shows the direction of influence of the factors.

Causes and motives

Strategic Social practice Opportunistic

Patterns of sexual violence Group ideologies and

(28)

28

2.4 Research Design

This section outlines the design of the research. Starting with the methodology which explains certain choices. Following the justification of the cases. Subsequently, the procedure and tools used for the analysis and lastly, the operationalisation which states the sub-questions of this research.

2.4.1 Methodology

The aim of the research is to explore the differences in patterns of sexual violence between Syria and Sri Lanka perpetrated by government forces experienced by the victims. According to Gerring (2017), exploratory studies are especially suited for a qualitative research design. A small number of cases also fits this type of design (Gerring 2017). Taking the aim of my research into account - to explore the similarities and differences concerning sexualised violence - and the fact that I have only two cases, I decided a qualitative approach best serves the aim of my thesis.

I follow the definition of Kaarbo and Beasley who define “a case study to be a method of

obtaining a ‘case’ or a number of ‘cases’ through an empirical examination of a real-world phenomenon within its naturally occurring context, without directly manipulating either the phenomenon or the context” (Kaarbo & Beasley 1999: 372). I chose a comparative case study,

as it is most suitable to examine, explore and describe variation between the selected cases. This type of case study can be understood as “the systematic comparison of two or more data

points (‘cases’) obtained through use of the case study method” (ibid). Since many types of

case studies exist I will explain which type serves the aim of this research the best. As the theoretical framework serves as a guideline to interpret the data yet the focus remains on the cases a disciplined-configurative case study is most suitable (Eckstein 1975).

Because the primary aim of this study is not to examine the frequency of sexual violence but rather the differences in its patterns. However, frequencies do matter to a certain extent, as they reveal information about the prevalence. Therefore, I provide a numeric description of the data at the beginning of the analysis which helps to gain insights in the prevalence of specific patterns of violence and the targeted victims.

Secondary data was used for the analyses for several reasons. First, ethical considerations played a decisive role because of my lack of training how to properly interview a victim of sexual violence. Consequently, it was not ethical to use primary data. Second, within the

(29)

29 relatively short time frame, it is extremely challenging to build a solid relationship of trust with victims of sexual violence. Without this relationship, it is difficult to conduct interviews, which might influence the data negatively and lead to biased data. Third, gaining access to the research field would have been problematic, especially with the Syrian case, as the conflict is ongoing.

2.4.2 Case Selection

As variation of conflict-related sexual violence is a large field, the topic must be narrowed down. Therefore, two cases were selected: Syria and Sri Lanka. The former was chosen for several reasons. First, researchers’ lack of access to the country resulted in an understudied area of research regarding sexual violence. Moreover, most of the research on sexual violence was conducted in neighbouring countries (Chynoweth 2017), but not Syria itself. Therefore, an analysis of material directly from Syria would be another angle of research and hence it may give different insights. Second, the levels of sexual violence perpetrated by the Syrian regime forces has not received a substantial amount of attention in international politics and media until a few years ago, especially compared to the attention ISIS received on their horrific atrocities. However, based on recently published reports by human rights organisations and UN affiliated institutions, sexual violence perpetrated by government forces is systematic, widespread, and used as a weapon of war. Third, the conflict shows several factors, such as the conflict characterised of an ethnic element, a gender gap, and patriarchy, which could indicate a higher probability of sexual violence. Fourth, a considerable amount of research of wartime sexual violence has focussed on armed rebel groups, which applies for the Syrian conflict (Bitar 2015). This discrepancy between research on rebel groups and state forces makes researching Syrian security forces in relation to sexual violence interesting because this study can contribute to the variations of sexual violence perpetrated by regime forces. Hence, it could provide valuable insights of the prevailing patterns of violence, what role group ideologies play and how the violence is motivated.

Fifth, Syria is considered a collectivistic society, as it scores 35 on the individualism scale (Hofstede Culture Compass). This means that the collective needs matter more than individual needs and therefore every member of the group is responsible for the collective. Because there is joint responsibility for the group as whole, offence and shame touches the collective, that is, the family, extended family or community (Hofstede Culture Compass). Therefore, family honour could play a significant role in handling sexual violations. For instance, victims could

(30)

30 be subjected to ‘victim blaming’ and other forms of humiliation. Consequently, it is probable that both victims and their families will try to hide the sexual abuse. The aspect of how honour relates to violence in the Middle-East is explained in the following statement by an MSF coordinator on Lesbos on cases of sexual violence “proportionally we see many more cases

reported coming from Sub Saharan African communities vs. the Middle East and Central Asia. This can be explained culturally, as the stigma is heavier in more conservative communities”

(Awad 2017).

For the selection of a suitable comparative case it is important that the case permits to have variation on the variables. Therefore, Sri Lanka was chosen to compare with Syrian based on the differing character (Gerring 2006: 89). Both cases could be considered as outliers due to the high levels of sexual violence perpetrated by the government. The cases also have some similarities. First Sri Lanka has the exact same score of individualism, which indicates that Sri Lanka is also a collectivistic society. Second, Sri Lanka’s power distance is similar to Syria’s, meaning that in both societies power relations are important. Third, both conflicts have a sectarian character. Fourth, the population size was similar at the start of the Syrian conflict.

The cases differ in several respects too. First, Sri Lanka has a low score on masculinity in contrast to Syria. Second, the countries are located far apart from each other and have different religions. Third, Sri Lanka was a democracy at the time of the conflict whereas Syria was a stark autocracy. Fourth, the type of civil war differs between Syria and Sri Lanka. The former involves a “Type 3. Internationalised armed conflict” which is otherwise referred to as an IAC while the latter involves a “Type 4. Non-internationalised armed conflict” or an NIAC (Gersovitz & Kriger 2013: 163). As the names reveal, the primary difference concerns international involvement. Although, the aim of the research is not to establish a causal relationship the differing traits could be used for further research as the explanatory variable could point out variations of sexual violence.

2.4.3 Analyses

To compare the two cases a comparative case analysis was conducted based on the selected secondary data. The data analysed for Syria originates from the Women Under Siege reports and reports by International Federation of Human Rights (FDIH). For the Sri Lankan case, a Human Rights Watch and an International Truth and Justice report were used with testimonies of victims of sexual violence by government forces.

Referenties

GERELATEERDE DOCUMENTEN

where CAR is the cumulative abnormal return and the dependent variable in the regres- sion, POL is a dummy variable of the announcement of appointment of a former politi- cian to the

De huidige studie onderzoekt het verband tussen externaliserend probleemgedrag, positieve én negatieve kwaliteiten van de leerkracht-leerling relatie en het schools

General factor scores are estimated, via first-order factors (second- order model) or directly from the items (bifactor model), and domain- specific factor scores have a

Keeping Clark’s argument as a central theme, this paper explores the importance of physicality in the field of computer supported cooperative work (CSCW).

In the Rome Statute of the permanent International Criminal Court (ICC), sexual violence crimes are now recognized as crimes against humanity and war crimes, and they encompass

The measures proposed in Annex 2 relating to special measures for women, children and LGBT persons in criminal proceedings are recommended to be included in this future directive,

The LOO cross-validation results are reported in Table 2. In each iteration of the N ¼ N 0 þ N 1 LOO itera- tions, variables were selected based on p-values derived from the null

Forse daling inkomens Omdat de kosten minder dalen dan de bruto productiewaarde daalt de bruto toegevoegde waarde in 2009 met 10% tot 7,2 miljard euro.. Uit deze waarde moeten,