Do sacred incompatibilities explain the use of genocidal sexual violence? - A case study of ISIS’ targeting of the Yazidis

This paper seeks to offer a solution to a paradox existing in the literature regarding the genocidal intent of the use of sexual violence in armed conflicts. Indeed, according to data highlighted by Cohen (2013), genocidal intent does not hold as a cause of the use of sexual violence in armed conflicts at a macro level. However, since there have been numerous conflicts where the use of sexual violence was genocidal in its intent, there must be a more precise explanation that could account for this phenomenon. To do so, this paper focuses on an in-depth case study of ISIS’ targeting of Yazidi women using theory-building process tracing methods. I propose the argument that sacred incompatibilities, a concept derived from Hassner (2003) and Svensson (2007), offer a plausible cause for the use of genocidal sexual violence. The case study – selected using most likely case selection – supports the argument, though the theory should be tested in other case studies.


Introduction
"They told me the latest genocide, by ISIS, was the seventy-fourth.
There had been so much violence against Yazidis that they had a word for attempted extermination -ferman-long before its English equivalent, genocide" (Lamb, 2020, p19) In a groundbreaking paper based on a new dataset on sexual violence in armed conflict, Cohen (2013) found that there did not appear to be a correlation between genocidal intent and the use of sexual violence at the macro level.Her findings create a paradox: genocidal intent by itself may not be deterministic of the use of sexual violence.However, we can observe strong instances of genocidal intent in mass-scale events of sexual violence.Thus, there must be a more precise explanation that is not present in all conflicts with genocidal intent which explains the use of sexual violence.In this paper, I will propose a new theoretical argument based on Hassner's (2003) concept of sacred incompatibilities and seek to provide evidence for how it can explain my dependent variable, the use of genocidal sexual violence. 1In order to propose such a new argument, I will use methods from process tracing to test my theory on a most-likely scenario: the Islamic State's targeting of Yazidi women in Syria and Iraq from 2014 to today.Hence, my research question is: How do sacred incompatibilities explain the use of genocidal sexual violence?
In this paper, I propose a causal mechanism that could offer some evidence to support the idea that in some conflicts with genocidal intent, there is a more plausible deterministic factor of the presence of genocidal sexual violence than combatant socialization.My proposed causal mechanism is that ISIS' sacred incompatibility with the existence of the Yazidis leads to the dehumanization of the Yazidis.This dehumanization has as a consequence the rightor perceived dutyof ISIS members to possess and use Yazidi people as mere property; thus creating systematic institutions of genocidal sexual violence against the Yazidis.
1 Later, I will define more precisely all of the aforementioned concepts.

2.1
Understanding ISIS' targeting of the Yazidis The Yazidis are a predominantly Kurdish-speaking people living mostly in Iraq, but also in Syria, Turkey and Armenia (Açıkyıldız, 2010).The Yazidis are an ethnoreligious group, meaning that their religion -Yazidismis symbiotic with their ethnic identity (Açıkyıldız, 2010).Although they are sometimes considered a subgroup of the Kurds, their non-Abrahamic monotheist religion distinguishes them from other Kurds (Açıkyıldız, 2010).
As of 2023, the inhumane suffering the Yazidis endure at the hands of the Islamic State2 is still ongoing.Born from the Iraqi branch of Al-Qaeda, the Islamic State expanded quickly amidst the on-going civil war in Syria and into the West of Iraq.ISIS was born to the world when it declared itself a Caliphate at the end of June 2014 (Fishman, 2021).Within weeks, it reached the Sinjar district where most Iraqi Yazidis lived (Nicolaus and Yuce, 2017).While a majority of Yazidis managed to flee, those who did not met a cruel fate.
Men and women were immediately separated (Nicolaus and Yuce, 2017).The men that were too old were either sent away or shot, those of a younger age were forcibly converted or killed.If they converted, then they were either forced into slavery or brainwashed into becoming fighters.The same separation was made with the women, with all deemed to be aged less than around forty years old being forced into sexual slavery.In this paper, sexual violence against the Yazidis will be understood as the practice of mass rapes, sexual slavery, and forced marriages against women.That is not to invalidate other forms of sexual violence that ISIS has committed (against men, notably), but the practice of such extreme forms of sexual violence may be what differentiates the treatment of the Yazidis from other non-Sunni groups living within ISIS territory.
Indeed, the scale and frequency at which Yazidi women were targeted differ from that of women belonging to other groups.Most reports state that the enslavement of women (according to their age and whether or not they were pregnant at the time of capture) was systematic for the Yazidis (UN Human Rights Council, 2016;Lamb, 2020).Women from other groups also became victims of sexual violence from time to time, and their lives were precarious, but they did not face similar levels of systematic targeting (UN Human Rights Council, 2016).(Sharlach, 2000;Kaiser and Hagan, 2015;Temoney, 2016).
As mentioned, Cohen (2013) demonstrated the lack of correlation between genocidal intent and the use of sexual violence at the macro level.She argued that the main motivation behind the use of sexual violence in times of conflict must then be something else: according to her, combatant socialization.In the following section, I will hence propose another argument, namely that sacred incompatibilities can help explain why genocidal sexual violence occurs in cases with similar prerequisites.

Theoretical argument
In his 2003 paper, Hassner discussed the concept of sacred space and incompatibilities.
He argued that sacred incompatibilities can be defined with three central concepts: "First, integrity […] The parties must hold that the issue cannot be parceled out or subdivided without significantly diminishing its subjective value.
Second, boundaries.The parties must mean the same thing when they refer to the issue they are bargaining over.If there is no overlap between the issue boundaries as they perceive them then the issue is, at least in part, divisible.Finally, nonfungibility.The parties must believe that the issue cannot be substituted for or exchanged for something of equal value" (2003, p12-13).
While Hassner's sacred incompatibilities were limited to sacred space, Svensson (2007) built on it, removing the geographical aspect, to explore the impact of sacred incompatibilities on conflict resolution.He found that conflicts in which one party makes explicit religious claims are significantly less likely to be resolved through settlement.
ISIS' self-proclaimed objective has always been to establish a universal caliphate where all would live under the rules of their version of Islam (Bozorgmehri, 2019).Here, I will argue that ISIS' core principle of pan-Islamism creates a sacred incompatibility with non-Abrahamic religious groups existing within ISIS' reach, such as the Yazidis.While one may rightly argue that ISIS' universalist objectives were never realistic, it does not actually hinder the extent to which they attempted to impose it on the people who lived within the borders of the caliphate.
Secondly, the issue is indivisible and unsubstitutable because for non-Abrahamic religious groups, complying to these objectives would quite simply mean eradication, as the ISIS' Islamic rules do not allow non-Abrahamic religions to be tolerated (Raben, 2018).Hence, ISIS' ambitions meet the three criteria of sacred incompatibilities as defined by Hassner.This definition is my independent variable.It can be understood as a part of Wood's concept of 'sexual violence as instrumental for the group' (Wood, 2008) which is a loosely defined notion of sexual violence being an outright goal of the attacker.
My hypothesis is thus the following: H1: Sacred incompatibility is used as a justification for the use of genocidal sexual violence.
Hence, the null hypothesis, if we were to think counterfactually, should be: H0: Alternative explanations are used instead of sacred incompatibility as the justification of genocidal sexual violence.
My proposed causal mechanism is that ISIS' sacred incompatibility with the existence of the Yazidis leads to t he dehumanization of the Yazidis.This dehumanization has as a consequence the rightor perceived dutyof ISIS members to possess and use Yazidi people as mere property; thus creating systematic institutions of genocidal sexual violence against the Yazidis.The scope condition is that the population of cases should only be armed conflicts in which at least one party holds a sacred incompatibility with another party on the basis of the mere existence of that other party.My assumptions are that this party has sufficient resources to attack, and that attacking does not cause short-term costs that are higher than the benefits of the attack.More importantly, I also assume that not all groups with genocidal intent hold sacred incompatibilities with the groups they target, because Cohen (2013) has already shown that it may not hold at the macro-level.

Research Design
In this paper, I will use theory-centric process-tracing methodology to test my proposed causal mechanism.As Beach and Pedersen put it, "the goal is to evaluate whether evidence shows that the hypothesized causal mechanism linking X and Y was present and that it functioned as theorized" (2019, p11).The reasons for using theory-building process-tracing is that we know the outcomegenocidal sexual violencebut not the cause (Centre for Development Impact, 2015).Now, while sacred incompatibilities may be seductive as a concept, they appear to be limited by that same conceptuality.That is to say that they areunlike sacred spacescompletely intangible, meaning they cannot be seen or touched.They also do not represent the outcome of a strategy, unlike Cohen's combatant socialization theory (2013).Indeed, theoretically, you could measure the level of combatant socialization before and after the use of sexual violence and observe a potential difference.
The same cannot be done for sacred incompatibilities.Sacred incompatibilities are in a Manichean state of being: they either are, or aren't.That is not to say that they could not develop and appear during a conflict, but it does remain a binary variable.An incompatibility either matches Hassner's (2003) three criteria, or it is resolvable through negotiation.
Yet, sacred incompatibilities are measurable.Indeed, they may be binary, but they are also observable in the behaviors, speech, and knowledge production of the Islamic State.They can be measured by using qualitative analytical tools.
Using process tracing, I will look for pattern and account evidence of structures and and testimonies of victims published in newspapers, books, and academic articles.This operationalization should make this paper's hypothesis easily falsifiable.
The internal validity of this paper is complicated, because there is no way of efficiently controlling for confounding variables apart from systematically dissociating them from sacred incompatibilities when looking at evidence.The external validity should however be quite strong as the scope conditions of my theoretical argument are quite demanding and therefore limiting: the few cases that meet the scope conditions are likely to show similar mechanisms as this case (see Sharlach, 2000;Kaiser and Hagan, 2015;Temoney, 2016).
Process tracing methods are usually utilised with typical case studies (Bennett and Checkel, 2015).However, since the purpose of this paper is to attempt to challenge the findings of Wood on the lack of correlation between genocidal intent and the use of sexual violence, a most-likely scenario allows to explore whether alternative causal mechanisms can exist at all or not.If no causal mechanism holds in a most-likely scenario, then Wood's findings will be strengthened.If a proposed causal argument proves useful in explaining the use of sexual violence as a consequence of genocidal intent, then more research will be needed to figure out whether such a result would hold in less likely cases.Only if the causal argument holds in such cases could it be useful to explain further cases.
One key limitation to my sources is a language barrier, as a lot of that content production is done in Arabic or Kurdish, which I do not speak.I will thus have to rely on translations and work done by other academics and journalists.Besides, since Syria and Iraq are some of the least accessible places in the world for academics, particularly when it comes to the Islamic State, it should be noted that we are facing a severe lack of data in multiple areas.For instance, we still lack a definite number of Yazidis that remain under ISIS control to this day (Lamb, 2020).
Another limitation is that my qualitative analysis may very well describe patterns and behaviors that do exist within the Islamic State but at the same time, cannot empirically quantify the extent to which those are the practical norm.Just because sexual violence is official policy does not mean that it is practiced, and vice-versa (Wood, 2018).Theoretically, we could thus see sacred incompatibilities as the motivation of ISIS leadership without the sacred incompatibilities being what pushes ISIS soldiers to commit sexual violence against the Yazidis particularly.Because of that, there are reasons to believe that this paper's reliability may be somewhat limited.

Analysis
I am going to analyze each causal argument in its own right.Firstly, the existence of a sacred incompatibility.Secondly, how it leads to the dehumanization of the Yazidis.Thirdly, how that justifies the use of sexual violence against them.Finally, I will outline how that justification at a macro scale has led to the institutionalization of genocidal sexual violence.
The sacred incompatibility that ISIS harbours with the Yazidis first stems from their religious differences; or rather, from the impossibility for ISIS to tolerate Yazidism, one the main religion of the Yazidis (UN Human Rights Council, 2016).This impossibility comes from several subreasons: Firstly, they consider the Yazidis, wrongly, to be polytheists.For the Islamic State, guarantees should be given to the Jews and Christiansthe dhimmi, the People of the Bookwho get to pay the Jizyathe religious taxin exchange for their right to live under the Caliphate.However, polytheists are not given such guarantees (Filipec, 2020).This depiction as polytheists was premeditated by ISIS theologians as it was made prior to the attacks on the Sinjar region (Callimachi, 2015).
Secondly, the Yazidis are seen and portrayed as "devil worshippers" (Callimachi, 2015;Lamb, 2020, p22).This categorization goes further than simply removing the possibility of tolerance, as it defines the whole of Yazidism as inherently hostile to Islam.This possibly explains why the Yazidis continued to be treated so inhumanely even after most of them were forced to convert to Islam by their torturers.However, the more probable reason for why the violence continued is that according to ISIS' reading of the Shari'a, conversion does not affect the slave's legal status (Nicolaus and Yuce, 2017).
Another smaller reason is that ISIS is inherently opposed to the existence of any form of non-Sunni community, whether religious, ethnic, or national (Ahram, 2015).This means that besides their religion, the existence of the Yazidis as members of an ethnic community is in itself intolerable.This may be yet another reason for the continued abuse of women forced to convert, as even after their conversion, they remain part of a group that ISIS rejects and cannot control.
Altogether, ISIS has found the existence of the Yazidis to be incompatible with their dreams of radical fundamentalist readings of the Shari'a.On the basis of their sacred incompatibility, they have successfully created a vast program of dehumanization of the Yazidis.
The most basic form of ISIS' dehumanization of the Yazidis is probably best understood with the word they attributed to the women in captivity: Sabaya, slave.Multitudes of women report being categorized with it (Callimachi, 2015;Lamb, 2020).Sabaya is not just a slurit represents the mark of the loss of the Yazidi women's humanity as it is systematically added to their names, sometimes even replacing them.On photographs, for example, taken for the purpose of human trafficking it is written next to the pictures: "Sabaya n.1, Sabaya n.2" (Callimachi, 2015).
The renaming of Yazidi women is already dehumanizing, and is exacerbated by the atrocious conditions that they were put in.After being separated on the basis of their gender and age, they were taken to processing centers.There, they sometimes stayed for days with little food and drinking water, and often no access to bathrooms or any form of hygiene (Lamb, 2020).In those processing centers, they were systematically groped, undressed, and touched.
Women and girls as young as six or seven (Lamb, 2020) were either given to high-ranking ISIS members or sold at auctions on slave markets (Revkin and Wood, 2021).These auctions were extremely bureaucratic: contracts of ownership were written and said girls were given to ISIS members as their property (Callimachi, 2015).ISIS fixed official pricesalthough there is doubt on the extent to which those were respectedbased on their ages: the younger the girl, the higher the price (Besenyõ, 2016).
Repeated rapes consolidated the dehumanization of Yazidi women.As Nadia Murad, Nobel Peace Prize winner and survivor points out: "The rape was the worst part.It stripped us of our humanity and made thinking about the futurereturning to Yazidi society, marrying, having children, being happy -impossible.We wished they would kill us instead" (Murad, 2017).What Murad describes is largely confirmed by mental health professionals that have been attending to survivors who managed to escape ISIS (Kizilhan, Steger and Noll-Hussong, 2020).Yazidi women going through this were denied basic rights and worse, were denied ownership over themselves, with their bodies having become the property of their owners.
Indeed, their bodies now belonged to ISIS, and ISIS gave itself the right, or the "duty", to abuse them.There are multiple reports of victims testifying that their rapists were praying before and after abusing them; there are also reports of the rapists explaining to the enslaved women that their rape is a religious duty, a way of worshipping (Callimachi, 2015).They gave themselves the right to do so on the basis of the sacred incompatibility, that the Yazidis were kafirah, infidels or mushrik, polytheists.ISIS members actually have a "duty" to have sex with their slaves, or else they must dispose of them, according to a question-and-answers published by ISIS' Research and Fatwa Department (Ali, 2016, pp. 53-54).
Yazidis who had been enslaved were thus punishable at the will of the ISIS members.
ISIS itself recommended in the same Q&A that enslaved Yazidi women be beaten if their owner found a reason to, even if they were underage girls (Ali, 2016).Multiple survivors described how they were often, and severely, beaten for any reason at all.Some for instance described how they were beaten not just by the ISIS member but by members of their families, or soldiers' wives who could get jealous of their husband's raping of the enslaved woman (Murad and Krajeski, 2017;Lamb, 2020).
Besides beating the Yazidi women, ISIS aimed to destroy their bodies and their spirits by raping them repeatedly.As soon as the women would arrive at the processing centers they became subjected to the possibility of being raped by any man at any time (Murad and Krajeski, 2017).From that point on, the threat of rape was omnipresent.Yazidi women were groped at auctions, raped on a daily basis by their captors who ensured that they could continue to do so by forcing them to use contraceptives (Callimachi, 2016).
The sexual violence to which Yazidi women were subjected was clearly institutionalized.And the source of this institutionalization was publications and directives emanating from the ISIS leadership on the basis of the sacred incompatibility and a voluntarily distorted reading of Islamic religious texts (Ali, 2016;Mirza, 2017).Indeed, ISIS scholars justified their right to put women into slavery even before attacking Sinjar by using medieval hadiths seemingly allowing the practice while willingly omitting the passages within those hadiths that refuted it (Mirza, 2017).Hadiths are "prophetic reports", or events and discourse attributed to Muhammad and passed on through generations (Brown, 2018).
ISIS institutionalized the right for fighters to sell, exchange or give Yazidi women at their own will, while setting conditions like the illegality to sell a Yazidi woman back to her own family (Revkin and Wood, 2021).While women were given contraceptivesor provided with abortionsit was not just because the fighters needed to continue to be able to rape them.
It is because according to the religious justification that they had published, a slave who would bear the child of an ISIS member could no longer be sold, and both the slave and the child would become free upon the death of the captor (Nicolaus and Yuce, 2017).The rules on how to treat Yazidi women were stated in a published document named "Questions and Answers on Taking Captives and Slaves".For instance, it established the legality of the rape of prepubescent Yazidi girls (Nicolaus and Yuce, 2017).
What the institutionalization of sexual violence also did was to ostracize Yazidi women from their communities.At least before ISIS attacks in 2014, inter-religious marriages between Yazidis and other groups were more than frowned upon.Women who engaged in interreligious relationships were ostracized if not attacked.Several survivors recalled the same story about a young woman who had been lynched after she had gotten close with a young Muslim man (Lamb, 2020).ISIS knew of these circumstances and used said information to forcibly ensure that Yazidi women would not return to their communities (Murad and Krajeski, 2017).
By systematically removing the women, ISIS underlined their intention to destroy the Yazidi as a people.Because ISIS decided that this should be their policy before their first attacks on Sinjar and the Yazidi community, I have reasons to believe that the temporality of the theory is respected.It is not that they used religious bases as a justification for their crimes a-posteriori but rather created a predefinite plan to systematically abuse and destroy the Yazidi as a people.

Discussion
While I have presented evidence in support of my hypothesis from a variety of sources, there are important alternative explanations that need to be discussed because they may directly hinder the strength of my argument.As I mentioned earlier, Cohen (2013) argued that combatant socialization was a credible explanation.It is entirely possible that this strategy has been used by the Islamic State as a way to build up morale, reward its fighters, or increase recruitment (Revkin and Wood, 2021).The occurrence of gang rapes, for example, was described as such.However, gang rapes were only tolerated, not officially sanctioned, as one is not allowed to sleep with a slave that does not belong to them (Revkin and Wood, 2021).
While it is sometimes difficult to distinguish between a tolerated practice and policy, the practice of gang rapes does not appear to be premeditated.It is entirely plausible that Cohen's theory on sexual violence as a strategy for combatant socialization can coexist with my proposed theory of sacred incompatibilities.It would possibly explain why we have seen non-Yazidi women also be targeted for rapes and forced marriages (Besenyõ, 2016;Revkin and Wood, 2021).However, it does not suffice in explaining why the Yazidis were so particularly targeted to a much larger extent than any other community.If the use of sexual violence was mostly explained by group socialization, then we should not observe specific targeting of some groups rather than others.
An alternative theory that could explain this difference is opportunity: the simple idea that no one cared enough to protect the Yazidis, and that other groups would have met the same targeting if they had not been backed by stronger actors (Wood, 2008).Indeed, as soon as ISIS approached Sinjar, Peshmergas withdrew from the region, leaving the Yazidis defenceless (Nicolaus and Yuce, 2017).Other communities such as the Druze in Syria and the Turkmen in Iraq found support in respectively Israel and the Iraqi government (Filipec, 2020).One may think that this difference may explain why only the Yazidis were targeted to such an extent.But other groups, such as Christian communities, similarly had little protection.In 2014, ISIS also murdered hundreds of Shia Muslims while it was expanding into Iraq (Filipec, 2020).Yet, again, the extent to which other unprotected groups found themselves to be systematically enslaved, even when ISIS had the capacity to do it, does not match that of the Yazidis.
Therefore, it is certain that the lack of protection of the Yazidi community largely simplified ISIS' plans, but it does not prove that the basis for ISIS' treatment of the Yazidis was simply because no one dared to stop them.
Another possibility is incentives: that the slave trade was an important source of income of the Islamic State (Wood, 2008;Besenyõ, 2016).In that logic, it would make sense for the Islamic State to kidnap and sell women, either to their own fighters, abroad, or back to their own families.There is some strength to this argument as we have seen that Yazidi families often paid tens of thousands of US dollars for the liberation of their daughters (Lamb, 2020).
However, in a report, the Financial Action Task Force stated that the income of those sales was unlikely to ever reach a level at which it would constitute a major source of income for ISIS (Financial Action Task Force, 2015).Besides, if it did, then we would again presumably see ISIS enslave and sell more women from other communities.Furthermore, ISIS cannot have designed its enslavement of Yazidi women on the premise of being able to sell them back to their families for a ransom as they outlawed it formally and punished members who were found to have breached the rule (Filipec, 2020).
ISIS' treatment of the Yazidis cannot be explained by more classical theories of violence against civilians such as loyalty to the other party (Schwartz and Straus, 2018) because there is no evidence that ISIS perceived the Yazidis to be part of the conflict or supporting a side in it.
It also did not provide evidence of a logic of imposing support to the civilian population, and the violence was far from being indiscriminate.Furthermore, it did not consist of sexual violence as a practice (Wood, 2018) but rather as an institution.
Therefore, it seems like my proposed hypothesis is the most suited to explain why the Yazidis were particularly targeted by the Islamic State.It respects temporality, can partially control for and is not invalidated by confounding variables, and proposes a logical causal mechanism.

Conclusion
In this paper, I proposed a new conceptualisation of sacred incompatibilities as a cause of ISIS' use of genocidal sexual violence against the Yazidis.I argued that on the basis of this incompatibility, ISIS systematically dehumanized Yazidis and then treated them accordingly, alongside the institutionalization of sexual violence with the purpose of eradicating the Yazidis from the territory controlled by the Islamic State.I found supporting evidence in the official publications of the Islamic State, declarations made by its members, and testimonies of Yazidi survivors.Alternative explanations lacked in their capacity to explain why the Yazidis were targeted so specifically, even though they could explain a part of why sexual violence was so prevalent.
However, the purpose of this paper was only to find whether a causal mechanism could possibly explain why genocidal intent would lead to the use of sexual violence.It proved so with a most-likely case scenario.In order to confirm the usefulness of the theoretical argument, further research should be conducted which utilises this causal argument in less likely cases.
behaviors.Empirically, it should look different depending on the origin of the source, whether an official publication, the declaration of a high-ranking officer, or the testimony of a low-level member.For official publications, it should consist of systematically dehumanizing speech targeting the Yazidis on the basis of their religion and justifying the use of sexual violence on the basis of it.For ISIS members, it should look like religious justification made to Yazidi women to legitimate the use of sexual violence against them.In practice, this means analyzing their official publications like their journals: Dabiq, Dar al-Islam, Rumiyah; other published content -notably published by ISIS' Research and Fatwa Department; interviews of fighters; There is a growing consensus around defining what the Yazidis are going through as genocide, defined as "acts committed with intent to destroy, in whole or in part, a national,