The Yale Law Journal

February 2017

Prosecuting Gender-Based Persecution: The Islamic State at the ICC

Emily Chertoff

abstract.Reports suggest that Islamic State, the terrorist “caliphate,” has enslaved and brutalized thousands of women from the Yazidi ethnic minority of Syria and Northern Iraq. International criminal law has a name for what Islamic State has done to these women: gender-based persecution. This crime, which appears in the Rome Statute of the International Criminal Court (ICC), has only been charged once, and unsuccessfully, in the Court’s two decades of existence. The case of the Yazidi women presents a promising opportunity to charge it again—and, potentially, to shift the lately unpromising trajectory of the Court, which has been weakened in recent months by a wave of defections by former member states.

This Note uses heretofore unexamined jurisprudence of the ICC’s Pre-Trial Chamber to elaborate—element by element—how the Prosecutor of the Court could charge gender-based persecution against members of Islamic State. I argue that the prosecution of Islamic State would not just vindicate the rights of Yazidi survivors of Islamic State violence. It would help to consolidate an international norm against gender-based persecution in armed conflict—a norm that, until now, international law has only incompletely realized. This Note argues that only by prosecuting the crime of gender-based persecution can international criminal law cognize violence, like the attacks on Yazidi women, that is motivated not just by race, ethnicity, or gender, but by the victims’ intersecting gender and ethnic or racial identities. I conclude by reflecting on the role that a series of prosecutions against perpetrators of gender-based persecution might have in restoring the legitimacy of the ailing ICC.

author. Yale Law School, J.D. expected 2017. The author would like to thank Harold Koh for his thoughtful suggestions and support throughout the drafting process, and Oona Hathaway and Ali Miller for formative insights and conversations. Megan McGlynn provided extremely helpful and professional editorial feedback, and I thank her as well as Hilary Ledwell, Urja Mittal, and the editors of the Yale Law Journal. I am grateful to friends and colleagues who read early drafts of this piece and offered their comments and encouragement, among them Zak Manfredi, Hava Mirell, and Noah Rosenblum, and to my parents for their support. This Note is dedicated to Yazidi survivors of violence, especially the estimated 2,000 Yazidi women who remained in Islamic State captivity as of this August.


Over the past two years, media reports from Syria and Iraq have documented how the terrorist group Islamic State has systematically enslaved and raped women and girls from the Yazidi ethnic minority group.1 One emblematic New York Times report details how Islamic State fighters have forced Yazidi women to take birth control in order to avoid violating a supposed religious mandate to abstain from sex with pregnant women slaves.2 Women who have escaped this torture are fighting back. Some have spoken out against their captors in the press, to human rights observers, and at the United Nations (UN).3 A group of them has even taken up arms against their former tormenters: a unit of women fighters based in Iraqi Kurdistan is training to take on members of Islamic State alongside the Kurdish peshmerga.4 “We want justice,” the commander of this battalion told a reporter.5 “We want the men who did this to go to court.”6 Yet despite the public outcry and sustained advocacy efforts, justice has remained out of reach.

Of particular concern to many outside observers is the apparently premeditated nature of the Yazidi women’s enslavement and the sexist ideology—evident in Islamic State pamphlets and articles—that underlies it.7 France’s Minister for Family, Children, and Women’s Rights echoed other commentators when she delivered a speech to the United Nations proposing a new international crime—“femicide”8—and insisting that countries take steps to prosecute Islamic State at the International Criminal Court (ICC).9 Unfortunately, there has never been a successful international trial for gender-based persecution, despite the long and tragic history of gender-based targeting in conflict. Indeed, persecutors in several recent conflicts have singled out women and girls,10 and sometimes men and boys,11 for vicious, differential treatment.

This Note argues that Islamic State can be charged under the “gender-based persecution” provision of Article 7(1)(h) of the Rome Statute, the multilateral treaty that codifies the ICC’s jurisdiction, substantive crimes, and major procedures.12 The provision—a triumph, at the time, for feminist advocacy groups13—has unfortunately failed to realize its full potential. Since the enactment of the Rome Statute, gender-based persecution has been charged only once, in a case that failed to move forward for evidentiary reasons.14

This Note fills a gap both in the literature on potential prosecutions of members of Islamic State and in the broader body of scholarship on gender-based crimes. I make two descriptive contributions and one normative contribution to the existing literature. On the descriptive front, this is the first piece to address the possibility of a gender-crimes prosecution against Islamic State. In addition, this Note is the first to analyze the persecution-related jurisprudence of the Pre-Trial Chamber of the ICC and to use this jurisprudence to assess the feasibility of bringing a charge of gender-based persecution in a particular case. Other scholars have tried to give Article 7(1)(h) meaning by examining the statute’s drafting history,15 analyzing the events of its drafting conference from the perspective of advocates,16 and analogizing to refugee law.17 But thus far, none has examined the Pre-Trial Chamber’s jurisprudence in this context. On the normative front, this Note makes novel use of the concept of intersectionality18 to analyze why international courts have failed to charge and develop the crime of gender-based persecution, a lapse that may be partly responsible for the failure to create an international norm against such persecution.

The Note proceeds in four Parts. Part I details what is known about Islamic State’s crimes against Yazidi women by drawing on evidence unearthed by journalists, advocacy groups, and others. Part II addresses why only a charge of gender-based persecution can address the harm in crimes of this type—not just in response to the atrocities against Yazidis, but in all cases of gender-based targeting of civilians. While aspects of gender-based persecution could be prosecuted as other crimes, gender-based persecution should constitute a separate international crime for the same reasons that genocide does: where evidence indicates that a group has committed widespread criminal acts with a specific intent to target a distinct social group, we consider those crimes distinctly blameworthy. Part II develops this intuition and its doctrinal implications. Pursuing members of Islamic State for violence against Yazidi women—or members of any other group engaged in gender-based persecution—would express an overall international commitment to prevent or respond to gender-based atrocities.

Part III elaborates how a prosecution of Islamic State members for gender-based persecution could proceed. I assess the elements of the crime of gender-based persecution. To help illustrate how the ICC might apply the gender-based persecution statute, this Note makes use of the jurisprudence of the ICC’s Pre-Trial Chamber, a judicial panel that decides the viability of indictments for the crimes of ethnic and political persecution. This Part subsequently explains how the alleged crimes against Yazidi women would satisfy the definition of gender-based persecution. In Part IV, I suggest that a series of prosecutions for gender-based persecution, in this case and in others, could help to bolster the legitimacy of the ICC at a critical moment in the institution’s troubled history. Undertaking a broadly popular and clearly justified prosecution against a nonstate actor rather than a state could give the ICC an opportunity to pivot from recent unpopular prosecutions to a legitimate new course.

i. evidence of gender-based crimes against yazidi women

Islamic State is a nonstate militant group that started as a little-known offshoot of the terrorist group al-Qaeda.19 Over the past five years, Islamic State has taken control of broad swathes of territory in Syria and Iraq as well as smaller areas in Libya and Afghanistan.20 It gained widespread notoriety in the West in 2014 for executing captured journalists and disseminating graphic videos of their murders.21 Since 2014, Islamic State has also taken responsibility for and inspired a number of terror attacks in North Africa, Europe, and the United States.22

As a matter of internal structure, evidence suggests that Islamic State has all of the characteristics of a de facto governmental authority: it provides services, collects taxes, manages local resources, and engages in extensive recordkeeping to fulfill its governance responsibilities and combat objectives.23 Under the ultimate control of Abu Bakr al-Baghdadi, the self-declared “caliph” of Islamic State,24 Islamic State’s bureaucracy consists of several governmental units, including a cabinet in charge of a number of ministries and regional commands.25 A “governor” who reports to Al-Baghdadi directs each of the regional commands.26 Centralized under the supervision of the cabinet members in Raqqa are Islamic State’s diwan, or departments.27 Those bureaucracies coordinate Islamic State’s affairs in areas such as education, health, public security, tribal outreach, and public relations.28 A cache of Islamic State documents translated by the terrorism researcher Aymenn al-Tamimi reveals administrative documents regulating matters from maternity care pricing and the proper format for Islamic State birth certificates to the examination schedule at Mosul’s pharmacology college.29 This same collection contains documents threatening Christians and a fatwa (or religious edict)introducing the death penalty for blasphemy.30

Islamic State operates “war spoils” departments to handle captured resources and slaves. One war spoils department coordinates the use of natural resources like oil; another coordinates the sale and movement of slaves.31 This department of slaves is the organ of Islamic State known to be directly responsible for the treatment of women captives,32 along with Islamic State’s Research and Fatwa Department, the body responsible for issuing religious edicts, including those that deal with women slaves.33

The Islamic State campaign has appeared to focus particularly on persecuting Yazidis, a religious minority of 300,000 to 700,000 people centered primarily in Northern Iraq.34 Islamic State’s particular brutality toward Yazidis is based in part on its view of their religion. While Islamic State ideology affords minimal protections to members of monotheistic religions, the Yazidis’ Zororastrian-derived form of polytheism affords them no such defense against Islamic State violence. Thousands of Yazidi civilians have been driven from their villages, killed, or enslaved by Islamic State.35 Most infamously, in 2014, tens of thousands of Yazidis were trapped for weeks by Islamic State fighters on Mount Sinjar,36 a traditional home of Yazidis and the location of many of their holy sites.37 Islamic State persecution of Yazidis comes in the context of a systematic military campaign in which the group has captured Mosul, the second-largest city in Iraq, as well as dozens of other towns and cities.38 The Office of the UN High Commissioner for Human Rights estimates that Islamic State-related violence in Iraq alone killed 3,855 people in a six-month period in 2015.39 And on March 17, 2016, U.S. Secretary of State John Kerry declared that Islamic State attacks on the Yazidi people and other ethnic minorities constituted genocide.40

The most horrifying aspect of Islamic State’s treatment of Yazidis is its systematic sexual and domestic enslavement of thousands of Yazidi women. Survivors, many of whom were captured during the attack on Mount Sinjar, have told journalists that they were bussed to the Islamic State-held city of Mosul and other Iraqi towns and then held in large buildings, including a wedding hall and a prison, with hundreds of other captives.41 “It was 100 percent preplanned,” one Yazidi activist told the New York Times. “I spoke by telephone to the first family who arrived at the Directory of Youth [one of the holding sites] in Mosul, and the hall was already prepared for them. They had mattresses, plates and utensils, food and water for hundreds of people.”42 Islamic State fighters took some Yazidi women directly from these holding sites; others were moved repeatedly from town to town until they had been auctioned off or sold.43

Journalists and other observers have reported on the treatment of Yazidi women captives. These captives have been raped repeatedly, subjected to sexual torture, forced to perform household labor in their capacity as “slaves” or “captives,” bought and sold like chattel,44 denied access to medical care, and gruesomely and publicly executed for failing to comply with their captors’ sexual demands.45 In an interview with the Middle East Eye, UN Special Representative on Sexual Violence in Conflict Zainab Bangura told a reporter:

After attacking a village, [Islamic State] splits women from men and executes boys and men aged 14 and over. The women and mothers are separated; girls are stripped naked, tested for virginity and examined for breast size and prettiness. The youngest, and those considered the prettiest virgins fetch higher prices and are sent to Raqqa, the IS stronghold . . . . At slave auctions, buyers haggle fiercely, driving down prices by disparaging girls as flat-chested or unattractive . . . . We heard about one girl who was traded 22 times, and another, who had escaped, told us that the sheikh who had captured her wrote his name on the back of her hand to show that she was his “property.”46

There is written evidence that Islamic State not only condones, but also actively encourages, the rape, torture, and enslavement of Yazidi women captives. Indeed, the central Islamic State bureaucracy has issued multiple fatwas dealing with the treatment of women slaves.47 A manual on women captives issued at the end of 2014 answers a number of questions about relations between Islamic State members and women slaves. It states that sex with a woman slave is permissible (quoting as support a passage from the Qu’ran48), but if the slave is not a virgin, “her uterus must be purified” beforehand to ensure that she is not pregnant.49 Sale of women captives, too, is permitted, “for they are merely property.”50 The manual also lays out various rules governing, among other things, whose slaves a fighter may have sex with (only one’s own, with co-owned slaves off-limits), and the appropriate forms of beating (“disciplinary beatings” are acceptable, but “breaking beatings,” “beating[s] for the purpose of achieving gratification,” and “torture beating[s]” are not).51

The Research and Fatwa Department, and the Islamic State bureaucracy more generally, purport to ground their rulings on the treatment of women slaves in dictates of Islamic law.52 An article published in Islamic State’s magazine Dabiq states, after citing various religious texts, that “enslaving the families of the kuffār [infidels] and taking their women as concubines is a firmly established aspect of the [Shariah] that if one were to deny or mock, he would be . . . apostatizing from Islam.”53 A New York Times journalist who interviewed escaped Yazidi survivors reported that fighters prayed before and after raping the women,54 and that the fighters referred to the assaults as “ibadah,” a term from Islamic scripture meaning worship.55

The shocking treatment of Yazidi women under Islamic State has drawn international condemnation and has naturally led some commentators to examine the prospect of criminal accountability.56 But thus far, no commentator has offered a comprehensive analysis of how a prosecution would grapple with the gendered nature of the atrocities. I argue that an appropriate accounting for these crimes is possible only if the international legal regime recognizes the gendered nature of these acts of violence—an argument I take up in the following Part.

ii. gender-based persecution and the expressive value of international law

Not only has Islamic State targeted the Yazidis as a group in general; it has also singled out Yazidi women for a particularly prolonged and brutal form of persecution. As this Part explains, we would expect members of Islamic State to experience legal condemnation at the international level given the scope and gravity of their crimes against Yazidis. We would also expect criminally punished Islamic State members to face a regime of justice that recognizes the motivations behind their targeted treatment of Yazidi women. That is, to the extent criminal law takes into account biased intent—as it does in hate-crime proceedings at the national level, and in genocide and persecution proceedings at the international level—we would expect it to recognize the maltreatment directed at the Yazidi women as women. Yet, while international criminal law theoretically recognizes a charge of gender-based persecution, in practice this crime has almost never been charged, and never once successfully.

This Part contends that charging the international crime of gender-based persecution against members of Islamic State should be a top priority. The moral harm that results from targeting on the basis of gender is akin to the moral harm that stems from targeting on the basis of ethnicity and ought to be recognized as an atrocity of comparable severity. While infrequent, where gender-based persecution does take place, it is particularly shocking. The bureaucratized slavery of Yazidi women is only the most recent example of gender-based persecution. It was the sex-selective execution of men and boys at Srebrenica in 1995, for instance, that finally galvanized members of the North Atlantic Treaty Organization (NATO) to intervene in the Bosnian Genocide.57

First, I argue that Islamic State’s persecution of Yazidi women should be prosecuted in an international criminal trial both because of the magnitude and gravity of the crimes alleged and because developing a strong international norm against gender-based persecution is critically important. The especially serious harms at the intersection of ethnicity and gender deserve increased recognition in international criminal law, a recognition that should begin with the Yazidi victims and survivors of Islamic State.

Next, I argue that gender-based persecution is a composite crime structurally akin to genocide and that it deserves recognition for the same expressive reasons that genocide itself was first recognized as a crime. Like genocide, gender-based persecution is composed of a series of acts—in this case, torture, rape, and slavery—that could each be charged as separate domestic or international crimes. The reason prosecutors should be allowed to charge these acts as a single composite crime is that the whole—the intent to terrorize a particular group—is more morally blameworthy and damaging than the sum of its parts.58 Moreover, recognizing gender-based persecution as a crime would serve similar purposes as recognizing the crime of genocide: it would allow international criminal law to fulfill its expressive function, and, in turn, to help crystallize a new international human rights norm against gender-based atrocities.59

A. International Criminal Accountability

The ICC, the world’s permanent criminal court, was established in 1998 through the ratification of the Rome Statute.60 Though atrocities in Rwanda and Yugoslavia in the early 1990s most immediately spurred its creation, the idea of such a court has existed in the international arena since the aftermath of World War II and the Holocaust, when the Allied powers tried and, in some cases, executed Nazi and Imperial Japanese war criminals in trials at Nuremberg and Tokyo.61 International policymakers and lawyers recognized after the Nuremberg trials that certain acts—because of their scale, their severity, and, in some cases, their motivation—were so heinous that they aptly could be called crimes against the whole of mankind and the international order.62 Nazi Germany’s unprovoked and unjustified acts of war against other countries were one such crime;63 its large-scale ethnic cleansing of Jews, Roma, and other minority groups was another.64

The 160 states65 that participated in drafting the Rome Statute had this same idea in mind: to prosecute the worst war criminals, persecutors, and tyrants on the reasoning that some acts are so heinous that they damage us all;66 they cross-cut national boundaries, giving multiple societies claims to see justice done. The severity of the crimes makes a mockery of basic principles of humane treatment. Moreover, the perpetrators, often governments, are in many cases effectively above national law and, without intervention, might never face justice.67

Though commentators continue to debate the issue, Islamic State’s abuses of Yazidi women are appropriate subjects for international criminal accountability, and thus for trial before the ICC.68 To be sure, trying these crimes in a national court would offer some advantages over international proceedings, assuming the existence of a functioning judiciary in the state conducting the trial. Most obviously, international justice can be slow and its processes less well established. Certainly the United States, Britain, or another Western country could more quickly prosecute the accused. Moreover, a domestic prosecution would likely face fewer jurisdictional challenges than an ICC trial would.69

On balance, however, the advantages of an international trial over a domestic one in the case of Islamic State outweigh the disadvantages. Because Islamic State operates across multiple states70 and claims a membership of diverse nationalities,71 it is difficult to pinpoint a single appropriate national forum. One country might have trouble establishing jurisdiction over a substantial number of Islamic State members. Thus, while there may be dozens of countries with jurisdiction to prosecute their own nationals, fragmenting the prosecutions across many countries would impede the gathering and airing of a single body of evidence on Islamic State crimes. The ICC, on the other hand, is well suited to this task. Moreover, even if jurisdiction in a single national forum were legally feasible, a prosecution in the United States, United Kingdom, or other Western court would seem inappropriate given the relatively tenuous connection of Western states to events on the ground in Iraq and Syria.72 Although these countries have highly developed legal systems, this fact alone may not entitle them to prosecute international crimes with a highly attenuated domestic connection when there is an appropriate international forum.

But most importantly, international criminal liability is not only justified but critical in light of the severity and magnitude of Islamic State’s crimes.73 The United States recently acknowledged that Islamic State’s treatment of Yazidis amounts to genocide, a paradigmatic and extremely severe international crime.74 It requires no stretch of the imagination to recognize that Islamic State’s gendered crimes against Yazidi women are equally appropriate candidates for international prosecution. Indeed, an international trial would in many ways be more legitimate than a national one. It would not only avoid the pitfalls of a Western or regional trial, but would also recognize, consistent with the policies underlying the establishment of the ICC, that dozens of countries’ nationals have fought for Islamic State and that the gravity and magnitude of their alleged crimes demand worldwide condemnation.

The gendered nature of Islamic State’s crimes provides further incentive to bring the case in an international court: a well-structured prosecution of Islamic State for gender-based persecution would help to solidify an international norm against gender-based violence. While Article 7 of the Rome Statute criminalizes persecution on the basis of gender as a crime against humanity,75 the gender-based persecution provision has never been successfully charged. The Statute defines the crime as a “severe deprivation of [the] fundamental rights” of the targeted individuals, by reason of their gender, in the context of a “widespread or systematic attack directed against any civilian population,” and it requires that the perpetrator know of the attack.76 As other scholars have noted, the statute is not necessarily confined to sex crimes; it condemns all crimes that involve targeting on the basis of gender.77 For instance, sex-selective murder of all males of fighting age could be a form of gender-based violence.78 Similarly, slavery could be a form of gender-based persecution even without its sexual characteristics if only women were being enslaved.79

But the promise of the gender-based persecution charge has, until now, remained only a theoretical possibility. While the inclusion of the charge in the Rome Statute shows some progress toward recognizing how gendered harms and other forms of bias are interwoven in international crimes, successfully charging gender-based persecution is a crucial next step in the evolution toward recognizing and combatting intersectional crimes.

B. Creating an International Norm

The experience of Yazidi women illustrates that gender-based persecution is often intersectional—indeed, its intersectionality may be one reason legal discourse often overlooks it. Intersectionality refers to the effects of overlapping identities on individuals’ lived experiences, particularly of discrimination. In legal scholarship, Kimberlé Crenshaw introduced the idea in her analysis of the difficulty faced by African American women seeking to bring employment discrimination claims to court. These women were told that their claims were not cognizable on the ground that they were denied employment because they were both African American and women; as a result, they sometimes received no relief at all from the courts.80 Similarly, the available evidence suggests that Islamic State targets Yazidi women because they are both Yazidi and women. Just as domestic judges’ understanding of the nature of discrimination had to change in order to make African Americanwomen’s claims cognizable, so, too, does the international community’s understanding of persecution based on identity need to change in order to recognize the harm in targeting based on both gender and ethnicity. Only once the international system recognizes gender-based targeting as a distinctive harm will it begin to develop a robust norm against gender-based persecution.

The idea that some victims of international crimes may be targeted based on the intersection of their gender and ethnic identities (rather than their ethnic identity alone), as well as the belief that an intersectional perspective is required to fully comprehend the implications of those crimes, is not itself new.81 However, despite the progress made by including gender-based persecution in the Rome Statute, international criminal law is, in Crenshaw’s words, still fundamentally a “discourse[] shaped to respond”82 to only one aspect of the Yazidi victims’ (and other genocide victims’) identities: their race or their ethnicity. The skewing of the discourse is obvious. Consider the failure to charge gender-based persecution in all but one of the twenty-three cases that have come before the ICC, when ethnic and political persecution has been charged nine times,83 including in cases where there is strong evidence of mass rape of women (and no comparable evidence of rape of men).84

As a result, the discourse of international criminal law, to borrow Crenshaw’s language again,85 ignores the intersectional identity of victims. In so doing, the law marginalizes those victims by failing to recognize their experiences both as ethnic minorities and as women or men. Including the crime of gender-based persecution in the Rome Statute is a step forward, but failing to ever successfully charge it limits its value, both practically and symbolically. Turning a blind eye in this way to the totality of victims’ experiences would be a failure in domestic criminal law. Expressive theories of punishment recognize criminal law as a device for conveying social disapproval;86 failing to charge a crime once it is codified compromises its expressive value and fails to strengthen the related social norm.87 This failure is particularly acute when it comes to international criminal law since expressive condemnation is one of its paramount functions.88International criminal law signals to both state and nonstate actors that the overwhelming weight of international opinion vilifies the criminalized practices and that there is a norm against them. It signals that engaging in those practices may carry not just criminal, but also diplomatic, economic, and military consequences.89

Indeed, norm creation through its expressive value lies at the heart of international criminal law. Since the earliest days of its existence, international criminal law has helped to spur the formation of new international law norms and to crystallize existing norms.90 By articulating an international consensus that certain acts are morally wrong, the law can make clear that such acts are broadly unacceptable and can lead to increased awareness of their harms in the realms of diplomacy, international policy, and even national-level politics.91

The history of the international crime of genocide demonstrates the truth of this story. Mass atrocities that we would today call genocide, such as the extermination and deportation of ethnic Armenians living in Turkey,92 occurred before the Holocaust. The concept of genocide as a distinctive and morally blameworthy act did not begin to take shape until after World War II. After the world realized the scale of ethnic cleansing of Jews and Roma in German-occupied Europe, and thanks to the lobbying of international lawyers, the UN General Assembly voted to pass the 1948 Convention on the Prevention and Punishment of the Crime of Genocide.93 The Genocide Convention criminalized its namesake act94 and provided that responsibility for enforcing the treaty would lie with “a competent tribunal of the State in the territory of which the act was committed, or by such international penal tribunal as may have jurisdiction . . . .”95 International criminal law in this case helped to cap an initial process of norm formation that began with recognition of a morally outrageous act after an international incident, and progressed with a broad international consensus against that act.

International criminal law again crystallized the norm against genocide in the 1990s, after the genocides in Yugoslavia and Rwanda, which the Genocide Convention notably failed to prevent. Horrified by both conflicts, the UN Security Council passed resolutions establishing the International Criminal Tribunals for Yugoslavia (ICTY) and Rwanda (ICTR) in 1993 and 1994, respectively.96 Critically, the international criminal norm against genocide in these cases had effects beyond punishment for the perpetrators. The debates over the response to the genocides led to the international policy of “Responsibility to Protect,” which, flying in the face of international law’s conventional respect for sovereignty,97 expressly endorses military intervention in a sovereign state as a last resort for preventing or responding to mass atrocities.98 Embarrassment over the failure to intervene in Yugoslavia and Rwanda also led NATO to intervene in Kosovo—potentially preventing genocide—and led to a temporary increase in humanitarian support in Somalia99 before U.S. forces withdrew following the notorious Black Hawk Down incident.100 In those years, international criminal justice served not just as an expression of international support for an existing norm against committing an atrocity. Awareness of the genocides in Rwanda and Yugoslavia and of the ongoing trials at the ICTR and ICTY helped create, along with other forces, a new norm in favor of intervention in cases of mass atrocities, and led to concrete policy decisions in Kosovo and Somalia.101

The failure to make use of the prohibition on gender-based persecution, then, is not just a failure to hold perpetrators accountable on a retributive theory of justice, which suggests that we punish because a bad actor deserves it.102 It is also tantamount to a failure of the international system to signal its disapproval of gender-based persecution. Of course, there may be other ways of demonstrating disapprobation, such as Security Council declarations or diplomatic conferences. But talk is cheap. International prosecutions can require coming to a difficult consensus. Indeed, the very challenges that impede international prosecution can amplify their signaling effect. Because of these signaling costs, international prosecution sends a stronger message about shared human values and priorities. As a result, international criminal prosecution can play a stronger role in the norm evolution and crystallization process.103

C. The Insufficiency of Current Charges at the ICC

As the previous Section details, there is a strong positive case for charging gender-based persecution against Islamic State members for atrocities committed against Yazidi women. International criminal law can help solidify a norm against gender-based persecution and signal international willingness to mobilize against it. This Section deals with the negative case: the reasons that the current protocol for charging crimes at international courts cannot adequately deal with cases that show potential patterns of gender-based persecution in general, or with the atrocities against Yazidi women in particular. Where ethnic violence also displays patterns of gender-based targeting, international criminal tribunals have tended to subsume potential acts of gender-based persecution under genocide, charging these acts as simply components of the overall crime of genocide.104 Unfortunately, the failure to recognize the particular bias at work in a pattern of gender-based targeting that, in turn, accompanies patterns of political or ethnic targeting is tantamount to a failure to fully recognize the harm that intersectional persecution does to its victims.

1. Current Theoretical Insufficiencies

In the domestic context it seems obvious: victims of hate crimes can be victimized for multiple reasons. Say a gay man of color were mugged by a group of assailants, and bystanders heard the man’s attackers yell slurs about both his race and sexual orientation. In evaluating whether the act was a hate crime, would we only take into account evidence that the perpetrators attacked the man because of his race? On the contrary, in a jurisdiction where LGBTQ people are a protected class,105 and in the presence of appropriate evidence to make out the intent requirement, a prosecutor likely would present evidence that the victim had been subject to a hate crime on the basis of his intersecting identities—not just because of his race, and not just because of his sexuality. The perpetrators chose him because he was gay, because he was a person of color, and because he was gay and a person of color.106

Unfortunately, the two dominant working theories regarding the role gender-based targeting plays in conflict appear to erase the important intuition that the criminal law ought to recognize intersectionality. Indeed, neither of the two theories, which stem from previous international prosecutions and radical feminist107 legal scholarship, provides a strong rationale for devoting resources to prosecuting gender-based persecution in international criminal law. As to the first theory, in previous cases of genocide that also showed evidence of gender-based targeting (such as the genocide in Rwanda), courts and prosecutors explained the differential treatment of women and men as a component of the overall crime of genocide—a tactic in service of a strategic plan to harm a particular ethnic group.108 The crime of “genocidal rape” is paradigmatic109: women are raped as part of an overall plan to exterminate an ethnic group by forcing them to bear the children of their persecutors rather than children of their own ethnicity.110 The intent in the end has nothing to do with gender, but with ethnicity. For some international courts and commentators, then, gender is invisible, so that apparent patterns of singling out one gender for persecution must purely be in service of a broader goal to, for instance, prevent an ethnic group from reproducing.111

On the other hand, the radical feminist position implies that patriarchal structures are so pervasive, in every culture and at every time, that extreme forms of violence against women (and only women) during periods of instability are no aberration. Such acts are merely the logical outgrowth of pervasive misogyny in a context of violent ethnic conflict.112 Abuses in wartime may be different in magnitude from those in peacetime, but not in kind.113 This radical feminist tradition bears Marxism’s imprimatur: it equates different forms of violence against women by analyzing them as products of the same patriarchal social structures, which, the argument goes, are designed to keep women in line by arbitrarily inflicting violence or other physical restraints on them.114 This analysis intuitively leads to the conclusion that a legal remedy, international or otherwise, cannot solve the underlying problem. If pervasive and universal social structures are responsible for wartime persecution of women as well as peacetime domestic violence, then only wholesale subversion or overthrow of those structures can deal with the issue. Specifically singling out the treatment of women in a handful of armed conflicts not only loses meaning, but may be counterproductive.

Whichever of these perspectives one adopts, a charge of gender-based persecution is stripped of its significance. From both points of view, we lose any sense that gender-based targeting in a conflict is distinctly morally blameworthy, or that pursuing international prosecutions of gendered crimes has much value. Placed within the framework of a more totalizing evil—in the first view, genocide, and in the second, pervasive misogyny—the crime loses its significance as a discrete harm that international criminal law should directly and effectively address. Gender-based targeting of civilians is either a means to the end of genocide, or morally and causally no different than peacetime violence against women. In the context of an international legal order that does recognize other forms of identity-based persecution as distinctly morally blameworthy, however, failing to recognize gender-based persecution as such has the effect of making it seem somehow lesser.

Both perspectives, if adopted, also lead to a legal response that neglects the motivations behind the differential treatment that people of different genders receive in some (but not all) armed conflicts, and the magnitude of the moral horror of such treatment. This is not only counterintuitive, but empirically unfounded. Researchers have found that perpetrators of gender-based violence in some conflicts treat victims differentially because of their intersecting identities. Beyond the case of Islamic State, for instance, the Interahamwe in Rwanda systematically executed Tutsi men, boys, and even male infants during the genocide, but did not systematically execute women and girls.115 Differential treatment here resulted from the victims’ intersecting identities as Tutsis and as males, as well as the perception, based in gender stereotypes about male tendencies toward violence and score-settling, that any Tutsi male left alive could grow up to avenge his relatives’ deaths.116 The perceived threat was firmly rooted in gender roles, not in individual behavior: the génocidaires removed diapers from babies so that they could determine “which were boys to be killed.”117 A gender-neutral perspective, the one that most international courts have taken, presents only half the story: taking such a perspective, it is difficult to see how killing all Tutsi men, but not all Tutsi women, more effectively accomplished the goals of the génocidaires than killing all Tutsis regardless of gender. Neglecting the role that gender bias plays as it intersects with ethnic, political, or religious bias compromises the explanatory power of the gender-neutral perspective.

The same analysis applies to crimes against the Yazidi people. Ethnicity alone cannot explain why Yazidi men are driven out of Islamic State-controlled areas or summarily killed, while women are kept in protracted situations of torture and enslavement, in some cases before suffering a violent death.118 One Yazidi victim reported that her male relatives were taken away while she was held for seven months.119 Reflecting on the disparate treatment, she told a reporter, “It’s better that they are dead and not in prisons with these people. Even us, we were just wishing to die rather than stay with such people.”120 The groups share an ethnicity but are treated differently. It is notable that Islamic State does not appear to enslave Yazidi men121 and that the “bureaucracy of rape”122 that news reports describe does not include non-Yazidi women.123 In other words, the treatment that Yazidi women suffer appears to be a product of the intersection of their ethnicity and their gender. Chalking up this differential treatment purely to an effort to produce, for instance, Sunni Muslim children (besides the fact that this may not be true124) ignores the possibility that an atrocity can have more than one motivation.

That said, the same empirical evidence that undermines the gender-neutral approach to international atrocities also undermines the radical feminist approach: armed groups do not always treat different genders differently in ethnic, political, or religious conflicts. According to the radical feminist approach, we should expect misogynistic violence in all cases of ethnic conflict, just as it is supposedly universal in peacetime contexts. But this is far from the case. For instance, there is little evidence that members of the Liberation Tigers of Tamil Eelam have committed gender-based violence during their protracted conflict with the government of Sri Lanka.125 Similarly, the ongoing conflict between Israelis and Palestinians centers on ethnicity alone. Little evidence of gender-based violence has emerged.126 Thus, because gender-based violence does not appear to be as pervasive as some radical feminist scholars suggest, there appears to be a morally significant distinction between contexts where gendered brutality does appear and contexts where it does not. Radical feminists, in insisting that violence against women (and only women) is universal and pervasive, fail to cognize this distinction, just as they fail to cognize potential instances of gender-based persecution of men.

2. The Need for a Composite Crime

For the reasons stated above, the crimes of genocide and ethnic persecution could not, charged by themselves, fully encompass the crimes committed against Yazidi women in this case—or against Bosnian women in the former Yugoslavia or Tutsi men in Rwanda. Widely differential treatment of individuals of the same ethnicity suggests that an additional gender animus gives rise to the treatment suffered by particular groups, while the inconsistent appearance of gender-based targeting across conflicts suggests that this gender animus is present in some contexts but not in others. This evidence suggests that the gender-specific persecution that appears in these historical examples cannot be attributed just to ethnic hatred or to pervasive misogyny. Instead, it is the intersection of ethnic background with gender that produces particularly grievous harms.

Nonetheless, one might ask why crimes that display a pattern of gender targeting should be charged under the rubric of gender-based persecution when all of the crimes alleged against Islamic State in this situation could alternately be charged as individual crimes under the Rome Statute. Would it really be inadequate to recognize the issues at stake here by charging one individual with, for instance, several dozen rapes or acts of enslavement? Why should we charge gender-based persecution?

To understand the inadequacy of charging individual crimes, consider again the crime of genocide. Genocide—in some ways the paradigmatic international crime, given the historical development of international criminal justice127—is itself a composite crime, the individual elements of which could be charged separately as war crimes or crimes against humanity under the Rome Statute.128 Article 6 of the Rome Statute says that the following individual crimes constitute elements of genocide: killing members of a national, ethnic, racial, or religious group;129 inflicting on them “serious bodily or mental harm”;130 impeding the group’s fertility;131 “transferring children of the group to another group”;132 and “[d]eliberately inflicting on the group conditions of life calculated to bring about [the group’s] physical destruction in whole or in part.”133

All of these acts can also be charged as individual crimes under the Rome Statute’s crimes against humanity and war crimes provisions. Both statutory provisions prohibit murder (or, in the case of the war crimes provision, “[w]ilful killing”).134 The crimes against humanity provision of the Rome Statute prohibits intentionally inflicting serious bodily or mental harm;135 “enforced sterilization,”136 which is comparable to the crime of impeding fertility; and “forcible transfer of population,”137 which could be used to charge the transfer of children mentioned in Article 6.138 The only element of the statute without a clear analog is the Article 6 “conditions of life” provision;139 even so, many acts that could be characterized this way are covered by other elements of the statute, for instance the crimes against humanity provision covering “other inhumane acts . . . intentionally causing great suffering, or serious injury to body or to mental or physical health.”140

Raphael Lemkin, the Jewish legal academic who lobbied furiously for the passage of the 1948 Genocide Convention,141 best articulated the justification for maintaining genocide as an additional, composite charge. Acknowledging that genocide “consists of acts which are themselves punishable by most existing legislation,” like killing or false imprisonment, “[t]he main task” in enabling national courts to prosecute genocide is “to redraft existing provisions . . . based upon the specific criminal intent to destroy entire human groups.”142 This “specific intent,” or dolus specialis, characterizes not just genocide, but also the perhaps less totalizing crimes of political and ethnic persecution. It recognizes that there is a distinctive harm where bias against a particular social group spills over into violence aimed at destroying, degrading, or prostrating that group.143

Similarly, where a bias against one gender catalyzes a pattern of extreme violence against that gender, we may also say that this “special intent” exists. Gender is not fundamentally different from race, ethnicity, political belief, or religion. Like those other facets of identity, it creates distinct (and often publicly recognizable) groups of people within a single society. Moreover, unlike a favorite color or sports team preference, most people consider it to be a fundamental feature of identity, so that punishing people for having a particular gender would impose intolerable constraints on individuals and impoverish otherwise diverse societies. If gender or gender expression are fundamental characteristics of identity rather than peripheral ones, they should be treated like other fundamental characteristics of identity, such as ethnicity and religion.

Assuming this is the case, there is no coherent reason to ignore a specific intent to target individuals on the basis of gender—where there is evidence of this intent—and yet to take such intent into account with respect to ethnicity or political belief. Because women in many countries suffer violence and discrimination on the basis of their gender,144 it may be tempting to argue that instances of gender-based persecution are merely cases of ordinary cultural bias that happen to manifest in violence. By that logic, punishing such acts under international criminal law may create a slippery slope toward prosecuting common differential social treatment.145 Prosecuting such common differential social treatment in an international criminal forum, in turn, would violate principles of state sovereignty and likely be seen as an overreach for the ICC, which is designed to prosecute a handful of very grave and necessarily uncommon crimes.146

But in practice, the ICC is more than capable of drawing the required lines. In fact, the same argument could have been made against the ICC prosecuting crimes involving persecution of minority ethnic, religious, or political groups. Yet in all those cases, the ICC chose not to prosecute common differential social treatment precisely because of the court’s focus on grave crimes. For instance, many (if not most) societies also engage in practices that suggest “ordinary” bias against minority ethnic, religious, or political groups, without spilling over into widespread, violent purging of these groups. For instance, Uighurs in China, Jews and African immigrants in Russia, and Muslim Americans in the United States have all faced recent hostility, discrimination, and sporadic violence that is deplorable but does not come close to the level of genocide.147 None of these countries have been investigated by the ICC on the basis of bias against racial, ethnic, or religious minorities. Indeed, such an investigation would strain the institution’s credibility.

Genocide and gender-based persecution both recognize that where bias intersects with and spurs widespread, systematic, and grave crimes on the order of murder and torture, the result is particularly destructive and morally blameworthy. Both genocide and gender-based persecution require us to draw lines between courses of tolerable and criminal conduct based on severity. Fortunately, the fact that the Prosecutor of the ICC has been able to successfully charge genocide shows that such line-drawing is possible.148

iii. charging gender-based persecution under the rome statute: the elements of a case

If successfully charging gender-based persecution is essential to forming a norm against it in international law, then finding the right case is critical. The case of the Yazidi women is not just a paradigmatic example of gender-based persecution but may also be the perfect test case to begin to recognize the crime. While the lack of an international norm against gender-based targeting may have allowed their treatment to continue despite an outcry, a successful prosecution of their case at the ICC could bring justice not only to them, but to future victims of gender-based persecution. This Part argues that Islamic State’s treatment of Yazidi women presents a paradigmatic case of gender-based persecution. Moreover, because there appears to be a strong body of evidence supporting potential charges, it presents a unique opportunity to give content to the elements of the crime for use in future cases.

A. Threshold Questions: Procedural Matters and Individual Criminal Liability

While this Note primarily addresses the feasibility of charging gender-based persecution within an ongoing prosecution of Islamic State, one should nonetheless be aware of the procedural and other substantive criminal components of the broader case. Below, I sketch how the Prosecutor of the ICC could prove two important procedural elements of any criminal prosecution against Islamic State—jurisdiction and admissibility—as well as the challenges she would face in doing so. It also outlines how the Prosecutor might go about fulfilling the substantive requirement of individual criminal responsibility that applies in every ICC case. As the situation in Syria and Iraq stands today,149 the Prosecutor would face serious jurisdictional hurdles to prosecuting the top leadership of Islamic State, including Baghdadi and his direct lieutenants, not just for this crime but also for other atrocities. However, with the cooperation of states that are friendly to the court, she has a reasonable chance of successfully charging certain midlevel figures with gender-based persecution, as well as other crimes.

1. Jurisdiction

An important but complex question at the start of this analysis is whether the ICC could obtain territorial jurisdiction over Islamic State, which operates primarily within the territory of Syria and Iraq. A thorough evaluation of the ways the ICC might or might not be able to obtain territorial jurisdiction over Islamic State would be a separate project. Commentators have already opined on both sides of the argument and the answer remains far from clear.150 Nonetheless, I provide a brief overview of the paths by which the ICC might be able to secure jurisdiction. The ICC’s jurisdiction depends heavily on diplomatic developments151 and is thus quite fluid. There is therefore a reasonable chance that at some point, geopolitical alignments and interests will change and jurisdiction over major Islamic State figures will become easier to secure.

The ICC can obtain territorial jurisdiction over an individual or a situation in four ways: (1) where the acts occur on the territory of a country that has ratified the Rome Statute;152 (2) where the perpetrators are citizens of Rome Statute countries;153 (3) by resolution of the UN Security Council;154 and (4) where a non-State Party accepts jurisdiction.155 In this case, the first option is inapplicable because neither Syria nor Iraq, the locations of the Yazidi abuses, are members of the Court.156

This leaves three options for obtaining territorial jurisdiction. First, either Syria or Iraq could accept the jurisdiction of the ICC at some point in the future without ratifying the Rome Statute. Prosecution of the gender-based crimes detailed in this Note might then be possible if the countries were to choose to allow jurisdiction to apply retrospectively.157 Unfortunately this is unlikely, if not impossible, as long as Bashar al-Assad is in power, since the consent of Syria would also subject Assad to possible investigation and prosecution. However, if Syria were partitioned, the ICC could gain jurisdiction to investigate crimes in at least part of the territory.158 Gaining jurisdiction in Iraq is similarly complicated, since neither its government nor any other will necessarily be able to control the course of the investigation once opened. The leaders of Iraq may be uncooperative given recent violence stemming from sectarian divisions,159 while the United States is unlikely to support an investigation that might cover the Iraq War.160

Second, the Security Council could refer the case to the ICC. Like retrospective jurisdiction, this strategy, depending on how deployed, might capture much or all of Islamic State’s conduct toward Yazidi women. The major hurdle here is Syrian and Chinese opposition, which stems in part from their desire to protect Assad from ICC investigation161 as well as from their longstanding reluctance to weaken principles of state sovereignty.162 Commentators have suggested two possible ways around this hurdle (besides lobbying to change attitudes): referring only the situation on Iraqi territory (where Islamic State captured many of the women)163 and referring Islamic State as an organization.164

However, for our purposes, the bestimmediate hook for jurisdiction may be the ICC’s jurisdiction over nationals of countries that have ratified the Rome Statute.165 Islamic State has an extremely diverse membership by many accounts. Reports detail large numbers of Islamic State fighters hailing from, among other countries, Belgium, Tunisia, and Denmark, all of which are members of the ICC.166

This form of jurisdiction would not capture Baghdadi and his inner circle, who are Syrian and Iraqi, but it would at least capture some midlevel Islamic State commanders and allow the Court to establish the elements of the crime of gender-based persecution while winning a measure of justice for Islamic State’s most brutalized victims. Chief Prosecutor Fatou Bensouda has previously expressed unwillingness to open a case against members of Islamic State using this jurisdictional principle because, according to her, the ICC is best suited to prosecute only the top leadership of governments or organizations.167

However, Bensouda’s perspective does not recognize that those high-level individuals, whose roles are attenuated from operations on the ground, are not always more morally blameworthy or culpable than lower-ranking individuals in the same organization. While the leaders of an organization may make policies that permit or encourage crimes, it takes midlevel and lower-level officials to turn those policies into practice by ordering or directly committing crimes. This in itself makes them culpable. So, too, does their greater proximity to victims, with whom they may be more likely to interact than are high-level officials who set organizational policy. In practice, it seems that survivors sometimes blame lower-level perpetrators just as much as the high-level officials who oversaw their persecution. For example, victims’ groups participating in the ICC case Prosecutor v. Simone Gbagbo recently withdrew from the proceeding, questioning the wisdom of prosecuting a former first lady who had a role in planning election violence without also prosecuting the individuals who actually carried out the crimes.168

Moreover, Bensouda has demonstrated a strong interest in pursuing sexual and gender-based crimes within the jurisdiction of the Court,169 having recently issued a policy statement on prosecuting such crimes.170 Given her demonstrated desire to correct for the historical underprosecution of these offenses, and the clear moral culpability of midlevel Islamic State leaders, she would do well to reconsider her stated aversion to prosecuting them. Proceeding against these midlevel officials would give her the opportunity to develop the elements of gender-based persecution with a strong case.171

2. Admissibility

Once the ICC has established jurisdiction over a given situation, the Prosecutor still faces the hurdle of admissibility. Whether a case will be admissible at the ICC hinges, in turn, on the application of the principle of complementarity. According to this principle, codified at Article 17(1)(a) of the Rome Statute, the ICC may only take cases when the state where the alleged acts occurred, or the nationals of which perpetrated the acts, is “unwilling or unable” to handle the perpetrators through its own criminal justice system.172

The application of this principle depends on which theory the ICC uses to obtain jurisdiction. For the purposes of the theory of jurisdiction over nationals—the most immediate path to prosecuting at least some individuals involved in the Yazidi atrocities—the case would be admissible if an ICC-friendly state with Islamic State-member nationals agreed to cede its prosecution to the Court by claiming itself “unwilling or unable” to handle the case. This seems at least plausible in light of the policy arguments mentioned above for trying members of Islamic State at the ICC rather than in national courts. As former U.S. State Department Legal Adviser John Bellinger noted, “[T]he United States and other governments should not limit themselves to potential prosecutions of Islamic State members under domestic criminal laws. The group is engaging in widespread and systematic attacks against civilians in Iraq and Syria that constitute grave international crimes (including genocide).”173 These “grave international crimes” are precisely the offenses that the ICC was established to handle.174

3. Individual Criminal Responsibility

As in U.S. federal and state criminal law, it is possible (indeed, common) to charge defendants at the ICC with crimes that they did not perpetrate directly. However, in order for any criminal charge to proceed, international criminal law, like domestic criminal law, requires that prosecutors prove individual criminal responsibility.175 The prosecutors must show that the defendant was sufficiently connected to the commission of the crime at issue to bear moral and legal responsibility for it. Articles 25 and 28 of the Rome Statute articulate the theories of individual criminal responsibility that form a basis for conviction at the ICC.176 Article 25 of the Rome Statute provides for liability on the basis of the same theories of individual criminal liability that appear in U.S. domestic criminal law, including direct liability, solicitation, aiding and abetting, and conspiracy.177 Additionally, Article 28 imposes a separate, more capacious form of individual criminal responsibility on military commanders based on a theory of command responsibility. Commanders who “knew or . . . should have known” their forces were committing or about to commit Rome Statute crimes and nonetheless failed to take “all necessary and reasonable measures” to avert those crimes may be individually criminally liable under this theory.178

The challenges of proving individual criminal responsibility were on full display in the only prior case at the ICC to charge gender-based persecution, Prosecutor v. Mbarushimana.179 Luis Moreno Ocampo, the first Chief Prosecutor of the ICC,180 brought the Mbarushimana case in 2010. It ended in 2011 when the Court declined to confirm the charges due to a lack of evidence.181 The case stemmed from the activities of a paramilitary group in the Democratic Republic of Congo (DRC). The Prosecutor’s Document Containing Charges182 identified some evidence that the group had exhibited a pattern of acute violence targeting women civilians. However, as the Pre-Trial Chamber’s Decision on the Confirmation of Charges noted, there were deep flaws in the Prosecutor’s case against the Defendant.183 The Pre-Trial Chamber found, with consternation, that the Prosecution had utterly failed to substantiate many of the crimes charged including murder, pillaging, and assault, asserting no evidence for some of the counts.184 For instance, the Document Containing Charges charged twelve different incidents of murder, yet only made factual allegations as to four.185 Moreover, the Pre-Trial Chamber also found that the Prosecution had failed to establish a substantial contribution by the Defendant to the acts alleged, which is perhaps unsurprising since the Defendant apparently functioned as the paramilitary group’s France-based press secretary rather than as a military commander or political leader.186

A potential case against Islamic State presents the Office of the Prosecutor—which increasingly has turned toward gender-based and sexual crimes187—with an opportunity that has evaded it until now. It may be able to make out a stronger case against some members of Islamic State, so long as it puts together its evidence carefully and makes considered decisions about which officials to charge. Reports emerging from Islamic State-controlled territory suggest that proof of instructions or knowledge of the acts of persecution, often lacking in the case of higher-ups, may be somewhat clearer in the case of Islamic State. Islamic State keeps extensive records on the identities of its fighters.188 These individuals fight in units that “rotat[e] between active frontline duty, days off in ‘liberated’ areas[,] and other deployments ‘on base.’”189 Reports also suggest the organization generates paper records that could be used to establish levels of hierarchy, including the documentation of reallocation and payroll costs, and memoranda and reports to superiors on expenditures.190 In a possible prosecution, these might be useful in establishing the identity of perpetrators, the existence of chains of command (both military and civilian), and the requisite mens rea.

In addition to admitting the reports of journalists and rapporteurs, the Prosecutor of the ICC may be able to obtain testimony from Yazidi survivors of violence. Several Yazidi survivors have addressed the UN and international media outlets,191 suggesting that they may be willing to testify publicly against Islamic State members. Since Islamic State is both a military and a political bureaucracy, multiple theories of liability could apply to its senior leaders, including Baghdadi, should the path to jurisdiction over his actions become clear.192 The particular theory of liability for an Islamic State member—direct liability, solicitation, aiding and abetting liability, conspiracy, and command responsibility—will depend on that member’s role in the organization.193

The ICC admits diverse forms of evidence, including NGO publications and news reports,194 and tends to accept most evidence offered when making admissibility determinations195 on an article-by-article basis. This Note cites reports from reputable news outlets, Islamic State press releases and media materials, and NGO and think tank reports. All of these sources would be viable sources of evidence at an ICC trial. While we would need to know more before assessing the strength of the overall evidence, the evidence that appears to be available is promising.

B. Charging Islamic State with Gender-Based Persecution Under Article 7(1)(h) of the Rome Statute

The ICC has never decided a gender-based persecution case on the merits. Nor does the Rome Statute alone tell us how the Court would apply this charge. Like many domestic criminal statutes, judges must interpret its terms in order to give it sufficient specificity. Nonetheless, several potential sources of information shed light on the gender-based persecution charge, including the drafting history of the Rome Statute,196 the practices of international criminal tribunals that antedate the Rome Statute,197 and other bodies of international law.198 This Note is the first to analyze and apply one of these sources: the statements of the ICC’s Pre-Trial Chamber on pending cases charging political199 and ethnic persecution,200 two charges with elements analogous to those of gender-based persecution. Because it represents the views of ICC judges on the proper definition of crimes under the Rome Statute, the jurisprudence of the Pre-Trial Chamber is a particularly valuable source of information for the Islamic State case.

The Pre-Trial Chamber is a panel of three ICC judges that decides the validity of the Prosecutor’s initial indictments.201 A Pre-Trial Chamber proceeding’s closest analogue in U.S. criminal law is the preliminary hearing, where a prosecutor presents evidence to a judge who then decides whether to allow a trial to proceed following a defendant’s arrest.202 As at a preliminary hearing, the standard of proof that the Prosecutor must meet in front of the Pre-Trial Chamber is lower than that which is required to convict an individual of the same crime.203 However, the Prosecutor must meet this lower evidentiary burden for each element of the crime charged.204

Because the Pre-Trial Chamber must assess the evidence for every element of a given crime before allowing an indictment to proceed, it necessarily enumerates and clarifies the elements of those crimes, including crimes of persecution, in its rulings. For instance, in order to assess whether a charge of rape may proceed at the ICC, the Pre-Trial Chamber would clarify that rape requires both penetration and force or coercion.205

Using the Chamber’s jurisprudence on the analogous charges of political and ethnic persecution, this Section breaks down the charge of gender-based persecution into its constituent elements and assesses each element in terms of the Pre-Trial Chamber’s interpretations of its requirements. I then consider how each element, as construed by the Pre-Trial Chamber, might be satisfied by the evidence we have of Islamic State’s actions against Yazidi women. While I use the Pre-Trial Chamber’s jurisprudence to analyze the particular case of Islamic State’s crimes against Yazidi women, the Pre-Trial Chamber case law could also apply to assessing possible charges of gender-based persecution in other situations. The next Part considers how the Prosecutor might bring the charge in other situations.

Article 7(1)(h) of the Rome Statute details the crime of persecution, defining it as “[p]ersecution against any identifiable group or collectivity on political, racial, national, ethnic, cultural, religious, gender . . . or other grounds,” “in connection with . . . any crime within the jurisdiction of the Court,” “as part of a widespread or systematic attack directed against any civilian population, with knowledge of the attack.”206

According to the UN’s Elements of Crimes,207 a nonbinding guide to the Rome Statute that is frequently cited at the ICC,208 the Prosecutor must show six elements to prove persecution.209

1.The widespread or systematic element: The persecutory conduct must have occurred “as part of a widespread or systematic attack directed against [a] civilian population.”210

2.The mens rea element: The persecutor must have known the persecutory conduct to be part of a widespread or systematic attack directed against a civilian population.211

3.The severe deprivation of fundamental rights element: The persecutory conduct must have led to a “severe deprivation” of “rights” that are “fundamental” in international law.212

4. The nexus element: The individuals must have been deprived “by reason of” their membership in a covered category.213

5.The group element: The deprivation must have occurred on “political, racial, national, ethnic, cultural, religious, [or] gender” grounds,214 with gender limited explicitly by the statutory text to its binary conception (deprivation by reason of being “male” or “female”).215

6.The other criminal acts element: The targeting must have occurred in connection with one or more individual persecutory acts criminalized under Article 7 of the Rome Statute—such as rape, murder, or torture—or in connection with “any [other] crime within the jurisdiction of the Court.”216

Because this Note is concerned with the crime of gender-based persecution, the fifth element—whether the alleged persecution targeted a particular group—is already satisfied for purposes of my analysis. The Note will only address evidence of targeting based on gender, which is one of the listed impermissible grounds. For purposes of my argument, then, I combine the nexus element and the group element to ask whether the victims of gender-based persecution were targeted “by reason of” their gender.

1. The Widespread or Systematic Element

“Widespread or systematic” is an intensity criterion; it refers to how intense the “attack directed against any civilian population” must be before it constitutes a crime against humanity. The “widespread or systematic” requirement “exclud[es] random or isolated acts of violence from the notion of crimes against humanity.”217 Instead, the “established jurisprudence of the ICC” requires it to inquire into the “large-scale nature of the attack and the number of targeted persons.”218Islamic State’s attack on Yazidi populations and minority populations in Syria and Iraq more generally, during which Islamic State took the Yazidi women captive, satisfies this element.

The “widespread” component was elaborated in Prosecutor v. Saif Al-Islam Gaddafi, where the Court found that an attack on Libyan opposition protesters was “widespread” because it occurred in the cities “where more than 50% of the Libyan population resides,” and because there were grounds to believe that hundreds of civilians were killed, hundreds injured, and hundreds arrested by the security forces.219 Similarly, in Prosecutor v. Laurent Gbagbo, the Pre-Trial Chamber found that an attack that left 316 members of the Ivorian opposition dead and wounded was “widespread” because it “involved a large number of acts; . . . targeted and victimized a significant number of individuals; . . . extended over a time period of more than four months; and . . . affected the entire city of Abidjan, a metropolis of more than three million inhabitants.”220

The attack on the Yazidis and others falls within the meaning of “widespread” as articulated in Gaddafi and Laurent Gbagbo. At one point, Islamic State controlled a territory up to the size of Jordan, far larger than the entire city of Abidjan in Laurent Gbagbo.221 A UN report indicates that thousands of people have been killed by Islamic State,222 far more than the hundreds killed and injured in the attacks at issue in Gaddafi and Laurent Gbagbo.223 And Islamic State’s campaign of attacks has extended over a period of years, much longer than the four months in Laurent Gbagbo.224 The organization held thousands of square miles of territory by 2014 and began its drive to amass territory in Iraq as early as 2007.225 If the attacks in Gaddafi and Laurent Gbagbo constitute “widespread” attacks, then Islamic State’s must, too.

The Pre-Trial Chamber has found attacks “systematic” where there is evidence of either a policy to attack civilians226 or of a “common pattern” to the attacks.227 The existence of a policy may be inferred from coordinated actions that appear designed to achieve a particular goal. For example, in Gaddafi, the Pre-Trial Chamber accepted evidence of a series of coordinated actions as proof of a “State policy” to quell demonstrations. These actions included acts of the Libyan security forces, speeches of Said and Muammar Gaddafi denouncing the opposition, and a threatening mass text message sent via the state-owned telecommunications system.228

Islamic State behavior satisfies the “systematic” element, too: evidence suggests Islamic State has persecuted Yazidis as part of both a policy and a pattern of systematically attacking civilians. Islamic State publications on polytheistic minorities and women slaves suggest that its attacks on Yazidis and captivity of the Yazidi women are part of a “policy” to target both.229 For instance, Islamic State’s Western-facing propaganda magazine Dabiq published a long justification of its treatment of Yazidis in 2014.230 Dabiq referred to enslavement, concubinage, and the impregnation of women slaves as “firmly established aspect[s] of the [Shariah].”231 Reports from Syria and Iraq as well as UN Special Representative Zainab Bangura’s comments to the Middle East Eye also confirm that the attacks on Yazidis and the enslavement of Yazidi women form a pattern and suggest a state policy.232 Islamic State has captured and sold thousands of Yazidi women and has displaced thousands of Yazidi villagers from their homes—evidence of a pattern of enslavement and attacks.233 Moreover, the rigorously organized captivity of the Yazidi women itself suggests pre-planning—and, hence, a policy. According to news reports, Islamic State maintains a highly developed bureaucratic system for transporting, selling, and keeping track of the roughly three thousand Yazidi women it currently holds captive.234

2. The Mens Rea Element

Like most domestic crimes in the United States, international crimes include a mens rea element. As Article 7(1)(h) indicates, individual criminal responsibility requires that the individual have “knowledge” of the civilian “attack” that gave rise to the persecution.235 In other words, the individual defendant need not intend the overall attack on civilians for a finding of liability, but he or she must be aware of it. The Elements of Crimes also suggests that the statute be read to require that the individual defendant be aware that the course of conduct constituting persecution was “widespread or systematic.”236 Article 7(2)(a) of the Rome Statute further clarifies that the existence of a policy to attack civilians is necessary to find knowledge of a civilian attack.237 In the particular case of Islamic State, then, the same evidence that suffices to show a “systematic” Islamic State policy to target civilians, as discussed above, would also be required to satisfy the mens rea element of the crime.

In some cases, a policy to attack civilians is explicit, whereas in other cases it may be inferred. In Prosecutor v. Ruto, for instance, the Pre-Trial Chamber found that speeches by officials and orders by midlevel organizers urging members of a political party to attack the opposition constituted evidence of an official policy to attack civilians.238 Where the existence of such a policy is not overt, the Pre-Trial Chamber has inferred knowledge of an attack from patterns of behavior by official organs subject to centralized control.239 In Gaddafi, for instance, the Court did not mention the mens rea requirement in its persecution analysis. Instead, the Court found “reasonable grounds to believe that Muammar Gaddafi, either directly or through the State apparatus, ordered the Security Forces to ‘discipline’ opposition members,” on the basis of evidence that the Forces “were deployed throughout Libya,” recruited new members including mercenaries, and took steps to cover up the murders and other crimes complained of in the application.240 In other words, the Court found that there was evidence of intent to attack a civilian population—a higher level of mens rea than necessary to sustain the charge. Gaddafi suggests that evidence of a widespread and coordinated pattern of behavior by direct perpetrators along with evidence of the defendant’s command or control of those perpetrators may be enough to establish that the defendant possessed the requisite mens rea.

However, while Gaddafi suggests that it is possible to infer the existence of an official policy from a pattern of behavior,241 this inference is unnecessary in the case of Islamic State. Islamic State has issued fatwas that authorize its members to brutalize women captives, and victims have indicated that rape is part of the organization’s pseudo-religious ideology.242 Moreover, Islamic State’s own public relations materials acknowledge that the organization promotes attacks on “infidels.”243 The content of these official communications and the victims’ representations indicate an organizational policy to target civilians that is comparable to the policy expressed in the speeches in Ruto.

Because both Ruto and Gaddafi deal with charges against high-level officials who presumably had a role in setting the policy of targeting civilians in conflict, it is not precisely clear how the mens rea requirement as articulated in those cases would map on to midlevel officials of Islamic State. However, it is fair to assume that a midlevel official who administers an organizational policy to attack civilians (for instance, by commanding a group of fighters on the orders of superiors) would have knowledge of this organizational policy. At the same time, the mens rea requirement might exclude low-level provincial collaborators who are unaware of Islamic State organizational policies to attack civilians.

3. The Severe Deprivation of Fundamental Rights Element

Moving on from mens rea, persecution is an “intentional and severe deprivation of fundamental rights contrary to international law by reason of the identity of the group or collectivity.”244 “Intentional and severe” is, like “widespread and systematic,” an intensity requirement. The complained-of deprivation of rights must meet this threshold to be cognizable.245 In addition, while commentators have not fully settled the question of which rights are “fundamental” in international law,246 many of the rights contained in the Universal Declaration of Human Rights, such as the rights to life, liberty, and security of person, as well as the rights to be free from slavery and torture,247 are generally considered “fundamental.”248

In Prosecutor v. Kenyatta, the Pre-Trial Chamber found that a series of “killings, displacement[s], rape[s], serious physical injuries, and acts causing serious mental suffering . . . constitute[d] severe deprivations of fundamental rights.”249 The Prosecutor’s application had alleged that the violence at issue in the case “resulted in more than 1,100 people dead, 3,500 injured, approximately 600,000 victims of forcible displacement, [and] at least hundreds of victims of rape and sexual violence.”250 Thus “severity of deprivation” appears to assess both the magnitude of the population whose rights were deprived and the gravity of those deprivations as to each individual.251

Yazidi women have experienced deprivations of a severity akin to what the pleading in Kenyatta alleged Kenyan opposition members had experienced. The Yazidi women in this case have suffered similar harms in similar numbers. A UN report indicates that, as of August 2016, nearly 2,000 of these women were still held in slavery, with approximately 1,000 having escaped.252 The women are subjected to repeated rapes and beatings253 and often murdered, in unknown numbers, for resisting their captors.254 The Pre-Trial Chamber has already found mass rapes to constitute a violation of fundamental rights in its opinion in Kenyatta.255

The treatment of Yazidi women violates many of their fundamental rights. Their confinement, enslavement, and repeated sexual victimization are deprivations of the fundamental rights to life and liberty as well as the rights to be free from enslavement and torture.256

4. The Nexus with Group Element

According to the Elements of Crimes, proving gender-based persecution requires evidence of a “nexus.”257 The complained-of deprivations of rights must occur “by reason of” the group’s identity, here, gender identity.258 Nexus requirements can be particularly challenging to make out because they contain a complicated element of mens rea. They demand proof not just that a certain act occurred, or even that the perpetrator intended both the act and its result, but that the perpetrator had a biased motive for undertaking the act—or, in other words, targeted individuals because of their identities as members of a certain group.259 The Court has suggested that two types of evidence may show nexus in persecution cases: evidence of an official policy to victimize a Rome-Statute-covered group, and evidence of a pattern of targeting that group.

Pre-Trial Chamber case law provides a guide for how nexus can be found on the basis of an explicit, official organizational policy to target a covered group.260 Though evidence of an overt policy to persecute a particular group because of bias can be difficult to obtain, the Pre-Trial Chamber has inferred nexus from a pattern of acts committed against a covered group. For instance, in the Pre-Trial Chamber’s ruling on the confirmation of charges in Prosecutor v. Harun, it found “reasonable grounds to believe” that the Sudanese government had persecuted the ethnic Fur people of Darfur on the basis of evidence “that the Sudanese Armed Forces and the Militia/Janjaweed launched attacks against specific localities believing that they were predominantly inhabited by the Fur population.”261 Similarly, in its ruling on the confirmation of charges in Laurent Gbagbo, the Pre-Trial Chamber found “substantial grounds to believe” that at least 316 victims of killings, rapes, and injuries by pro-government forces had been targeted because the victims were members of the opposition.262 The evidence for this claim was, again, a pattern of attacks. Government forces had targeted participants at pro-opposition demonstrations, as well as “inhabitants of areas perceived as supporting” the opposition.263

In the case of Islamic State, the organization appears to have both a policy to attack and a pattern of targeting Yazidi women because of their gender. The incorporation of sexual and gender-based violence into Islamic State ideology strongly suggests that Islamic State has adopted a policy of persecuting women. Publicly available fatwas and publications attributed to Islamic State extensively detail rules for the treatment of women slaves in particular.264 These statements explicitly condone the sale of women and refer to intercourse with women slaves and to permissible and impermissible forms of beating.265 Press reports indicate that perpetrators told victims that rape is not only condoned but encouraged by Islamic State religious leaders.266 And a large bureaucracy, based at Islamic State headquarters in Raqqa, exists to coordinate the transportation and sale of the women (and not men). These facts all strongly suggest an official policy of targeting women for sale, and likely also for rape and slavery specifically.267 But even if that evidence were not sufficient, the ICC could alternatively infer a nexus from an apparent pattern of rapes, beatings, and sales of women captives.268

5. The Other Criminal Acts Element

The Rome Statute notes that persecution may only be found “in connection with” some act criminalized elsewhere in the Rome Statute.269 An analogy to the domestic law of hate crimes proves useful here as well: a hate crime may only be charged where some other underlying crime is also charged, like murder or battery.270 The requirement that gender-based persecution occur “in connection with” one of these other Rome Statute crimes seems designed to ensure that relatively trivial acts of discrimination, like forcing women to remain in the home or conscripting only men in a draft, do not form the basis for international criminal liability.

In addition to criminalizing certain composite crimes that originated in international law, such as gender-based persecution and genocide, the Rome Statute outlaws numerous acts that also appear in most countries’ domestic criminal laws, including murder and willful killing,271 serious bodily and mental harm,272 enslavement,273 and torture.274 It also contains other non-composite crimes that emerge more often in the context of international law, such as inhuman treatment,275 sexual slavery,276 forced pregnancy,277 enforced sterilization,278 and “[d]eliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part.”279 In the cases that have been brought before the Pre-Trial Chamber, Rome Statute crimes that have been used to satisfy the “other criminal acts” requirement for an ethnic or political persecution charge include murder,280 rape,281 forced displacement,282 torture,283 severe deprivation of physical liberty or prolonged arbitrary detention,284 and pillaging.285

Islamic State persecution of Yazidi women easily satisfies this “in connection with”286 requirement. The Yazidi women themselves have been raped, tortured, subjected to prolonged arbitrary detention, enslaved, and forcibly displaced, all of which are Rome Statute crimes under Articles 7 and 8.287

This Part aims to show that there is a plausible road to prosecuting at least some members of Islamic State for gender-based persecution. Though any trial will raise complicated issues of individual criminal liability, prosecutors may eventually be able to gather clear and systematic evidence about Islamic State’s membership and hierarchy and may also be able to obtain evidence of a policy of targeting civilians because of the nexus of their gender and ethnicity. Though the full extent of evidence against Islamic State will not become obvious until more of the group’s documents are obtained, the case presents an opportunity to develop international criminal law in a necessary direction. While it might be difficult to establish jurisdiction over top Islamic State leaders, there may be sufficient evidence to link members of the organization at all levels to the persecution of Yazidis. Although the Chief Prosecutor has opposed charging midlevel operatives with crimes thus far,288 the evidence emerging from Syria and Iraq suggests that a case against Islamic State members is a strong and urgent one in which the Court can establish the elements of the crime of gender-based persecution.

iv. gender-based persecution and the evolution of the icc

A prosecution against Islamic State members for crimes against Yazidi women not only might succeed at the ICC but also would come at a critical time. In October 2016, after years of complaints that the Office of the Prosecutor had disproportionately and inappropriately targeted African leaders,289 three African states—Burundi,290 South Africa,291 and the Gambia292—announced their intention to withdraw from the Court. (Russia, which had signed but never ratified the Rome Statute, subsequently withdrew its signature in a largely ceremonial gesture apparently designed to undermine the Court.293) While the leaders of Burundi and the Gambia at the time of withdrawal might well be reasonable subjects for international criminal liability themselves,294 the loss of South Africa, which played a pivotal role in negotiating the Rome Statute during the Mandela government,295 is a major blow to the Court.

Thus, the ICC faces a defining moment for its legitimacy.296 Criticisms of its excessive focus on African countries and African leaders were linked to particular charging decisions297 that led some observers to the conclusion that the Court was politicized or even biased, undermining its status as a neutral arbiter.298 If the ICC is to emerge as a legitimate institution, it must pivot its focus, redefining the scope of its own action from a narrow focus on African states to a broader focus on the world’s worst crimes.

In this context, bringing a popular and widely desired case against Islamic State could serve as the course-correction that the Court needs. The time for an accounting for Islamic State’s acts may be drawing nearer—though any possible criminal trial is still years off, given the slow and painstaking nature of international criminal justice.299 Recent advances by U.S.-backed forces in Iraq and Libya liberated the key cities of Falluja and Surt from Islamic State control, and Iraqi fighters are gradually retaking the city of Mosul.300 In May 2016, two months after the Obama Administration declared Islamic State treatment of Yazidis to be genocide, Politico reported that the White House and State Department had begun to discuss potential paths to international legal accountability for Islamic State, such as a trial at the ICC or the establishment of a dedicated international criminal tribunal,301 though legal accountability may be deprioritized in the Trump Administration. The news that international human rights lawyer Amal Clooney had agreed to represent Yazidi survivors in pressing for a case at the ICC confirms that transnational forces of diplomacy and publicity are aligning to lobby for an international trial.302 In a context of increasing dissatisfaction with the ICC, the failure of the Prosecutor to take steps against Islamic State in the face of so much outrage has arguably further harmed the Court’s legitimacy.

Bringing a case against members of Islamic State would put the Prosecutor in the relatively powerful position of consolidating a norm against gender-based persecution after international outrage. As observed earlier, international law can play a role in norm formation at different points in the process of consolidation: sometimes it begins a process of norm formation, and sometimes it caps a process that already has occurred.303 The sharp international outcry against the treatment of Yazidi women suggests that a norm against gender-based persecution by nonstate actors is already in the process of forming. The ICC might be able to capitalize on this momentum to embark on a relatively uncontroversial prosecution. Moreover, a prosecution of members of Islamic State —a nonstate armed group—does not run up against the norm of respect for state sovereignty that has been so problematic for the ICC.

The Office of the Prosecutor can and should take advantage of the political demand for a trial and the developed evidentiary record of Islamic State atrocities to hold Islamic State officials accountable for their crimes against Yazidi women.304 Questions about the ICC’s neutrality and the scope of its activity indicate that the Court may soon undergo what Steven Koh calls a “Marbury moment,” a point in its early history when it is called on to “rul[e] on the nature of its own authority” in order to become legitimate.305 Doing so in the context of a case broadly viewed as justified could help the Court to establish this authority.306

A prosecution of Islamic State members could set the ICC on a new path. The first successful prosecution to be brought under the gender-based persecution portion of the Rome Statute is critical for the same reason that a first prosecution under any domestic criminal statute would be. The judgment will create precedent, and the clear precedent that emerges from a strong first case will likely make subsequent cases easier to bring. But the Office of the Prosecutor should not stop there. This Note’s analysis of Pre-Trial Chamber jurisprudence sheds new light on how the gender-based persecution charge could be employed to hold other aggressors accountable for their gender-based crimes (though the evidentiary record in other cases may place some limits on the charge’s usefulness). The Office of the Prosecutor might seize the opportunity to change the direction of the Court with a series of prosecutions signaling a change in focus, once it has established gender-based persecution as a viable charge.

For instance, A. Widney Brown and Laura Grenfell have suggested that the Afghan Taliban’s targeting of women for harsh restrictions on their behavior, dress, and activity constitutes gender-based persecution.307 Members of the Taliban have stoned women to death for “adultery,”308 have beaten women for violating extremely restrictive laws about dress and public comportment,309 and have thrown acid on the faces of girls who attend school.310 In 2012, members of the Taliban infamously boarded a school bus and shot fourteen-year-old Malala Yousufzai in the head in retribution for her advocacy of children’s rights.311 Drawing on the jurisprudence of the International Criminal Tribunals for Yugoslavia and Rwanda, Brown and Grenfell suggest that the violations of victims’ rights in these cases may be severe enough to form the basis for a charge of gender-based persecution.312

The Office of the Prosecutor could also consider charging gender-based persecution in the DRC. The conflict there has smoldered for roughly two decades313 and displays high levels of gender-based targeting of women for what doctors in the region have called rape with extreme violence, a particularly brutal form of rape that often involves either gang rape or penetration with objects designed to damage the internal organs.314 It remains unclear what the incidence of rape among the male population is in these situations; a very high incidence of male rape could indicate that the targeting is not gender-based.315 However, Denis Mukwege, a human rights activist and gynecologist in the DRC, has said that the high incidence of rape with extreme violence in the country is partly the result of the country’s hierarchical gender roles and patterns of discrimination against women.316


International criminal law must recognize and punish the systematic abuse of women, men, girls, and boys by reason of their gender, just as it recognizes and punishes systematic abuse grounded in ethnicity or political views. Because of the particular horror of the acts described in this Note and the clear evidence emerging from seized documents and press reports, a prosecution of Islamic State members could be a paradigm case for developing the crime of gender-based persecution. Though the path to jurisdiction and admissibility is by no means clear, the international legal community should prioritize a prosecution of Islamic State members in order to secure justice for Yazidi victims and survivors of both genocide and gender-based persecution. If our promises to each other are to have any meaning at all, we must firmly declare our disapproval of those who doubly persecute other human beings and offer our support for those who are doubly victimized.